United States        Office of          EPA-600/8-83-016AF
             Environmental Protection     Research and Development    July 1984
             Agency          Washington, DC 20460


             Research and Development
v>EPA      The Acidic Deposition
             Phenomenon and-
             Its Effects

             Critical Assessment
             Review Papers

             Volume I  Atmospheric Sciences


-------
    THE ACIDIC  DEPOSITION PHENOMENON AND ITS EFFECTS

              CRITICAL  ASSESSMENT  REVIEW  PAPERS



                                 VOLUME I
 Aubrey  P. AltshuHer, Editor
    Atmospheric Sciences
 Rick A. Linthurst, Editor
     Effects Sciences
   Production

 Clara B. Edwards
 Wanda Frazier
 Elizabeth McKoy
 Benita Perry
                                Project Staff

                          Rick A. Linthurst-Z>w>ec£or
                          Betsy A. Hood-Coordinator
                       Gary  B. Blank-MznwscHpt Editor
            Graphics

           Mike Conley
           David Urena
         Steven F.  Vozzo
        C.  Willis Williams
                             Advisory Committee

                         David A. Bennett-U.S. EPA
                              Project Officer
John Bachmann-U.S.  EPA
Michael Berry-U.S.  EPA
Ellis B. Cowling-NCSU
J. Michael  Davis-U.S. EPA
Kenneth Demerjian-U.S. EPA
  J.  H. B.  Garner-U.S. EPA
  James L.  Regens-U.S. EPA
  Raymond Wilhour-U.S. EPA
     This document has been prepared through  the NCSU Acid Deposition Program,
a cooperative agreement between the United States Environmental Protection
Agency,  Washington, D.C. and North Carolina State University, Raleigh,  North
Carolina.   This work was conducted as part of the National Acid Precipitation
Assessment Program and was funded by U.S.  EPA.
                         U.S.  Ervir
                         Region V,
                         230  Soio
                         Chicago,  l
  Agency

-------
                                   DISCLAIMER





     This document has  been  reviewed in  accordance  with  U.S.  Environmental



Protection Agency policy and approved for  publication.   Mention  of  trade



names or commercial  products is  not intended to  constitute  endorsement  or



recommendation for use.
   UjS. Environment -   ^ion Agency

-------
                                   AUTHORS


                          Chapter A-l  Introduction


Altshuller, Aubrey Paul, Environmental Sciences Research Laboratory,  U.S.
     Environmental Protection Agency, MD 59,  Research Triangle Park,  NC,
     27711.

*Nader, John S., 2336 New Bern Ave., Raleigh, NC  27610.

*Niemeyer, Larry E., 4608 Huntington Ct., Raleigh, NC  27609.


           Chapter A-2  Natural and Anthropogenic Emission Sources


Homolya, James B., Radian Corp., P. 0. Box 13000, Research Triangle Park,  NC
      27709.

Robinson, Elmer, Civil and Environmental Engineering Dept.,  Washington  State
     University, Pullman, WA, 99164.


                      Chapter A-3  Transport  Processes


*Gillani, Noor V., Mechanical Engineering Dept., Washington  University,
     Box 1185, St. Louis, MO  63130.

Patterson, David E., Mechanical Engineering Dept., Washington  University,
     Box 1124, St. Louis, MO  63130.

Shannon, Jack D., Bldg. 181, Environmental  Research Div.,  Bldg.  181,  Argonne
     National Laboraory, Argonne, IL  60439.


                    Chapter A-4  Transformation Processes

Gillani, Noor V., Mechanical Engineering Dept., Washington University,
     Box 1185, St. Louis, MO  63130.

Hegg, Dean A., Atmospheric Sciences, AK-40, University of  Washington,
     Seattle, WA  98195.

Hobbs,  Peter V., Dept. of Atmospheric Sciences,  AK-40,  University  of
     Washington, Seattle, WA  98195.

*M111er, David F., Desert Research Institute,  University of  Nevada, P. 0.  Box
     60220,  Reno, NV  89506.


*Served as co-editor.

                                    iii

-------
Whltbeck, Michael, Desert Research  Institute,  University of Nevada, P. 0. Box
     60220, Reno, NV   89506.


           Chapter A-5  Atmospheric Concentrations and Distributions
                            of Chemical  Substances


Altshuller, Aubrey Paul, Envlromental  Sciences Research Laboratory, U.S.
     Environmental Protection  Agency,  MD 59, Research Triangle Park,
     NC  27711.
               Chapter A-6   Precipitation  Scavenging Processes


Hales, Jeremy M., Geosciences Research  and Engineering, Battelle, Pacific
     Northwest Laboratories, P.  0.  Box  999,  Richland, WA  99352.


                    Chapter A-7   Dry Deposition Processes


Hicks, Bruce B., NOAA/ERL,  Atmospheric  Turbulence and Diffusion Div., ARL,
     P. 0. Box E, Oak Ridge, TN   37830.


                     Chapter A-8 Deposition Monitoring


Hicks, Bruce B., U.S. Dept. of Commerce, National Oceanic and Atmospheric
     Administration, Environmental  Research  Laboratories, P. 0. Box E,
     Oak Ridge, TN  37830.

Lyons, William Berry, Dept. of Earth Sciences, James Hall, University of New
     Hampshire, Durham, NH   03824.

Mayewski, Paul A., Dept. of Earth Sciences,  James Hall, University of New
     Hampshire, Durham, NH   03824.

Stensland, Gary J., Illinois State  Water Survey, 605 E. Springfield Ave.,
     P. 0. Box 5050, Station A,  Champaign, IL  61820.


                       Chapter A-9  Deposition Models


Bhumralkar, Chandrakant M., Atmospheric Science Center, SRI International,
     333 Ravenswood Ave., Menlo  Park, CA   94025.

Ruff, Ronald E., Atmospheric Science Center, SRI International, 333
     Ravenswood Ave., Menlo Park, CA  94025.
                                    IV

-------
                          Chapter  E-l   Introduction


Unthurst, Rick A., Kilkelly Environmental Associates, Inc., P. 0. Box 31265,
     Raleigh, NC  27622.


                     Chapter E-2   Effects  on Soil Systems
Adams, Fred, Dept. of Agronomy and Soils,  Auburn University, Auburn, AL
     36849.

Cronan, Christopher S., Land and Water Resources Center, 11 Coburn Hall,
     University of Maine, Orono, ME  04469.

Firestone, Mary K., Dept. Plant and Soil Biology, 108 Hilgard Hall,
     University of California, Berkeley, CA  94720.

Foy, Charles D., U.S. Dept. of Agriculture,  Agricultural Research Service,
     Plant Stress Lab-BARC West, Beltsville, MD  20705.

Harter, Robert D., College of Life Sciences  and Agriculture, James Hall,
     University of New Hampshire, NH  03824.

Johnson, Dale W., Environmental Sciences Div., Oak Ridge National Laboratory,
     Oak Ridge, TN  37830.

*McFee, William W., Natural Resources and  Environmental Sciences Program,
     Purdue University, West Lafayette, IN  47907.


                     Chapter E-3   Effects on Vegetation


Chevone, Boris I., Dept. of Plant Pathology, Virginia Polytechnic Institute
     and State University, Blacksburg, VA  24060.

Irving, Patricia M., Environmental Research  Div., Bldg. 203, Argonne
     National Laboratory, Argonne, IL  60439.

Johnson, Arthur H., Dept. of Geology D4, University  of Pennsylvania,
     Philadelphia, PA  19104.

*Johnson, Dale W., Environmental Sciences  Div., Oak  Ridge  National
     Laboratory, Oak Ridge, TN  37830.

Lindberg, Steven E., Environmental Sciences  Div., Bldg. 1505, Oak Ridge
     National Laboratory, Oak Ridge, TN 37830.

McLaughlin, Samuel B., Environmental Sciences Div.,  Bldg.  3107, Oak Ridge
     National Laboratory, Oak Ridge, TN 37830.

-------
Raynal, Dudley J., Dept. of Environmental  and Forest  Biology, College of
     Environmental Science and Forestry,  State University of New York (SUNY),
     Syracuse, NY  13210.

Shriner, David S., Environmental  Sciences Div., Oak Ridge National
     Laboratory, Oak Ridge, TN 37830.

Sigal, Lorene L., Environmental Sciences  Div., Oak Ridge National Laboratory,
     Oak Ridge, TN  37830.

Skelly, John M., Dept. of Plant Pathology, 211 Buckhout Laboratory,
     Pennsylvania State University,  University Park,  PA  16802.

Smith, William H., School of Forestry and Environmental Studies, Yale
     University, 370 Prospect Street, New Haven, CT   06511.

Weber, Jerome B., Dept. of Crop Science,  North Carolina State University,
     Raleigh, NC  27650.


                  Chapter E-4  Effects  on Aquatic Chemistry


Anderson, Dennis S., Dept. of Botany and  Plant Pathology, University of
     Maine, Orono, ME  04469.

*Baker, Joan P., NCSU Acid Deposition Program, North  Carolina State
     University, 1509 Varsity Dr.,  Raleigh,  NC  27606.

Blank, G. B., School of Forest Resources,  Biltmore Hall, North Carolina State
     University, NC  27650.

Church, M. Robbins, Corvallis Environmental  Research  Laboratory, U.S.
     Environmental Protection Agency, 200 SW 35th Street, Corvallis, OR
     97333.

Cronan, Christopher S., Land and  Water Resources Center, 11 Coburn Hall,
     University of Maine, Orono,  ME   04469.

Davis, Ronald B., Dept. of Botany and Plant Pathology, Univeristy of Maine,
     Orono, ME  04469.

Dillon, Peter J., Ontario Ministry of the Environment, Limnology Unit, P. 0.
     Box 39, Dorset, Ontario, Canada, POA 1EO.

Driscoll, Charles T., Dept. of Civil Engineering, 150 Hinds Hall, Syracuse
     University, NY  13210.

*Galloway, James N., Dept. of Environmental  Sciences, University of Virginia,
     Charlottesville, VA  22903.

Gregory, J. D., School of Forest  Resources, Biltmore  Hall, North Carolina
     State University, NC  27650.

                                    vi

-------
Norton, Stephen A., Dept. of Geological  Sciences,  110 Boardman Hall,
     University of Maine, Orono,  ME   04469.

Schafran, Gary C., Dept.  of C1v1l Engineering,  150  Hinds Hall, Syracuse
     University, Syracuse, NY  13210.


                   Chapter E-5  Effects  on Aquatic  Biology


Baker, Joan P., NCSU Add Deposition Program, North Carolina State
     University, 1509 Varsity Dr., Raleigh,  NC  27606.

Drlscoll, Charles T., Dept. of Civil  Engineering,  150 Hinds Hall, Syracuse
     University, Syracuse, NY  13210.

Fischer, Kathleen L., Canadian Wildlife  Service, National Wildlife Research
     Centre, Environment  Canada,  100 Gamelin Blvd., Hull, Quebec, Canada,
     K1A OE7.

Guthrie, Charles A., New  York State  Department  of  Environmental Conservation,
     Div. of Fish and Wildlife, Bldg. 40,  SUNY-Stony Brook, Stony Brook, NY
     11790.

*Magnuson, John J., Laboratory of Limnology, University of Wisconsin,
     Madison, WI  53706.

Peverly, John H., Dept. of Agronomy, University of  Illinois, Urbana, IL 61801

*Rahel, Frank J., Dept. of Zoology,  Ohio State  University, 1735 Neil Ave.,
     Columbus, OH  43210.

Schafran, Gary C., Dept.  of Civil Engineering,  150  Hinds Hall, Syracuse
     University, Syracuse, NY  13210.

Singer, Robert, Dept. of  Civil Engineering,  150 Hinds Hall, Syracuse
     University, Syracuse, NY  13210.


                   Chapter E-6  Indirect Effects on Health


Baker, Joan P., NCSU Acid Precipitation  Program, North Carolina State
     University, 1509 Varsity Dr., Raleigh,  NC  27606.

Clarkson, Thomas W., University of Rochester School of Medicine, P. 0. Box
     RBB, Rochester, NY   14642.

Sharpe, William E., Land  and Water Research  Bldg.,  Pennsylvania State
     University, University Park, PA  16802.
                                    Vll

-------
                      Chapter E-7   Effects on Materials


Baer, Herbert S., Conservation Center of the Institute of Fine Arts,
     New York University, 14 East  78th  Street, New York, NY  10021.

Kirmeyer, Gregory, Economic and Engineering Services, Inc., 611 N. Columbia,
     Olympia, WA  98507.

Yocom, John E., TRC Environmental  Consultants, Inc., 800 Connecticut Blvd.,
     East Hartford, CT  06108.
                                   vm

-------
                                    PREFACE
The Acidic Deposition  Phenomenon  and  Its  Effects:  Critical  Assessment  Review
Papers was written  at  the suggestion, in  the  summer of  1980,  of  the Chairman
of the  Clean  Air   Scientific  Advisory  Committee  of  EPA's  Science  Advisory
Board.  The document was  prepared for EPA through the  Acid  Deposition  Program
at North  Carolina  State  University.   This document  is the  first of  several
documents of  increasing  sophistication   that   assess  the  acidic  deposition
phenomenon.  It will  be succeeded  by assessment documents in  1985,  1987,  and
1989, based largely  on research  of the National  Acid  Precipitation  Assessment
Program.

The document's original  charge was to prepare  "a comprehensive document  which
lays out the  state  of  our  knowledge  with regard to  precursor  emissions,  pol-
lutant transformation  to  acidic  compounds,  pollutant  transport,   pollutant
deposition and the  effects  (both measured  and  potential)  of   acidic  deposi-
tion."  The editors  provided the  following  guidelines  to the  authors  writing
the Critical  Assessment Review Papers  to  meet this overall charge:


      1.  Contributions  are  to  be written  for  scientists  and  informed  lay
          persons.

      2.  Statements are to  be explained  and supported by  references;  i.e.,  a
          textbook  type of approach, in an objective style.

      3.  Literature  referenced   is  to   be  of  high   quality  and  not  every
          reference available is  to be included.

      4.  Emphasis  is to be placed on  North American  systems  with  concentrated
          effort  on U.S. data.

      5.  Overlap  between this document and the  SOx  Criteria  Document is  to be
          minimized.

      6.  Potential  vs known processes/effects  are to be clearly noted to  avoid
          mi si nterpretati on.

      7.  The certainty  of  our knowledge  should  be quantified, when  possible.

      8.  Conclusions are to be drawn  on  fact only.

      9.  Extrapolation beyond the available  data is to be avoided.

     10.  Scientific  knowledge  is  to be included  without  regard  to  policy
          implications.
                                       ix

-------
     11.  Policy-related options  or recommendations  are  beyond the  scope  of
          this document and are not to  be included.
The reader, to avoid  possible  misinterpretation of the  information  presented,
is advised to consider and understand these directives  before  reading.

Again, the document has  been designed to address our  present  status of  know-
ledge of  the  acidic  deposition phenomenon  and  its  effects.    It  is  not  a
Criteria Document; it is  not  designed to  set  standards  and no  connections  to
regulations should be  inferred.  The literature is  reviewed  and  conclusions
are drawn  based  on the  best evidence available. It  is  an authored document,
and as  such,  the  conclusions  are those  of the  authors  after  their review  of
the literature.

The success  of  the  Critical   Assessment  Review Papers  has  depended  on  the
coordinated efforts of  many  individuals.   The  document  involved  the  partici-
pation of  over 60 scientists  contributing  material  on their special  areas  of
expertise under the broad headings of  either  atmospheric processes or effects.
Coordination within these  two  areas  has  been  the  responsibility of A.   Paul
Altshuller and Rick A.  Linthurst, the  atmospheric and  effects  section editors,
respectively.  Overall  coordination  of the project  for  EPA is  under  David  A.
Bennett's direction.   Dr. Altshuller  is  an atmospheric chemist,  past recipient
of the  American  Chemical  Society's  Award  in  Pollution  Control,  and  recently
retired director  of  EPA's Environmental   Sciences  Research  Laboratory;  Dr.
Linthurst is  an   ecologist and  served  as Program  Coordinator  for the  Acid
Precipitation Program at  North  Carolina  State  University.  He  is currently  at
Kilkelly Environmental  Associates,  Inc.    Dr.  Bennett  is  the  Director of  the
Acid Deposition Assessment Staff  in  EPA's Office of Research  and Development.

The written materials that follow  are contributions from  one to eight  authors
per chapter,  integrated by the editors.   Approximately  75  scientists,  with
expertise in the fields being  addressed, reviewed early drafts  of the chapters.
In addition,  200  individuals  participated  in  a public  workshop  held  for  the
technical review  of  these  materials  in  November  1982.   Numerous  changes
resulted from  these   reviews,  and this  document reflects  those  comments.   A
public  review draft of  this document  was  distributed in  June  1983 for a 45-day
comment period.   During that  period,  130 sets  of  comments from  53 reviewers
were received.  These comments were summarized and evaluated by a technical  and
editorial panel,  and  then provided   to  the  authors  who addressed  them  by
revision and  rewriting   to produce  this  final  document.   In  response to  the
comments  received,  revisions   were   made  to  all chapters including  a  major
revision of  Chapter  E-4, Effects on  Aquatic  Chemistry,  and the  addition  of a
section on  corrosion  in  water piping  systems  in Chapter  E-7,  Effects  on
Materials.

-------
             ACKNOWLEDGMENTS FROM NORTH CAROLINA STATE UNIVERSITY
The  editorial  staff wishes  to  extend special thanks  to  all the  authors of
this  document.   They have  been patient  and tolerant of  our  changes,  re-
commendations,  and deadlines,  leading  to this  fourth and  final  version of
the  document.   These  dedicated scientists  are to  be  commended  for their
efforts.

We also wish  to acknowledge  our  Steering Committee, who has been patient with
our  errors  and deadline delays.  These  people  have made  major contributions
to this product, and  actively  assisted  us with  their recommendations on pro-
ducing this document.  Their objectivity, concern for  technical  accuracy, and
support is appreciated.  Dr. J.  Michael Davis of EPA deserves special thanks,
as  he directed the  initial  draft of the  document  in December  of  1981.  His
concern for clarity  of thought and writing in the  interest of  communicating
our  scientific knowledge  was most  helpful.   Dr.  David Bennett  of  EPA is
specifically  recognized  for his  role  as a scientific reviewer, and  an  EPA
staff member who buffered  the editorial staff and the  authors from the public
and policy  concerns associated  with this document.  Dr.  Bennett's tolerance,
patience, and  understanding  are  also appreciated.

All  the  reviewers, too  numerous to  list,  are gratefully  acknowledged   for
helping us  improve the quality  and  accuracy  of  this  document.  These people
were from private, state,  federal, and special-interest organizations in both
the  United  States  and Europe.    Their  concern  for  the  truth,  based  on  the
available data, is a  compliment to  all  the individuals and organizations who
were willing to deal objectively  with this most important topic.  It has been
a  pleasure  to  see all groups,  independent  of their  personal  philosophies,
work together  in the interest of  producing a technically accurate document.

Dr. Arthur Stern is  acknowledged for his contribution as a technical  editor
of the atmospheric sciences  early in the document's preparation.  He has made
an important contribution  to the  final product.

Finally, EPA  is acknowledged for its willingness  to  give  the  scientists an
opportunity to prepare this  document.  Its interest, as expressed through the
staff and authors,  in  having this document be an  authored  document to assist
in research planning, is most appreciated.  Rarely does a  group of scientists
have  such  a  free  hand in  contributing  independently to  such   an  important
issue and  in   such a  visible way.   Although  coordinating  the  efforts  of so
many scientists can be a difficult and  lengthy process, we  feel the authored
scientific  product makes  a valuable contribution  to the  acidic  deposition
issue.

The entire staff of  the NCSU Acid  Deposition Program and  several  part-time
workers have been  involved in  the production  of  this document since it began
in 1981.   In addition to the people listed on  the title page, these include:

William R. Alsop -  Program Assistant
Ann Bartuska - Program Coordinator

-------
Jody D. Castleberry - Receptionist/Secretary
Connie S. Harp - Secretary
Jeanie Hartman - Librarian
Helen Koop - Library Assistant
                                     XII

-------
                      THE ACIDIC DEPOSITION PHENOMENON AND  ITS EFFECTS:
                             CRITICAL ASSESSMENT REVIEW PAPERS

                                    Table of Contents

                                         Volume I
                                   Atmospheric Sciences
                                                                                   Page

AUTHORS 	   111

PREFACE 	    1 x

ABBREVIATION-ACRONYM LIST 	  xxlx

GLOSSARY 	 xl 111


A-l  INTRODUCTION

     1.1  Objectives 	  1-1
     1.2  Approach—Movement from  Sources to Receptor 	  1-1
          1.2.1  Chemical Substances of  Interest 	  1-1
          1.2.2  Natural  and Anthropogenic Emissions Sources  	  1-1
          1.2.3  Transport Processes  	  1-1
          1.2.4  Transformation Processes 	  1-2
          1.2.5  Atmospheric Concentrations and Distributions of Chemical
                 Substances 	  1-2
          1.2.6  Precipitation Scavenging Processes  	  1-2
          1.2.7  Dry Deposition Processes 	  1-3
          1.2.8  Deposition Monitoring  	  1-3
          1.2.9  Deposition Models 	  1-4
     1.3  Addle Deposition 	  1-4


A-2  NATURAL AND ANTHROPOGENIC EMISSIONS SOURCES

     2.1  Introduction	  2-1
     2.2  Natural Emission Sources 	  2-1
          2.2.1  Sulfur Compounds  	  2-1
                 2.2.1.1   Introduction  	  2-1
                 2.2.1.2   Estimates of Natural Sources 	  2-2
                 2.2.1.3   Blogenlc Emissions of Sulfur Compounds	  2-3
                 2.2.1.4   Geophysical Sources of Natural Sulfur Compounds  	  2-15
                          2.2.1.4.1 Volcanlsm 	  2-17
                          2.2.1.4.2 Marine sources of aerosol particles and
                                    gases 	  2-19
                 2.2.1.5   Scavenging Processes and Sinks  	  2-21
                 2.2.1.6   Summary  of Natural Sources of Sulfur Compounds 	  2-22
          2.2.2  Nitrogen Compounds 	  2-23
                 2.2.2.1   Introduction  	  2-23
                 2.2.2.2   Estimates of Natural Global Sources and Sinks 	  2-24
                 2.2.2.3   Blogenlc Sources of NOX Compounds 	  2-28
                 2.2.2.4   Tropospherlc and Stratospheric Reactions  	  2-30
                 2.2.2.5   Formation of NOX by Lightning 	  2-30
                 2.2.2.6   Blogenlc NOX Emissions Estimate for the United States  ...  2-32
                 2.2.2.7   Blogenlc Sources of Ammonia 	  2-33
                 2.2.2.8   Oceanic  Source for Ammonia 	  2-36
                 2.2.2.9   Blogenlc Ammonia Emissions Estimates for  the United
                          States  	  2-37
                 2.2.2.10 Meteorological  and Area Variations  for NOX and Ammonia
                          Emissions 	  2-38
                 2.2.2.11 Scavenging Processes for NOX and  Ammonia  	  2-38
                 2.2.2.12 Organic  Nitrogen Compounds 	  2-39
                 2.2.2.13 Summary  of Natural NOX and Ammonia  Emissions 	  2-39
                                         xm

-------
Table of Contents (continued)

                                                                                     Page

          2.2.3  Chlorine Compounds	   2-39
                 2.2.3.1  Introduction 	   2-39
                 2.2.3.2  Oceanic Sources 	   2-40
                 2.2.3.3  Volcanlsm	   2-44
                 2.2.3.4  Combustion 	   2-44
                 2.2.3.5  Total Natural  Chlorine Sources 	   2-45
                 2.2.3.6  Seasonal Distributions 	   2-45
                 2.2.3.7  Environmental  Impacts of Natural  Chlorides 	   2-45
          2.2.4  Natural Sources of Aerosol  Particles 	   2-45
          2.2.5  Precipitation pH In Background Conditions  	   2-48
          2.2.6  Summary 	   2-52
     2.3  Anthropogenic Emissions 	   2-53
          2.3.1  Origins of Anthropogenlcally Emitted Compounds and
                 Related Issues 	   2-53
          2.3.2  Historical Trends and Current Emissions of Sulfur Compounds 	   2-57
                 2.3.2.1  Sulfur Oxides 	   2-57
                 2.3.2.2  Primary Sulfate Emissions 	   2-62
          2.3.3  Historical Trends and Current Emissions of Nitrogen Oxides 	   2-68
          2.3.4  Historical Trends and Current Emissions of Hydrochloric Acid (HC1)   2-72
          2.3.5  Historical Trends and Current Emissions of Heavy Metals Emitted
                 from Fuel Combustion 	   2-76
          2.3.6  Historical Emissions Trends In Canada 	   2-84
          2v3.7  Future Trends 1n Emissions 	   2-93
                 2.3.7.1  United States 	   2-93
                 2.3.7.2  Canada  	   2-93
          2.3.8  Emissions Inventories 	   2-96
          2.3.9  The Potential for Neutralization of Atmospheric
                 Acidity by Suspended Fly Ash  	   2-97
     2.4  Conclusions 	   2-102
     2.5  References 	   2-106


A-3  TRANSPORT PROCESSES

     3.1  Introduction  	  3-1
          3.1.1  The Concept of Atmospheric Residence Times 	   3-2
     3.2  Meteorological Scales and Atmospheric Motions 	  3-3
          3.2.1  Meteorological Scales 	  3-3
          3.2.2  Atmospheric Motions  	  3-4
     3.3  Pollutant Transport  Layer:  Its Structure and Dynamics  	  3-10
          3.3.1  The Planetary Boundary Layer  (Mixing Layer) 	  3-10
          3.3.2  Structure of  the Transport Layer (TL)  	  3-12
          3.3.3  Dynamics of the  Transport Layer  	  3-16
          3.3.4  Effects of Mesoscale Complex  Systems on Transport Layer Structure
                 and Dynamics  	  3-27
                 3.3.4.1  Effect  of Mesoscale  Convectlve Precipitation Systems
                           (MCPS)  	  3-27
                 3.3.4.2  Complex Terrain Effects 	:	  3-31
                          3.3.4.2.1   Shoreline environment effects	  3-31
                          3.3.4.2.2   Urban effects 	  3-34
                          3.3.4.2.3   Hilly terrain effects  	  3-35
     3.4  Mesoscale PIume Transport and D11ut1on  	  3-38
          3.4.1  Elevated Point-Source Emissions  (Power Plant  Plumes)  	  3-38
          3.4.2  Broad  Areal Emissions Near Ground (Urban Plumes)  	  3-60
     3.5  Continental and Hemispheric Transport  	  3-65
     3.6  Conclusions  	  3-88
     3.7  References  	  3-92
                                            XIV

-------
Table of Contents (continued)

                                                                                    Page

A-4  TRANSFORMATION PROCESSES

     4.1  Introduction 	  4-1
     4.2  Homogeneous Gas-Phase Reactions 	  4-3
          4.2.1  Fundamental Reactions 	  4-3
                 4.2.1.1  Reduced Sulfur Compounds 	  4-3
                 4.2.1.2  Sulfur Dioxide 	  4-4
                 4.2.1.3  Nitrogen Compounds 	  4-11
                 4.2.1.4  Halogens 	  4-17
                 4.2.1.5  Organic Adds 	  4-17
          4.2.2  Laboratory Simulations of Sulfur Dioxide and Nitrogen Dioxide
                 Oxidation 	  4-17
          4.2.3  Field Studies of Gas-Phase Reactions 	  4-21
                 4.2.3.1  Urban Plumes 	  4-21
                 4.2.3.2  Power PI ant PIumes 	  4-24
          4.2.4  Summary 	  4-29
     4.3  Solution Reactions 	  4-31
          4.3.1  Introduction  	  4-31
          4.3.2  Absorption of Add 	  4-32
          4.3.3  Production of HC1 1n Solution 	  4-38
          4.3.4  Production of HN03 In Solution 	  4-38
          4.3.5  Production of H2S04 In Solution 	  4-42
                 4.3.5.1  Evidence from Field Studies 	  4-42
                 4.3.5.2  Homogeneous Aerobic Oxidation of S02'H20  to  H2S04  	  4-43
                          4.3.5.2.1  Uncatalyzed 	  4-43
                          4.3.5.2.2  Catalyzed 	  4-45
                 4.3.5.3  Homogeneous Non-aerobic Oxidation of S02'H20 to H2S04  ...  4-47
                 4.3.5.4  Heterogeneous Production of H2S04 In Solution 	  4-52
                 4.3.5.5  The Relative Importance of the Various H2S04
                          Production Mechanisms 	  4-53
          4.3.6  Neutralization Reactions 	  4-61
                 4.3.6.1  Neutralization by NH3 	  4-61
                 4.3.6.2  Neutralization by Particle-Add Reactions 	  4-62
          4.3.7  Summary 	  4-63
     4.4  Transformation Models 	  4-63
          4.4.1  Introduction  	  4-63
          4.4.2  Approaches to Transformation Modeling 	  4-66
                 4.4.2.1  The  Fundamental Approach 	  4-66
                 4.4.2.2  The Empirical Approach 	  4-68
          4.4.3  The Question of Linearity 	  4-71
          4.4.4  Some Specific Models and Their Applications 	  4-74
                 4.4.4.1  Detailed Chemical  Simulations 	  4-74
                 4.4.4.2  Parameterized Models 	  4-67
          4.4.5  Summary 	  4-81
     4.5  Conclusions 	  4-82
     4.6  References 	  4-86


A-5  ATMOSPHERIC CONCENTRATIONS AND DISTRIBUTIONS OF CHEMICAL SUBSTANCES

     5.1  Introduction 	  5-1
     5.2  Sulfur Compounds 	  5-2
          5.2.1  Historical Distribution Patterns 	  5-2
          5.2.2  Sulfur Dioxide 	  5-3
                 5.2.2.1  Urban Measurements 	  5-3
                 5.2.2.2  Nonurban Measurements 	  5-4
                 5.2.2.3  Concentration Measurements at Remote Locations 	  5-12
                 5.2.2.4  Comparison of Sulfur Dioxide Emissions and Ambient
                          A1r Concentration 	  5-12
                                           XV

-------
Table of Contents (continued)

                                                                                    Page

          5.2.3  Sul fate 	  5-13
                 5.2.3.1  Urban  Concentration Measurements  	  5-13
                 5.2.3.2  Urban  Composition Measurements  	  5-15
                 5.2.3.3  Nonurban  Concentration Measurements  	  5-16
                 5.2.3.4  Nonurban  Composition Measurements  	  5-19
                 5.2.3.5  Concentration  and Composition Measurements at Remote
                          Locations 	  5-22
                 5.2.3.6  Comparison  of  Sulfur Oxide Emissions and Ambient Air
                          Concentrations of Sulfate  	  5-23
          5.2.4  Particle Size Characteristics of Partlculate Sulfur Compounds  ....  5-24
                 5.2.4.1  Urban  Measurements  	  5-24
                 5.2.4.2  Nonurban  Size  Measurements 	  5-27
                 5.2.4.3  Measurements at Remote Locations  	  5-27
     5.3  Nitrogen Compounds  	  5-28
          5.3.1  Introduction  	  5-28
          5.3.2  Nitrogen Oxides 	  5-28
                 5.3.2.1  Historical  Distribution Patterns and Current
                          Concentrations of Nitrogen Oxides  	  5-28
                 5.3.2.2  Measurements Techniques-Nitrogen Oxides 	  5-29
                 5.3.2.3  Urban  Concentration Measurements  	  5-29
                 5.3.2.4  Nonurban  Concentration Measurements  	  5-30
                 5.3.2.5  Measurements of Concentrations  at  Remote Locations  	  5-34
          5.3.3  Nitric  Acid  	  5-38
                 5.3.3.1  Urban  Concentration Measurements  	  5-38
                 5.3.3.2  Nonurban  Concentration Measurements  	  5-40
                 5.3.3.3  Concentration  Measurements at Remote Locations  	  5-44
          5.3.4  Peroxyacetyl Nitrates 	  5-45
                 5.3.4.1  Urban  Concentration Measurements  	  5-45
                 5.3.4.2  Nonurban  Concentration Measurements  	  5-48
          5.3.5  Ammonia 	  5-50
                 5.3.5.1  Urban  Concentration Measurements  	  5-50
                 5.3.5.2  Nonurban  Concentration Measurements  	  5-51
          5.3.6  Partlculate Nitrate  	  5-51
                 5.3.6.1  Urban  Concentration Measurements  	  5-53
                 5.3.6.2  Nonurban  Concentration Measurements  	  5-55
                 5.3.6.3  Concentration  Measurements at Remote Locations  	  5-56
          5.3.7  Particle Size Characteristics of Partlculate Nitrogen Compounds  ..  5-56
     5.4  Ozone 	  5-58
          5.4.1  Concentration Measurements Within the Planetary Boundary Layer
                 (PBL)  	  5-60
          5.4.2  Concentration Measurements at Higher Altitudes 	  5-63
     5.5  Hydrogen Peroxide  	  5-63
          5.5.1  Urban Concentratlon  Measurements 	  5-64
          5.5.2  Nonurban Concentration  Measurements 	  5-64
          5.5.3  Concentration Measurements In Rainwater  	  5-65
     5.6  Chlorine Compounds  	  5-65
          5.&.1  Introduction  	  5-65
          5.6.2  Hydrogen Chloride  	  5-66
          5.6.3  Partlculate Chloride 	  5-66
          5.6.4  Particle Size Characteristics of Partlculate Chlorine Compounds  ..  5-67
     5.7  Metallic Elements  	  5-68
          5.7.1  Concentration Measurements and Particle  Sizes In Urban Areas 	  5-68
          5.7.2  Concentration Measurements and Particle  Sizes In Nonurban Areas  ..  5-71
     5.8  Relationship of Light  Extinction and Visual Range Measurements to Aerosol
          Composition 	  5-73
          5.8.1  Fine Particle Concentration and Light Scattering Coefficients  	  5-73
          5.8.2  Light Extinction or  Light Scattering Budgets at Urban Locations  ..  5-74
          5.8.3  Light Extinction or  Light Scattering Budgets at Nonurban
                 Locations 	  5-76
          5.8.4  Trends  1n Visibility as Related to Sulfate Concentrations 	  5-78
     5.9  Conclusions 	  5-78
     5.10 References  	  5-84
                                         XVI

-------
Table of Contents (continued)

                                                                                     Page

A-6  PRECIPITATION SCAVENGING PROCESSES

     6.1  Introduction 	  6-1
     6.2  Steps In the Scavenging Sequence 	  6-2
          6.2.1  Introduction 	  6-2
          6.2.2  Intermixing of Pollutant and Condensed Water (Step  1-2)  	  6-5
          6.2.3  Attachment of Pollutant to Condensed Water Elements (Step 2-3)  ...  6-6
          6.2.4  Aqueous-Phase Reactions (Step 3-4)  	  6-13
          6.2.5  Deposition of Pollutant with Precipitation (Steps 3-5 and 4-5)  ...  6-13
          6.2.6  Combined Processes and the Problem  of Scavenging  Calculations ....  6-16
     6.3  Storm Systems and Storm Climatology 	  6-16
          6.3.1  Introduction 	  6-16
          6.3.2  Frontal  Storm Systems 	  6-17
                 6.3.2.1   Warm-Front Storms 	  6-19
                 6.3.2.2   Cold-Front Storms 	  6-23
                 6.3.2.3   Occluded-Front Storms 	  6-23
          6.3.3  Convectlve Storm Systems 	  6-23
          6.3.4  Additional Storm Types:   Nonideal Frontal  Storms, Orographlc
                 Storms and Lake-Effect Storms 	  6-27
          6.3.5  Storm and Precipitation Climatology 	  6-28
                 6.3.5.1   Precipitation Climatology  	  6-28
                 6.3.5.2   Storm Tracks 	  6-28
                 6.3.5.3   Storm Duration Statistics  	  6-31
     6.4  Summary of Precipitation-Scavenging Field  Investigations 	  6-31
     6.5  Predictive and  Interpretive Models of Scavenging  	  6-41
          6.5.1  Introduction 	  6-41
          6.5.2  Elements of a Scavenging Model  	  6-50
                 6.5.2.1   Material  Balances 	  6-50
                 6.5.2.2   Energy Balances 	  6-52
                 6.5.2.3   Momentum Balances 	  6-52
          6.5.3  Definitions of Scavenging Parameters 	  6-53
          6.5.4  Formulation of Scavenging Models:   Simple  Examples
                 of Microscopic and Macroscopic Approaches  	  6-58
          6.5.5  Systematic Selection of Scavenging  Models:
                 A Flow Chart Approach 	  6-61
     6.6  Practical  Aspects of Scavenging Models:  Uncertainty  Levels and  Sources
          of Error 	  6-64
     6.7  Conclusions 	  6-68
     6.8  References	  6-71


A-7  DRY DEPOSITION  PROCESSES

     7.1  Introduction 	  7-1
     7.2  Factors Affecting Dry Deposition 	  7-1
          7.2.1  Introduction 	  7-1
          7.2.2  Aerodynamic Factors 	•;	  7-6
          7.2.3  The Quasi-Laminar  Layer  	  7-9
          7.2.4  Phoretlc Effects and Stefan Flow  	  7-13
          7.2.5  Surface  Adhesion 	  7-14
          7.2.6  Surface  Biological  Effects 	  7-15
          7.2.7  Deposition to Liquid Water Surfaces 	  7-16
          7.2.8  Deposition to Mineral  and Metal Surfaces 	  7-17
          7.2.9  Fog and  Dewfall  	  7-19
          7.2.10 Resuspenslon and Surface Emission 	  7-20
          7.2.11 The Resistance Analog  	  7-21
     7.3  Methods for Studying Dry  Deposition	  7-27
          7.3.1  Direct Measurement 	  7-27
          7.3.2  Wind-Tunnel  and Chamber  Studies  	  7-29
          7.3.3  Mlcrometeorological  Measurement Methods  	  7-33
                                        xvn

-------
Table of Contents (continued)

                                                                                    Page

     7.4  Field Investigations of Dry  Deposition  	  7-37
          7.4.1  Gaseous Pollutants  	  7-37
          7.4.2  Partlculate Pollutants  	  7-44
          7.4.3  Routine Handling 1n Networks  	  7-50
     7.5  M1crometeorolog1cal  Models of  the  Dry Deposition Process  	  7-51
          7.5.1  Gases 	  7-51
          7.5.2  Particles 	  7-53
     7.6  Summary 	  7-54
     7.7  Conclusions 	  7-58
     7.8  References 	  7-60


A-8  DEPOSITION MONITORING

     8.1  Introduction 	  8-1
     8.2  Wet Deposition Networks 	  8-2
          8.2.1  Introduction and Historical Background  	  8-2
          8.2.2  Definitions 	  8-3
          8.2.3  Methods, Procedures and Equipment  for Wet Deposition Networks  ....  8-5
          8.2.4  Wet Deposition Network  Data Bases  	  8-7
     8.3  Monitoring Capabilities for  Dry Deposition  	  8-12
          8.3.1  Introduction 	  8-12
          8.3.2  Methods for Monitoring  Dry  Deposition  	  8-18
                 8.3.2.1  Direct Collection  Procedures  	  8-19
                 8.3.2.2  Alternative  Methods  	  8-20
          8.3.3  Evaluations of Dry  Deposition Rates  	  8-22
     8.4  Wet Deposition Network Data  With Applications  to Selected Problems  	  8-31
          8.4.1  Spatial Patterns 	  8-31
          8.4.2  Remote Site pH Data 	  8-50
          8.4.3  Precipitation Chemistry Variations Over Time  	  8-60
                 8.4.3.1  Nitrate Variation  Since 1950's 	  8-60
                 8.4.3.2  pH Variation Since 1950's 	  8-63
                 8.4.3.3  Calcium Variation  Since the 1950's  	  8-67
          8.4.4  Seasonal Variations 	  8-67
          8.4.5  Very Short Time Scale Variations 	  8-69
          8.4.6  Air Parcel Trajectory Analysis  	  8-69
     8.5  Gladochemical Investigations  as a Tool 1n  the Historical Delineation of
          the Acid Precipitation Problems 	  8-71
          8.5.1  Glaciochemlcal Data 	  8-72
                 8.5.1.1  Sulfate -  Polar Glaciers  	  8-73
                 8.5.1.2  Nitrate -  Polar Glaciers  	  8-73
                 8.5.1.3  pH and Acidity - Polar  Glaciers 	  8-74
                 8.5.1.4  Sulfate -  Alpine Glaciers 	  8-74
                 8.5.1.5  Nitrate -  Alpine Glaciers 	  8-74
                 8.5.1.6  pH and Acidity - Alpine Glaciers  	  8-75
          8.5.2  Trace Metals - General  Statement	  8-75
                 8.5.2.1  Trace Metals - Polar Glaciers  	  8-76
                 8.5.2.2  Trace Metals - Alpine Glaciers 	  8-77
          8.5.3  Discussion and Future Work  	  8-78
     8.6  Conclusions 	  8-80
     8.7  References 	  8-85


A-9  LONG-RANGE TRANSPORT AND ACIDIC DEPOSITION MODELS

     9.1  Introduction 	  9-1
          9.1.1  General Principles  for  Formulating Pollution  Transport and
                 Diffusion Models 	  9-1
          9.1.2  Model Characteristics 	  9-3
                 9.1.2.1  Spatial and  Temporal Scales 	  9-3
                 9.1.2.2  Treatment  of Turbulence 	  9-3
                                         xvm

-------
Table of Contents (continued)

                                                                                    Page

                 9.1.2.3  Reaction Mechanisms 	  9-5
                 9.1.2.4  Removal  Mechanisms 	  9-5
          9.1.3  Selecting Models for Application 	  9-6
                 9.1.3.1  General  	  9-6
                 9.1.3.2  Spatial  Range of Application  	  9-6
                 9.1.3.3  Temporal Range of Application  	  9-6
     9.2  Types of LRT Models	  9-9
          9.2.1  Eulerlan Grid Models 	  9-9
          9.2.2  Lagranglan Models 	  9-9
                 9.2.2.1  Lagranglan Trajectory Models  	  9-9
                 9.2.2.2  Statistical Trajectory Models  	  9-11
          9.2.3  Hybrid Models 	  9-13
     9.3  Modules Associated with Chemical  (Transformation)  Processes  	  9-13
          9.3.1  Overview	  9-13
          9.3.2  Chemical Transformation Modeling 	  9-14
                 9.3.2.1  Simplified Modules 	  9-14
                 9.3.2.2  Multlreaction Modules 	  9-15
          9.3.3  Modules for NOX Transformation 	  9-16
     9.4  Modules Associated with Wet and Dry Deposition 	  9-17
          9.4.1  Overview 	  9-17
          9.4.2  Modules for Wet Deposition 	  9-20
                 9.4.2.1  Formulation and Mechanism 	  9-20
                 9.4.2.2  Modules Used 1n Existing Models  	  9-21
                 9.4.2.3  Wet  Deposition Modules for Snow	  9-23
                 9.4.2.4  Wet  Deposition Modules for NOX 	  9-23
          9.4.3  Modules for Dry Deposition 	  9-24
                 9.4.3.1  General  Considerations 	  9-24
                 9.4.3.2  Modules Used In Existing Models  	  9-25
                 9.4.3.3  Dry  Deposition Modules for NOx 	  9-26
          9.4.4  Dry Versus Wet Deposition 	  9-26
     9.5  Status of LRT Models as Operational  Tools 	  9-26
          9.5.1  Overview	  9-26
          9.5.2  Model Application 	  9-27
                 9.5.2.1  Limitations In Applicability  	  9-27
                 9.5.2.2  Regional Concentration and Deposition Patterns  	  9-27
                 9.5.2.3  Use  of Matrix Methods to Quantify  Source-Receptor
                          Relationships 	  9-28
          9.5.3  Data Requirements	  9-33
                 9.5.3.1  General  	  9-33
                 9.5.3.2  Specific Characteristics of Data Used In Model
                          Simulations 	  9-36
                          9.5.3.2.1   Emissions 	  9-36
                          9.5.3.2.2   Meteorological  Data 	  9-37
          9.5.4  Model Performance and Uncertainties 	  9-37
                 9.5.4.1  General  	  9-37
                 9.5.4.2  Data Bases Available for Evaluating Models 	  9-39
                 9.5.4.3  Performance Measures 	  9-39
                 9.5.4.4  Representativeness of Measurements 	  9-40
                 9.5.4.5  Uncertainties 	  9-40
                 9.5.4.6  Selected Results  	  9-40
     9.6  Conclusions 	  9-46
     9.7  References 	  9-48
                                          XIX

-------
                      THE ACIDIC DEPOSITION  PHENOMENON AND  ITS EFFECTS:
                             CRITICAL  ASSESSMENT  REVIEW PAPERS

                                    Table of Contents

                                         Volume  II
                                    Effects Sciences
                                                                                   Page

AUTHORS 	   111

PREFACE 	     1x

ABBREVIATION-ACRONYM LIST 	  xxlx

GLOSSARY 	  xl 111


E-l  INTRODUCTION

     1.1  Objectives 	   1-1
     1.2  Approach 	   1-1
     1.3  Chapter Organization and General Content	   1-3
          1.3.1  Effects on Soil Systems 	   1-3
          1.3.2  Effects on Vegetation 	   1-4
          1.3.3  Effects on Aquatic Chemistry 	   1-5
          1.3.4  Effects on Aquatic Biology  	   1-5
          1.3.5  Indirect Effects on Health  	   1-6
          1.3.6  Effects on Materials  	   1-6
     1.4  Acidic Deposition 	   1-6
     1.5  Linkage to Atmospheric Sciences 	   1-7
     1.6  Sensitivity 	   1-7
     1.7  Presentation Limitations 	   1-7


E-2  EFFECTS ON SOIL SYSTEMS

     2.1  Introduction 	   2-1
          2.1.1  Importance of Soils to Aquatic  Systems  	   2-1
                 2.1.1.1  Soils Buffer Precipitation Enroute to  Aquatic  Systems  ...   2-2
                 2.1.1.2  Soil as a Source of Acidity for Aquatic  Systems  	   2-2
          2.1.2  Soil's Importance as  a Medium for  Plant Growth  	   2-2
          2.1.3  Important Soil Properties 	   2-2
                 2.1.3.1  Soil Physical Properties  	   2-3
                 2.1.3.2  Soil Chemical Properties  	   2-3
                 2.1.3.3  Soil Microbiology  	   2-3
          2.1.4  Flow of Deposited Materials Through Soil Systems  	   2-3
     2.2  Chemistry of Acid Soils 	   2-5
          2.2.1  Development of Acid Soils 	   2-5
                 2.2.1.1  Biological Sources of  H+  Ions  	   2-6
                 2.2.1.2  Acidity from Dissolved Carbon  Dioxide  	   2-6
                 2.2.1.3  Leaching of  Basic  Cations 	   2-7
          2.2.2  Soil Cation Exchange  Capacity 	~.	   2-8
                 2.2.2.1  Source of Cation Exchange Capacity in  Soils 	   2-8
                 2.2.2.2  Exchangeable Bases and Base Saturation 	   2-8
          2.2.3  Exchangeable and Solution Aluminum in Soils 	   2-9
          2.2.4  Exchangeable and Solution Manganese In  Soils 	   2-12
          2.2.5  Practical Effects of  Low pH 	   2-12
          2.2.6  Neutralization of Soil Acidity  	   2-13
          2.2.7  Measuring Soil pH 	   2-14
          2.2.8  Sulfate Adsorption 	   2-15
          2.2.9  Soil Chemistry Summary 	   2-18
     2.3  Effects of Acidic Deposition on Soil Chemistry and Plant Nutrition 	   2-18
          2.3.1  Effects on Soil pH 	   2-19
          2.3.2  Effects on Nutrient Supply  of Cultivated Crops  	   2-24
          2.3.3  Effects on Nutrient Supply  to Forests 	   2-28
                                           XX

-------
Table of Contents (continued)

                                                                                    Page

                 2.3.3.1  Effects on Cation Nutrient Status  	  2-28
                 2.3.3.2  Effects on S and N Status  	  2-31
                 2.3.3.3  Acidification Effects on Plant Nutrition  	  2-33
                          2.3.3.3.1  Nutrient deficiencies  	  2-33
                          2.3.3.3.2  Metal ion toxlcities 	  2-33
                                     2.3.3.3.2.1  Aluminum  toxidty  	  2-34
                                     2.3.3.3.2.2  Manganese toxidty 	  2-35
          2.3.4  Reversibility of Effects on Soil  Chemistry  	  2-35
          2.3.5  Predicting Which Soils will be Affected Most 	  2-36
                 2.3.5.1  Soils Under Cultivation  	  2-36
                 2.3.5.2  Uncultivated, Unamended  Soils 	  2-36
                          2.3.5.2.1  Basic catlon-pH changes 1n  forested  soils  ....  2-37
                          2.3.5.2.2  Changes in aluminum concentration 1n soil
                                     solution in forested soils  	  2-40
     2.4  Effects of Acidic Deposition on Soil Biology 	  2-40
          2.4.1  Soil Biology Components and Functional Significance 	  2-40
                 2.4.1.1  Soil Animals 	  2-40
                 2.4.1.2  Algae 	  2-40
                 2.4.1.3  Fungi 	  2-41
                 2.4.1.4  Bacteria 	  2-41
          2.4.2  Direct Effects of Acidic Deposition on Soil Biology 	  2-42
                 2.4.2.1  Soil Animals 	  2-42
                 2.4.2.2  Terrestrial Algae 	  2-42
                 2.4.2.3  Fungi 	  2-43
                 2.4.2.4  Bacteria 	  2-43
                 2.4.2.5  General Biological Processes 	  2-44
          2.4.3  Metals—Mobilization Effects on Soil Biology 	  2-45
          2.4.4  Effects of Changes in Microbial Activity on Aquatic Systems 	  2-46
          2.4.5  Soil Biology Summary 	  2-46
     2.5  Effects of Acidic Deposition on Organic  Matter Decomposition 	  2-47
     2.6  Effects of Soils on the Chemistry of Aquatic Ecosystems 	  2-52
     2.7  Conclusions 	  2-54
     2.8  References 	  2-57


E-3  EFFECTS ON VEGETATION

     3.1  Introduction  	  3-1
          3.1.1  Overview	  3-1
          3.1.2  Background 	  3-1
     3.2  Plant Response to Acidic Deposition  	  3-3
          3.2.1  Leaf Response to Acidic Deposition 	  3-3
                 3.2.1.1  Leaf Structure and Functional Modifications 	  3-5
                 3.2.1.2  Foliar Leaching - Throughfall Chemistry 	  3-8
          3.2.2  Effects of Acidic Deposition on Lichens and Mosses 	  3-13
          3.2.3  Summary 	  3-16
     3.3  Interactive Effects of Acidic Deposition with Other Environmental
          Factors on Plants 	  3-17
          3.3.1  Interactions with Other Pollutants 	  3-17
          3.3.2  Interactions with Phytophagous Insects 	  3-20
          3.3.3  Interactions with Pathogens  	  3-20
          3.3.4  Influence on Vegetative Hosts That Would Alter Relationships
                 with Insect or Microbial Associate 	  3-23
          3.3.5  Effects of Acidic Deposition on Pesticides 	  3-23
          3.3.6  Summary 	  3-25
     3.4  Blomass Production  	  3-26
          3.4.1  Forests 	  3-26
                 3.4.1.1  Possible Mechanlslms of Response  	  3-27
                 3.4.1.2  Phenologlcal Effects  	  3-29
                          3.4.1.2.1  Seed germination and seedling establishment ..  3-29
                          3.4.1.2.2  Mature and reproductive stages 	  3-32
                 3.4.1.3  Growth of Seedlings and Trees In  Irrigation
                          Experiments  	  3-32
                                           XXI

-------
Table of Contents (continued)

                                                                                    Page

                 3.4.1.4  Studies of Long-Term Growth of Forest Trees  	  3-33
                 3.4.1.5  Dleback and Decline 1n  High Elevation Forests  	  3-36
                 3.4.1.6  Recent Observations on  the German Forest Decline
                          Phenomenon 	  3-39
                 3.4.1.7  Summary 	  3-41
          3.4.2  Crops 	  3-41
                 3.4.2.1  Review and Analysis of  Experimental  Design  	  3-42
                          3.4.2.1.1   Dose-response determination 	  3-42
                          3.4.2.1.2   Sensitivity  classification 	  3-44
                          3.4.2.1.3   Mechanisms 	  3-44
                          3.4.2.1.4   Characteristics of precipitation  simulant
                                     exposures 	  3-45
                          3.4.2.1.5   Yield criteria 	  3-45
                 3.4.2.2  Experimental  Results 	  3-46
                          3.4.2.2.1   Field studies 	  3-46
                          3.4.2.2.2   Controlled environment studies 	  3-50
                 3.4.2.3  Discussion 	  3-58
                 3.4.2.4  Summary 	  3-61
     3.5  Conclusions 	  3-61
     3.6  References 	  3-64


E-4  EFFECTS ON AQUATIC CHEMISTRY

     4.1  Introduction 	  4-1
     4.2  Basic Concepts Required to Understand the Effects of
          Acidic Deposition on Aquatic  Systems 	  4-2
          4.2.1  Receiving Systems 	  4-2
          4.2.2  pH, Conductivity, and  Alkalinity 	  4-3
                 4.2.2.1  pH 	  4-3
                 4.2.2.2  Conductivity  	  4-4
                 4.2.2.3  Alkalinity 	  4-5
          4.2.3  Acidification 	  4-6
     4.3  Sensitivity of Aquatic  Systems  to Acidic Deposition  	  4-7
          4.3.1  Atmospheric Inputs  	  4-7
                 4.3.1.1  Components of Deposition 	  4-7
                 4.3.1.2  Loading vs Concentration 	  4-8
                 4.3.1.3  Location of the Deposition 	  4-8
                 4.3.1.4  Temporal Distribution of Deposition  	  4-9
                 4.3.1.5  Importance of Atmospheric Inputs  to  Aquatic  Systems 	  4-9
                          4.3.1.5.1   Nitrogen (N), phosphorus  (P),  and
                                     carbon (C) 	  4-9
                          4.3.1.5.2   Sulfur 	  4-10
          4.3.2  Characteristics  of  Receiving Systems Relative to  Being Able to
                 Assimilate Acidic Deposition	  4-13
                 4.3.2.1  Canopy  	  4-13
                 4.3.2.2  Soil  	  4-14
                 4.3.2.3  Bedrock 	  4-16
                 4.3.2.4  Hydrology  	  4-17
                          4.3.2.4.1   Flow paths 	  4-17
                          4.3.2.4.2   Residence times 	  4-22
                 4.3.2.5  Wetlands 	  4-23
                 4.3.2.6  Aquatic 	  4-24
                          4.3.2.6.1   Alkalinity as an Indicator of  sensitivity	  4-24
                          4.3.2.6.2   International  production/consumption
                                     of ANC 	  4-28
                          4.3.2.6.3   Aquatic  sediments 	  4-31
          4.3.3  Location of Sensitive  Systems 	  4-32
          4.3.4  Summary—Sensitivity 	  4-35
     4.4  Magnitude of Chemical Effects of Acidic  Deposition on
          Aquatic Ecosystems 	  4-38
                                         XX11

-------
Table of Contents (continued)

                                                                                    Page

          4.4.1   Relative Importance  of HN03 vs HaSCH  	  4-39
          4.4.2   Short-Term Acidification  	  4-45
          4.4.3   Long-Term Acidification  	  4-48
                 4.4.3.1   Analysis  of Trends based on  Historic Measurements of
                          Surface Water Quality 	  4-53
                          4.4.3.1.1   Methodological problems with the evaluation
                                     of historical trends  	  4-53
                                     4.4.3.1.1.1  pH  	  4-54
                                                 4.4.3.1.1.1.1  pH-early method-
                                                                ology 	  4-54
                                                 4.4.3.1.1.1.2  pH-current method-
                                                                ology 	  4-55
                                                 4.4.3.1.1.1.3  pH-comparabH1ty
                                                                of early and cur-
                                                                rent measurement
                                                                methods 	  4-56
                                                 4.4.3.1.1.1.4  pH-general
                                                                problems 	  4-57
                                     4.4.3.1.1.2  Conductivity 	  4-60
                                                 4.4.3.1.1.2.1  Conductivity
                                                                methodol ogy 	  4-60
                                                 4.4.3.1.1.2.2  Conductivity-com-
                                                                parability of
                                                                early and current
                                                                measurement
                                                                methods 	  4-60
                                                 4.4.3.1.1.2.3  Conductivity-gen-
                                                                eral problems ....  4-61
                                     4.4.3.1.1.3  Alkalinity 	  4-61
                                                 4.4.3.1.1.3.1  Alkalinity-early
                                                                methodology 	  4-61
                                                 4.4.3.1.1.3.2  Alkalinity-current
                                                                methodology 	  4-62
                                                 4.4.3.1.1.3.3  Alkalinity-compar-
                                                                ability of early
                                                                'and current meas-
                                                                urement methods ..  4-63
                                     4.4.3.1.1.4  Sample storage 	  4-63
                                     4.4.3.1.1.5  Summary  of measurement
                                                 techniques 	  4-63
                          4.4.3.1.2   Analysis of trends 	  4-64
                                     4.4.3.1.2.1  Introduction 	  4-64
                                     4.4.3.1.2.2  Canadian studies 	  4-66
                                     4.4.3.1.2.3  United States studies 	  4-74
                          4.4.3.1.3   Summary—trends  1n historic data 	  4-98
                 4.4.3.2   Assessment  of Trends Based  on Paleollmnologlcal
                          Technique 	  4-99
                          4.4.3.2.1   Calibration and  accuracy of paleolimnologlcal
                                     reconstruction of pH  history 	  4-100
                          4.4.3.2.2   Lake  acidification determined by
                                     paleolimnologlcal reconstruction 	  4-100
                 4.4.3.3   Alternate Explanations for  Acidification-Land Use
                          Changes  	  4-105
                          4.4.3.3.1   Variations in the groundwater tabje 	  4-105
                          4.4.3.3.2   Accelerated mechanical weathering or
                                     land  scarification 	  4-105
                          4.4.3.3.3   Decomposition of organic matter 	  4-106
                          4.4.3.3.4   Changes In vegetation 	  4-106
                          4.4.3.3.5   Chemical amendments 	  4-107
                          4.4.3.3.6   Summary—alternate explanations for
                                     acidification 	  4-107
                                        xxm

-------
Table of Contents (continued)

                                                                                    Page

          4.4.4   Summary—Magnitude of Chemical Effects of Acidic Deposition	  4-109
     4.5  Predictive Modeling of the Effects of Addle Deposition
          on Surface Waters  	  4-113
          4.5.1   Aimer/Dlckson Relationship 	  4-114
          4.5.2   HenMksen's Predictor Nomograph 	  4-119
          4.5.3   Thompson's  Cation Denudation Rate Model (CDR) 	  4-121
          4.5.4   "Trickle-Down" Model 	  4-122
          4.5.5   Summary  of  Predictive Modeling 	  4-125
     4.6  Indirect Chemical  Changes Associated with Acidification
          of Surface Waters  	  4-128
          4.6.1   Metal s 	  4-128
                 4.6.1.1   Increased Loading of Metals From Atmospheric
                          Deposition  	  4-129
                 4.6.1.2  Mobilization of Metals by Acidic Deposition 	  4-130
                 4.6.1.3   Secondary Effects of Metal Mobilization 	  4-131
                 4.6.1.4   Effects of Acidification on Aqueous Metal  Speclatlon ....  4-132
                 4.6.1.5   Indirect Effects on Metals 1n Surface Waters 	  4-132
          4.6.2   Aluminum Chemistry In Dilute Acidic Waters 	  4-132
                 4.6.2.1   Occurrence, Distribution, and Sources of Aluminum 	  4-132
                 4.6.2.2   Aluminum Speclatlon  	  4-136
                 4.6.2.3   Aluminum as a  pH Buffer  	  4-136
                 4.6.2.4   Temporal and Spatial Variations 1n Aqueous
                          Level s of Al um1 num  	  4-137
                 4.6.2.5   The Role of Aluminum 1n Altering Element Cycling
                          Within Acidic  Waters 	  4-140
          4.6.3   Organlcs 	  4-141
                 4.6.3.1   Atmospheric Loading of Strong Acids and Associated
                          Organic Mlcropollutants  	  4-141
                 4.6.3.2   Organic Buffering Systems 	  4-142
                 4.6.3.3   Organo-Metalllc Interactions  	  4-142
                 4.6.3.4   Photochemistry 	  4-143
                 4.6.3.5   Carbon-Phosphorus-Alumlnum Interactions 	  4-143
                 4.6.3.6   Effects of Acidification on Organic Decomposition
                          1n Aquatic Systems  	  4-144
     4.7  M1t1gat1ve Strategies for Improvement of Surface Water Quality  	  4-144
          4.7.1   Base  Additions 	  4-144
                 4.7.1.1   Types of Basic Materials  	  4-144
                 4.7.1.2   Direct Water Addition of Base 	  4-148
                          4.7.1.2.1  Computing base dose requirements 	  4-148
                          4.7.1.2.2  Methods  of base application 	  4-152
                 4.7.1.3   Watershed Addition  of Base 	  4-154
                          4.7.1.3.1  The concept of watershed
                                     application of base 	  4-154
                          4.7.1.3.2  Experience 1n watershed liming  	  4-156
                 4.7.1.4   Water Quality  Response to Base Treatment  	  4-158
                 4.7.1.5   Cost Analysis, Conclusions and Assessment of Base
                          Addition  	  4-160
                          4.7.1.5.1  Cost analysis  	  4-160
                          4.7.1.5.2   Summary—base additions  	  4-162
          4.7.2   Surface  Water  Fertilization  	  4-162
                 4.7.2.1   The Fertilization Concept  	  4-162
                 4.7.2.2   Phosphorous Cycling In Acidified Water 	  4-164
                 4.7.2.3   Fertilization  Experience and  Water
                          Quality Response to Fertilization  	  4-164
                 4.7.2.4   Summary—Surface Water Fertilization  	  4-166
     4.8  Conclusions 	  4-166
     4.9  References  	  4-169
                                        XXIV

-------
Table of Contents (continued)

                                                                                    Page

E-5  EFFECTS ON AQUATIC BIOLOGY

     5.1  Introduction 	  5-1
     5.2  Biota of Naturally Acidic  Waters  	  5-3
          5.2.1  Types of Naturally  Acidic  Waters  	  5-3
          5.2.2  Biota of Inorganic  Acldotrophlc Waters  	  5-4
          5.2.3  Biota In Acidic Brownwater Habitats  	  5-5
          5.2.4  Biota In Ultra-OHgotropMc Waters  	  5-7
          5.2.5  Summary 	  5-9
     5.3  Benthlc Organisms 	  5-14
          5.3.1  Importance of the Benthlc  Community  	  5-14
          5.3.2  Effects of Acidification on Components  of the Benthos 	  5-16
                 5.3.2.1  Mlcroblal  Community 	  5-16
                 5.3.2.2  Perlphyton 	  5-17
                          5.3.2.2.1   Field  surveys  	  5-17
                          5.3.2.2.2   Temporal  trends  	  5-18
                          5.3.2.2.3   Experimental  studies  	  5-20
                 5.3.2.3  Mlcrolnvertebrates 	  5-21
                 5.3.2.4  Crustacea  	  5-22
                 5.3.2.5  Insecta 	  5-24
                          5.3.2.5.1   Sensitivity of different groups	  5-24
                          5.3.2.5.2   Sensitivity of Insects from different
                                     mlcrohabltats  	  5-29
                          5.3.2.5.3   Acid sensitivity of Insects based on food
                                     sources 	  5-29
                          5.3.2.5.4   Mechanisms of effects and trophic
                                     Interactions  	  5-29
                 5.3.2.6  Mollusca 	  5-30
                 5.3.2.7  Annelida 	  5-31
                 5.3.2.8  Summary of Effects of Acidification on Benthos 	  5-32
     5.4  Macrophytes and Wetland PI ants 	  5-37
          5.4.1  Introduction  	  5-37
          5.4.2  Effects on Acidification on Aquatic Macrophytes 	  5-41
          5.4.3  Summary 	  5-43
     5.5  Plankton 	  5-44
          5.5.1  Introduction  	  5-44
          5.5.2  Effects of Acidification on Phytoplankton 	  5-45
                 5.5.2.1   Changes In Species Composition 	  5-45
                 5.5.2.2   Changes In Phytoplankton Blomass and Productivity 	  5-52
          5.5.3  Effects of Acidification on  Zooplankton 	  5-55
          5.5.4  Explanations  and Significance 	  5-67
                 5.5.4.1   Changes In Species  Composition 	  5-67
                 5.5.4.2   Changes 1n Productivity  	  5-70
          5.5.5  Summary  	  5-72
     5.6  Fish	  5-74
          5.6.1  Introduction  	  5-74
          5.6.2  Field Observations  	  5-75
                 5.6.2.1   Loss of Populations  	  5-75
                          5.6.2.1.1   United  States 	  5-75
                                     5.6.2.1.1.1   Adirondack Region of
                                                 New York State 	  5-75
                                     5.6.2.1.1.2   Other regions of the eastern
                                                 United States 	  5-79
                          5.6.2.1.2   Canada  	  5-79
                                     5.6.2.1.2.1   LaCloche Mountain Region  of
                                                 Ontario 	  5-79
                                    5.6.2.1.2.2   Other areas of Ontario  	  5-83
                                    5.6.2.1.2.3   Nova Scotia 	  5-83
                         5.6.2.1.3  Scandinavia  and Great Britain 	  5-88
                                    5.6.2.1.3.1   Norway 	  5-88
                                    5.6.2.1.3.2   Sweden 	  5-93
                                    5.6.2.1.3.3   Scotland 	  5-93
                                        XXV

-------
Table of Contents (continued)

                                                                                    Page

                 5.6.2.2  Population  Structure  	  5-93
                 5.6.2.3  Growth  	  5-98
                 5.6.2.4  Episodic F1sh Kills  	  5-99
                 5.6.2.5  Accumulation of Metals  1n F1sh  	  5-101
          5.6.3  Field Experiments  	  5-101
                 5.6.3.1  Experimental Acidification of Lake 223 Ontario  	  5-102
                 5.6.3.2  Experimental Acidification of Norrls
                          Brook,  New  Hampshire  	  5-104
                 5.6.3.3  Exposure of Fish  to Acidic Surface Waters  	  5-104
          5.6.4  Laboratory Experiments  	  5-108
                 5.6.4.1  Effects of  Low pH 	  5-109
                          5.6.4.1.1   Survival  	  5-109
                          5.6.4.1.2   Reproduction 	  5-112
                          5.6.4.1.3   Growth 	  5-119
                          5.6.4.1.4   Behavior  	  5-119
                          5.6.4.1.5   Physiological responses  	  5-120
                 5.6.4.2  Effects of  Aluminum and Other Metals  In  Acidic  Waters  ...  5-122
          5.6.5  Summary	  5-125
                 5.6.5.1  Extent  of Impact  	  5-125
                 5.6.5.2  Mechanism of Effect  	  5-127
                 5.6.5.3  Relationship Between  Water Acidity  and F1sh
                          Population  Response  	  5-128
     5.7  Other Related Biota 	  5-129
          5.7.1  Amphibians 	  5-129
          5.7.2  Birds 	  5-134
                 5.7.2.1  Food Chain  Alterations  	  5-134
                 5.7.2.2  Heavy Metal Accumulation 	  5-134
          5.7.3  Mammals 	  5-135
          5.7.4  Summary 	  5-136
     5.8  Observed and Anticipated Ecosystem Effects  	  5-139
          5.8.1  Ecosystem Structure  	  5-139
          5.8.2  Ecosystem Function  	  5-141
                 5.8.2.1  Nutrient Cycling  	  5-141
                 5.8.2.2  Energy  Cycling  	  5-141
          5.8.3  Summary 	  5-142
     5.9  Mitlgatlve Options Relative to Biological Populations at Risk  	  5-143
          5.9.1  Biological Response  to  Neutralization 	  5-143
          5.9.2  Improving F1sh Survival in Acidified  Waters  	  5-145
                 5.9.2.1  Genetic Screening 	  5-145
                 5.9.2.2  Selective Breeding 	  5-145
                 5.9.2.3  Acclimation 	  5-146
                 5.9.2.4  Limitations of Techniques to Improve  Fish  Survival  	  5-148
          5.9.3  Summary 	  5-149
     5.10 Conclusions	  5-149
          5.10.1  Effects of Acidification  on Aquatic  Organisms 	  5-149
          5.10.2  Processes and Mechanisms  by Which Acidification
                  Alters Aquatic  Ecosystems 	  5-155
                  5.10.2.1  Direct  Effects  of Hydrogen Ions 	  5-155
                  5.10.2.2  Elevated  Metal  Concentrations 	  5-156
                  5.10.2.3  Altered Trophic-Level Interactions  	  5-156
                  5.10.2.4  Altered Water Clarity 	  5-157
                  5.10.2.5  Altered Decomposition of  Organic Matter  	  5-157
                  5.10.2.6  Presence  of  Algal Mats 	  5-157
                  5.10.2.7  Altered Nutrient Availability 	  5-157
                  5.10.2.8  Interaction  of  Stresses  	  5-157
          5.10.3  Biological Mitigation  	  5-158
          5.10.4  Summary  	  5-159
     5.11  References 	  5-160
                                         XXVI

-------
Table of Contents (continued)

                                                                                    Page

E-6  INDIRECT EFFECTS ON HEALTH

     6.1  Introduction 	  6-1
     6.2  Food Chain Dynamics  	  6-1
          6.2.1  Introduction  	  6-1
          6.2.2  Availability  and  Bloaccumulatlon of  Toxic Metals  	  6-2
                 6.2.2.1  Speclatlon (Mercury)  	  6-2
                 6.2.2.2  Concentrations  and  Speclatlons  In Water  (Mercury)  	  6-4
                 6.2.2.3  Flow of  Mercury 1n  the Environment  	  6-4
                          6.2.2.3.1   Global cycles  	  6-4
                          6.2.2.3.2   Blogeochemlcal cycles of mercury  	  6-5
          6.2.3  Accumulation  In F1sh 	  6-10
                 6.2.3.1  Factors  Affecting Mercury Concentrations In  Fish  	  6-10
                 6.2.3.2  Historical  and  Geographic Trends 1n Mercury  Levels 1n
                          Freshwater F1 sh 	  6-20
          6.2.4  Dynamics and  Toxlclty 1n Humans (Mercury) 	  6-22
                 6.2.4.1  Dynamics In Man (Mercury)  	  6-22
                 6.2.4.2  Toxlclty In Man 	  6-23
                 6.2.4.3  Human Exposure  from Fish  and Potential for Health
                          Risks 	  6-27
     6.3  Ground, Surface and  Cistern Waters  as Affected  by Acidic Deposition 	  6-31
          6.3.1  Water Supplies 	  6-32
                 6.3.1.1  Direct Use of Precipitation (Cisterns) 	  6-32
                 6.3.1.2  Surface  Water Supplies 	  6-34
                 6.3.1.3  Groundwater Suppl1es  	  6-37
          6.3.2  Lead 	  6-39
                 6.3.2.1  Concentrations  1n Noncontamlnated Waters 	  6-39
                 6.3.2.2  Factors  Affecting Lead Concentrations
                          1n Water,  Including Effects of  pH 	  6-39
                 6.3.2.3  Speclatlon of Lead  In Natural Water 	  6-41
                 6.3.2.4  Dynamics and Toxlclty of  Lead 1n Humans  	  6-41
                          6.3.2.4.1   Dynamics of lead In  humans  	  6-41
                          6.3.2.4.2   Toxic effects  of lead on humans 	  6-42
                          6.3.2.4.3   Intake of lead 1n water  and potential for
                                     human health effects 	  6-49
          6.3.3  Aluminum 	  6-51
                 6.3.3.1  Concentrations  In Uncontamlnated Waters  	  6-53
                 6.3.3.2  Factors  Affecting Aluminum  Concentrations In Water 	  6-53
                 6.3.3.3  Speclatlon of Aluminum In Water 	  6-54
                 6.3.3.4  Dynamics and Toxlclty 1n  Humans 	  6-54
                          6.3.3.4.1   Dynamics of aluminum 1n  humans 	  6-54
                          6.3.3.4.2   Toxic effects  of aluminum In  humans  	  6-55
                 6.3.3.5  Human Health Risks  from Aluminum In Water 	  6-55
    6.4  Other Metals 	  6-55
    6.5  Conclusions 	  6-56
    6.6  References 	  6-58


E-7  EFFECTS ON MATERIALS

     7.1  Direct Effects on Materials 	  7-1
          7.1.1  Introduction  	  7-1
                 7.1.1.1  Long Range and  Local Effects 	  7-2
                 7.1.1.2  Inaccurate Claims of Acid Rain  Damage to Materials 	  7-2
                 7.1.1.3  Complex  Mechanisms  of Exposure  and Deposition 	  7-5
                 7.1.1.4  Deposition  Velocities 	  7-6
                 7.1.1.5  Laboratory vs Field Studies 	  7-6
          7.1.2  Damage to Materials by Acidic Deposition 	  7-8
                 7.1.2.1  Metals 	  7-9
                          7.1.2.1.1   Ferrous  Metals	  7-11
                                     7.1.2.1.1.1  Laboratory Studies 	  7-13
                                     7.1.2.1.1.2  Field Studies  	  7-14
                                        xxvn

-------
Table of Contents (continued)

                                                                                    Page

                         7.1.2.1.3  Nonferrous Metals  	  7-17
                                    7.1.2.1.2.1  Aluminum 	  7-17
                                    7.1.2.1.2.2  Copper  	  7-19
                                    7.1.2.1.2.3  Z1nc  	  7-19
                 7.1.2.2  Masonry  	  7-20
                         7.1.2.2.1  Stone  	  7-20
                         7.1.2.2.2  Concrete  	  7-26
                         7.1.2.2.3  Ceramics  and Glass 	  7-27
                 7.1.2.3  Paint	  7-27
                 7.1.2.4  Other Materials  	  7-31
                         7.1.2.4.1  Paper  	  7-32
                         7.1.2.4.2  Photographic Materials  	  7-32
                         7.1.2.4.3  Textiles  and Textile Dyes  	  7-32
                         7.1.2.4.4  Leather  	  7-34
                 7.1.2.5  Cultural Property 	  7-34
                         7.1.2.5.1  Architectural Monuments  	  7-34
                         7.1.2.5.2  Museums,  Libraries and  Archives  	  7-34
                         7.1.2.5.3  Medieval  Stained Glass  	  7-35
                         7.1.2.5.4  Conservation and Mitigation Costs	  7-35
                7.1.2.6  Economic  Implications 	  7-37
                7.1.2.6  M1t1gat1ve  Measures  	  7-38
     7.2  Potential Secondary Effects  of Acidic Deposition on Potable Water
          Piping Systems 	  7-39
          7.2.1  Introduction 	  7-39
          7.2.2  Problems Caused by  Corrosion  	  7-39
                 7.2.2.1   Health 	  7-39
                 7.2.2.2  Aesthetics 	  7-40
                 7.2.2.3  Economics  	  7-40
          7.2.3  Principles of Corrosion 	  7-40
          7.2.4  Factors Affecting Internal Piping Corrosion 	  7-42
          7.2.5  Corrosion of Materials Used In Plumbing  and Water
                 Distribution Systems  	  7-48
                 7.2.5.1  Corrosion  of Iron Pipe  	  7-48
                 7.2.5.2  Corrosion  of Galvanized Pipe  	  7-49
                 7.2.5.3  Corrosion  of Copper Pipe  	  7-49
                 7.2.5.4  Corrosion  of Lead Pipe  	  7-49
                 7.2.5.5  Corrosion  of Non-Metallic  Pipe  	  7-50
          7.2.6  Metal Leaching 	  7-50
                 7.2.6.1  Standing vs  Running Samples  	  7-50
                 7.2.6.2  Metals Surveys 	  7-51
          7.2.7  Corrosion Control Strategies 	  7-53
          7.2.8  Economics 	  7-53
     7.3  Conclusions  	  7-54
     7.4  References  	  7-58
                                        xxvm

-------
                        ABBREVIATION-ACRONYM LIST
6-ALA
ACHEX
ADI
Ag
AI
Al
A1203
AL
A1(OH)2H2P04
A1(OH)3
ANC
APN
ARL
ARS
As
ASTRAP

AWWA
B
BCF
BLM
BLMs
6-aminolevulinic acid
Aerosol Characterization Experiment
Acceptable daily intake
Silver
Aggresiveness index
Aluminum
Aluminum ion
Aluminum oxide
Aluminosilicate
Aeronomy Laboratory, NOAA
Varascite
Aluminum hydroxide
Acid neutralizing capacity
Air and Precipitation Monitoring Network
Air Resources Lab, NOAA
Agricultural Research Service, DOA
Arsenic
Advanced Statistical Trajectory Regional Air
  Pollution Control Model
American Water Works Association
Boron
Biconcentration factor
Bureau of Land Management, DOI
Boundary layer models
                                  xx ix

-------
BM
BNC
BNC aq
BOD
Br
BS
BSC
BUREC
BWCA
CB
Ca
CaCl
CaC03-MgC03
CaO
Ca(OH)2
CaS04
CaS04-K2S04-H20
CAMP
CANSAP
CAPTEX
CCN
Cd
CDR
Bureau of Mines, DOI
Base neutralizing capacity
Aqueous base neutralizing capacity
Biologic oxygen demand
Bromine
Base saturation
Base saturation capacity
Bureau of Reclamation, DOI
Boundary Water Canoe Area
Base cation level
Calcium
Calcium ion
Calcium chloride
Calcium carbonate or crystalline caldte - limestone
Dolomite
Calcium bicarbonate
Calcium oxide - Hme
Calcium hydroxide - 11me
Calcium sulfate, sulfate  salt
Syngenite
Continuous  Air  Monitoring Program
Canadian  Network for  Sampling Acid Precipitation
Cross-Appalachian Transport  Experiment
Cloud condensation nuclei
Cadmium
Cation denudation rate
                                    xxx

-------
 CEC
 CEQ
 CH3Br
 CH3C1
 CH3COOH
 (CH3)2Hg
 CH3Q
 (CH3)2S
 (CH3)2S2
 CH3SH
 CH4
 cr
 C12
 cm-* molecule'*  s~
 cm
 cm  s"l
 cm  yr~l
 CO
 C02
 -COOH
 COS
 Cr
 CS2
CSI
CSRS
Cu
 Cation exchange capacity
 Council  on  Environmental  Quality
 Methyl  bromide
 Methyl  chloride
 Acetic acid
 Dimethyl  mercury
 Methoxy  radical
 Dimethyl  sulflde (also  CH3SCH3)
 Dimethyl  dlsulflde
 Methyl  sulflde (or methyl  mercaptan)
 Methane
 Chloride  ion
 Elemental chlorine
 Cubic  centimeters per molecule per  second
 Centimeter
 Centimeters per  second
 Centimeters per year
 Carbon monoxide
 Carbon dioxide
 Carboxyl
 Carbonyl  sulflde
 Chromium
 Carbon disulfide
 Calcite saturation index
 Cooperative States Research Service, DOA
Copper
                                  xxxi

-------
DEC                          Department of Environmental Conservation, NY
DFI                          Driving  force index
DO                           Dissolved oxygen
DOA                          Department of Agriculture
DOC                          Dissolved organic  carbon
DOD                          Department of Defense
DOE                          Department of Energy
DOI                          Department of Interior
DOS                          Department of State
ELA                          Experimental  Lakes Area
emf                          Electromotive force
ENAMAP                       Eastern  North America Model of Air Pollutants
EPA                          Environmental Protection Agency
EPRI                         Electric Power Research Institute
eq                           Equivalent
eq ha"1 y1                  Equivalents per hectare per year
ERDA                         Energy Research and Development Agency  (defunct)
ESRL                         Environmental Sciences Research Laboratory, EPA
F"                           Fluoride ion
FA                           Fulvic acid
FDA                          Flourescein diacetate
FDA                          Food and Drug Administration
Fe                           Iron
FeS2                         Pyrite
                             01ivine  (and
                                 xxxn

-------
                             Ferrous sulfate
FEP                          Free erythrocyte protoporphyrin
FGD                          Flue gas desulfurization
FS                           Forest Service, DOA
FWS                          Fish and Wildlife Service, DOI
g                            Gram
9  r1                        Grams per liter
g dry wt m~2                 Grams dry weight per square meter
9 ro~2                        Grams per square meter
g m~2 s-1                    Grams per square meter per second
g m-2 yr-l                   Grams per square meter per year
g ha~* hr'l                  Grams per hectare per hour
GAMETAG                      Global Atmospheric Measurement Experiment of
                               Tropospheric Aerosols and Gases
GTN                          Global Trends Network
H                            Hydrogen
H+                           Hydrogen ion
                             Carbonic acid
                             Hydrogen peroxide
H2o                          Water
H2S                          Hydrogen sulfide
                             Sulfuric acid
H3P04                        Phosphoric acid
ha                           Hectare
HAOS                         Houston Area Oxidant Study
HC                           Hydrocarbon

                                  xxxiii

-------
HC1
HC03"
HCOH
HCOOH
HF
Hg
HIVOL
HgCl2
HgS
HHS
HN02
HN03
H02
H02N02
HO
MONO
HOS02
hr
ILWAS
IRMA
K
K+
KC1
K2S04
keq ha'1
Hydrochloric add
Bicarbonate 1on
Formaldehyde
Formic acid
Hydrogen fluoride
Mercury
High-volume
Mercuric ion
Mercuric chloride
Mercuric sulfide
Department of Health and Human Services
Nitrous acid
Nitric acid
Peroxy radical
Pernitric acid
Hydroxyl
Nitrous add
Bisulfite
Hours
Integrated Lake Watershed Acidification Study
Immission rate measuring apparatus
Potassium
Potassium Ion
Potassium chloride
Potassium sulfate, sulfate salt
Kiloequlvalents per hectare
                                  XXXIV

-------
keq ha-1 yr-1                Klloequlvalents per hectare per year
kg                           Kilogram
kg ha-1                      Kilograms  per  hectare
kg ha-1 wk-l                 Kilograms  per  hectare  per week
kg km-2 yr-1                 Kilograms  per  square kilometer per year
kg ha-1 yr-1                 Kilograms  per  hectare  per year
KHM                          Kol-Halsa-Miljo Project
KJ mol-1                     Kilojoule  per  mole
km                           Kilometer
km2                          Square kilometer
km hr-1                      Kilometers per hour
KMn04                        Potassium permanganate
 £                           Liter
 (£)                          Liquid phase
 A m-3                        Liters per cubic meter
 LAI                          Leaf area index
 LI                           Langelier's index
 LIMB                        Limestone Injection/Multistage Burner
 LR                          Larson's  ration
 LRTAP                        Long-Range Transport  of Air Pollutants
 LSI                          Langelier Saturation  Index
 m2                          Square meter
 m3 yr-1                     Cubic meter per year
 yeq                         Microequivalent
 yeq A-l                     Microequivalents  per  liter
                                   xxxv

-------
 yg 2"1
 yg 100 ml-1
 yg dH
 yg HP3
 ym
 ym £-1
 yM
 pm yr-1
 ymho cnr-*-
 m
 M
 m s-1
 m yr"1
 MAP3S

 mb
 MCC
 MCL
 MCPS
 ME
 meq JT*
 meq 100 g-1
 meq m~2 yr-1
 METROMEX
Mg
 Micrograms
 Micrograms per liter
 Micrograms per 100 milliliters
 Micrograms per decaliter
 Micrograms per cubic meter
 Micrometer
 Micrometers per liter
 Micromolar
 Micrometers per year
 micromhos per centimeter (conductivity)
 Meter
 Molar
 Meters per second
 Meters per year
 Multi-State Atmospheric Power Production
  Pollution Study
 Millibars
 Mesoscale convective complex
 Maximum contaminant level
 Mesoscale convective precipitation systems
 Momentary excess
 Milliequivalents per liter
 Milliequivalents per 100 grams
 Milliequivalents per square meter per year
 Metropolitan Meteorological Experiment
Magnesium
      xxxvi

-------
Mg2+                         Magnesium ion
mg                           Milligram
mg £-1                       Milligrams per liter
mg nr3 hr'l                  Milligrams per cubic meter per  hour
MgC(>3                        Magnesium carbonate
Mg2Si04                      Oil vine and (FegSiO^
MgS04                        Magnesium sulfate,  sulfate salt
mho cnrl                     mhos per centimeter (conductivity)
MISTT                        Midwest Interstate  Sulfur  Transport and
                                Transformations
mm                           Millimeter
mm hr'1                      Millimeters per hour
mm s-1                       Millimeters per second
mm yr'1                      Millimeters per year
mM                           Millimolar
Mn                           Manganese
Mo                           Molybdenum
MOI                          Memorandum of Intent on  Transboundary Air Pollution
mol                          Mole
mol £~1                      Moles per liter
mol £-1 atm"1                Moles per liter per atmosphere
mT                           Metric ton
mT yr-1                      Metric tons per year
MW                           Megawatt
N204                         N02  dimer
N205                         Nitrogen pentoxide
                                  xxxvii

-------
N20
(-NH)
N
N(III)
Na
Na+
Nad
NaN02
NADP
NAS
NASA
NASN
NATO
NBS
NCAC
NCAR
NECRMP
NEDS
ng £-1
ng kg'1
ng m-3
NH3
NH4+
Nitrous oxide
Imi de
Nitrogen
Liquid phase nitrogen
Sodium
Sodium ion
Sodium chloride
Sodium carbonate
Sodium nitrite
Sodium sulfate, sulfate salt
National Atmospheric Deposition Program
National Academy of Sciences
National Aeronautics and Space Administration
National Air Sampling Network
North Atlantic Treaty Organization
National Bureau of Standards, DOC
National Conservation Advisory Council
National Center for Atmospheric Research
Northeast Corridor Regional Modeling Program
National Emissions Data System
Nanograms per liter
Nanograms per kilogram
Nanograms per cubic meter
Ammonia
Ammonium ion
                                 xxxviii

-------
NH4C1
NH40AC
(NH4)2HP04
(NH4)2S04
NH4OH
Ni
nm
NMAB
N02
N03-
NO
NOX
NOAA
NFS
NRCC
NSF
NSPS
NTN
NWS
0
°2
03
(-OH)
Ammonium chloride
Ammonium acetate
Letorlclte
Ammonium phosphate
Ammonium nitrate
Ammonium sulfate
Ammonium hydroxide
Nickel
Nanometer
National Materials Advisory Board
Nitrogen dioxide
Nitrate Ion
Nitric oxide
Nitric oxides
National Oceanic and Atmospheric Administration
National Park Service, DOI
National Research Council Canada
National Science Foundation
New Source Performance Standards
National Trends Network
National Weather Service, NOAA
Oxygen
Elemental  oxygen
Ozone
Phenol
                                  xxx ix

-------
OECD                         Organization for Economic Cooperation  and
                               Development
OH                           Hydroxyl
OMB                          Office of Management and Budget
ORNL                         Oak Ridge National  Laboratory
OSM                          Office of Surface Mining, DOI
P                            Phosphorus
PAH                          Polycyclic aromatic hydrocarbons
PAN                          Peroxyacetyl nitrate
Pb                           Lead
Pb2+                         Lead ion
PBCF                         Practical bioconcentration  factor
PBL                          Planetary boundary layer
PbS04                        Lead sulfate
PCB                          Polychlorinated biphenyl
P6F                          Pressure gradient force
PHS                          Public Health Service
904^-                        Phosphate ion
ppb                          Parts per billion
ppm                          Parts per million
RAM                          St. Louis Regional Air Modeling Study
RAPS                         St. Louis Regional Air Pollution Study
RI                           Ryznar index
RSN                          Research Support Network
 s                           Second
S cm-1                       Seconds  per centimeter
                                    xl

-------
S                            Sulfur
$2-                          Sulfide
S(IV)                        Gas-phase sulfur, an oxidation state
SAC                          Sulfate adsorption capacity
SAES                         State Agricultural Experiment Station, DOA
Sb                           Antimony
SCS                          Soil Conservation Service, DOA
Se                           Selenium
Si                           Silicon
Si02                         Silicon dioxide
SMA                          Swedish Ministry of Agriculture
S02                          Sulfur dioxide
S032-                        Sulfite
S042~                        Sulfate ion
STP                          Standard temperature and pressure
SURE                         Sulfate Regional Experiment, EPRI
IDS                          Total dissolved solids
TFE                          Total fixed endpoint alkalinity
Tg                           Teragram (1012 gram)
Tg yr-1                      Teragrams per year
TIC                          Total inorganic carbon
TIP                          Total inflection point alkalinity
IPS                          Tennessee Plume Study
TSP                          Total suspended particulates
                                  xli

-------
TVA                          Tennessee Valley  Authority
USGS                         United States Geological  Survey, DOI
V                            Vanadium
V205                         Vanadium pentoxide
V cm-1                       Volts per centimeter
VDI                          Verein Deutcher Ingenieure
VOC                          Volatile organic  compounds
WHO                          World Health Organization
WMO                          World Meteorological Organization
yr                           Year
Zn                           Zinc
ZnS                          Zinc sulfide
                                   xlii

-------
                                 GLOSSARY

Acceptable daily Intake (ADI) - rate of safe consumption of a particular
substance or element In human food or water, as determined by the U.S. Food
and Drug Administration.
Acidic deposition - the deposition of acidic and acidifying substances from
the atmosphere.
Acid neutralizing capacity (ANC) - equivalent sum of all bases that can be
titrated with a strong acid; also known as alkalinity.
Adiabatic - occurring without gain or loss of heat by the substance
concerned.
Adsorption - adhesion of a thin layer of molecules to a liquid or solid
surface.
Advection - horizontal flow of air to the surface or aloft; one of the means
by which heat is transferred from one region of the Earth to another.
Aerosols - suspensions of liquid or solid particles in gases.
AH quoting - dividing into equal parts.
Alkalinity - measure of the ability of an aqueous solution to neutralize acid
(also known as acid neutralizing capacity or ANC).
Allochthonous inputs - substances introduced from outside a system.
Ambient - the surrounding outdoor atmosphere to which the general  population
may be exposed.
Ammonium - cation (NH4+) or radical (NH4) derived from ammonia by
combination with hydrogen.  Present in rainwater, soils, and many commercial
fertilizers.
Anion - a negatively charged ion.
Aqueous phase - that part of a chemical transformation process when
substances are mixed with water or water vapor in the atmosphere.
Antagonistic effects (less-than-additive) - results from joint actions of
agents so that their combined effect 1s less than the algebraic sum of their
individual  effects.
Anthropogenic - manmade or related to to human activities.
Artifact -  a spurious measurement produced by the sampling or analysis
process.
                                   xliii

-------
Atmospheric residence time - the amount of  time  pollutant  emissions are held
1n the atmosphere.

Autochthonous Inputs - Indigenous,  formed or  originating within  the system.

Autotrophic - able to synthesize nutritive  substances  from Inorganic
compounds.

Background measurement - pollutants in ambient air  due to  natural  sources;
usually taken 1n remote areas.

Base neutralizing capacity - equivalent sum of all  adds that  can  be  titrated
with a strong base.

Base saturation (BS) - the fraction of the  cation exchange capacity satisfied
by basic cations.

Benthic organisms - life forms  living  on the  bottoms of bodies of  water.

Bioaccumulation - the phenomenon wherein toxic elements are progressively
amassed in greater quantities as individuals  farther up the food chain ingest
matter containing those elements.

Biconcentration factor (BCF) - the ratio of  the concentration of  a substance
in an organism to the concentration of the  substance in the surrounding
habitat.

Bioindicators - species of plants or animals  particularly  sensitive to
specific pollutants or adverse conditions.

Biomass - that part of a given habitat consisting of living matter.

Biosphere - the part of the Earth's crust,  waters,  and atmosphere  where
living organisms can subsist.

Brownian diffusion - spread by random movement of particles suspended in
liquid oV gas, resulting from the impact of molecules  of the fluid
surrounding the particles.

Brownwater lakes and streams -  acidic waters  associated with peatlands,
cypress swamps; acidity is caused by organic  acids  leached from  decayed plant
material and from hydrogen ions released by plants  such as Sphagnum mosses.

Budget - a summation of the inputs and outputs of chemical substances
relative to a given biological  or physical  system.

Buffer - a substance in solution capable of neutralizing both  acids and bases
and thereby maintaining the original pH of  the solution.

Buffering capacity - ability of a body of water  and its watershed  to
neutralize introduced acid.
                                    xliv

-------
Bulk sampling - method for collecting deposition that does  not  separate dry
and wet deposition (see Chapter A-8).

Calcareous - resembling or consisting of calcium carbonate  (lime), or growing
on limestone or lime-containing soils.

Calclte saturation Index (CSI)  - measure of the degree of saturation of water
with respect to CaCOa, Integrating alkalinity,  pH,  and Ca concentration.

Cation - a positively changed 1on

Cation exchange capacity (CEC)  - the sum of the exchangeable  cations,
expressed In chemical equivalents, 1n a given quantity of soil.

Chemoautotrophlc - having the ability to synthesize nutritive substances
using an Inorganic compound as  a source of available energy.

Colorimetric - a chemical analysis method relying on measurement  of the
degree of color produced In a solution by reaction of the compound of
interest with an indicator.

Conductivity - the ability to conduct an electric current;  this is a function
of the individual mobilities of the dissolved ions in a solution, the concen-
trations of the ions, and the solution temperature; measured  in mho cm~l.

Continental scale - measurement of atmospheric  conditions over  an area the
size of a continent.

Coriolis effect - an effect caused by the Earth's eastward  rotation in which
the speed of the movement falls off as the circumference of the Earth gets
progressively smaller at higher latitudes; this results in  the movement of
winds, and subsequently ocean currents, to the  right in the northern
hemisphere and to the left in the southern hemisphere.

Cosmic ray - a stream of ionizing radiation of  extraterrestrial origin,
chiefly of protons, alpha particles, and other  atomic nuclei  but  including
some high energy electrons and  protons, that enters the atmosphere and
produces secondary radiation.

Coulomb - a meter/kilogram/second unit of electric charge equal to the
quantity of charge transferred  in one second by a steady current  of one
ampere.

Coarse particles - airborne particles larger than 2 to 3 micrometers 1n
diameter.

Cultivar - cultivated species of crop plant produced from parents belonging
to different species or different strains of the same species,  originating
and persisting under cultivation.

Cuticular resistance -  the resistance to penetration of a  leaf cuticle.
                                    xlv

-------
Cyclone track - the path of a low pressure system.

Denitrification - a bacterial process  occurring  in  soils,  or  water,  in which
nitrate is used as the terminal  electron  acceptor and  is  reduced  primarily to
Ng.  It is essentially an anaerobic process;  it  can occur  in  the  presence
of low levels of oxygen only if the microorganisms  are metabolizing  in an
anoxic microzone (an oxygen-free microenvironment within  an area  of  low
oxygen levels).

Deposition velocity - rate at which particles from  the atmosphere contact
surfaces and adhere.

Detritus - loose material resulting directly  from disintegration.

Diffusiophoresis - an effect created when particles approaching an
evaporating surface are impacted by more  molecules  on  the  side nearer the
surface.

Dissolved organic carbon (DOC)  - the amount of organic carbon in  an  aqueous
solution.

Dissolved inorganic carbon (DIC) - the amount of inorganic carbon in an
aqueous solution.

Dose - the quantity of a substance to  be  taken all  at  one time or in
fractional amounts within a given period; also the  total  amount of a
pollutant delivered or concentration.

Dose-response curve - a curve on a graph  based on  responses occurring  in a
system as a result of a series of stimuli intensities  or  doses.

Edaphic differences - soil differences.

Eddies - currents of water or air running contrary  to  the main current.

Eddy diffusities - dispersive movements of particles,  caused  by circular
motions in air currents.

Ekman layer - a layer of the atmosphere typically  extending between  1  and  3
kilometers above the surface; see Section A-3.2.2  for  detailed discussion.

Electromotive force (emf) - the amount of energy derived  from an  electrical
source per unit quantity of electricity passing through  the source (as  a cell
or generator).

Entrainment - the process of carrying  along or over (as  in distillation or
evaporation).

Epifaunal - organism living on an animal.

Epilimnion - the upper layer of a lake in which the water temperature  is
essentially uniform.
                                   xlvi

-------
Episodic precipitation event - a period during which rain, snow,  etc.,  1s
occurring.

Erlcaceous - heathllke or shrubby; a member of the Ericaceae family.

Eucaryotlc algae - algae composed of one or more cells with visibly evident
nuclei.

Eulerlan models - models with reference frames fixed on the source or at the
surface.

Eurytopic - having a wide range of tolerance to variation of one  or more
environmental factors.

Eutrophic - relating to or being 1n a well  nourished condition; a lake  rich
in dissolved nutrients but frequently shallow and with seasonal oxygen
deficiency in the hypolimnion.

Eutrophlcation - the process of becoming more eutrophic either as a natural
phase  in the maturation of a body of water  or artificially, as by
fertilization.

Exposure level - concentration of a contaminant with which an individual  or
population is in contact.

Extinction coefficient - a measure of the space rate of diminution, or
extinction of any transmitted light; thus,  it is the attenuation  coefficient
applied to visible radiation.

Fine particles - airborne particles smaller than 2 to 3 micrometers in
diameter.

Fly ash - fine, solid particles of noncombustible ash carried out of a  bed  of
solid  fuel by a draft.

Foliar - referring to plant foliage (leaves).

Fumigate - to subject to smoke or fumes.

Gas-phase mechanism - a process occurring when pollutants are in  a gaseous
state, as opposed to being combined with moisture.

Geostrophic - of or pertaining to the force caused by the Earth's rotation.

Global scale - measurement of atmospheric conditions on a world-wide basis.

Ground loss - the effect of deposition of pollutant from atmposhere to
Earth's surface.

Ground sink - the Earth's surface, where airborne substances may  be
deposited.
                                   xlvii

-------
Haze - an aerosol that impedes vision and may consist of a combination of
water droplets, pollutants, and dust.

Hemispheric scale - measurements of activity covering half of the Earth.

Heterotrophic - obtaining nourishment from outside sources, requiring complex
organic compounds of nitrogen and carbon for metabolic synthesis.

Humic acid - any of various organic acids that are insoluble in alcohol and
organic solvents and that are obtained from humus.

Hydrocarbons - a vast family of compounds containing carbon and hydrogen  in
various combinations; found especially in fossil  fuels.

Hydrologic residence time - the amount of time water takes to pass from the
surface through soil to a lake or stream.

Hydrometeor - a product of the condensation of atmospheric water vapor (e.g.,
raindrop).

Hydrophilic - of, relating to, or having a strong affinity for water;  readily
wet by water.

Hydrophobic particles - particles resistant to or avoiding wetting;  of,
relating to, or having a lack of affinity for water.

Hydroxyl radical - chemical prefix indicating the [OH] group.

Hygroscopic particles - absorbing moisture readily from  the atmosphere.

Hypolimnion - the lowermost region of a lake, below the  thermocline, in which
the temperature from its upper limit to the bottom is nearly uniform.

Hysteresis - the failure of a property to return  to its  orginal  condition
after the removal of the causal external agent (i.e., irreversibility).

Infauna - population of organisms living in sediments.

Inorganic acidotrophic lakes - waters associated  with geothermal  areas or
lignite burns;  extremely acidic, often heated, and frequently containing
elevated metal concentrations.

Interstitial water - water in the space between cells.

Isopleth - 1. a line of equal or constant value of a given quantity with
respect to either space or time, also known as an isogram; 2. a line drawn
through points on a graph at which a given quantity has  the same numerical
value as a function of the two coordinate variables.

Labile - readily or continually undergoing chemical  or physical  or biological
change or breakdown.
                                   xlviii

-------
Lacustrine sediments -  deposits  formed in  lakes.

Lagrangian models - models with  reference  frames  fixed  on  the  puff of
pollutants.

Langmuir equations - empirical  derivations from kinetic treatment of the
physical adsorption of gases or solids by  soils;  relating  to the  relative
adsorption capacity of a soil  for a specific anion.

Leaf area index (LAI) - ratio of the total foliar surface  area to the  ground
surface area that supports it.

Lentic - of, relating to, or living in still waters.

Lidar - a laser-radar system operated from a mobile  van.

Ligands - those molecules or anions attached to the  central  atom  in  a
complex.

Limnological - of or relating to the scientific study of physical,  chemical,
meteorological, and biological  conditions  in freshwaters,  especially ponds
and lakes.

Lipophilicity - the strong affinity for fats or other lipids.

Liquid-phase mechanism - a process occurring when pollutants are  combined
with moisture, as opposed to being in a purely gaseous state.

Littoral - the shore zone between high and low watermarks.

Loading rate - the amount of a nutrient available to a unit area  or body of
water over a given period.

Long-range transport - conveyance of pollutants over extensive distances,
commonly referring to transport over synoptic and hemispheric scales.

Macrophytes - higher plants.

Manometer  - an instrument for measuring pressure of gases or work.

Mean (arithmetic) - the sum of observations divided by sample size.

Median  - a value  in a collection of data  values which  is exceeded in
magnitude  by one-half the entries in the  collection.

Mesoscale  - of or relating to meteorological phenomena  from 1 to 100
kilometers in horizontal extent.

Metalimnion - the thermocline.

Microbial  pathogens - microscopic organisms capable of  producing disease,
such as viruses,  fungi, etc.


                                   xlix

-------
Microflora - a small or strictly localized plant.

Micrometeorological - referring to conditions specific to a very small  area,
such as a surface, a particular site, or locale.

Mist - suspension of liquid droplets formed by condensation of vapor or
atomization; the droplet diameters exceed 10 micrometers  and in general  the
concentration of particles is not high enough to  obscure  visibility.

Mixing layer - also called the planetary boundary layer (PBL); usually  the
domain of microscale turbulance.

Mobile sources - automobiles, trucks, and other pollution sources that  are
not fixed in one location.

Mole - The mass, in grams, numerically equal  to the molecular weight of a
substance.

Morphology - structure and form of an organism at any stage of its life
history.

Mycorrhizal - relating to symbiotic association of a fungal mycelium with  the
roots of a seed plant.

Nitrification - the principal natural source of nitrate,  in which ammonium
(NH4+) ions are oxidized to nitrates by specialized microorganisms.
Other organisms oxidize nitrites to nitrates.

Nocturnal jet - phenomenon in the atmosphere of a high-velocity air stream
occuring at night above the nocturnal inversion layer.

Non-humic lakes - lakes without significant inputs of humic acid.

Ohm's law - a law in electricity:  the strength or intensity of an unvarying
electrical current is directly proportional to the electromotive force  and
inversely proportional to the resistance of the circuit.

Oligochaete worms - an annelid worm of the class  Oligochaeta, i.e., having a
segmented body.

Oligotrophic - a body of water deficient in plant nutrients; also generally
having abundant dissolved oxygen and no marked stratification.

Ombrotrophic peat bog - a peat bog fed solely by  rain water.

Oxic condition - the presence of oxygen.

Oxidant - a chemical compound that has the ability to remove electrons  from
another chemical species, thereby oxidizing it; also a substance containing
oxygen which reacts in air to produce a new substance, or one formed by the
action of sunlight on oxides of nitrogen and hydrocarbons.

-------
Palearctlc lake - a lake in the biogeographlc  region  that  includes  Europe,
Asia north of the Himalayas, northern Arabia,  and  Africa north  of the  Sahara.

Particle morphology - the structure and form of substances
suspended in a medium.

Particulates - fine liquid or solid particles  such as dust,  smoke,  mist,
fumes, or smog found in air or in emissions.

Ped surfaces - surfaces of natural soil aggregates.

Pelagic - of, relating to, or living in the open sea.

Periphyton - organisms that live attached to underwater surfaces.

Photoautotrophic organisms - autotrophic organisms able to use  light  as  an
energy source.

Photochemical oxidants - primarily ozone, N02, PAN with lesser  amounts of
other compounds formed as products of atmospheric  reactions  involving  organic
pollutants, nitrogen oxides, oxygen, and sunlight.

Phytophagous insects - insects feeding on plants.

Phytoplankton - autotrophic, free-floating, mostly microscopic  organisms.

Planetary boundary layer (PBL) - first layer of the atmosphere  extending
hundreds of meters from the Earth's surface to the geostrophic  wind level ,
including, therefore, the surface boundary layer and the Ekman  layer;  above
this level lies the free atmosphere.

Plume - emission from a flue or chimney, normally distributed streamlike
downwind of the source, and which can be distinguished from surrounding  air
by appearance or chemical characteristics.

Plume touchdown - point of a plume's contact with the Earth's surface.

Podzol - any of a group of zonal soils that develop in a moist  climate,
especially under coniferous or mixed forests.

Point source - a single stationary location for pollutant  discharge.

Precipitation scavenging - a complex process composed of four distinct but
interactive steps:  intermixing of pollutant and condensed water within  the
same airspace, attachment of pollutant to the  condensed water,  chemical
reaction of pollutant within the aqueous phase, and delivery of
pollutant-laden water to surfaces.

Precursor - a substance from which another substance is formed, specifically
one of the anthropogenic or natural emissions  or atmospheric constituents
that reacts under sunlight to form secondary pollutants comprising
photochemical smog.
                                    li

-------
Primary particles (or primary aerosols)  -  dispersion  aerosols  formed  from
particles emitted directly into the air  that do not change  form  in the
atmosphere.

Quasi-laminar layer - the internal  viscous boundary layer above  non-ideal  or
natural surfaces; it is frequently neither laminar nor constant  with  time.

Rayleigh scattering - spread of electromagnetic radiation by bodies much
smaller than the wavelength of the radiation;  for visible  wavelengths, the
molecules constituting the atmosphere cause Rayleigh  scattering.

Secondary particles (or secondary aerosols) - dispersion aerosols that  form
in the atmosphere as a result of chemical  reactions,  often  involving  gases.

Sensitivity - the degree to which an ecosystem or organism  may be affected by
inputs or stimuli.

Sequential sampling - repeated, periodic collection of data concerning  a
phenomenon of interest.

Sinks - reactants with or absorbers of substances; collection  surfaces  or
areas where substances are gathered.

Steady state exposure - exposure to air  pollutants whose concentration
remains constant for a period of time.

Stefan flow - results from injection into a gaseous medium  of  new gas
molecules at an evaporating or subliming surface; Stefan flow  is capable of
modifying surface deposition rates by an amount that  is larger than the
deposition velocity appropriate for many small particles to aerodynamically
smooth surfaces.

Stokes1s law - a law in physics:  the force required  to move a sphere through
a given viscous fluid at a low uniform velocity is directly proportional to
the velocity and radius of the sphere.

Stoma - opening on a leaf surface through which water vapor and  other gases
diffuse; often term applies to the entire stomatal apparatus including
surrounding specialized epidermal cells, guard cells.

Stream order - positions a stream in relation to tributaries,  drainage  area,
total length, and age of water.  First-order streams  are the terminal twigs
(headwaters or youngest segments of a stream system,  having no tributaries).
Second-order streams are formed by the junction of two first order  streams,
and so on.  At least two streams of any given order  are required to form the
next highest order.

Sub-optical range - particles too small  to be seen with the naked  eye.

Surfactant - a substance capable of altering the physiochemical  nature  of
surfaces, such as one used to reduce surface tension  in a liquid.
                                    lii

-------
Symbiotic - a close association between two organisms  of  different  species,
1n which at least one of the two benefits.

Synerg1st1c effects (more-than-addltive) -  result from joint  actions of
agents so that their combined effect 1s greater than the  algebraic  sum of
their Individual  effects.

Synoptic scale -  relating to or displaying  atmospheric and weather  conditions
as they exist simultaneously over a broad area;  the scale of  weather maps.

Teragram (Tg) - one million metric tons, 1012 grams.

Thermocline - the stratum of a lake below the epilimnion  1n which there  is a
large drop in temperature per unit depth.

Thermophoresis -  a force near a hot surface that drives small  particles  away
from that surface.

Throughfall - precipitation falling through the canopy of a forest  and
reaching the forest floor.

Titratlon - the process or method of determining the concentration  of a
substance in solution by adding to it a standard reagent  of a known
concentration in carefully measured amounts until a reaction  of  definite and
known proportion is completed, as shown by a color change or  by  electrical
measurement, and then calculating the unknown concentration.

Total fixed endpoint alkalinity (TFE) - a measure of acid neutralizing
capacity involving acidimetric titrations performed to an endpoint  of pH 4.5
determined electrometrically or to an endpoint determined by  either a
colorimetric indicator or mixed indicators.

Total inflection point (TIP) - a measure of acid neutralizing capacity,
involving acidimetric titration to the HC03-H+ equivalence point of the
titration curve.

Total suspended particulates (TSP) - solid and liquid  particles  present  in
the atmosphere.

Toxicity - the quality, state, or relative degree of being poisonous.

Trajectory - a path, progression, or line of development, as  from a plume of
pollutant carried through the atmosphere from a source to a receptor area.

Transport layer - the layer between the earth's surface and the  peak mixing
height of the day; for any given instant, it is made up of the current mixing
layer below and the relatively quiescent layer above.

Troposphere - that portion of the atmosphere in which  temperature decreases
rapidly with altitude, clouds form, and mixing of air  masses  by  convection
takes place; generally extending to about 11 to 17 km  above the  Earth's
surface.
                                    liii

-------
Ultra oUgotrophlc lakes - lakes 1n  areas where gladatlon has removed
calcareous deposits and exposed weather  resistant granitic and siliceous
bedrock; such lakes have little carbonate-bicarbonate buffering capacity and
are very vulnerable to pH changes; they  tend to be small and have low
concentrations of dissolved Ions.

Variance - a measure of dispersion or variation of a sample from Its expected
value.

Washout - the capture of gases  and particles by falling raindrops.

Wet deposition - the combined processes  by which atmospheric substances are
returned to Earth 1n the form of rain or other precipitation.

Wind shear - a sudden shift In  wind  direction.

X-ray diffraction - technique by which patterns of diffraction can be used to
Identify a substance by Its structure.

Zooplankton - minute animal life floating or swimming weakly 1n a body of
water.
                                   liv

-------
               THE ACIDIC DEPOSITION PHENOMENON AND ITS EFFECTS

                              A-l.  INTRODUCTION

              (A. P. Altshuller, J. S. Nader, and L. E. Niemeyer)


 1.1  OBJECTIVES

 This portion  of  the Critical  Assessment Review Papers  addresses the various
 atmospheric processes  starting  with emissions  to  the  atmosphere from natural
 and anthropogenic  sources,  leading up to  the  presence of acidic  and  acidi-
 fying  substances in the  atmosphere,  and  concluding  with the  deposition  of
 these  substances from  the  atmosphere  to the surfaces of  manmade and natural
 receptors.  The  objective is to provide an  understanding of these phenomena
 and the latest technical data base supporting this understanding.

 1.2  APPROACH—MOVEMENT FROM SOURCES TO RECEPTORS

 1.2.1  Chemical Substances of Interest

 The approach begins by identifying the acidic and acidifying substances emit-
 ted from  natural and anthropogenic sources.  The  chemical species of princi-
 pal concern  are the  hydrogen  ion (H+),  ammonium  ion  (NH4+),  sulfate  ion
 (S042-),  and   the  nitrate  ion  (NOs").    Chloride,  in   the  form of  hydro-
 gen chloride,  may  be  of  concern, particularly  downwind of  some types  of
 anthropogenic emission  sources.   A number of  metal cations  are of  interest
 because they  affect material  balances or  cause unique biologicial  effects.
 Weaker acids  such  as nitrous  acid,  formic acid,  and dibasic  acid have been
 identified  in  the  atmosphere  but do not contribute  significantly  to  the
 acidic deposition phenomenon.

 1.2.2  Natural and Anthropogenic Emissions  Sources

 Natural sources are classified as geophysical and biological.  The former in-
 cludes volcanic and sea spray contributions, the  latter,  soil  and vegetation
 contributions.   The anthropogenic source  categories  include  electric  utili-
 ties,   industrial   combustion  sources,   commercial/residential   combustion
 sources,  highway  (mobile)  vehicles,   and  miscellaneous  sources.    Emission
 patterns are given for spatial,  seasonal,  and  temporal  variations.   Although
 data are  given  for the United  States and  Canada,  the  main focus is  on  the
 area east  of  the Mississippi,  where  acidic deposition levels appear  to  be
 greatest.

 1.2.3   Transport Processes

The movement  of emissions from  sources  to receptors involves  atmospheric
transport and transformation  processes.  The transport process  is discussed


                                     1-1

-------
with  regard to the  structure  and dynamics of the  planetary  boundary layer.
The  impact of  the  source's  physical  configuration,  elevated point  source
(power  plant  plume),  and  broad  area!  emissions  near  ground level  (urban
plumes) on  the  transport  and dilution processes are reviewed.  Transport is
treated on  the  mesoscale,  the  continental  scale,  and the hemispherical  scale
and allows  for  the effects of complex terrain and shoreline environment.

1.2.4  Transformation Processes

Atmospheric transformation  processes  account for  the  chemical and  physical
changes  in some  of  the  emissions  (precursors)  into  acidic   and  acidifying
species that ultimately result  in the  presence and  deposition  of  atmospheric
acid matter.   In  relatively  dry,  cloudless atmospheres,  these  changes can be
the  result of  homogeneous  gas-phase  reactions  between  radicals  (such  as
hydroxyl) and sulfur dioxide and nitrogen dioxide to form sulfuric and nitric
acids.   Ammonia can  subsequently partially or  completely neutralize these
acids.  Solution reactions can occur also in water droplets on  vegetation, in
cloud droplets, in fog, and in dewdrops.   The oxidation of sulfur  dioxide can
involve, to various extents, other chemical-reacting atmospheric constituents
such as oxygen, ozone,  hydrogen peroxide,  and ammonia.    In addition, cata-
lytic metal  constituents  such as iron and manganese may  participate  in the
oxidation  process  in low-lying clouds  or  fogs  over highly polluted  areas.
The products of these transformation  reactions  add  to the primary acid  ori-
ginally emitted from anthropogenic sources, and  the net amalgam of substances
continues downwind.

1.2.5  Atmospheric Concentrations  and Distributions of  Chemical  Substances

Acidic and  acidifying substances in  the  atmosphere prior to  deposition  on
natural  and manmade  receptors include both  those  emitted directly into  the
atmosphere  (primary  pollutants) and those  resulting from  atmospheric  trans-
formations  (secondary pollutants).   Transport on various  scales,  as well  as
emissions  that  vary  temporally with  seasons and time of day  and that  vary
spatially  with  meteorology  and distribution of  emission  sources  and  geo-
graphic locations, provide a complex picture of  concentrations  of these  sub-
stances of  interest  prior to deposition.   Urban and  nonurban  concentration
data  on   sulfur compounds,  nitrogen  compounds,  chlorine compounds, basic
substances, metals, and particle size characteristics of  particulate constit-
uents of  these compounds  are  reviewed.    Available  information is given  on
geographic distribution,  seasonal  and diurnal  variations, and variations  with
elevation above ground level.

1.2.6  Precipitation  Scavenging Processes

The complex process of precipitation scavenging  depends upon a  host of inter-
active physical  and  chemical  phenomena  that  occur prior  to  and  during  the
precipitation process.   Cloud  droplets  form and evaporate, airborne  pollu-
tants  are  incorporated   into   and  released  by  condensed water,  chemical
reactions occur, ice crystals form  and melt, energy  is exchanged,  and hydro-
meteors are  created  and  evaporate.   These and  a  multitude  of   additional
processes create  a continually changing  environment for  pollution elements
within a  storm system.  The final  stage of these  complex  scavenging processes


                                     1-2

-------
is the  actual  wet delivery of pollutants to  the  ground.   A large number of
models  have  been  developed,  but their very  number is an  indication  of the
work remaining before a satisfactory modeling capability  is  possible.

1.2.7  Dry Deposition Process

In addition to deposition of acidic and acidifying substances  from  the  atmos-
phere by  wet scavenging  with  rain, snow, and  fog, dry  deposition  plays  a
similar role with respect to the same substances of interest in  the gas phase
and  as  solid particulate  matter.   The  dry  deposition  processes  take  into
account  aerodynamic  factors, the  surface-boundary layer,  phoretic effects,
dewfall, surface effects,  and deposition  to  water surfaces.  The  concept of
resistance analog  provides a model for identifying process parameters asso-
ciated with the transfer of substances from the  atmosphere to  the vicinity of
the  final receptor surfaces.

Methods for measuring dry  deposition consist  of direct measurement with col-
lection vessels and with surrogate surfaces specific to various  receptor sur-
faces of  interest.   Laboratory  studies have been  conducted under  controlled
conditions to provide an  understanding of  the relative importance  of various
factors  in  the processes.   These  include chamber and wind-tunnel work,  and
they address  resistances  to  deposition of selected trace gases onto various
substrate  surfaces  and deposition  velocities  of  different size  particulate
matter  to  a  variety  of surfaces.   Micrometeorological  techniques are  also
discussed  and  consist  of  eddy-correlation  methods,  gradient  measurement
techniques,  and  other  new developments.   Field investigations  are providing
data  on the  impact of  the diurnal  cycle on dry  deposition rates  of gaseous
pollutants  on different  surfaces.   Data  are  also  available on  deposition
velocities of submicron particles.  Results of many of these studies have led
to the  development  of micrometeorological  models of the  dry  deposition pro-
cesses  for gases and for particles.

1.2.8   Deposition Monitoring

Deposition  monitoring  networks  have been established to  collect  wet  deposi-
tion  data during  periods of precipitation  and  dry  deposition data  during
periods of  no precipitation.  Networks have  been designed to  collect  data on
various spatial,  temporal,  and  density scales.   These data bases  are  essen-
tially  wet deposition  monitoring  networks.   Dry  deposition  monitoring net-
works  exist  to  a limited  extent  if  any and  are primarily of   a research
nature.

Wet  deposition network  data  have  been  analyzed  and interpreted  to  provide
maps of the  United States  and Canada  with sampling site  locations  and  median
concentration data for specified  sampling  periods for  sulfates, nitrates,
ammonium  ion, calcium,  chloride,  and pH.  Spatial  patterns are generated by
isopleths  identifying regions of  high  and low values.   Temporal  variations
are  also analyzed and include seasonal variations and changes  over both short
and  long time scales.

Glaciochemical  investigations are  being  conducted and are shown to provide a
tool  in the  historical  delineation of  acid precipitation problems.    These


                                     1-3

-------
studies  also  provide  a  bench   mark  on  the  natural  background  void  of
anthropogenic pollution and contamination.

1.2.9  Deposition Models

Developing suitable models for acidic  deposition  is  a  difficult undertaking.
The models have to have  algorithms  that  take  into account natural  emissions,
anthropogenic emissions,  transport  processes,  transformation,  precipitation
scavenging  processes,  and  dry  deposition  processes  on  scales  from a  few
millimeters to thousands of  kilometers.   Moreover, the results must  be  com-
pared to measurements made on a variety  of  scales for  a variety of purposes.
Therefore, in terms  of  the detail inherent in  the models,  there is  a  large
variation from the simple  to the  complex.  All  need verification,  and  while
progress has been made in the acquisition of  data bases,  more information is
needed for a proper evaluation of  long-range transport  models.

1.3  ACIDIC DEPOSITION

Atmospheric pollutants  consist  of both  acidic  and basic substances  and  in-
clude both primary and secondary pollutants.   The acidity in depositions from
the atmosphere onto  natural  and manmade  receptors such as soils,  vegetation,
bodies of water,  pavements,  and buildings is  the  net acidity after neutrali-
zation in  the atmosphere of  the  acidic  substances  by  the  basic  substances.
Acidity measurements  are usually expressed on a pH scale  where  pH  is  defined
as the  negative  logarithm of the hydrogen  ion concentration.  The  pH  scale
extends from 0 to 14.  A neutral  pH  in water at 25 C is 7.0.   Solutions with
a pH below 7.0 are considered acid;  those with a pH above 7.0 are considered
alkaline or basic.   The  logarithmic  pH scale  means that a  whole  unit change
in pH corresponds  to a 10-fold change in acidity or hydrogen ion concentra-
tion.  A pH of 6.0 is ten times  more acidic  than a pH of 7.0.

Atmospheric water droplets are in  equilibrium with the  geophysical  concentra-
tions of carbon dioxide  in air.   This  equilibrium results in a pH of 5.6 for
such  droplets.    However,  even  this  pH value  applies only  to a  perfectly
"clean"  atmosphere.    Lower  pH  values  have  been measured  at remote  sites
although these pH values are still  well above  those  measured over  eastern
North America.   If  substantial  amounts  of  basic  particulate  substances  are
present, the pH may be greater than  5.6.

The acidity measured  in a manmade  collector  is not necessarily representative
of the acidity in soil  or water.    Most  deposition monitoring,  being limited
to collection of  rain or  snowfall,  does  not include monitoring of  dry  depo-
sition.   Acidic  or  basic  substances  can collect  on vegetation or  soil  sur-
faces and subsequently be washed  into  the soil  by rainfall.   Once substances
are within an ecosystem, additional   changes  in  acidity can  occur as a result
of processes  involving  plants and organisms.   Ammonia can  be  released  from
deposited particulate ammonium  salts.   Hydroxyl  ions  can be released as the
result of metabolic  processes.   These processes  may change  the  net  acidity
significantly.
                                     1-4

-------
               THE ACIDIC DEPOSITION PHENOMENON AND ITS EFFECTS

              A-2.  NATURAL AND ANTHROPOGENIC EMISSIONS SOURCES


2.1  INTRODUCTION (Eds.)

Acidic and acidifying substances in the atmosphere may  be  produced  by nature
or  by  human (anthropogenic)  activities.   In  either  case, emissions  become
available for transport to other locations,  for combination with other atmos-
pheric substances, and  for deposit  to  surfaces.   Chapter A-2  discusses where
acidic and  acidifying  substances  originate,  thus setting the  stage  for fur-
ther  examinations of  transport,  transformation,  and  deposition  processes;
concentrations and distributions;  and modeling efforts.   It considers natural
and anthropogenic sources separately and subdivides the discussions  among the
various substances of concern.

Numerous questions  arise  relative to  emissions  sources.   For  instance,  are
natural sources of sulfur, nitrogen, and chlorine compounds significant, and,
if  so,  where are they  and  how do  emission  rates vary  seasonally?    On  the
other  hand,  concerning  anthropogenic  sources, how have historical  trends in
fuel use changed emission rates and how are  future trends likely to  alter the
rates?  How are current emissions distributed between  stationary and mobile
sources, among geographic regions, between urban and rural  areas, seasonally,
and at various  heights?   Do  non-combustion,  anthropogenic  sources of sulfur,
nitrogen, and chlorine  compounds exist, or do  any  additional  materials emit-
ted anthropogenically affect acidic deposition, either by catalysis  or direct
reaction with sulfur,  nitrogen, and chlorine-containing compounds?    In  con-
trast to acidic or acidifying substances,  what sources exist for neutralizing
substances—including ammonia, soil-related or cement plant dusts,  and alka-
line  particles  from  combustion—and   how  do  these  vary geographically  and
seasonally?

In addition to addressing these issues, Chapter A-2 also presents information
concerning emissions of several heavy  metals  from  combustion  sources because
information  on  these  metals  may  be  useful  in  assessing  dispersion  from
specific sources.

2.2  NATURAL EMISSIONS SOURCES (E. Robinson)

2.2.1  Sulfur Compounds

2.2.1.1  Introduction—Sulfur  compounds,  including  sulfates and sulfur diox-
ide, are ubiquitous trace constituents of  the Earth's  atmosphere even in very
remote, natural  areas.  Thus, it is common to assume that these common, rela-
tively  reactive compounds result  from natural  sources in  the  unperturbed
environment.  Concentrations in most background situations  are low,  and samp-
ling and analysis problems are major factors  that  limit the determination of


                                     2-1

-------
the gaseous sulfur compounds.  Our present knowledge is strongly dependent on
the analytical tools  that  have been available to the  various  investigators.
It  will  be convenient  to  consider natural  sulfur  sources  in  terms of  two
general  classifications:    geophysical,  including  volcanic  and  sea  spray
contributions, and biological,  including soil and  vegetation  contributions.
This discussion  will  emphasize conditions appropriate  for  the area east  of
the Mississippi  River,  which seems to  be the area  of  eastern  North  America
most critically affected by acidic deposition.  In  this region  of  the  United
States natural sources  may act in two ways  to influence  conditions.   First,
natural sources  within  the region may be  contributors to the  local  concen-
tration patterns.  Second, natural  sources in areas remote  from this  region
may contribute to the global  background concentration,  and thus influence  the
total   mass  of the  natural  emissions  that are advected  across the  region.
Biogenic emissions from the soil, coastal wetlands,  and vegetation  are  poten-
tial local  sources  that can contribute  directly  to the sulfur cycle  in  the
local   region.  Volcanos and  the open ocean  are  examples  of natural  sources
that will impact on the local  northeast  United States  primarily by influenc-
ing the  general  level  of  sulfur  compounds in the  global  environment.   The
dilution and scavenging processes that regularly  take place  on  a global  scale
limit the impact of remote volcanic and  oceanic sources on  the  specific area
of  interest in  the  northeast  United  States.   In  the following  discussion
biogenic sources will  be considered in some  detail  because  of  their  possible
local   importance; the more distant sources that contribute  primarily  to  the
global  background will be considered in a more general  fashion.

2.2.1.2  Estimates of Natural  Sources—Estimates of the magnitude  of  natural
sulfur compound  sources usually reference  the initial  estimate  of  the  global
sulfur flux published by Eriksson in 1960.   Using the  global balancing  tech-
nique  described  below,  Eriksson (1960)  estimated natural  sulfur sources,  as
sulfur, to  be  77 x  10^ mT (77  Tg S)  from land areas  and 190  x 10^ mT (190
Tg  S)  from  the  oceans.   (The  unit Tg  S yr'1 is 1012  grams per year).   In
the two decades  since Eriksson's  first  estimate,  a  number of  variations  and
"improved"  global  estimates  have  been  made  by a number  of authors but  the
methods used have not undergone major  changes.   Some of the most  frequently
referenced  global  sulfur  circulation  models, which,  of  necessity,  include
estimates of natural   sources,  are those of Junge (1960, 1963),  Robinson  and
Robbins  (1970a),  Kellogg  et al.  (1972),  and Friend (1973).   More recently,
Granat et al.  (1976)  have  assembled  a  more detailed sulfur  budget and  esti-
mate of natural  sources by drawing  on  the rapidly expanding research  in this
area.

The methods used by  the  above-mentioned  authors employed   the  steady-state
balancing of  sources  against sinks or  removal  mechanisms averaged over  the
Earth as a whole.  On this scale, the sinks for sulfur compounds probably  can
be  estimated  with sufficient accuracy  in  terms  of  total  mass  to  estimate a
global  cycle.  The sulfur sinks are mostly accounted for by  wet and dry depo-
sition.  On this basis, they typically exceed the estimated sources.   Sources
of  sulfur compounds  include anthropogenic and natural  sources.   The  former
can be estimated using emission  factors  and the  magnitudes   of  production
activities.  Within the natural source area,  volcanic and  ocean spray sources
ha»e been estimated,  but until  recently, (Adams et al. 1980, 1981a),  the much
larger biological component  had to be estimated from  only  fragmentary data.


                                     2-2

-------
Thus,  in  the various  global  sulfur cycles, it  has  been common practice  to
balance the steady-state sulfur cycle, after quantifying the sources  and the
dry and wet  deposition sinks,  by assuming that  any  difference  was  accounted
for by biological  emissions processes.

Estimates of the  biogenic  flux of sulfur components  from  land  areas to the
atmosphere made using  this material balance approach  have  varied from 5  Tg S
yr-1  (Granat et al.  1976)  to  110  Tg  S yr-1 (Eriksson  1963).   To  place the
biogenic contribution  in perspective,  Granat et  al.  (1976)  estimated  anthro-
pogenic sulfur emissions to  be 65 Tg  S yr'1 and the total land and  oceanic
biogenic  sulfur emissions  to  be  32  Tg S yr-1,  so the global   biogenic  con-
tribution was estimated to  be roughly half the  global  anthropogenic  emission.
Earlier estimates had  the  biogenic fraction  equal  to  or greater than the
anthropogenic fraction (Eriksson  1960,  1963;  Robinson  and Robbins  1970a).
Extrapolation of  field data  to a global  cycle results in  a value of  64  Tg S
yr'1  (Adams  et  al. 1980),  and,  although this particular  estimate is  still
only  preliminary,  since it is based  on  detailed field data it seems likely
that  better  estimates  will  tend  toward a  value between previous  extreme
estimates rather than toward the high or low ends of  the range.

Estimating natural emissions from a steady-state  material  balance  can  readily
be seen as applicable to global considerations, but for continental  and  other
smaller areas,  the material  balance procedure is less successful.   This  is
because steady-state,  homogeneous  mixing  across  a limited area and a closed
cycle  of  sources  and  sinks  generally  cannot be  assumed.   To treat smaller-
than-global  areas,  such as the  United States,  one  must deal   with  specific
estimates for the natural sources.

Although,  as mentioned  above, there  may  be considerable  doubts  as to  the
total magnitude of  natural  sulfur compound  sources on both local  and global
scales, the analytical techniques probably now have sufficient sensitivity  to
measure the major  sulfur constituents  of  the  global  background.  Sze and  Ko
(1980), as part of  their photochemistry modeling studies  of atmospheric  sul-
fur compounds, tabulated tropospheric concentration data for these  compounds.
In  Table  2-1,  background concentration data  are presented from the  tabula-
tions of Sze and  Ko (1980) that  are considered to be generally  applicable  to
the  northeastern  U.S. conditions without anthropogenic  influences;  however,
the  S042~  value  appears to  be  significantly  lower   than  has  been  ascribed
frequently to remote background conditions.   For  the  most part,  these are not
the  same  as  measurements made at  sites  in  the  northeast  that  are  currently
designated  as  rural  or nonurban  because  these latter sites  can still  be
affected  by  pollutants  through  long-range  transport.   This  was  noted  by
Galloway et al. (1982) in as distant a location as Bermuda.

2.2.1.3   Biogenic  Emissions  of  Sulfur  Compounds—The initial  estimates  of
biogenic  emissions,  such  as  those by Eriksson   (I960),  assigned  the  total
biogenic  estimate to  hydrogen  sulfide  (H2S)  because this  gas  was easily
identifiable  by  its   odor  as  being   evolved  in  swamps   and  certain  other
anaerobic  situations  and because  there was  little evidence that  other  com-
pounds were  also  part of the  natural  background.   It should be noted,  how-
ever,  that all  of the authors dealing with the  sulfur  cycle recognized the
probable  complexity  of the natural emission  cycle,  and the assumption  that


                                     2-3

-------
         TABLE 2-1.   BACKGROUND  CONCENTRATIONS  OF  SULFUR  COMPOUNDS
                      (ADAPTED FROM SZE  AND  KO  1980)
Compound                      Concentration                Location
                                yg m~3
S02                            0.52^0.23           Western  U.S.  and
                                                     Canada  above
                                                     boundary layer
                               0.25 +_ 0.12           Western  U.S.  and
                                                     Canada  within
                                                     boundary layer

S042'                               0.05            Remote ocean  areas

COS                             1.26 +_ 0.15         67°N-57°S

H2S                           0.007 - 0.07           Southern Florida

(CH3)2S                             0.15            Wallops  Island, VA

CS2                               ~ 0-31            England
                                     2-4

-------
the total  emission  was HoS was  recognized  as a simplification of the  prob-
able  real   situation.    These  initial   evaluations  were  not  supported  by
measurements because there were no methods available  for these  measurements.

The obvious problem in measuring  the biogem'c component of  the sulfur cycle,
i.e., the emissions from natural  sources, was one of  having  suitable  analyti-
cal methodology.  It was not until the  1970 's that the  measurement technology
for  H2S  and  the  organic  sulfur  compounds   that  might be  expected  to  come
from natural sources was developed.  The nature  of potential biogem'c sulfur
emissions  had  emphasized H2S  as  the probable  compound (e.g., see  Eriksson
I960) although  earlier Conway (1942)  had concluded  that  non-sea-salt sulfur
in  precipitation  away  from  anthropogenic  sources  may be  due  to  volatile
sulfur  compounds   such  as  H2$  or possibly  mercaptans.   Lovelock  et  al .
(1972)  showed  that  ((^3)2$  (dimethyl   sulfide)  was  present  in  sea  water
and  given   off by  enclosed  soils, and they  proposed  ((^3)2$  as  an  im-
portant component of the natural  atmospheric sulfur cycle.  This proposal  was
supported  by  Hitchcock  (1975,   1976)   with  calculations  of  the  probable
emissions  from  the  turnover of biomass in  the  form  of leaves, soil organic
material,  and  marine  algae (Hitchcock 1975)  and by  evaluations of  seasonal
atmospheric sulfate concentrations  in  several  nonurban areas  of the eastern
United States (Hitchcock 1976).  Reliable measurements  were  made subsequently
of possible biogem'c emissions present in the atmosphere above  soil  and  water
surfaces  suspected of being  strong  sources  of  natural  sulfur  compounds.
Jaeschke et al . (1978) describe  one of  the  first  such studies using a  very
sensitive  sampling  and  analytical  technique for  H2$.   Maroulis and  Bandy
(1977)  used  gas  chromatographic  techniques  for  atmospheric   studies  of
((^3)2$.   Delmas  et  al .  (1980)   carried out  a number  of measurements  of
the rate of evolution  of H2S  from different soils  in France and at  a number
of  sites in  the Ivory Coast.  Atmospheric  concentrations were also  measured
by Delmas et al . (1980) at many of these  sites.

These research  studies provided  an initial   test  of  the global mass balance
estimates of biogem'c sulfur emissions, but comprehensive studies  of  biogenic
emissions were not carried  out until gas chromatographic  techniques  covering
a  wide  range  of compounds  were developed.   Aneja et  al . (1981)  applied gas
chromatography to soil emissions in the  form of  air  samples collected from a
small stirred chamber placed over selected soil  and water surfaces.   This gas
chromatographic analytical  technique was capable of  detecting six  potential
biogenic   sulfur  emissions   compounds:   H2$,   ((^3)28,   (^JpS?,   COS,
C$2, and  CH3SH.   In  the  sampling program  used  by  Aneja  et  al . (1981)  the
detectable emission rate for H2S, (^3)2$, and COS was 0.01 g  S m-2  yr-l   and
for  CS2, CH3SH,  and  (CH3)2$2 it was 0.05 g  S m-2   yr-l.   In their research
they carried out a program of  sampling  on a variety of soils, marshland, and
water surfaces in  the North Carolina  area in the  summer  and fall  of  1978.
The results  of  this study of  seven  types of surfaces showed  that  the  emis-
sions of most of the  likely biogenic  sulfur compounds from most of  the  test
surfaces were below the analytical detection limits  (Aneja  et  al .  1981,  Table
I).   In  particular,  studies of  "dry  inland soils"   showed  none  of the  com-
pounds to be  above  the detection limit while  "saline marsh mud  flat"  showed
detectable emissions only of H2S and COS.
                                     2-5

-------
Further improvements in sulfur gas  analysis  by gas chromatography were made
by Farwell et  al.  (1979)  and used  by  Adams  et al.  (1980,  1981a,b,c)  in an
extensive examination of the emissions  of sulfur compounds from  soil surfaces
in the eastern, midwestern,  and southeastern  United States.  This program was
part  of the  Electric  Power  Research  Institute  Sulfur  Regional   Experiment
(SURE) program  (Perhac 1978).   Because this study produced  the largest and
most  complete  set of experimental  data  available  at  this  time on biogenic
emissions of sulfur  gases and because  it includes a considerable  amount of
measurement data from the area of the  United States  affected by acidic depo-
sition, the  results  of  this  study  by  Adams et  al. as  reported  in several
available papers and reports will  be used as  a  basis for  the following evalu-
ation of biogenic sulfur gas emissions  in the United States.  In general, the
analytical techniques described by Farwell et al.  (1979)  were able  to show an
approximately  one-order-of-magnitude  improvement   in  detection  limits  over
those reported in the earlier studies by  Aneja  et  al.  (1981).  As a  result,  a
variety of sulfur  gases could be  identified as  being emitted  even by dry,
inland soils with low rates  of evolution. The performance of the sampling and
analytical system was evaluated by Adams  et al.  (1980)  as being  indicative of
a minimum sulfur flux from the soil  and water surfaces  rather than  an average
or maximum flux value because of  possible  nonquantifiable  losses  of sulfur
compounds within the system.

Table  2-2  shows the  average sulfur flux by compound for  the  various soil
orders  and  suborders (i.e.,  "types")  sampled  by  Adams  in the SURE region
(Adams et al. 1981a).  The results of 760 field samples gathered from 10 soil
types over a period  of  4 years were averaged  for this  table.   As shown in
this listing, six sulfur compounds were identified in a large fraction of the
samples.   H?S  typically  ranked  highest  in  the  various samples  with very
high values  in some of the samples taken  in  saltwater marsh areas.   Among the
other compounds, the emissions of carbonyl sulfide (COS)  and carbon  disulfide
(C$2)  were   typically  higher  than   those  of   dimethylsulfide   [(Cl^^S].
Dimethyldisulfide  [(CH3)2S2l  was  found  in  low  concentrations in  a  large
proportion  of  the  samples,  and  methylmercaptan  (CH3$H)  was  found  to be
primarily an emission from  saline marsh  areas.  Wide variations in  emissions
were  encountered and  statistical  methods  were  used to  establish average
emission rates (Adams et al. 1980, 1981c).

In this  research program on  soil  emissions,  variations  in sulfur  emissions
were  found  to be dependent not only on  the  soil  order,  but also  on ambient
temperature,  time  of day,  and  whether  there  was  vegetative cover or bare
soil.   Temperature was  a major variable through its control   of  biological
activity  in  the soil, and  relationships  were  developed  between soil  sulfur
emissions and average temperature data  (Adams et al. 1980).   Detailed statis-
tical  analyses of the  sampling  data  provided  a   basis  for  summarizing  the
experimental  data into  three general  soil  types—coastal  wetlands,  inland
high  organic,  and  inland  mineral—and  extending the emissions  estimate  over
an  annual  temperature cycle.   The  results  for the study area,  essentially
from  47°N to the Gulf Coast and  east of  the Mississippi  River, are shown  in
Table  2-3 (Adams et al.  1981a).   As  shown  at the bottom of the  table,  the
average  sulfur  flux over  the region  is  0.03  g S  nr2  yr-1,  and  it  is
associated  with a  total  SURE region  biogenic  emission  of  about  0.12 Tg  S
yr'1.


                                     2-6

-------
                   TABLE 2-2.   AVERAGE COMPOSITION OF SULFUR COMPOUND FLUXES AND TOTAL SULFUR FLUX

                           BY  SOIL ORDERS AND SUB-ORDERS (ADAPTED FROM ADAMS ET AL. 1981 a)
ro
i
                                             Average sulfur flux, g S m~2 yr'1
        Soil  types/locales
H2S
COS     CH3SH    (CH3)2S
CS2
(CH3)2S2
Saline Marshes
Cox's Landing, NC (11/77)
Cox's Landing, NC (7/78)
Cedar Island, NC (10/77)
Cedar Island, NC (5/78)
Cedar Island, NC (7/78)
E. Wareham, MA
Lewes, DL
Georgetown, SC
Wallops Island, VA
Everglades, N.P., FL
Sanibel Island W.R., FL
St. Marks W.R., FL
Rockefeller W.R., LA
Aransas W.R., TX
Non saline Swamp
Llba, NY
Brunswick Co., NC
Okefenokee, GA
Jeanerette, LA

139.5
502.9
0.02
0.02
0.16

0.096
0.94

74.61
601.6
1.31
0.09
0.06

0.16
0.09
0.001


6.36
0.88
0.002
0.01
0.02
0.004
0.013
0.05
0.03
0.04
0.002
0.06
0.001


0.006
0.024
0.005
0.0002

6.56
11.65


0.0003


0.006
0.22
0.22
23.45
0.08
0.001
0.002







1.77
0.007
0.04
1.57
0.60
0.48
0.47
1.87
0.26
0.81
1.23
0.008
0.07

0.004
0.005
0.021
0.029


0.97

0.009
0.060
0.028
0.07
0.22
1.38
0.39
1.10
1.05
0.02
0.38

0.006
0.022
0.022
0.001





0.003
0.026
0.004
0.001
0.90
0.09
22.29
0.01



0.003


0.002


0.073

0.0004
0.0005
0.006
0.0005
0.005
0.04
0.05
1.63
0.07
0.003
0.005



0.001


152.4
518.3
0.029
0.079
1.82
0.65
0.66
1.69
4.45
75.7
650.9
3.80
0.12
0.52

0.19
0.14
0.051
0.032

-------
 I
oo


TABLE 2-2.
CONTINUED


Average sulfur flux, g S nr2 yr"1
Soil types/locales
Histosols (peat, muck)
Dismal Swamp, NC (10/77)
Dismal Swamp, NC (5/78)
Laingsburg, MI
One Stone Lake, WI
Fens, MN
Celery ville, OH
Elba, NY
E. Wareham, MA
Brunswick Co., NC
Belle Glade, FL
Lakeland, FL
Jeanerette, LA
Fairhope, AL
Coastal Soil s
Georgetown, SC
Mollisols
Ames, LA
Linneus, MO
Yankeetown, IN
Stephenville, TX
H2S

0.018
0.046
0.044
0.084
0.042
0.047
0.158

0.09
0.005
0.069
0.01


0.008

0.147
0.104
0.073

COS CH3SH


0.008
0.011
0.024
0.01
0.012
0.023

0.007
0.002

0.001
0.001

0.008

0.017
0.009
0.023
0.002
(CH3)2S

0.0007
0.002
0.001
0.001
0.001
0.003
0.006
0.013
0.006
0.001
0.003
0.001
0.002

0.002

0.003
0.003
0.002
0.001
CS2 ??a (CH3)2S2

0.0001
0.002 0.0003
0.004
0.012
0.003
0.006 0.0004
0.136 0.002 0.003
0.0004 0.0002
0.017
0.004 0.0002
0.008 0.0005
0.003
0.014

0.005

0.016
0.005
0.021 0.0005
0.004 0.0015
S

0.019
0.058
0.056
0.121
0.056
0.068
0.33
0.014
0.12
0.012
0.08
0.014
0.017

0.023

0.18
0.12
0.12
0.008

-------
                                           TABLE 2-2.  CONTINUED
Average sulfur flux, g S nr2 yr-1
Soil types/locales
Alluvial Soils
Clarkedale, AR
Al f i sol s
Wadesville, IN
Kearnysville, WV
R.T.P., NC (Wooded)
R.T.P., NC (Cultivated)
Jeanerette, LA
Shreveport, LA
Stephenville, TX
Inceptisols
Philo, OH
Belle Valley, OH
Spodosols
W. Wareham, MA
Ul ti sol s
Calhoun, GA
Fairhope, AL
Hastings, FL
Freshwater Pond
Bel 1 e Val 1 ey , OH
H2S

0.0003

0.01
0.082

0.008
0.002



0.003
0.072



0.009
0.0005
0.001

0.07
COS CHaSH

0.001

0.002
0.029
0.004
0.003
0.0003
0.002
0.0002
0.002
0.004
0.002


0.003
0.001
0.001

0.02

(CH3)2S

0.0001

0.001
0.002

0.0005
0.0003
0.006
0.0003

0.0002
0.004

0.013

0.002
0.002
0.003

0.005
CS2

0.003

0.002
0.022
0.001
0.001
0.0004
0.005
0.003

0.001
0.010



0.011
0.005
0.002

0.028
??a (CHs)2S2



0.002
0.0001






0.0014
0.002

0.0002

0.0001
0.0001 0.0003
0.0003 0.0007

0.002
S
0.002

0.017
0.13
0.0
0.013
0.003
0.013
0.004

0.008
0.094


0.013

0.024
0.008
0.008

0.13
Unidentified  sulfur gases.

-------
TABLE 2-3.  SUMMARY OF ANNUAL SULFUR FLUX BY SOIL GROUPINGS
         WITHIN THE STUDY AREA (ADAMS ET AL. 1981a)
Soil grouping
Coastal wetlands
Inland high organic
Inland mineral
Total
Sulfur flux
g S yr"1
48,822 x 106
13,451 x 106
56,843 x 106
119,116 x 106
Land area
m2
2.56 x 1011
6.85 x 1011
27.26 x 1011
36.7 x 1011
Emission density
g S m"2 yr-1
0.191
0.020
0.021
0.032
                           2-10

-------
 In evaluating these results it must be remembered that the sample coverage of
 the test  area was  not  complete.   The  program considered a total of 32  sites
 mostly  in single visits of  about 5 days each.   Statistical  techniques  were
 used to select sites and to evaluate the data (Adams et al.  1980).   borne  sur-
 face soil  types  showed a high  degree  of variability,  especially the wetlands
 and  tide  marsh  areas,  and these  were  assessed in detail  by  this research
 program.   Adams  et al. (1980)  discusses in  detail  the problems of evaluating
 the biogenic sulfur flux from tide flat and wetland areas.   The major  conclu-
 sion was  that the very high emissions were from 1 percent or less of the  tide
 flat surface,  and  this was  an even  smaller fraction  of  the  total  coastal
 wetland soil type.  Thus the average biogenic emission from  this soil  surface
 is weighted according to the relative emission areas within  the soil type.

 In this analysis,  standard soil  classifications were  used  as  the basis  for
 the soil  identification.   These  soil  classifications are shown as  soil  type
 subheadings in Table 2-2.   In Table 2-3,  coastal  wetlands include  the saline
 and nonsaline marshes  or   swamps  and  the coastal soils; inland high  organic
 soils include the  Histosols, Mollisols,  and  the  Ultisol/Spodosol soil  orders
 and suborders; and inland  mineral  soils comprise the remaining drier soils of
 the region  (Adams  et al.   1980).   In  terms  of  a percentage of  the extended
 study area (essentially the area of the United States  east of the Mississippi
 River), coastal  wetlands are 7  percent of the area,  inland high organic soils
 are  19 percent,  and  the  inland  mineral  soils are  the  remainder,  or  74
 percent.

 Table 2-3 and Figure 2-1  illustrate several  features  of the  biogenic  sulfur
 flux.   First,  and probably most  important,  the total biogenic  or  soil  flux
 depends to a significant extent on the  inland soils,  even though their emis-
 sions density is  an order  of magnitude less than that  of the wetland soils.
 The much  larger  area  of inland soils, 93 percent of  the study  region,  more
 than makes up for the low  emissions density;  and, as shown in the figure, the
 inland  soils  account for   59  percent of  the sulfur emissions  in  the study
 area.   It is  of course recognized that  there is considerable variability in
 the soil  emission system and this  must  be allowed for in any  application  of
 these results.

 Figure 2-2  (Adams  et al.   1981a)  shows  the  results  of the estimates  of  bio-
 genic sulfur flux  measurements for the total SURE  grid plotted in terms  of
 the average  sulfur emissions in  metric tons per year  per  grid area  (6,400
 km2)  as a function  of  latitude from 47°N, about the  latitude of Duluth,  to
 25°N, the latitude of  the tip  of the  Florida  peninsula.    The relationship
 between annual  sulfur flux  per  6,400 km2 grid as a function  of latitude is:

                          log Y =  4.70212 -  0.035588X

where Y is  106  g  S  per 6,400  km2  and  X is the north-south grid  identifi-
 cation  number (Adams et al. 1981c).

This relationship  between  sulfur  flux and  latitude  shows  an approximate
exponential  increase toward  the  south, especially  south  of about 33°N,  the
latitude  of a line between Shreveport,  LA,  and Georgetown,  SC.  This rapid
increase  of  sulfur flux southward  is interpreted  as  being a  result of  an


                                    2-11

-------
                                           INLAND
                                        HIGH ORGANIC
                                            19%
                            INLAND MINERAL

                                 74%
                     RELATIVE  LAND  AREA  BY  SOIL TYPE
                                          COASTAL
                                         WETLANDS
                                            41%
INLAND MINERAL
     48%
                                   INLAND
                                   HIGH
                                   ORGANIC
                   RELATIVE SULFUR FLUX BY SOIL TYPE
Figure 2-1.  Comparison of relative land area and sulfur flux by soil
             type.
                                  2-12

-------
    2,000
    1,000
CVJ
 o
 o
 VO
•H-
w
 >-
 to
      500
      300
      100
            47°N
                                                                   25°N
                                  80  km  INTERVALS
Figure 2-2.   Total  natural  gaseous  sulfur emissions averaged across
             latitude zones in the  SURE study area, 47°N and 25°N,
             expressed as a function of latitude.   Emission rate as metric
             tons of sulfur per year per SURE grid area (6400 km?) (103 mT
             S yr~l equals  0.16 g S m~2 yr~l).   Adapted from Adams et al.
             (1981b).
                                  2-13

-------
increase in temperatures,  an increase in wetland areas,  and  a  higher  fraction
and a  higher fraction of  high organic  soils.    To  the  north  into  Canada,
biogenic emissions  would  be expected  to decrease as  shown by the  downward
trend toward higher latitudes in Figure 2-2.

Figure 2-2 has been used to estimate the potential biogenic sulfur flux  from
the State  of Florida,  as  an example of  a  high  bioaenic emission area.   For
Florida, the area along the northern border near 30 N  has  an indicated annual
flux density  in  units  of metric  tons  (103 kg)   of about  350  ml S per 6,400
km2,  or  about  0.05  g  S  nr2 yr-1;  in  southern  Florida,   at  25°N,   the
indicated annual  emission density is about 2,000 mT S  per SURE  grid  of 6,400
km2,  or about 0.3 g  S nr2  yr-1.    The  total  statewide  estimated  sulfur
flux for  Florida  is 16,980 mT  S  yr-1.   By comparison, the estimated state-
wide anthropogenic emissions of S02 for Florida  in  1978  were  about  606,000
mT  S02 yr-l  or  303,000  mT  S  yr'1   (Section  2.3.2.1).    Thus,  the esti-
mated biogenic emissions on a  statewide basis in Florida  are  about 5  percent
of the 1970 estimated anthropogenic  emission.

Hawaii, with  its  generally warm  and moist climate,  would  have a  relatively
high estimated  biogenic  sulfur  emission  density of  about  3,000  mT S  yr"1
per 6,400  km2.    For  an  area  of 16,500 km2,  the  biogenic  sulfur  emission
estimate  is  about 7,600 mT S  yr"1.   This  compares  with  a  1970 statewide
sulfur  emission  from  anthropogenic  sources of  about  29,000 mT S yr-1 (U.S.
EPA 1973).  For large areas in the Northeast the ratio of  biogenic  to anthro-
pogenic emissions would be  much  less than  for either  Florida  or Hawaii where
biogenic processes would be expected to be a maximum.

If areas smaller  than  a state  are considered, it is,  of  course, possible  to
find areas where  natural  sources exceed anthropogenic estimates.  The indi-
vidual Hawaiian Islands other than Oahu, with its concentration  of  population
and industry, probably have predominantly  natural emission  sources.   Rice  et
al. (1981) assessed the ratio  of  natural and  anthropogenic  sources in a  num-
ber  of sectors of about  104  km2 across the United  States.  They concluded
that  in rural and  nonindustrial  areas of  the   United  States  local   natural
sources may exceed local  anthropogenic sources.   However,  they also concluded
that  in  the eastern  United  States,  where  high S042'  concentrations   are
found,  the  natural sources of  sulfur  probably  make a minor  contribution  to
the airborne  sulfur compounds.   Galloway  and  Whelpdale  (1980)  estimated  that
northeastern U.S. and southeastern Canadian anthropogenic  emissions are about
16  Tg  S  yr'1,  which  supports  the  conclusion  that  biogenic  sources   are
unimportant on a regional  basis.

It is  not reasonable to evaluate  the biogenic versus  anthropogenic  ratio  over
a small area  relative  to  acidic  precipitation problems because of the rela-
tively  long  reaction  times required for  sulfate  formation  and  incorporation
in precipitating  storm systems.   These  processes  lead  to  longer travel  times
and  thus  considerable  mixing  of emanations from  over  a  relatively  large
source  area.

As  a  first approximation  to a  global  system,  Adams  et al. (1981c)  extended
their  model  beyond the midlatitude  zone  of measurement  shown  in  Figure 2-2
and concluded that, on a global basis, the biogenic sulfur emission flux  from


                                     2-14

-------
land  areas  is about  64 Tg  S yr-1.   This  may  be compared  with Granat et
al.'s  (1976)  estimate  mentioned  earlier,  of 32  Tg  S yr-1  for  land and
coastal  areas.    On  a  global basis,  the emission of 64 Tg  S  yr"1  is an
average  emission  density  of  about  0.43  g  S  nr2 yr"1  over  the  149  x  lO1^
m2 global land  area.   A similar  figure  for Granat1 s  estimate is about  0.22
g S  m"2  yr-1.  The model  shown  in  Figure 2-2  when  extended to equatorial
latitudes predicts an  emission value that is within  the range  of  the measure-
ments made by Delmas et al.  (1980).   Adams et al. (1981c) point out that the
sulfur emission  rates in tropical  areas are probably  at least  an  order of
magnitude higher than  those  found at 25°N—along the U.S. Gulf Coast.   Simi-
larly, as illustrated  by  Figure 2-2,  these latter  rates are about 10  times
higher than  those found at about 35°N.  The emissions  rates decrease further
by about another factor of two between 35°N and 47°N in the study  area.

A summary of  the  natural  or biological emissions rates  for sulfur compounds
in the United States  east of the Mississippi River can  be made  by applying
the  average  density  from  Table 2-3,  0.03  g S m"2 yr"1,  to  an area of  2.23
x 1012 m2  to yield an  estimated  natural  emission  flux of about 0.07  Tg S
yr'1.   if this  same  emission density is  extended  to the contiguous United
States,  an area  of 7.824 x 1012 m2,  the resulting  natural   source  is  0.23
Tg S yr"1.    This latter  figure assumes  sulfur  emission soil properties in
the more arid areas of the west to be similar to those measured in the  east.
This is not likely to  be the case.  Also, in the  west  there is no  counterpart
to the moist Gulf Coast and  its significant wetland  areas.

Figure 2-3 illustrates the results of  the  measurements  of biogenic emissions
of gaseous  sulfur compounds made over the EPRI  SURE  grid.   Figure 2-3 was
prepared from the individual grid estimates of annual  soil sulfur  flux (Adams
et al.  1980,  Figure 4-1).   The highest emission areas  are  found along the
coastal region from South Carolina north  to southern  New Jersey.  This  zone
appears to be about 100 km  wide,  although  the 80 x 80 km grid squares do not
permit a  detailed presentation.   In this  coastal  zone  the  average annual
emission  is  greater than 30  kg km"2 yr"1.   Another  region  with relatively
high  annual   grid  emissions   is  along  the  Mississippi River   south   from
Illinois.   Relatively  low  emissions are  found  along  the coast north  from
central New Jersey and over  most of  the interior  land  areas.   The  New England
states,  except  for  the  southern coastal  zone, and southern  Canada  fall
generally into the lowest soil emission category, an  annual  emission of  less
than  15  kg  knr2  yr"1.   Open  ocean  areas are estimated  by Adams  et al.
(1980) to  have  an  emission  of  less than  10 kg  knr2 yr-1,  although  open
ocean emissions were  not  measured.    South of the  SURE grid,  soil emissions
are  expected  to  increase  generally,  as indicated by the latitudinal distri-
bution of average emissions  shown in Figure 2-2.

2.2.1.4   Geophysical  Sources  of Natural  Sulfur  Compounds—Natural emissions
of sulfur from nonbiological  sources include two classes of sources that are
important to  the  northeastern United States mainly because  they  are part of
the global sulfur cycle:  sulfate aerosol  particles  produced by sea spray and
sulfur compounds  emitted  by volcanic activity.  In  the global cycles  esti-
mated  by material  balances,  both  of these  sources   are  determined  to be
relatively small  contributors to background sulfur  levels  over land  areas
(Eriksson 1960,  Robinson  and  Robbins  1970a,  Granat  et  al.  1976);  however,


                                     2-15

-------
                                                       >.3Q kg

                                                      22.5 - 29  kg  km'2 yr'1

                                                ili  15 - 22.5  kg  km-2 yr'1

                                                       <15 kg km'2 yr'1

                                                      OCEAN,  <10 kg km"2  yr"1
Figure 2-3.   Annual  biogenic sulfur emission pattern for the SURE grid
             over the northeastern United States.  Adapted from Adams
             et al.  (1980).
                                2-16

-------
more  recent estimates  by  Cadle  (1980)  may change  the evaluation  of  the
importance of volcanic emissions.

2.2.1.4.1   Voleanism.   Volcanic  eruptions  are  obvious  sources of  a wide
variety  of  materials  including sulfur compounds and,  as such, volcanos  can
make important contributions  to the  global  sulfur  background.  For example,
the Mt.  St. Helens eruption  in Washington  State  on May 18,  1980, contained
S02, H2S, COS,  S042-,  and  H2S04  as  well  as  chlorine-  and nitrogen-containing
compounds  (Pollack  1981).   Concentrations  of CS2  and  COS  in  the  Mt.  St.
Helens plumes  were  reported by Rasmussen et al.  (1982).   Although  Mt.  St.
Helens was  a major event locally,   its  total  impact  on the atmosphere  was
relatively  short  lived  and  its   contributions  to  global  back-   ground
concentrations  in  the   troposphere  are  not likely   to  have  caused  major
pertubations.  The April  1982  eruption  of El  Chichon  in southern Mexico  was
perhaps  20  times  as large as  that of Mt. St.  Helens  and injected a  massive
amount of sulfur gases into the middle atmosphere (Kerr  1982).  However,  the
southern latitude of  the  El Chichon  eruption, relative to the United  States,
prevented the  early  transport of most  of  the  El  Chichon plume  across  the
United States.  Significant northward spread of the stratospheric portion of
the plume was not expected until  the  seasonal  climatic  shifts occurred in  the
fall of 1982 (Kerr 1982).

Estimates of volcanic sulfur  compound contributions to the global atmosphere
vary greatly because the emissions of volcanos differ  in  gas  content,  volume,
and eruption  frequency;  each investigator  must make  a  number  of personal
judgments of the relative importance  of  these factors.  Granat et al.  (1976),
in  reviewing  emissions  data  up  to   about 1975, estimated the annual   global
volcanic emissions of sulfur  compounds  at about  3  Tg  S  yr~l,  or only  a  few
percent of the total estimated global sulfur cycle.

Since Granat's evaluation of  this emission  classification, several important
field programs have been carried out on the active volcanos  of St. Augustine
in Alaska and  Mt.  St. Helens in Washington.  At St. Augustine, Stith  et  al.
(1978)  estimated  S02  emissions at about  0.05 Tg  S yr-I  and lesser  amounts
of  H2$.   Emissions of  sulfur gases  from Mt. St.  Helens in Washington over
the year March 1980 to March 1981, which included the major eruptions in May
and June  1980,  were estimated by  Hobbs  et al. (1982)  to  be  0.15 Tg S yr-1
as  S02  and  0.02  Tg  S yr"1  as  H2S,  for  a  total  of  about  0.17  Tg  S
yr-1.  This  is  three to four times the estimate made by Stith et al. (1978)
for St. Augustine.

Cadle  (1980) has  summarized  volcanic  sulfur gas emissions and has commented
on  impacts  of these  emissions.   There  have  been a  number  of estimates  of
average annual volcanic emissions, and Cadle describes the hazards of making
the  various  assumptions  that are necessary  for a  volcanic  gaseous  flux
estimate.  A  number of estimates of volcanic sulfur  gas  emissions  cited  by
Cadle  (1980) are listed in Table 2-4.  Cadle's  (1980) conclusion  relative  to
this published data was that volcanic  emissions may  contribute as much as a
third  of the global  anthropogenic sulfur emission  of  about 65  Tg  S yr-1.
This  would   be  about  20  Tg S  yr~l.    However, Cadle   (1980)  calculated
volcanic sulfur gas  emissions from lava  flow data and the result was in the
range of 2 to 8 Tg  S  yr~l.   The  major sulfur compound from volcanic action,


                                     2-17

-------
        TABLE 2-4  ESTIMATES OF VOLCANIC SULFUR  GAS  FLUX VALUES
                 (ADAPTED FROM DATA IN  CADLE  1980)
                                                    Estimated  Flux
Authors                          Date                (Tg  s  yr-i}
Bartels                          1972                     17
Kellogg et al.                    1972                      0.8
Friend                           1973                      2
Stoiber and Jepsen               1973                      5
Naughton et al.                   1975                     24
Granat et al.                     1976                      3
                                     2-18

-------
as  noted by  Cadle,  is SOg.    Cadle  (1980)  also considered  the volcanic
emissions of  H?S,  COS,  and CS?  and concluded  that they  were  unimportant
on a global  scale relative to SO?.

Cadle (1980)  has suggested that precipitation scavenging around  volcanos  is
underestimated.  Thus, as more data on volcanic activity become  available,  it
might be more  reasonable  to  assign any  significant  increase  in volcanic
emissions to  the  precipitation  part of  the global sulfur cycle,  which  would
probably  leave relatively  unchanged  the biogenic  sulfur  estimates made  by
difference.    The  discussion   by   Cadle  (1980)  relative  to  precipitation
scavenging  of  volcanic  emissions  points   up   a  fact  that   should   be
reemphasized;  I.e., the  long-term effects  of  volcanic emissions  are  due
primarily to  the part of  the  eruption  cloud  that  reaches  the  stratosphere,
where it will  have a  residence time  long  enough to  cover a  considerable
distance  from  the  source.    Tropospheric  emissions,  while  they  can   be
devastating  in  the vicinity   of  the  mountain,   will  decrease  rapidly   in
importance with  distance and will  not be  contributors  to long-term, elevated
background emissions over large areas.

Although  it  was  stated  earlier that  the  volcanic  contribution should  be
considered  primarily on  a global   basis,  it  also might  be argued that  the
volcanic zones of North America could have an  important impact  on the United
States.   The  volcanic  activity in  both  Central  America and Alaska  can  at
times be  significant to the United States, at least  on a local  basis.   The
volcanic emissions  in Alaska are likely to be  important because  of the lower
tropopause and the wind circulation toward the "lower 48" associated with  the
polar jet stream.   A good example of pollutant transport over long distances
from northern  latitudes  is the  drift of Canadian forest fire smoke over  the
United States, which occurs from time to  time.  In Central  America,  the much
higher tropopause exposes  more  of  the volcanic emissions to rapid precipita-
tion  and cloud scavenging  processes  than  might be typical  in  Alaska.  Also,
wind  circulation systems  near  the equator are  not generally favorable  for
transport north  toward  the United States (Ratner  1957,  Kerr  1982).   The  Mt.
St. Helens eruptions spread a plume over large portions of the United  States;
however,  after several  months  of active emissions, the  rate  of  activity  has
decreased  to  low  levels.    Unless  Mt.  St.  Helens becomes more  or less
continuously   active,   it   can   probably   be   disregarded   as  an  important
background source both in  the United  States and on a global  scale.

2.2.1,4.2  Marine sources  of aerosol  particles and gases.  The oceans  contain
sulfur compounds  in the form of sulfate salts, and,  when  sea water droplets
evaporate  in  the  atmosphere,   some  sulfate-containing  particles  are  formed
(Junge  1963).   In  the formation  of marine  aerosol  particles,  the  larger
particles from wind-blown  waves and bursting bubbles rapidly fall back to  the
ocean surface and are of  little consequence to the large-scale  distribution
of  marine  aerosols.   Fine  particles  with  some  prospect  of  a  prolonged
atmospheric  residence  time  are formed  in  the spray  bubble process by  the
bursting of the bubble film or  "skin."  The numbers of particles, and  whether
they  will  remain airborne, will  depend on wind  and  sea surface conditions.
Quantitative  estimates of  these aerosol  formation conditions are difficult to
make.  Most  authors of  atmospheric sulfur  cycles  reference  Eriksson's (1960)
estimate  of  44  Tg   S yr"1 as  the sea  spray  contribution of  more  or less


                                      2-19

-------
persistent fine  particles  in the  atmosphere.   Of  this  total, he estimated
that about 10 percent, or 4 Tg S yr-1  of  sulfur,  would be carried over land
areas.  Since 90  percent  of sea spray remains in the  oceanic  regions  rather
than mixing into  continental  air masses,  it may  be  considered as playing  a
secondary role  in the overland phases of  the  global  sulfur cycle (Eriksson
1959, 1960;  Robinson and Robbins  1970a; Granat  et  al.  1976).

Another aspect of the oceanic contribution  to the  sulfur  cycle  is  the release
of  gaseous  sulfur compounds  from  the  ocean surface.   Because of the  large
area of the global oceans, even a relatively small emission rate may lead  to
a significant total  emission.   Sulfur  or sulfate that cannot  be  balanced  by
considering the  other common sea salt components such  as  sodium is  called
"excess"  sulfur  and has been  noted by a  number  of  authors.   For example,
Lodge  et  al.  (1960)  measured "excess sulfur" in  the  North  Pacific   Ocean
atmosphere.   Cadle  et  al.  (1968)  measured  trace levels of  SOa  at coastal
sites  in  Antarctica,  and  Lovelock  et al.  (1972) measured dimethylsulfide  in
the Atlantic.

In  global  sulfur balances,  the  "excess"  marine  sulfur  source is sometimes
identified as a  separate biogenic  source  needed to balance the total  sulfur
cycle  (Eriksson  1960,  Robinson  and  Robbins   1970a);  alternatively  it  is
considered a coastal  phenomenon and  is combined with  the biogenic land area
sources (Granat et al. 1976).

In  the United States, the transport  of background gaseous or  "excess"  sulfur
from oceanic areas should be considered along  the  Pacific and  Gulf of  Mexico
coasts where onshore winds  are predominant.   The  excess  oceanic  area  sulfur
is  due to both  sea  surface emissions  and volcanos.   The magnitude  of  this
onshore transport can be estimated  using an  average  onshore or westerly wind
of  8  m s'1  through a 3000-m mixing  depth  (Ratner 1957,  U.S.  DOC 1968).   On
an  annual basis  this  gives  an  onshore  transport of marine air of  about 1.2 x
1018  m3  yr'1   across the  Gulf  Coast  (about 1600   km)  and  about  1.5   x
1018  m3  yr-1  across  the  Pacific  Coast  (about  2000  km).    Background
sulfur  compound concentrations applicable  to  marine  air  masses, from  data
summarized by Sze and Ko (1980), have been given in Table 2-1.   In that list,
S02,   H2S,   (CHa^S,  and   S042'   have    atmospheric   residence   times   of
up  to  a  few days (Sze and Ko 1980)  and thus could contribute  to a background
loading  that  might  in  turn  participate  in precipitation  pH  reactions  and
acidic  dry deposition.   The  remaining  compounds,  COS  and  C$2, have  much
longer atmospheric residence times, several years or longer (Sze and  Ko 1980;
Ravishankara et al.  1980)  and, with this  slow reaction  rate,  probably exert
little  influence on  precipitation  pH or  acidic  deposition.    The  four more-
reactive  compounds provide  a  total  concentration of  about 0.2 ug S  nr-3  in
the marine  air  masses  that  could be  expected  to  participate  in  acidic
deposition processes.  Considering the estimated total annual  air mass  Y.olum§
transported across  the Gulf  and  Pacific  Coasts given above  (1.2 x lO1^  m
yr'1  across  the Gulf  Coast  and  1.5  x  1018 m3  yr'1  across  the  Pacific
Coast)  results  in  an estimated marine  air input  of about 0.36  Tg  S  yr"
across the  Pacific coast and about  0.24 Tg  S  yr-1  across the Gulf Coast for
a  total  background marine  air  mass contribution of  about  0.6 Tg S yr"1  to
the total United States.   We have not included  an  estimate  of the  possible
transport across  the Atlantic  Coast  because general  wind   climatology  is


                                     2-20

-------
unfavorable for  this  transport (Ratner  1957).   Local  winds and  individual
short-lived circulation systems could bring  some marine  S  across  the Atlantic
Coast, but  it  would not be a  persistent situation  such as occurs along  the
other coasts.   We previously estimated the biogenic  emissions  for the contig-
uous United States  at 0.23 Tg S yr-1, and  thus  it  would seem that incoming
marine air  masses may  be more  or  less  equivalent to biogenic sources  in
importance  to  background  sulfur  loading.   The  precision  of these several
estimates cannot be expected to be high,  but,  when  they are compared to  the
estimated  anthropogenic  emissions  of 12 to  15  Tg  S  yr"1,  these natural
sources would  still  seem  to  be less  than  10 percent  of  the  total  sulfur
burden.

2.2.1.5  Scavenging Processes  and  Sinks—Ultimately, reactive materials  such
as  the  sulfurcompoundsreturntothe  Earth's   surface  either  through
precipitation-related  mechanisms   or by  direct  attachment to  the Earth's
surface through  processes  known collectively  as  dry deposition.   Both gases
and aerosol particles participate  in  both deposition routes.

Sulfur compounds  also participate in  a  variety of reactions in the atmos-
phere, generally tending  toward  oxidation  to S042"  and  the  formation  of
sulfuric acid  or sulfate  particles.  Hydrogen  sulfide,  probably  the  most
common natural  sulfur emission to  the  atmosphere,   is  oxidized  to  S02  and
then to sulfate.  Graedel (1978),  Sze and Ko (1980), and others describe  this
reaction.  The initial reactant is probably  the hydroxyl  radical, OH, and the
average  lifetime of  H2$  is  given  usually  as only  a  few  days at typical
atmospheric concentrations.  Reactions of S02  in  the atmosphere due to  both
homogeneous and  heterogeneous reaction  processes  have been  estimated  by  a
number of  authors including  Granat et al.  (1976),  Graedel  (1978), Husar  et
al.  (1978),  Altshuller  (1979),  Sze  and Ko  (1980),  and  Rodhe  and Isaksen
(1980), to name  only  a few.   Although some  calculated S02  atmospheric life-
times  are  quite long  (e.g.,  Graedel [1978,  pp.  29-30] estimates about  430
days), the general consensus seems to  favor an atmospheric  residence time  of
only a few days  (e.g., Sze  and Ko  1980).  Altshuller (1979),  in  an extensive
set of chemical  model  calculations of S02 reactions in nonurban situations,
showed that the  rate  of  reaction  was more rapid in  summer  than  winter,  much
more significant at low  latitudes  than at high latitudes, and more rapid  at
low altitudes  than  in the middle or  upper  troposphere.    Altshuller (1979)
concluded  that the  most significant  reactant for  S02  was OH-   Rodhe  and
Isaksen (1980), on the basis of a  global  model, estimated the global average
residence  times  for   H2S,  S02,   and  S042"  to  be  about  1,   1.5,  and  5
days, respectively.

H2S  oxidation  in liquid  drops  is  also  possible   (Cox  and Sandalls 1974).
The product is sulfate,  with an intermediary  status as  S02«  The decay  rate
for H£$ via the  liquid droplet route  is  given by Granat et al.  (1976)  as  a
day  or more;  and  for  (^3)2$  the  reaction  rate  is  even  slower.     The
reaction  of (CH3)oS  apparently  goes directly  to   sulfate without  an  S02
intermediate step (Cox and Sandalls 1974).

Gaseous  reactions  of  the  organic sulfur compounds commonly  identified  in
natural emissions,  CS?,   COS,  (CH3)2S,   (CH3)2$2,   and  CH3$H,  are given by
(1978), Sze  and Ko  (1980), and  others.   These  reactions  proceed  to  HS04


                                     2-21

-------
and/or  sulfates,  but  not always  through  S02  as  an  Intermediate  compound.
The  common   sulfate   compound  in   the   atmosphere  is   ammonium   sulfate
C(NH4)2S04]  as a  result  of  the  reaction,  presumably  in liquid  droplets,
between the two common gases ammonia (NH3)  and S02-

As mentioned above, pollutants are deposited on the Earth's surface by either
wet or  dry processes  and  these topics are  discussed in detail  in other chap-
ters (Chapters A-6 and A-7) of this document.  However,  briefly  with regard
to  acidic  deposition, the  precipitation  scavenging mechanisms  are  directly
involved in the precipitation  pH or  acidic  deposition  controversy,  and it is
useful  to mention some aspects of deposition in this discussion.  Various au-
thors have pointed out that surface  waters may  be  affected by  deposited pol-
lutants, whether  they arrive  as part of  the precipitation chemistry or  are
deposited on  the  ground  in a  dry  state and  then  are incorporated  into  the
surface water.  Resuspension of sulfur compounds is probably minor because of
their general solubility  and thus  rapid incorporation  into the  soil.  Desert
areas and agricultural regions with exposed soils may create situations where
strong winds may cause blowing dust.  This  would resuspend both the deposited
material and natural  soil  constituents.

Granat  et  al.  (1976)  have  attempted to estimate the  relative  importance of
precipitation and dry  deposition processes.   They argue  that  dry deposition
increases in  relative importance for  situations  where the value of the  dry
deposition velocity,  V
-------
times the rate of  Inland  soils and account for about 40 percent of the bio-
genic sulfur emissions in  the area east of the Mississippi.  Figure 2-3 sum-
marizes, on  a grid basis,  the results of a  measurement program on gaseous
sulfur emissions  from  soils in the midwestern  and  eastern United  States.  The
more arid and alkaline soils in the west would be  expected  to have lower bio-
genie emissions than are found along the east coast, but actual  measurements
have not been made in  these  areas.   Nevertheless, extending the east coast
average emissions  rate  to  the 48 contiguous states, an  area of  about 7.8 x
1Q12 m2,  results  in  an estimated total biogenic  emission  of  about 0.23 Tq
S yr-1.   U.S. anthropogenic  sulfur  oxide  emissions are in  the  range of 12
to 15 Tg S yr-1.

The  compounds that are most important  in  the biogenic flux  are  H2S, COS,
and C$2.  Of secondary importance are ((^3)2$, (CH3)2$2,  and
Ocean areas may  also  make a contribution to  the  natural  sulfur burden  over
land  areas  through (1)  the  transport of particles  from  the evaporation  of
fine  seawater aerosol  particles formed  in  bubble-bursting  processes,  (2)
sea-surface-generated gaseous sulfur compounds,  and (3)  the  sulfate  particles
formed by atmospheric reactions  of  sea-surface-generated gaseous sulfur  com-
pounds.  Estimates of oceanic transported sulfur were made  using a  3-km mix-
ing depth,  an 8-m-sec"1  average onshore wind, and background sulfur  concen-
trations of 0.18 x  10-6 g $ m"3  for  gaseous  compounds  and 0.02  x 10~6  g
S nr3 for sulfate  particles.  The  results of this calculation  indicate  that
the  annual  sulfur input across the Pacific  Coast is  about  0.36  Tg S  yr-1
and  about  0.24  Tg  S yr~l  across  the  Gulf Coast.    Because large-scale
onshore winds do not dominate the east coast, no  attempt was made  to extend
this rough  estimation procedure  to  that area.  Thus, marine background  input
may  introduce about 0.6 Tg S  yr-1  across  the United  States  coastal  area;
this is about  three times  the amount estimated to  be generated  by biological
soil  processes.   As  marine air masses  travel  inland,  this  sulfur  compound
content would  be subject to a continuing process of scavenging reactions.

On  a  long-term basis,  volcanic  activity is not expected  to be a major  con-
tributor  to the levels  of natural  sulfur  in the contiguous United  States,
although special situations like the  1980 eruption of Mt.  St.  Helens or the
southern Mexico  volcano El Chichon could perturb  conditions for short  time
periods.

Thus,  in  total, the  potential  upper-limit background  sulfur burden of the
United  States  is  about  1.0  Tg S  yr-1, which  includes  contributions  from
biospheric  and oceanic  generation processes.   This  figure does not  include
any correction for amounts "exported"  by  air  masses  moving  across  the coasts
or  borders.   In  terms  of relative  importance, it may be compared to  anthro-
pogenic sulfur oxide emissions that are in the range of 12 to 15 Tg  S yr'1.

2.2.2  Nitrogen  Compounds

2.2.2.1   Introduction—Nitrogen  compounds are emitted  to  the atmosphere from
natural sources  in several  forms:   as  relatively  inert nitrous oxide (N2°)»
as  potentially acidic nitric oxide  (NO)  and nitrogen  dioxide  (NO?), and as
potentially  acid-neutralizing   ammonia  (NHs).     The  sources  for  these


                                     2-23

-------
compounds, other than anthropogenic  emissions,  are,  to a major extent, in the
terrestrial  biosphere  with some  injections  into  the  troposphere  from the
oceans, from  stratospheric photochemistry and  from atmospheric fixation by
lightning.

The estimation  of  natural  sources of  nitrogen oxides  and  ammonia  has been
severely restricted in the past by a lack of reliable data  on concentrations
of  these  compounds  in  the ambient  atmosphere.    Even  at  present, ambient
atmospheric measurements  in clean  or  background  areas  are  research  tasks
rather  than  routine  monitoring  with   continuous  instruments,  such as  is
carried out in urban area studies.   Thus,  the evaluation of  likely impacts of
natural sources of nitrogen compounds is subject to  considerable variability,
probably greater  than  is the case  for  estimates of natural sulfur  compound
emissions and their impacts.

Nitrous oxide  is  essentially inert  in  the troposphere and  plays  no role in
problems of precipitation pH; thus,  detailed consideration of its sources and
sinks can be omitted without affecting  the objective of  this document.

Table  2-5  lists  background  concentrations  of  NOx  and  NH3,   based  on
relatively recent  research,  which  are probably applicable to nonanthropo-
genically-affected locations.

2.2.2.2  Estimates of Natural Global  Sources and Sinks—A  first  approximation
of  the global  magnitude  of naturalsources  of nitrogen  compounds can be
obtained from a review of  two previously  published  nitrogen compound cycles,
one  by  Robinson   and  Robbins   (1970b)  and  one by Soderlund  and   Svensson
(1976).  Major differences between these two environmental cycles exist, with
the  more recent  one by  Soderlund  and Svensson  (1976)  proposing   signifi-
cantly smaller  fluxes  between reservoirs.   This reduction  in fluxes results
from  improved  estimates of atmospheric  concentrations, based on an  increased
number of better measurements of background  concentrations.  Table 2-6 lists,
as  a  starting  point  for  this  discussion, emission  and sink flux  estimates
adapted  from  Soderlund  and Svensson  (1976)  for NO,  NOe,  and NHs  or
NH4+.  The  nitrogen  oxides,  NO  and  N02,  were  combined  as  NOX  for  this
estimate,  and   the  NHs values  also  include   the  ammonium  ion  NH4+-   The
NOX  deposition  values  include   nitrate  (N03~)  compounds   also.     In the
original  reference by  Soderlund and  Svensson  (1976),  anthropogenic  emis-
sions  of  NOX  compounds  totaling  19  Tg  N yr-1 were  included in the NOx
flux  values,   and  the NH3  emission estimates included  the  emissions from
coal  combustion,   ranging  from  4  to 12  Tg  N  yr~l.   These  were  estimated
global emission values for 1970  (Soderlund  and Svensson 1976).    To empha-
size  the  natural  emission  cycle  in  Table 2-6,  we have  subtracted  these
anthropogenic  emissions  from the  original  values to arrive at  the  tabulated
values.   Emissions  and  gaseous  reactions  are given   in  terms of NHs  (N)
while deposition terms are shown in reference  to NH4+ (N).

In  a  detailed  paper  submitted for publication,  Logan (1983) derived a  nitro-
gen cycle with several  important  differences relative to  that  given in  Table
2-6.   Biogenic emission of  NOv  is estimated  by Logan  at 8 Tg N   yr"1 with
a  range  of 4  to  16  Tg  N yr-f  in  comparison   to  a value  of 21  to 89 Tg  N
yr"1  in  Table  2-6.   Logan  (1983)  also estimates  lightning  as a  potential


                                     2-24

-------
            TABLE 2-5.  ATMOSPHERIC BACKGOUND CONCENTRATIONS OF
                        NITROGEN OXIDES AND AMMONIA
Constituent
Concentration
   ug m"3
  Reference
NOX (afternoon)
as N02

NO (afternoon)


NH3 (land)


NH3 (ocean)
  0.4 - 0.5


 0.04 - 0.12


    1 - 8


     0.06
Kelly et al.
 (1980)

Kelly et al.
 (1980)

Hoell et al.
 (1980)

Ayers and Gras
 (1980)
                                     2-25

-------
             TABLE 2-6.  GLOBAL EMISSIONS OF NITROGEN COMPOUNDSa


                                           Total             Global  emission
                                         Tg N yr-1             density  .
                                                             g N m-2 yr--1


Terrestrial^

     NOX emissions0                       21-89            0.14  - 0.59
     NOX wet deposition                   13 - 30            0.09  - 0.20
     NOX dry deposition                   19-53            0.13-0.36
     NH3 emissions'*                      109 - 232           0.73  - 1.56
     NH4 wet deposition                   30 - 60            0.20  - 0.40
     NH4 dry deposition                   61 - 126           0.41  - 0.85
     Organic N wet deposition             10 - 100           0.07  - 0.67

Atmospheric Reactions (global)6

     NH3 loss via OH                       3-8              0.006  - 0.016
     NOX formation                         3-8              0.006  - 0.016
     NOX lightning formation                 ?g
     N0xfrom N20 + UV                       0.3                 0.0006

Oceanic^

     NOX wet deposition                    5-16            0.014  - 0.04
     NOX dry deposition                    6-17            0.017  - 0.05
     NH4 wet deposition                    8-25            0.022  - 0.07
     NH4 dry deposition                   11 - 25            0.03  - 0.07
     Organic N emissions                  10 - 20            0.03  - 0.05

River flow to ocean

     NOX                                   5-11
     NH4                                   < l
     Organic N                             8-13


aAdapted from Soderlund and Svensson (1976).

bTotal  land area:  1.49 x 1014 m2-

C0riginal reference includes 19 Tg  N yr'1  anthropogenic emissions.
 Deposition terms include anthropogenic contributions.

Original reference includes 4 to 12 Tg N  yr'1 from coal  combustion.
 Deposition terms include anthropogenic contributions.

6Global  area:  5.13 x 1014 m2.

fOcean  area:  3.64 x 1014 m2-

^Recent data indicate a possible value of  5-10 Tg N yr-1  (see Section
 2.2.2.5).
                                     2-26

-------
 source  of 8  Tg N  yr-1  (range  2  to  20  Tg  N  yr'l).   Logan estimated  NOv
 from  fossil  fuel  sources at  21  Tg N  yr"1  plus an additional  12 Tg  N  yr~T
 from  biomass  burning (slash  and  burn  agriculture,  land clearing,  forest
 fires).    If  these latter  sources were considered  man-caused  sources  then
 Logan's  anthropogenic sources would  total  33 Tg N yr'1  with a  range  of 18
 to 52 Tg N yr-1.

 An estimate of  the  wet deposition  of  organic  nitrogen compounds, e.g., amino
 acids,  amines,  and  proteins,  is included  in  the  above-noted  estimate.
 Soderlund  and Svensson  (1976)  include some  generation  of organic  nitrogen
 compounds  at  the ocean surface,  but this process is  not well  known, as indi-
 cated by the  order  of magnitude  range for the  estimate of terrestrial  depo-
 sition.   Other sources  or  sinks (e.g.,  dry  deposition)  of organic nitrogen
 compounds are not identified in  Table  2-6,  nor  is  the organic nitrogen cycle
 balanced.

 Table 2-6  also  includes  estimates  of  the global  emission  density in units of
 g N m~2  yr"1.  These figures  were calculated  from  the  values  of  the total
 fluxes shown  in the table,  using values  from  Butcher  and  Charlson (1972)  for
 global  land  and ocean areas  without  attempting  to   correct  for surface  or
 climatic effects expected to change emissions in polar regions, deserts,  etc.

 Gal bally  (1975) has  made  separate  estimates  of NOv  and NH$   sources  and
 sinks, based  on a  boundary  layer gradient  method analogous to a calculation
 of dry  deposition.   For the  Northern Hemisphere, he obtained  an  NOv  emis-
 sion  of  30 Tg  N  yr-1 and  a   value  of 130  Tg  N  yr-1  for NH4+.   Galbally
 (1975)  also  considered  differences between  tropical  and  temperate  latitude
 conditions in background  concentrations and between land  and ocean conditions
 1n making  his estimates.  His  estimates may  be  converted  to average emission
 densities  of  0.32  g N  nr2   yr-1 for  NOX  and  0.55  g  N  m-2  yr-1   for
 NH4+.    These  values   are  comparable  to   those  derived  from  Soderlund
 and Svensson (1976) and listed in Table 2-6.  Galbally's  estimating  procedure
 would appear  to be relatively insensitive to  local  high  concentrations  of
 anthropogenic emissions.    In  Table 2-6 natural  NOX  emission  densities  of
 0.14  to 0.60  g N  nr2  yr~l  are  indicated.   More  recent estimates  (e.g.,
 Logan 1983) arrive  at lower values of natural emissions because  they  relate
 to newer and lower ambient background  NOX concentrations.

 The  nitrogen  compounds  N02  and  NH4+  return   to  the  Earth's  surface  by
 both dry and  wet deposition mechanisms.  Dry  and wet deposition  rates would
 be expected to vary between  being of about  equal  importance in areas general-
 ly removed from industrial  source  areas  (Granat et al. 1976)   and situations
 where dry  deposition  was perhaps twice the magnitude of  wet  deposition  near
major source regions  (Garland  and Branson 1976).  As  pointed  out by  Gal bally
 (1975),   the  natural  sources  of  NOX   and  NH4+   appear  to  be of sufficient
magnitude  to  explain  the observed  global  deposition of  these  compounds  in
 precipitation; but this would  not necessarily  be true  for  individual  regional
areas because of the tendency  for anthropogenic  sources to be  concentrated in
 relatively small areas (with  reference to  a global scale).   It  is  generally
assumed that  natural   sources  are distributed  more  or less  uniformly over
relatively large  areas  of  the globe, with  their emission fluxes   changing
gradually in  response  to  temperature,  moisture,  and soil conditions.


                                    2-27

-------
2'2.2.3   Biogenic Sources  of  NOX Compounds--It  seems  to be generally  con-
cluded that  the  major natural  sources  of  NOx are  found  in the  terrestrial
biosphere  (Junge   1963,   Galbally  1975,   Soderlund  and   Svensson   1976),
although one  set of observations  indicating  a  tropical  ocean  source of  NO
will  be  described subsequently  (Zafiriou  et  al. 1980).   A wide variety  of
experiments have been carried  out on  nitrogen compound losses from soils  of
various types because of the impact such losses  may have  on the  availability
of fertilizer nitrogen to crops.

Altshuller (1958) pointed out that NO  production  can be quite large  and rapid
under  certain  conditions.    He described how N02 concentrations of  several
hundred parts  per million  occurred  in silos  shortly after  the storage  of
silage.   These  concentrations  occurred under anaerobic conditions  with  high
moisture content in an all-organic environment.

In this assessment of terrestrial sources it will not be  possible to  present
a  comprehensive  review of  all  work  in the  soil  sciences  that relates  to
nitrogen compound releases from the soil, but work  that can  be related to  an
NOX  source  for  precipitation  chemistry  will  be   reviewed.   In  the past  few
years, interest has been renewed  in nitrogen emissions from soil  triggered  by
nitrogen  fertilizer  because N;>0  is a significant  fraction  of  this  release
(Nelson and Bremner 1970)  and  its impact on the  stratospheric ozone layer  is
of great concern.

Nelson and Bremner  (1970),  as  a  result of  laboratory experiments,  concluded
that  soil  or fertilizer nitrite  can  be a   source of significant amounts  of
N02-    Although   the  amounts  of  NOg  released   in  these  experiments  were
inversely related to soil pH,  significant  amounts of N02 were released  from
soils with pH greater than 7.0, i.e.,  from  alkaline  soils. Some of the exper-
iments  were  consistent  with  the  hypothesis that   atmospheric  N02  results
from the breakdown of nitrous acid to  NO and the  atmospheric oxidation of the
NO to  N02.   However, they  did not have  the  capability  of measuring NO  in
their experiments.

Nelson and Bremner  (1970)  found  that  in the laboratory,  the organic  content
of the soil  had  an  important effect on  the  amount of nitrite that  was fixed
to  N2J  however,   the  proportion   of the nitrite  that was  recovered   as  N02
was not dependent on the organic content.   In many  of their experiments,  the
evolution of N02 represented  the largest   fraction  of  the nitrite added  to
the  soil;  however,  the  total  nitrogen  recovered was divided among nitrate,
nitrite,  Np,  N20,   and  N02*     In  experiments   on  five  soils  in  the  pH
range of 4.8 to  6.0, held for  2  days  at 25  C, the evolved  N02 accounted  for
55 percent of  the applied nitrite. At near neutral   pH (6.6  to 7.0),  28  per-
cent of the  nitrite  was  evolved  as N02-  As  indicated above, at least  part
of  this  N02 was  released  as  NO and  was  subsequently oxidized  to  N02«
Experiments  with  completely closed systems showed  that  N02 reacted  further
and was recovered as nitrate.

As mentioned, these  experiments  were  done  in the laboratory  under  a  variety
of conditions and cannot be translated to flux rate  values under  field condi-
tions.   However, they do indicate clearly  the evolution  of NO and N02  from
                                     2-28

-------
soils under  a variety of  conditions  and the  probable  dominant role of  NOX
in the spectrum of soil  emissions.

The work  of Nelson and  Bremner (1970)  cited  above dealt  with NOg  evolved
from  nitrite  applied  to the  soils as NaNC^.   Prior experiments by  Makarov
(1969) were  related to  applications  of  nitrate  as NH4N03  and the  results
showed a  decrease in  the evolution of N02  from these field soils when mic-
robiological processes were reduced by the  addition  of  inhibiting  substances
to  the   test  soil  field plots.   Thus  it was  hypothesized  that  NOa soil
emissions  were  related  to  microbiological  activity.    Perhaps  the most
interesting  data  for  our  considerations   were  produced by the  conditions
reported by Makarov (1969)  for  his unfertilized control plots.   His  control
plot  tests  with  a  Sod-Podzolic  soil  in  the  U.S.S.R.  showed  that  NOe
evolution  during  one  experimental period  averaged  0.6 g ha~l  hr~l from
May 31  to  September  26 (119  days).   This N02  production is  0.17  g m-2,
which is equivalent to 0.05 g N m~2,  for the  experimental  period.  A second
experiment  in the same soil  over the  88-day  period from  24 June to   20
September  averaged 1.06 g N02  ha"1   hr-1, which  converts  to a  total   of
0.07  g  N  m"2 for the  period  of the  experiment.   An  experiment   using  a
different  soil,   Chernozem,  was  shorter in  duration  and   not  reported   in
detail,  but it appears  that significant  N02 emissions were  produced  similar
to those shown in the  other tests.

Because  gaseous nitrogen evolution decreases with temperature  (Keeney et  al.
1979), it  is  likely that these summer NOg emissions can serve  as at  least a
first approximation of  an  annual   emissions rate for higher latitude areas.
Thus  we  can  compare Makarov's  results,  which approximate 0.06  g  N nr2, with
the global  cycle results shown in Table 2-6.    In  this tabulation,  natural
NOv emissions were estimated  to have  an  emission density of 0.14 to  0.6  g N
m-2 yr-l.    The   two  sets  of   results  seem compatible because  the  global
estimate would be  increased by  the effect of warmer,  low-latitude areas with
longer warm seasons.  This  has been shown to be the  case with biogenic sulfur
emissions  where   field  experiments   have   identified  a  strong  temperature
relationship (Adams et al.  1980).

Field experiments  on  NO evolution from  grazed and ungrazed grassland  areas
were carried out by Galbally and Roy  (1978).  They were able to show,  through
the use  of improved  instrumentation,  that NO  is continuously evolved from
natural   grassland soils, and that N02  is  a negligible  fraction of  the  NO*
flux  from  the soil.   In  the  atmosphere,  the NO emission is  rapidly  oxidized
to  N02  by the ambient ozone  (03)  concentration.   This  emission of  NO fol-
lowed by  an atmospheric reaction  to  form  N02  was  hypothesized earlier  by
Robinson  and Robbins  (1970b).     In  the  Australian  field  measurements   by
Galbally  and  Roy, the observed  NO emission  density,  if  integrated over  a
year, amounted  to a value  of  0.1  g  N nr2  yr"1.   If  this  rate  is  extended
to a global land area  value, it produces  a total  nitrogen emission of 10 Tg N
yr'1.

Bulla et  al.  (1970) also reported that  the emission of NO  from  soil  is  not
dependent  on microbiological action.   Their experiments were done on Oregon
soils in  the  laboratory.   In these experiments,  as  with those  of Nelson  and
                                     2-29

-------
Bremner  (1970),  NO as  a fraction  of  added nitrite  dominated  the nitrogen
emissions over both N2  and  N20.

The  generation  of  NOX  in  oceanic  atmospheres  has  not  been  considered a
significant  feature of  the  global nitrogen  cycle  by  most  investigators
(Galbally   1975,   Soderlund   and   Svensson   1976).      However,   in   an
investigation in  the central  equatorial  Pacific (7°N-10°S, 170°W), Zafiriou
et   al.   (1980)   found  that   nitrite   photolysis   in   seawater  produced
concentrations of NO.  They showed  that in  these tropical  areas,  the  buildup
of  NO  in  the surface  water  layers occurred  in  daylight  and  disappeared
quickly at night.  From partial  pressure comparisons of  the water  samples  and
atmospheric  NO  concentrations,  Zafiriou   et  al.  (1980)  and  Zafiriou   and
McFarland  (1981)  concluded  that tropical  ocean  areas, especially  areas rich
in  nitrite,  may  be sources of  atmospheric NO,  but  on  a global  scale   the
source is  less  than 1  Tg  N  yr'1  and thus is  insignificant in  the global
nitrogen  oxide cycle.

2.2.2.4  Tropospheric and Stratospheric Reactions—A small  transport of  N02
into  the  troposphere  from  the  stratosphere  probably  occurs.   Soderlund
and Svensson  (1976) estimate  this flow at 0.3  Tg N  yr~l,  which on a global
basis  is  0.0006  g  N  m~2  yr'1,  a negligible  part  of  the  cycle.  This
stratospheric formation  results  from  reactions of  N20  with  0('D), which
occur at altitudes where wavelengths below 2500 nm  are present  to form 0('D)
(Bates and  Hays  1967).   Robinson and  Robbins  (1970b)  give  some  additional
comments  on this stratospheric NOX source.

As a result of improved measurement techniques,  Kley  et al.  (1981) have been
able to  develop  observational  data of  vertical  NOX profiles  through   the
troposphere.   These profiles  show  that  the concentrations of  NOX change
from 0.19  yg  m~3  as  N02  in  surface  air  to   about 0.38  yg   nr3   at   the
tropopause.   They  attribute this  increase  in  concentration to  the intrusion
of NOX into the  troposphere from  the stratosphere,  which is consistent with
a  flux of about  1  Tg  N yr~l  (Kley et al.  1981).    This stratospheric   NOX
flux is consistent with  other transtropopause source estimates  (Johnston et
al. 1979).   The  NOX  source  may  be  the stratospheric  photochemical reactions
of  N20 or  the  NOX  emissions  of  subsonic aircraft flying  in   the upper
troposphere and lower stratosphere (Kley  et al.  1981).  There have been some
questions  raised  relative  to  the  importance   of  this   stratospheric   NOX
source to the tropospheric  global  nitrogen  cycle  (Fishman  1981).

Atmospheric reactions of NH3  in  the troposphere  involving reactions  with OH
radicals   have been  proposed  as another   source  of  NOx.   Soderlund   and
Svensson   (1976),  using  reaction systems  suggested  by Crutzen  (1974)   and
McConnel   (1973),  estimated  a   formation   rate  of  NOX   from  NH3   in   the
atmosphere of 3  to 8 Tg N  yr-1.    As indicated  in  Table  2-6, this is equal
to  a  global  source emission  density  of  0.006 to 0.016  g  N  nr2 yr"1-
Thus, this is also an inconsequential source of NOX.

2.2.2.5   Formation of  NOX  by Lightning—The question  of nitrogen fixation
by  lightning  has been  studied  tor  more  than 150  years,  and no  definitive
answer is  yet  at hand.    Soderlund and   Svensson  (1976) leave  the possi-
bility of  lightning fixation  as  still   a questionable  atmospheric source


                                     2-30

-------
of NOY,  as Indicated in  Table 2-6.   They note one  reference  on the ques-
tion of  lightning  fixation  of nitrogen, dated  1827  and authored  by  J.  von
Liebig.

Junge  (1963)  stated  that the  consensus of opinion at  that time  (1963)  was
that  the  evidence  for  lightning  formation  of  N02  was  marginal,  and
referenced Viemeister's  studies  of thunderstorms  (Viemeister  1960)  and the
NO?  concentration  measurements done  on the  Zugspitz by  Reiter and  Reiter
(1958).   Georgii  (1963),  in  reviewing  the evidence to  1963  and including
Visser's  detailed  analysis  of  rain  chemistry  in  Uganda  (Visser   1961),
concluded that lightning was not a  factor in nitrogen  oxide concentrations.

Although Noxon  (1976,  1978)  was able  to observe  enhanced N02 patterns  near
thunderstorms, confirming the  information  of NOX  by  lightning,  it is still
apparent that  observational  evidence  linking  atmospheric  NOX to  electrical
discharge  is   for  the  most part  still  lacking.1    However,  modeling  and
theoretical analyses done since the early  1960's indicate  more  strongly  that
lightning  or  electrical  discharges in  the  atmosphere could be  a source of
NOX.

One  of the more recent  assessments  of lightning  fixation of nitrogen is by
Hill et  al.  (1980)  who  conclude  that  lightning  may  cause a  maximum  N02
production  rate  of 14.4  Tg  yr"1  or  4.4 Tg N  yr-1.    Dawson  (1980),  in an
article  published  back-to-back  with  Hill  et al.  (1980),  concluded  that
lightning  may  produce about 3 Tg  N yr-1.   Dawson  also used Noxon's  (1976,
1978)  data   on  solar   spectral   measurements  of  enhanced   N0£    around
thunderstorms  to  deduce  a  global  annual  N0£ production rate  of  7  Tg  N
yr~l   but   commented,   "with   considerable   uncertainty"   (Dawson    1980).
Finally,  the   laboratory  studies  of  nitrogen  fixation  by spark  discharges
(Levine et al. 1981) can  be  mentioned,  which,  when extended to  a  global NOx
budget,  result  in  an  estimated production of 1.8  Tg yr-1  of NO  or about 0.8
Tg N yr"1.  Logan (1983)  has  reevaluated the  lightning  NOX formation  data
and  concludes  that a reasonable annual  global  source  is  about  8 Tg  N  yr-1
with a range of between 2 and 20 Tg N yr-i.

On the basis of  the available  assessments  of  nitrogen fixation  by lightning,
it is  probably realistic  at  this time  to assign a  production rate of  5 to 10
Tg N yr-1  to  this  source in place  of  the  question mark shown in  Table  2-6.
This production  would translate to  an  emission  density  nitrogen  flux  of  0.01
g  N  nr2  yr"1  on a   global  basis,  although  lightning  and   thunderstorm
distributions are geographically skewed toward warm, humid areas and  seasons.

If further research can link lightning discharges  more directly  with  signifi-
cant  NOX  formation,  the  frequent  occurrence of  thunderstorms  and their
accompanying  lightning  in the midwestern  and  eastern regions of  the  United
iNote added  after  external  review:   Drapcho  et al.  (1983)  report  increased
 N0£  and  NO at  a  ground station  during and after  passage of a midwestern
 thunderstorm.     Their   extrapolation   of   these  measurements  and   other
 references  indicate  that global  nitrogen  fixation  by  lightning  is in  the
 range of 1 to 40 Tg N yr'1.


                                     2-31

-------
States could be  an  important consideration with regard to  acidic  deposition
in the northeastern states and southeastern Canada.

2.2.2.6   Biogenic NOX Emissions  Estimate for  the  United  States—Quantita-
tive  measurementsofRfixemissionsTorawidevarietyof  biospheric
situations, such  as  were  made for biogenic  sulfur  emissions, have not  been
made  for  NOX.    Nevertheless  there  is   little  doubt that  there  are  NOX
emissions from the biosphere, as described in the previous discussions.  Thus,
in  order  to arrive  at some  estimate of  biogenic emission  rates it will  be
necessary to use  secondary methods of estimate.  The material  balance  pro-
cedure has already been described, and, as noted in  Table  2-6, the  nonanthro-
pogenic global  emission  of NOX has  been  estimated  to  range between  21  and
89  Tg N  yr"1.    If  this  NOX  emission  is  assumed  to come  only  from  land
area processes in the nonpolar  regions, an average  calculated biogenic  emis-
sion  density  is  then  in  the  range of 0.16 to 0.68  g N  m"2 yr"1  for  the
131 x  1012  m2 of  global  nonpolar land area  (70°N  to 55°S).  Applying  these
global  emission  rates derived  from material  balance considerations  to  the
contiguous  United  states,  7.8  x  1012   m2,  and   the  area  east  of  the
Mississippi  River,  2.23   x  1012  m2,  results  in  an  annual  biogenic  NOX
emission estimate  of 1.25  to 5.30 Tg N yr"1  for the  United  States and  0.36
to  1.52 Tg  N  yr"1 for the area east of  the Mississippi River.  The lack of
precision and the large possibility for error in this  very simple calculation
is obvious, but it still  can  be used as a guide for  further discussion.

Galbally  (1975)  has  taken  another approach  in making  an  estimate  of natural
emissions  by  using  the  diffusivity  and  concentration  gradient.  With  this
calculation procedure  and a  surface layer  average concentration  of  4  ppb,
Galbally  (1975)  estimates the  Northern  Hemisphere  natural  emission  of  NOX
to  have  an  upper limit  of  30 Tg  N yr"1  or  0.31  g N m"2  yr"1  for  the
nonpolar regions of the Northern Hemisphere (equator to 70°N).  Applying this
emission  density  to  the United States  results in  an  estimated  maximum bio-
genic  NOX emission  of 2.4 Tg  N  yr"1 and 0.69 Tg  N yr"1  for the contig-
uous  United States and the area east of  the Mississippi River, respectively.
These  values are  about midway  in  the values derived from the  range given by
Soderlund and Svensson (1976) and given in Table 2-6.

More  recently Logan  (1983),  using  NO and N02  emission measurements  from
pasture plots of  Galbally  and  Roy (1978), has estimated  the  global NOX bio-
genic  source  to  be  8 Tg  N  yr"1.   This   is  a  value  of  about 0.06  g  N  m"2
yr"1  or  about  20 percent of  the  emission  density   calculated above  from
Galbally  (1975).  Applying this value to the contiguous United States  and the
area  east  of  the Mississippi River  results  in annual  biogenic  NOX emission
estimates of 0.47 and 0.13 Tg N yr"1, respectively.

Measurement techniques  for NOX that are  applicable to background  situations
have  been available  only  in recent years and  it  appears  that general  MOX
background  concentrations  may  be  significantly  lower  than the values  used by
Galbally  (1975)  and  Soderlund  and  Svensson (1976).   This may  be  expecial-
ly  true  for midlatitude  areas  such as the United States.   For example, Kelly
et  al. (1980),  after a  program of  background  measurements  in  the  Colorado
Rockies,  concluded  that  the  NOX  concentration  in the  boundary  layer  was
about  0.39 pg m"3,  as shown  in  Table 2-5.   This is  very  much  lower than


                                     2-32

-------
the 6  yg m-3  used by  Gal bally (1975)  as the  basis for  his  NOX  biogenic
emission estimate.   Thus,  even the  relatively  low annual emissions  derived
for the  United States  from  Logan's (1983) global  emission estimate may  be
high by about a factor of 3  or so.

Table 2-7  summarizes these  several  estimates of  the biogem'c  NOx  emission
source as they may relate  to the contiguous United States and to  the region
east of  the Mississippi  River.   The  1978  estimates of anthropogenic  NOX
emissions for these two areas is also shown (see this  chapter, Sections 2.3.1
and 2.3.3, Figures  2-4  and 2-7).   On  the basis  of  Logan's estimate  or  the
modified data based  on  the  ambient air measurements  of Kelly et al.  (1980),
the biogenic  estimates  are  about  7 percent  of the  estimated anthropogenic
emissions in the contiguous United States and  4  percent in the region east of
the Mississippi River.

2.2.2.7  Biogenic Sources of Ammonia—The identification  of a biogenic source
for  ammonia  and  ammonium  compounds that  are part of both atmospheric  and
precipitation trace chemistry is more or less circumstantial.  Dawson (1977)
summarizes the evidence by which a  surface emission of ammonia can be infer-
red.   First,  ammonium is found in relatively  high  concentrations  in  rain-
water, and, because it can be presumed that there are  no  major sources in  the
atmosphere  (except  of  course  the  reactions  to form  NH4+  from  NH3),  a
surface  NH3  source  can probably  be inferred.    Second,  concentrations  of
NH3 in the  air are directly  related  to  the pH  of the underlying  soil,  in-
creasing with  soil  temperature, and are  higher over land than water areas.
These  factors  favor  an  alkaline  land  source.    Furthermore,   atmospheric
ammonia  concentrations  decrease rapidly  with  altitude above the ground sur-
face but are trapped and tend to increase under  an inversion layer.

Dawson  (1977)  provides  a  number of  references  that support  these  various
features  of  the  atmospheric  NH3/NH4+  distribution.    He  further  states
that "the evidence thus indicates that the soil   is the primary source of  the
world's  ammonia,  though emission  from  uncultivated,  unfertilized vegetated
land  has never  been  measured."   This latter  statement  still   seems  to  be
correct, as of late 1982, although there  have  been a large number  of investi-
gations  by soil  scientists  and agronomists examining  NH3 losses   as  a  func-
tion of  added  fertilizer (Smith and Chalk 1980).  Also,  there is  one set of
measurements from Korean forest  and grass  soils by Kim (1973).    In  this
study, Kim  measured the evolution  of NH3 and  NOv by placing small  plastic
hoods  over  areas  of  topsoil  in  pine-,  oak-,  ana grass-sod-covered  areas.
During his field  test periods, 22 May to 27 July 1971,  the  average emission
of  NH3  was  3.41  kg  ha-1   wk-1   for  topsoil   in  a  pine  stand,  2.62   kg
ha-1  wk"1  for  topsoil   in  the oak forest,  and 1.84  kg ha-1  wk'1  for  an
adjacent  grass sod  area.    If an  average  of   3 kg ha*1  wk~l   as  NH3  is
taken for the  forest soil  emissions rate, it would translate into an annual
nitrogen flux  of  about  13  g N  nr2  yr~l,  a figure  about an order  of magni-
tude  higher than that  estimated  for  ammonia  emissions by  Soderlund  and
Svensson (1976)  and listed in  Table  2-6.   Even  if  the  NH3 emissions  esti-
mate by Kim is considered as a peak seasonal  value, which it probably was, it
is  still significantly  greater than  the NH3  emissions  factors  listed  in
Table  2-6.   However,  because  the  emissions  measured by  Kim  are  from soil
surfaces within  vegetated  canopies, they may indicate an emissions  density


                                     2-33

-------
                TABLE 2-7.   SUMMARY OF BIOGENIC NOX  ESTIMATES
                             FOR THE UNITED  STATES
Author
Soderlund and Svensson
(1976)
Galbally (1975)
Logan (1983)
Boundary Layer Cone.
= 0.25 ppb (see text)
1978 Anthropogenic (this
chapter)
Contiguous
U.S.
Tg N yr'1
1.25 - 5.30
2.4
0.4
0.15
5.7*
U.S. east of
Mississippi River
Tg N yr"1
0.36 - 1.52
0.69
0.12
0.04
3.2b
aAdapted from Table 2-32.

bAdapted from Table 2-21.
                                     2-34

-------
that  needs  to be corrected for some  significant  amount  of canopy or vegeta-
tion  reabsorption.    This  factor  of  canopy  interaction  has been  discussed
briefly by  Dawson (1977) who  cites the  research of Denmead et al. (1976)  and
Porter et al. (1972).

To compensate for the fact that  applicable,  generalized  flux measurements of
NH3  from  soils  or  the   land  surface  were  not available,  Dawson  (1977)
developed a  "simplified" model  for the  production  and emission of  NHs  from
soil,  based on  "unsophisticated  physical chemistry  and microbiology."   In
this  model,  soil   NH4+  concentrations   were   derived  from  comparisons  of
biomass decomposition and  nitrification rates.  After calculating equilibrium
concentrations of NH3 in  the soil, Dawson incorporated  a  diffusion  equation
to  generate  the  flux of  NH3 to the  atmosphere.    Model  input  parameters
allowed for  effects  of  soil  moisture  as  determined by rainfall  and  evapora-
tion,  soil   temperature  as  inferred  from air temperature,  and  biomass  or
primary  productivity.   Soil  pH  was  also  a  major  model  parameter.    The
necessary model parameters were estimated on a global  basis for 10°  latitude
zones  from  70°N  to  60°S,  and the zonal  flux  of NH3  to the atmosphere  was
estimated  and then  totaled.   The result was  32.5  Tg  NH3 yr-1  (27  Tg  N
yr-1)  from  the  Northern  Hemisphere  and  14  Tg NH3  yr-1  from  the  Southern
Hemisphere for a total of  about 47 Tg  NH3 yr-1, or 39  Tg  N yr-1.

The  latitudinal   pattern   showed  essentially  zero emissions  in  the  polar
regions, a  relative  maximum  in  the midlatitudes,  and a  relative minimum  in
the  tropics.   The  tropical  minimum  may  be  surprising  at first, but  it  is
explained by low pH  values  in  the  soil,  which limit  NH3  release,  accom-
panied by excessively high temperatures, which also are not conducive to high
NH3  emission.   NH3  emissions are modeled  as  having  a  maximum emis-  sions
rate  in  a  temperature  range from about  18  to  24  C.   These  model  calcu-
lations  agree well   with   the latitudinal emissions  pattern for NH4+  that
Dawson (1977)  obtained  from  Eriksson's  (1952)  rain  chemistry data  and  with
Eriksson's  total  global   estimate  of  42.5   Tg  NH4+  yr-1.    However,  the
value  calculated  by Dawson (1977)  is only 16  to 35  percent of  the ammonia
emissions estimate  of Table  2-6  from  Soderlund and  Svensson  (1976),  and,
although  it may closely approximate  a  precipitation deposition  pattern,  it
does  not  account for  any  dry deposition of  either  gaseous or  particulate
components.

According  to Soderlund  and  Svensson (1976),  dry  deposition  processes  are
estimated to  be  about twice  as effective an ammonia sink  as precipitation.
Dawson (1977) discounts dry  deposition  onto the  soil  because,  as he states,
"there is no reason  for ammonia to be  significantly absorbed by  soils."   This
is a  questionable  assumption considering  the  solubility of  ammonia  and  the
wide distribution of moist vegetation  and moist and acidic soil.  A number of
investigators  have  argued that  ammonia  will  be readily  absorbed  in a  dry
deposition  process   similar   to  that  for sulfur dioxide  and  other  gases
(Robinson and Robbins 1970a, McConnel  1973,  Soderlund  and  Svensson  1976).
Experiments  on  plants in  growth  chambers  has shown  significant uptake  of
ammonia through the  leaves (Hutchinson et al. 1972).

The  global   nitrogen  cycle  proposed   by   Soderlund  and  Svensson  (1976)
mentions, in  particular, the  ammonia  produced from animal  urea and  excreta.


                                     2-35

-------
The  total  amounts  of NH3 On  a  global  basts from  wild  and domestic animals
and  humans  is estimated  to be  between  22  and  41  Tg  N yr-1  or 17  to 19
percent of the total  emissions estimate for ammonia.  The  remainder, about 80
percent of the total  (about 4 Tg N yr"1 is  attributed  to coal combustion),
is  assigned  to  ammonia  emissions  from  the  decomposition  of  dead  organic
matter,  but   presumably  this  could  include  the  sort  of  microbiological
emissions modeled  by Dawson  (1977).    The estimate of  ammonia losses  from
animal and human waste is based  to  a  significant extent on the measurements
by Denmead et al. (1974)  of ammonia  losses  to the atmosphere from  an actively
grazed sheep pasture in Australia.  Emission densities this pasture averaged
0.25  kg  N   ha-1  day1  (9.5  g  N m-2  yr-1)   for  a  3-week, late   summer
period.  If this  very large  emission rate is  assumed, the  ammonia  losses  from
the  global  animal  and  human populations  could  play a  role   in  the   global
nitrogen  balance.    It  is   still  less  than  75  percent  of  the  forest  soil
emissions of Kim (1973) described above.   Interestingly,  however, the  grazed
pasture emission rate of Denmead et al.  (1974)   is  larger than Kim's  (1973)
estimated rate  from ungrazed grass sod of  8 g N  iir*  yr-1.    Harriss and
Michaels  (1982)  have  shown  that animal   wastes  and  other man-caused NH3
sources are significant NH3  emission sources  in the United States.

The soil emission estimates  by Galbally (1975) have already been mentioned in
the discussion of  NOX sources.    He has  also applied his gradient transfer
methods  to  make  an  ammonia soil source estimate.   In  his  calculation, he
assumes  ammonia  concentrations   in  the atmospheric  boundary  layer of 5 yg
m"3  in  temperate  zones,   13 yg  m"3  in  tropical  areas,   and   3  yg m"3
over oceanic  areas.   His resulting ammonia  emissions  estimate is 130 Tg N
yr'1 for  the Northern Hemisphere.  If this  value  were  doubled to about 260
Tg  N yr"1  to approximate  a global  ammonia emissions  estimate,  it   would
approximately equal  the source estimate for ammonia given  in Table 2-6.

Since Galbally (1975) made  his  global  source estimates  for ammonia, further
improvements have  been  made  in  measurement  techniques   and  indications are
that  actual  boundary layer concentrations  are  probably  significantly  lower
than those used  by Galbally  in  his calculations.   For temperate latitudes
Galbally  used an  ammonia  concentration  of  5  yg  nr3  whereas  more   recent
data  indicate  a  range  from  less  than  0.7  yg  nr3  to  around  1.4  yg nr3
(e.g., Braman and Shelly 1981).   For ocean  areas  Galbally  used  a value  of 3.5
yg  m"3;  more  recent  data  indicate   that about  0.07   yg nr3  is  a  more
realistic concentration  (Ayers   and Gras  1980).    Although recent  data are
apparently not available for tropical  areas, it  seems likely that Galbally's
value of 13  yg m"3  is  also high.  Thus,  global concentration patterns may
be  only 10  percent  or less  of  those  that  Galbally used  in  his emission
estimate  and as  a  result  it may  be  appropriate  to reduce  his  global NH3
emission estimates by this  factor or to about 13  Tg N yr"1  for the Northern
Hemisphere.

2.2.2.8  Oceanic Source for Ammonia—For the most  part,  investigators  of the
ammonia  cycle  tend  to  consider  the  ocean  surface  as  being  an  improbable
source  of  ammonia  because of the latter's solubility.    However, these  con-
clusions fail to recognize  that  a steady ammonia background concentration of
about 0.9  yg nr3 has been  observed over  the Atlantic  Ocean  by  Georgii and
Gravenhorst  (1977)   and  that  in  the  area  of  the  Sargasso  Sea and the


                                    2-36

-------
Caribbean,  ammonia concentrations  of 3.5 to 7  yg nr3  were observed  over
relatively large  areas.   Also, in  Panama, where  air trajectories have  some
ocean  fetch,  Lodge and Pate (1966) measured ammonia concentrations of  14  yg
nr3,  and  Junge   (1963)  reported   marine  air  concentrations  of  ammonia  in
Florida  and  Hawaii   of  5  yg  nr3  and  2  yg  nr3, respectively.    In the
Southern  Hemisphere  (Tasmania),   Ayers  and  Gras  (1980)   found that NHs
averaged  about 0.06  yg  m~3  in   air  that  had  not had a  recent overland
trajectory.   In  discussing  ammonia emissions from  the  ocean,  Junge  (1963)
pointed  out that  nitrate  reduction  by  plankton   in the surface layers may
provide a marine source of ammonia.

Using  their low measured concentrations of ammonia over marine areas,  Georgii
and  Gravenhorst  (1977) calculated  an average ammonia  emission density  from
the  sea to  the  atmosphere  of only 0.05 yg nr2  hr~l   as  ammonia.    This
converts  to an  annual emission  density of  about 0.0004 g  nr2 yr~l  or  a
total  global ammonia emission of 0.15 Tg N yr~l.

Graedel  (1979) approached  the  problem of the  trace chemistry of ammonia  on
the  basis  of  a photochemical reaction system.  He  considered organic,  inor-
ganic, and halogenated compounds in the marine atmosphere and  in  particular a
set  of precursor  compounds.   His  selection was based  on limiting considera-
tion to  those  compounds that were  potential  natural  emissions; thus,  obvious
anthropogenic  compounds such as the Freons or CCL4 were  not  included  in the
study.  Tabulated data on the  trace  constituents  in  the  atmosphere were  used
along  with an extended  set of reactions  and rate  constants  to estimate  a
steady-state trace chemical  composition  of  the marine  atmosphere.  For  this
consideration, a  set  of 13  precursor compounds  (e.g.,  ozone and  hydrochloric
acid)  were introduced  into the  computation system.   The  photochemical model-
ing  system,  including scavenging  processes,  was  run   along  with   typical
diurnal  changes  in meteorological  conditions such  as  solar  flux and mixing
depth.   Emission  fluxes into  the  atmosphere  must be added  to the system  to
establish  a  steady-state  situation;  these calculated  emission  rates  for  a
steady-state  situation are  one  product of  the  model.   For  NH3,   Graedel
(1979)  starts  with an average  marine atmosphere  concentration  of about 0.7
ug m~3,  probably  significantly higher  than  is now considered  realistic  on
the  basis of the  newest measurement techniques.   Thus,  his estimated  global
ammonia  emission  from the  ocean   of 3.2 Tg  (NHs)  y*"1   or  about 2.6  Tg   N
yr'1 is  probably  high.  It  is also  significantly  larger than  the estimate
of Georgii and Gravenhorst (1977).   However,  even  this value  is  only  a  small
percentage of  the estimated global  ammonia  emissions given in  Table  2-6.
Thus,  although  the ocean  probably  is a  net  source of  ammonia to the atmos-
phere,  it  would  not  be expected  to  play  a  significant  role  in the global
ammonia cycle.

2.2.2.9  Biogenic Ammonia  Emission  Estimates  for  the  United States—In the
previous discussion  of biogenic NOX emissions,  procedures  based on  atmos-
pheric  concentration  estimates  were used to  estimate biogenic emissions for
the  United States.  Similar  procedures  can be used for estimates of  ammonia
or biogenic emissions.  Applying Galbally's  (1975)  estimate of the natural  or
biogenic  NH3   emission  density of  0.55  g  N nr2  yr"1  to   the contiguous
United  States  (7.82  x 1012 m2)   and to the  area east  of  tne  Mississippi
River  (2.23 x  1012 m2) results in  estimated  biogenic ammonia  emissions  of


                                     2-37

-------
4.3 Tg  N yr-1  and  1.2 Tg  N  yr'1, respectively.   However,  as noted  above,
    concentrations  in  the atmosphere are  now  believed to be  only about  10
percent of  the concentrations used  by Galbally  (1975).   These changes,  of
course, are  the result of  major improvements  in  measurement techniques  in
recent years and  not  of any errors on the part of Galbally  or  other  authors
of previous  studies.    A  proportionate change  in  Galbally1 s estimate  would
result  in  an   indicated  global   emission rate  of  13  Tg  N yr-1  for  the
Northern Hemisphere, and if this  is  assumed to be  essentially a  nonpolar land
area  (0°  to 70°N)  source,  the average  emission  density  is  about 0.14 g  N
m-2 yr~l.    Applying  this  emission  value to  the contiguous United  States
(7.8  x 1012  m2)   and  tne  region  east   of  the  Mississippi River  (2.23  x
1012  m2)  results   in  estimated  annual   NHs  emissions  of  1.1  Tg  N  yr"1
and 0.3 Tg N yr-1, respectively.

This  biogenic  emission  source can  be compared  to  manmade  sources  in  the
United States,  which  are  a summation  of  the  emissions from  livestock  waste
products, fossil  fuel combustion, and  agricultural fertilizer usage  (Harriss
and Michaels  1982).  The  total  emission  for the United  States from  these
three  sources  is  estimated  by Harriss and Michaels (1982) to be 3.4  Tg yr-1
as  NH3 or  3.0 Tg  N yr-1.    Of this  total, 62  percent is from  domestic
livestock, 21 percent from fossil fuel  combustion,  12  percent from  fertilizer
usage,  and  the   remainder  from   various  industrial  sources.     From   a
state-by- state  tabulation   by  Harriss and Michaels  (1982)   of  the  ammonia
emission through the upper Mississippi  Valley  and  the  Ohio  Valley,  the states
of  Iowa,  Illinois, Indiana,  and Ohio are shown  to be  a  region of  maximum
ammonia emissions  density of  about  1  g  N nr2 yr"1.   This  is  about  seven
times  the  biogenic emission  density  of  0.14 g N  m-2 yr-1  estimated  above.
Harriss  and  Michaels  (1982)  concluded  that  emissions  from   natural   or
undisturbed  soil  surfaces  were insignificant compared to  their  summation  of
anthropogenic ammonia sources.

2.2.2.10  Meteorological  and Area Variations  for  NOy and Ammonia  Emissions
--The  natuTFI  emissions of NOX and  ammonia  are  both related   primarily  to
microbiological and physical  processes in the  soil.   These  processes  are
enhanced  by warm weather  and rainfall.    Thus,  warm, moist  summer  weather,
such  as  that found in the  eastern and southern parts  of the United  States,
would  be expected to maximize natural  emissions of both NOX and  ammonia.

On an  area  basis,  soil  pH  tends  to  affect emissions  for both compounds, with
NOX  emanations being  higher  with more  acidic soils.   On  the  other  hand,
ammonia emissions probably tend to increase in alkaline soils.  However, soil
moisture plays  a  role in both situations; thus,  a simple area  distribution
approximation  should  not  be made in which ammonia emissions are assigned  to
alkaline western  areas and NOx  to  tne  more  acidic  midwest  and east.   For
one  thing,  the desert soils of the west may be too dry and  too  hot  for high
ammonia production, as would be inferred from Dawson  (1977).

2.2.2.11   Scavenging  Processes  for N0y  and  Ammonia— The  previous  discus-
sions  have  indicated that  both  dry  and  wet  deposition processes  are impor-
tant  sinks  for  NOx  and  ammonia gases  and their reaction products.   In  their
global  model,  Soderlund  and  Svensson  (1976)  estimate the dry  deposition
processes  as  being about  twice as  important as  precipitation  scavenging


                                     2-38

-------
mechanisms.   This  seems  to be  a  reasonable estimate,  although  significant
variation  in this  ratio  could be  expected  on the  basis  of local  rainfall
frequencies  and characteristics.  In desert areas, dry deposition may be even
more important than usual,  while in  periods  or  regions  of  persistent rain or
showers, the balance could  shift toward precipitation scavenging.

2.2.2.12   Organic Nitrogen Compounds--For a complete nitrogen  cycle through
the atmosphere, the generation,  transfer,  and  deposition of  organic  nitrogen
compounds  should  be considered.   These compounds  may  be either  gaseous or
particulate  materials  and include amines, ami no  acids, and proteins.   Some
investigators have  found  strong  evidence that  the organic  nitrogen compounds
are  gaseous.   Denmead et  al.  (1974), for  example, found  in  samples  over
grazed  pasture  that,  at times, as much as 50  percent of the total  collected
nitrogen  compounds was  not  ammonia;  the  excess  has been  attributed  to
volatile amines.

On   the  basis   of  organic   nitrogen   concentrations  in   precipitation,
Soderlund  and  Svensson  (1976)   postulated  an  annual   deposition  over  land
of  10  to 100  Tg  N yr-l.   This wide  range  is  indicative  of the fact  that
little  is  known  about these  compounds.     Because  at  least some  are  not
participate  compounds  when  emitted  to   the  atmosphere,  a  gaseous  cycle
involving  reactions and  further scavenging  mechanisms may be  present  in
addition to  the fine particle/precipitation scavenging mechanisms.

2.2.2.13   Summary  of  Natural  NOX  and Ammonia Emissions--The  environmental
effect  of  naturalemissions  of  the  nitrogen  compounds,  NOX  and  ammonia,
will be seen primarily as a part of the pattern  of precipitation  chemistry.
The NOX  component,  if  it occurs as  HN03  after  atmospheric reactions,  may
lower precipitation pH, while  ammonia, when absorbed  into  liquid drops  as
NH4+,  will  act  as   a   weak   neutralizing  compound   for   absorbed   acidic
factors.  Because the  natural sources  are  spread  over wide  areas  in  patterns
that change only slowly with distance,  impacts  from natural  sources would  not
change markedly from place to place  in  a given  regional  area.

Although our data  on  natural   sources  of  both  NOX and ammonia  within  the
United States are inadequate, estimates of natural emissions have been made.
These comparisons  indicate  that natural  NOX  emissions in  the contiguous
United States  likely  range between  0.1  and 2.4  Tg N  yr-l.   For  NHa,  the
natural  emissions for the contiguous United  States  is of the order of  1.0  Tg
N yr-1.  For the area  east of  the  Mississippi River,  the  range of natural
NOX emissions  is  between  0.04  and  0.7  Tg  N  yr-1.   In this  same region,
the estimated natural  ammonia  emission  is of  the  order of 0.3 Tg N yr-l.

2.2.3  Chlorine Compounds

2.2.3.1   Introduction—Part of  the acidity  of precipitation is contributed  by
chlorides"It is  hypothesized  by many investigators that hydrochloric acid
important sinks for NOX and ammonia gases  and their reaction  products.    In
their global model, Soderlund  and  Svensson (1976)  estimate the  dry (HC1)
and elemental chlorine  (Cl2) are the precursor compounds.    In  terms of  its
contribution to  precipitation  chemistry,  chloride  is   generally  much less
significant  than  sulfate.   Richardson  and Merva  (1976)  list precipitation


                                     2-39

-------
chloride at about half that of sulfate on an  annual basis  in rural Michigan.
Long-term (1964-74) records  of  precipitation at  Hubbard Brook Experimental
Forest in New Hampshire indicate that, on the average, chloride accounts for
about 13 percent of the total anion content  (Likens et al. 1976).  Although
there are  some pollutant emissions  of Cl~  or  Cl2»  especially as  a result
of fossil  fuel  combustion (see  Section  2.3.4),  a  significant part  of the
total atmospheric  burden  of chlorine  compounds is due  to natural  sources.
Cicerone (1981)  has described  the  atmospheric  chlorine compound  cycle in
detail.

There are three major natural sources of chlorine compounds to consider: the
ocean with  emissions  of  sea salt  (primarily NaCl)  and  organic  chloride as
CH3C1, volcanic  emissions,  and  forest  fires.   The sea  salt processes will
be shown to be dominant.   This was  also  Cadle's  (1980)  conclusion.  Table 2-8
shows the atmospheric  background concentrations  of several  chlorine compounds
as summarized mainly by Cicerone  (1981).

This  discussion  mentions  both  Cl2  and  HC1   as  gaseous atmospheric chlorine
compounds  to  be  considered  because  they  were  considered in  the original
references; however,  as   pointed  out  by  Eriksson  (1959), the  only  stable
gaseous chlorine compounds likely to  be formed  in the atmosphere are hydro-
chloric  acid  and ammonium  chloride,  NfyCl.   Gaseous chlorine,  Cl2»  would
not be expected because of the relatively large concentration of  atmospheric
hydrogen.

2.2.3.2   Oceanic  Sources—The  production  of  sea  salt spray  is  the  largest
source of atmospheric chloride.   Eriksson  (1959)  has estimated the production
of fine salt particles resulting from the evaporation  of sea spray particles
to  be on  the  order of  103 Tg yr"1.   The  chloride  fraction of  103 Tg of
sea salt would be 550 Tg.   Eriksson (1959) made a further  estimate, based on
river chemistry, that about  10  percent of  the ocean-generated  spray particles
are carried over land areas. Thus,  on a global  basis, the ocean is a poten-
tial  source  of about 55  Tg  Cl  yr"1 over land  areas.   This aerosol   will be
deposited on land  areas by  both precipitation and dry deposition  processes.
It was Eriksson's estimate that  dry deposition processes  would be  about  twice
as  important  as  precipitation  over land  areas;  a one-third  to  two-thirds
division  of 55  Tg Cl  yr-1 allots about  18 Tg  Cl   yr"1  to precipitation
deposition processes and  36  Tg  Cl yr-1 to  dry deposition  on a  global basis.

The  deposition of  chloride  over land  areas is  biased toward  the  coastal
zones.   Eriksson  (1960)  gives examples  of patterns  in  Australia,  South
Africa, Europe, and the United States.   In  each of these  areas the gradient
inland from  the  coast is  marked, with chloride.concentrations decreasing by
an  order  of  magnitude  or  more  at  inland  sites  as compared  to  coastal
stations.   U.S.  data  cited by  Eriksson  (I960) were  gathered  by Junge  and
Werby (1958) from an extensive, nationwide rain chemistry  network.  The data
show  a  range  of annual   deposition  rates from  a high  of 32 kg ha"1  yr"1
(3.2  g  Cl  m"2  yr"1)  in   the Pacific  Northwest to  low  values of less  than
0.5  kg  ha"1 yr"1  on the west  and  east slopes  of the  Rocky  Mountains  in
the  area  from about Utah and  New  Mexico to  Nebraska  and eastern  Colorado.
Along the  Gulf Coast,  precipitation chloride  is about  16 kg  ha"1  yr-1.
Eastward  from  the  Pacific  Coast,   chloride  concentrations decrease  rapidly


                                     2-40

-------
              TABLE 2-8.  ATMOSPHERIC BACKGROUND  CONCENTRATIONS
                        OF NATURAL CHLORINE  COMPOUNDS
Compound                        Concentration                   Reference
                                   yg m-3


Inorganic gaseous CT             1.4 -  2.8                  Cicerone  (1981)

Aerosol Cl-                          1-10                    Cicerone  (1981)

                                   -1.2                    Rasmussen et  al
                                                              (1980)
                                  2-41

-------
into the Great Basin.   Along  the Gulf and East Coasts, most of the chloride
in precipitation  falls south and east of  the  Appalachian  Mountains.   In the
northeastern states,  except for immediate  coastal  locations, precipitation
chloride deposition  is  less  than  3  kg  ha-1  yr-1  (0.3  g  Cl   nr2  yr-1).
At Hubbard Brook, Likens et al.  (1976)  report an annual chloride deposition
rate of  0.47 x  10"3 g  a"1,  which  is  about  one-third  of what  would have
been inferred from Junge and Werby's  (1958)  data.

As  a first  approximation, it  would  appear  that  the chloride  content of
precipitation over  the  northeastern  United States  can be explained  by the
rainout and washout of transported sea salt aerosol particles that had  their
origin  in sea spray  generated  at  the  ocean surface.

All  of  the  airborne  chlorine is  not  in the  form of  chloride  particles.
Gaseous chlorine  compounds, either  as Cl2  or  HC1,  are also reported (Junge
1963, Cicerone 1981).    Ryan  and Mukherjee  (1975)  summarize  the  admittedly
scanty   gaseous  chlorine  compound   data  as   indicating   a  global  average
concentration of  about  1 ppb  Cl  in  the  form  of  HC1  and/or  Cl2-   Eriksson
(1959)  considered Cl2 as an unlikely atmospheric  constituent because of its
reactivity.

The  natural  source  of  atmospheric gaseous  chlorine is  frequently  given as
being a  product  of  atmospheric  reactions of  sea  salt particles  with  other
species.    Eriksson  (1960)  proposed  a  reaction  process  involving  the
absorption  of  $03  or  ^$04,  produced  originally in  the  atmosphere  from
SOe, and the release of chlorine from  the  particle.    Eriksson  (1960) also
suggested  that NO could act in  a similar manner to produce gaseous chlorine
from a  sea  salt aerosol.    Robbins  et  al.  (1959) carried  out  laboratory
experiments  on  sea  salt (NaCl)  reactions with NOg.   As  a  result of  these
experiments,  these  authors  proposed  a  reaction  system  involving  the
hydrolysis  of  N02   to   HN03  vapor,  followed  by  HN03  absorption  by dry
NaCl or  into NaCl  solution droplets, followed  by  the  reaction between HN03
and  NaCl leading to the release  of HC1.

A more complex chemical  reaction model  for  HC1  production  in  clouds has been
proposed by Yue et al. (1976).  This model  includes the  initial oxidation of
S02  to   H?S04   and  competing   reactions  with  NHs   for  H2$04   in   a
mechanism  that  produces  HC1  from  the  NaCl-H2S04  reaction.    The   model
proposed by  Yue  et  al.  (1976) includes cloud  parameters  such  as  temperature
and  liquid water content.  In  many respects  it is  a more complete  development
of  the  basic system  proposed by Eriksson  (1960).   Yue  et  al.  (1976)  used
their model  to estimate the annual  global  HC1  production  with more or  less
typical  background  concentrations  and cloud parameters.   The result was an
HC1  production  of about 2 x  102 Tg yr-1.    Duce  (1969)  has estimated the
production of HC1 in the marine  atmosphere to  be about 6 x 102  Tg  yr-1.

In  assessing the possibility of a  sea salt  source  for  chloride, Ryan and
Mukherjee  (1975)  suggest that  about 3 percent of the sea  salt  aerosol may be
converted  to gaseous chlorine compounds.   Using Eriksson's (1959) sea  spray
production  estimate of  103  Tg  yr"1  or 550  Tg  Cl   yr-1,  this  3  percent
estimate gives an estimated gaseous  Cl  production  rate of 17  Tg  yr"1.   This
lower value  compared to the  200 to  600  Tg yr-1,  quoted above  for  gaseous


                                     2-42

-------
chlorine from the work of Yue et  al.  (1976)  and  Duce  (1969)  would  seem  to  be
more reasonable.  Junge  (1963)  found  particulate and  gaseous chloride  to  be
in about equal proportions in marine  air  in  Florida.   Chlorine  production  in
the range  of  200 to 600  Tg  yr"1 would  consume  essentially all  of the sea
salt spray produced, as estimated by Eriksson (1959).   Although  each of  these
estimates of chlorine production may be in error, they can  be used  as a  basis
for a consistent estimate of the atmospheric  transport of chlorine.

Eriksson's  (1959)  estimate  of  sea  salt aerosol   of 1000 Tg yr-1  translates
to  550  Tg  Cl  yr'1  as  aerosol  particles  and 17  Tg  Cl yr-1,  converted  to
gaseous chloride.  For land area  impact,  10  percent of the aerosol, or  55  Tg
Cl yr'1 is  estimated to  be carried  over  the  coast (Eriksson 1959)  while the
gaseous chlorine appears over the land in  proportion  to  the  fraction of land
over the Earth, 29 percent, or  5  Tg Cl  yr-1, assuming that  gaseous  chloride
will have  a significantly longer residence  time in  the atmosphere  than the
sea salt spray aerosol  particles.  The total   ocean contribution to  land area
deposition  is  thus  about  60  Tg  Cl  yr'1  or  0.6 g m~2  yr"1,  averaged  over
the global  land area.

An  additional  natural   source  of  atmospheric gaseous  Cl2  °r  HC1  involves
atmospheric reactions of CHsCl,  which is biogenically produced in  the  ocean
and released to the  atmosphere.   Measurements from aircraft over  the United
States, the north  and  south Pacific,  and in Antarctica (Cronn et  al.  1977,
Rasmussen et al. 1980)  indicate generally  uniform concentrations through the
troposphere.   A concentration  of  about  1.2 yg  nr3  is indicated by  these
measurements as  an  appropriate  average  concentration.    Although  CHsCl  is
not highly  reactive  in the troposphere,  it  does undergo  a  reaction  process
involving oxidation by OH with  the  potential   production  of gaseous  chlorine.
Graedel (1978)  lists an atmospheric  lifetime of about 1.5 years  for C^Cl-
Using this  lifetime estimate with  an average concentration of  1.2  vg m~6
gives a  CH3C1  emissions rate of 2.6  Tg  yr-1 or  1.8 Tg  Cl yr"1.   This  is
much less  than any  of  the  estimates  of  chlorine production  from sea  salt
particles.

Graedel  has carried out an  extensive chemical   and  photochemical  modeling
study of  the  marine atmosphere  (Graedel  1979) during which he  was able  to
estimate the  flux  of various trace  atmospheric  constituents from  the  ocean
into the  atmosphere.  From  this study,  he  estimated  a CHaCl  flux to the
atmosphere  of  1.8  Tg yr-1  and  an  HC1  flux   of  2.0  Tg  yr'1.   His  combined
flux of  gaseous chlorine is  3.2 Tg  Cl  yr-1.   The  generation of CHaCl  is
presumed to be a biogenic process (Rasmussen  et al. 1980) while  the formation
of HC1  can  result from reactions  involving CHsCl or  sea salt,  as  previously
mentioned.

Although  the  chemical  release  of  chlorine  from sea   salt particles can  be
supported  experimentally  (Robbins et al.  1959), theoretically  (Yue et al.
1976),  and by the  decrease  in  Cl/Na  ratios in  precipitation with  distance
from the ocean (Eriksson 1960),  this  oceanic  HC1   generation  mechanism is not
consistently supported  by  field  measurements.   Valach  (1967),  using  a de-
tailed analysis of  the  gaseous  and aerosol  chlorine  data  gathered by  Junge
(1956,  1963) in  Florida,  and  the analysis of the atmospheric chlorine  cycle
by Eriksson (1959, 1960), argued  for  a volcanic  source  for  gaseous  chlorine


                                     2-43

-------
compounds  in  the atmosphere.   Lazrus et  al.  (1970), after  carrying out  a
program  of cloud  water  analyses,  concluded  that  excess  chloride  in  the
atmosphere  does not  originate  from  sea  salt.    They also  concluded  that
volcanic  emissions  could  be  the  source of  gaseous  chlorine  compounds  in
marine  atmospheres   since,  in  their  cloud water   experiments,  they  found
neither    depletion  nor  enhancement  of the  cloud  water  chloride  ratios
compared to seawater mixtures.

2.2.3.3   Volcam'sm—The chlorine compound  emissions to  the atmosphere  from
volcanic activity have been estimated  by  several  authors.   Ryan  and  Mukherjee
(1975), using estimates of particle and lava production coupled  with probable
gas and chlorine ratios, estimated the volcanic  source of  atmospheric gaseous
chlorine  at 0.25 Tg  Cl yr-1.   Lazrus  et  al.  (1979) have  reported on  the
changes in  stratospheric chlorine compound concentrations caused by  a  number
of Western  Hemisphere  volcanos  that were active in  the 1976-78 time period.
Johnston (1980), after an examination  of  data from Alaskan  volcanos,  proposed
ash degassing as a significant  source of atmospheric  chlorine in addition to
the magma  outgassing  processes  considered by other  investigators.   For  St.
Augustine in Alaska, Johnston (1980) estimated a Cl  emission of about 0.5 Tg
during the  January  to  April  1976  eruptions.    About 16  percent of this  Cl
entered  the stratosphere  (Johnston 1980).   Cadle  (1980), in  a  summary  of
information from a variety of sources, has estimated  the annual global  emis-
sion of  HC1 from volcanos at 7.8  Tg  yr-1 with  the  comment that this  value
may still be "somewhat low."   It represents  a tenfold increase in his earlier
estimate  (Cadle  1975).   Measurements  of Cl" particles  and  acidic   vapor  in
the Mt.  St. Helens  plume by  Gandrud and  Lazrus  (1981)  indicate  that  Cl"
concentrations  were  significantly  less than  for  SO^-.   Although  flux
values  were not calculated,  one  may  infer  from  this  and  the   S02  an<*
S042-   data  of Hobbs  et al.  (1982) that Mt. St.  Helens's Cl  contributions
to  the atmosphere would  be less  than  0.15 Tg  yr-1.   The  usual   expected
change  in  atmospheric  chemistry  would  be  an  increase as  more sources  and
longer  periods  of eruptive  activity  are assessed.    Anticipating  this  and
recognizing Cadle's  evaluation  of  his  7.8  Tg  yr-1   figure,  it is  probably
realistic to estimate  volcanic chlorine  emissions to  the atmosphere  at  about
10 Tg  yr-1, with a range  of  at least plus or  minus a factor of 2,  perhaps
more.   Volcanic emissions are estimated  to be deposited uniformly in  oceanic
and land areas, in proportion  to total  area.

2.2.3.4  Combustion—Other possible sources of atmospheric chlorine  are  com-
bustion  processes because of the  production  of CH3C1  in these operations.
Although combustion  is usually considered an anthropogenic  source, it is also
reasonable  to consider  some fraction  as  a  natural  source because a  signifi-
cant fraction of combustion is nonindustrial.   Falling more or less  logically
into this natural source category is fuel wood  combustion,  agricultural  waste
burning, forest  residue combustion,  and  wildfires.    Palmer (1976)  estimated
that in  the United  States, combustion in the "natural" categories  accounted
for a  total  emission  of 0.13  Tg yr-1 of  CHaCl,   the  typically  observed
chlorine combustion effluent.   Wildfires are  about one-third of this  total.
If  it  is  estimated  that these  natural  combustion   sources of  CH3C1 in  the
United States are perhaps 5 percent of the world's total  in these categories
(probably  an  overestimate),  a  potential  emission  of about 2 Tg  Cl yr-1  is
indicated for combustion  sources.   This  is  a minor  global  source  of Cl  and


                                     2-44

-------
does not seem to justify further detailed treatment.  It is assumed that this
source will be mainly a contributor to land area deposition.

2.2.3.5  Total Natural Chlorine Sources—In Table  2-9  these estimates of the
several  proposed naturalchlorine  sources  are  listed  in  terms of  global
totals and  in  terms of the  estimated deposition on land  areas.     As indi-
cated, sea salt aerosols are the source for all but a small percentage of the
atmospheric chlorine,  either directly through  salt deposition or  following
reactions  in  the atmosphere  to  form  gaseous  chlorine compounds.   The land
area  deposition  of  chlorine,  65  Tg Cl  yr-1,  averages to about 0.4 g  Cl
m-2 yr-l  -,-f it  were to be  deposited evenly on  the  total land area.   This
is not an unreasonable value for combined wet and dry depositions  considering
Eriksson's (1960) findings that, away from coastal areas, chloride in precip-
itation is generally 0.5 g m~2 yr-1 or less.

In summary, it seems that  the  recognized sources of atmospheric chlorine are
generally comparable to the identified sinks.

2.2.3.6   Seasonal  Distributions--As shown in  Table 2-9,  chlorides  in  the
atmosphere  are  due  primarily to  sea salt  aerosols  or  chloride compounds
derived  from  sea salt.   The  airborne  sea salt  has  its  origin  in  aerosols
lifted away  from the ocean  surface  after their  formation, either  as wind-
blown  spray or  in  the bubble-bursting  process.    Rain  and clouds  over  the
ocean might be expected  to increase  the local  scavenging   rate and  decrease
the air mass transport of  sea  salt aerosols,  although  there does  not seem to
be any data on this subject.   In the  absence of  storms and strong  winds,  the
aerosol generation  processes may be reduced  but the particle  residence time
might be expected to increase.  From  arguments such as these,  it  is  apparent
that a significant  seasonal cycle  in chloride  transport  and deposition would
not be expected.   Rainfall chemistry  data  gathered  by Johannes  et  al.  (1981)
in the  Adirondack  region  of New York do  not show any  clearly identifiable
seasonal  cycle for  chloride.   In  this  area  of the United States,  a trend
toward a winter minimum for marine  aerosols could  be expected  because of the
increasing exposure to polar continental  air  masses during  this season rather
than the maritime tropical  air masses typical  of much of  the summer.

2.2.3.7    Environmental   Impacts   of Natural  Chlorides—Chloride compounds
transported from  oceanic  areas toland  areas  occur primarily in  very  low
concentrations,  probably in  the  range of fractions of a microgram per cubic
meter for both gases and aerosol  particles  at areas away  from the  coast.  The
chloride ions may contribute  10  to 15 percent or so of  the total  anion con-
tent in precipitation at stations  in the  northeastern United States.   As such
they would  be  relatively  unimportant  in altering  precipitation pH  by them-
selves.

2.2.4  Natural  Sources of  Aerosol  Particles

The atmosphere near the surface over  land  areas  probably has a  concentration
of particulate materials at  all  times except under some very  unique circum-
stances.   Natural sources  produce  materials  that  are  blown up from  exposed
soil  surfaces by wind and  remain suspended in  the  atmosphere for  a period of
time.     These  solid  particles  may  be  removed  from  the   atmosphere   by


                                     2-45

-------
        TABLE 2-9.  ESTIMATED ANNUAL CHLORINE COMPOUND (AS Cl)
                 EMISSIONS AND LAND DEPOSITION - Tg Cl yr'1
Source
Sea salt aerosol
Gaseous Cl from
NaCl particles
Biogenic C^d
Volcanos
Combustion CH3C1
Total
or approximately
Global
emission
550
17
2
10.0
2.0
581
580
Land
Deposition3
55
5
0.5
3.0
2.0
65.5
65
aSee text for details.
                                    2-46

-------
 gravitational  settling,  impaction onto exposed surfaces,  or  they may become
 incorporated  in  cloud  and  precipitation  particles  and  fall  out with  the
 precipitation.  These materials form the natural atmospheric dust loading and
 result  from a  variety of  soil  surfaces  being  exposed  to  wind and  other
 impacts  that cause the particles  to become airborne.   These dust particles
 are caused  by breaking and other natural comminution processes.  As described
 by  Whitby  and  Cantrell  (1976), dust particles of this type  are  classed as
 "coarse  particles"  and would  normally  be in  the 2  to  10 ym  diameter  size
 range.   Although dust storms  and periods of  strong  winds over dry,  exposed
 soil  surfaces  may  produce  periods of  spectacular soil  movement  and excep-
 tional  atmospheric  transport,  in  the  normal  situation  dust  sources  and
 atmospheric dust  concentrations are local source problems.

 In the eastern  part of  the  United States, the  National  Air Sampling  Network
 has  had  a  number of  HIVOL  sampling stations  in  rural  or nonurban locations
 (Spirtas and Levin  1970).  During  the 10-year period from 1957 to 1966 in the
 area  east of the Mississippi,  12  nonurban sampling  stations  were in opera-
 tion.   The average total suspended particle concentration for these stations
 for  this  period was  36 yg nr3.    The range  was  from  a  high  of 57  yg
 nr3  in  Kent Co.,  DE, to a  low of  18  yg m~3  in Coos  Co.,  NH.  This  aver-
 age,  nonurban  particle concentration can  be  used to estimate  the  regional
 emission rate of  this material if we make several  assumptions.  First, we can
 assume that these larger dust particles are uniformly mixed to a depth of 500
 m, or through about the  lower  half of the  mixing   layer.   Since these parti-
 cles  are relatively  large  and we  are  considering an  average concentration
 over both day and night, this  seems  to  be a  reasonable  assumption.  Next, we
 will  assume that these  dust particles have an  average  atmospheric residence
 time of  1  day.   This  seems  reasonable considering the size of the particles
 and  the  effectiveness of  scavenging processes  for  larger-sized  particles.
 Using these values,  an  annual  emission density  results from  the following
 calculation:

               36 vg nr3 x 500 m x 365 = 6.6 g nr2 yr'1.

 Applying  this  annual  emission  density rate  of  6.6   g  m-2  yr-l  to  the
 United States  east  of the  Mississippi  River,  about  2   x  1012 m2, gives  an
 estimated emission  of dust  into the nonurban  atmosphere of about 13  Tg  yr-1
 or 13 x 10° mT yr'1.

 Of the total natural dust loading in the atmosphere,  probably  the most impor-
 tant  constituent for precipitation  chemistry is  its calcium  and magnesium
 content  (Stensland  and Semonin  1982).    These elements make  up about  3.6
 percent and 2.1 percent, respectively, of the  Earth's crust (Weast 1973). If
 the composition of  the dust  aerosol  is  representative of  the  crustal  compo-
 sition as  studies  have  indicated (Lawson and  Winchester 1979), then  the
 background concentration and annual emissions can  be  estimated for Ca  and Mg.
 The  results for  Ca are:    1.3 yg m-3  for  an estimated  average  concentra-
 tion  and  0.5  Tg yr"1 for an estimated  annual  emission.   For Mg the esti-
 mated values are:    0.8  yg nr3  for the  average concentration  and  0.3  Tg
yr'1 for the annual  emission.
                                     2-47

-------
The extension of these estimates of dust  particle  emissions  and  chemistry  to
an estimate of  the  concentrations  of these constituents  in  rainfall in the
region is not within the framework  of this section.  However, it  can  be  noted
that  Hidy  (1982)   has  tabulated  some  summer  particle  concentration and
chemistry data  along with concurrent precipitation  chemistry data  at  three
western Pennsylvania rural stations from  Pierson  et  al.  (1980).    It appears
from  the  analysis  by  Hidy   (1982)  that  both  Ca2+  and  Mg2+  appear  at
greater  ratios  relative  to  sulfate  in   rainwater  than   in  dry   atmospheric
particles.  These are only limited  data  from a short  summer period and should
not  be  considered  definitive.   The  topic  of precipitation  scavenging  is
considered in detail in Chapter A-6.

2.2.5  Precipitation pH in Background  Conditions

The pH of precipitation under conditions  not  affected by  air pollutant  emis-
sions is  an important consideration  for  acidic  deposition  situations.   We
will  examine  briefly some of  the  aspects of  natural  pH variations  in  this
section.  Because  these  pH variations can most reasonably be linked to the
effects of  natural  emissions on precipitation, it is reasonable  to  consider
them as part of the discussion  of natural  emissions.

A completely neutral  precipitation pH would be a  value of 7.0.    However  it
has  long  been  realized that natural  precipitation would likely  be  slightly
acidic because  the precipitation would  tend  to  come  into  equilibrium  with
atmospheric trace  constituents, which when  absorbed into the precipitation
would lower the pH value.  Probably the most  common  assumption has been that
an equilibrium  would  be  set  up with the  C0£  concentration in  the atmosphere
and that  this would produce  a  controlling natural pH value  of 5.6.   Likens
and  Butler  (1981)  and a  great  many  other investigators  have used this C02-
equilibrium pH value of 5.6 as  a criterion  to  separate natural precipitation
pH,  any  value  equal  to  or higher  than  5.6,  and  acidic precipitation, any
value lower  than  5.6.   Since  there had  not  been a considerable amount  of
precipitation pH data  from locations  that could not have  been influenced  by
anthropogenic  pollutant  sources,   this   assumption   of   a  C02-ecluil ibrated
limiting value seemed reasonable.

Two  types of research  investigations have  now been  undertaken  that  raise
considerable doubt about  whether  a  limiting pH  value  of  5.6   is  in  fact
realistic and,  as will  be described below,  there is considerable  evidence
that, at  least  in  some  natural  situations,  the pH  of  precipitation can  be
significantly lower than 5.6.   First,  atmospheric  chemists have begun to look
more  carefully  at  the  factors  in  addition  to C02  that  affect  the pH  of
precipitation (Charlson and Rodhe  1982).   These assessments  show that  there
are  a number of  factors  in the natural  or  background  atmosphere  that can
cause precipitation  pH  to be  lower than 5.6.  Second, under the auspices  of
the  Global  Precipitation Chemistry Project,  a program  of measurements has
been  started  at five remote sites  in the Northern  and  Southern Hemispheres
(Galloway et  al.  1982).    The  findings  of  Charlson  and Rodhe  (1982) and
Galloway et al. (1982) will be described briefly below.

Charlson and Rodhe  (1982)  have  taken  the  chemist's view  of the precipitation
pH  situation  and  have  considered  the  impact of  natural compounds of the


                                  2-48

-------
atmospheric  sulfur  cycle on  pH.   In  the  absence of common  basic  compounds
such  as  NH3 and  CaC03 in the  atmosphere, it  is shown that  pH values  due
to  natural  sulfur compounds  could be expected  to be  about  5.0.   Because the
atmospheric  concentrations  resulting  from  natural   emissions  are  highly
variable,  these  authors conclude that even  in background  situations  the  pH
may  range  from  pH 4.5 to 5.6.   Sulfur compound data  for a variety  of back-
ground  situations have been summarized by Sze and Ko  (1980),  and  they con-
clude  that S04   concentrations  in  very  remote, clean areas  can   be  about
0.05   yg   m  .   This   very-  low  S042~   concentration  with   a  background
S02  value  of  0.26  yg  m"J  and  0.66  g  m"J for  C02  will  result  in  a
cloud  water  pH  value of 5.4 in  a  cloud  of 0.5 g m"3  liquid  water  according
to  Charlson  and  Rodhe (1982).   This  is a  moderate  density for  cloud  liquid
water  content.    Higher  concentrations  of  S042~  would  lead  to   lower  pH
values,  as  would  lower cloud  water  content.    Situations  where  HN03  was
present  in the  atmosphere would  also  reduce  the  pH.    Concentrations  of NH3
or  CaC03  in the  atmosphere  would raise  the  precipitation pH.   Thus,  over
land  areas  where  biogenic  NH3  and  dust  containing   CaC03  could  be  ex-
pected,  a  higher pH  than  5.4 might  be  expected if  the SO^-  were  as  low
as 0.05 yg m-3 and no other acids were present.

Remote area  precipitation  chemistry  data  have  been  reported  by  Galloway  et
al.  (1982)  as  the  initial  results from  the  Global   Precipitation  Chemistry
Project  have become  available.    The stations  in this program  are:   St.
Georges,  Bermuda;  Poker  Flat,  Alaska  (Fairbanks  area);  Amsterdam  Island
(South  Indian  Ocean),  Katherine, Australia (northern part);  and  San Carlos,
Venezuela  (Amazon jungle).  Although  the  results  of  the first year  or so  of
measurements  cannot  be  considered   conclusive  the  results  are  certainly
important factors in the total  acidic deposition picture.

In  summarizing   the  data  from  these  stations  for  the  available  rainfall
events,  the  number  of which  ranged  from 14  for San Carlos  to 67 for  St.
Georges,  Galloway et  al.  (1982)  concluded  that all  stations  experienced
acidic precipitation, on the average,  as a result of  varying  combinations of
strong  ^04 and  HN03,  and   weak,  probably  organic,   acids.    The  higher
acidities  were   primarily  due  to ^$04.    Especially   in  the  case of  St.
Georges, Bermuda, the higher acidic events were shown  to be due to  air mass
transport  from   the  United  States.    These  transports caused  the  average
precipitation pH  at  Bermuda  to  be 4.8.    When trajectories were considered
that apparently had not been influenced by North  America,  the average  pH was
5.0.  At Poker Flat, Alaska, the  average  pH  for  16 precipitation events  was
5.0, but since these  events  included periods when pollutants  from  Fairbanks
or arctic haze pollutants were present, the  "background" pH at this site  is
believed to be  greater than 5.0.

The  precipitation events  at  San  Carlos,  Venezuela;  Amsterdam Island;  and
Katherine, Australia,  were  much  less  likely  to  be  influenced by  pollutant
emissions,  although  Katherine  may   have  been   influenced  by  agricultural
burning  at the  beginning  of  the  rainy  season.    At  San Carlos,  the  14
available precipitation events averaged a  pH  of 4.8, with  a  relatively high
contribution from organic  acids compared  to the other  stations.   At  Amsterdam
Island (37°47IS-77031'E) in the remote Southern Indian  Ocean,  the average  pH
for 26 rainfall  events was 4.9.  Galloway et al.  (1982)  speculated  that some


                                     2-49

-------
pollutant  transport  from the heavy  industrial  areas of  South Africa might
have  influenced  this remote  station also  and  so  they concluded  that  the
natural pH was likely to be greater than 5.0.

As  a  result  of  the detailed  chemical  analyses of  the precipitation event
samples, Galloway et al.  (1982)  were able  to  estimate  the relative  contri-
bution   of   the   three   acids,   H2S04,   HN03,   and   "others"   (probably
organic), to the free acidity.  The  results  for the  three stations with  the
least  probable  influences of pollutants, Amsterdam  Island,  San Carlos,  and
(Catherine, are shown in  Table 2-10.

Although each of the sites in this Global  Precipitation  Chemistry Project  was
remote  in  location,  each had a  different combination of compounds that  de-
termined the  precipitation  chemistry.   Furthermore,  none  was  located in an
area that was apparently similar to eastern  North America  in vegetation, soil
and climate.  Thus,  care should be taken  in applying  these results  to  United
States locations.

In  the United States there  are no  long-term  measurements of background pH
that are directly applicable  to the northeastern area  that  is presently of
concern because  of  frequent low pH  values.   Likens  and Butler (1981), how-
ever,   have  approximated  the  pH patterns  over  much  of the  eastern   United
States  in  1955-56  on  the  basis  of  detailed  precipitation  chemistry data
obtained by C. E. Junge  and  his colleagues  (Junge  and Gustafsm  1956, Junge
1958,  Junge and Werby 1958).  These  calculations of  pH  indicate that most of
the Mississippi  Valley  and the Gulf Coast states had  average pH values  of  5.6
or perhaps higher in the time period 1955-56 (Likens  and Butler 1981).  These
results  are more  alkaline  than   the  background station  data  reported  by
Galloway  et al.  (1982);  the  influence  of  NH3 from soil areas  and CaC03
content in soil  dust could be an explanation.

A different interpretation of  the  Junge (1958)  precipitation  chemistry data
with  regard  to  indications of background pH  was  developed by Stensland  and
Semonin (1982).   They concluded  that the Junge samples  in general   indicated
greater than  normal  pH  because  the sampling period was during a general  low
rainfall  or drought period  and,  as a  result of  this  droug i,   excessive
amounts of  soil  dust containing alkaline salts were  present i,i the precipi-
tation  samples.   By  comparisons  with  more recent  precipitation   analyses,
Stensland  and Semonin   (1982)  developed  dust  correction  factors for   the
1955-56 Junge data and estimated pH  values  after  removing  the  effect due to
anomalously high values  of calcium and  magnesium.   The result,  as might be
expected, was a  set  of  significantly lower  pH values in nonindustrial areas
of the Midwest and Gulf  Coast.   In most of  the  areas where Likens and  Butler
(1981) had estimated the pH to be 5.6 or higher, Stensland  and  Semonin  (1982)
estimated  pH  values  to  range between 4.4 and 5.2.    From  these  results  and
considering the  fact that  some  pollutant emission  impacts were  probably a
factor  in  the 1955-56 Junge  data,  the conclusions of Galloway et al.  (1982)
indicating naturally acidic precipitation with  a pH somewhat greater than  5.0
may also be applicable to the eastern parts  of  the United States.
                                     2-50

-------
         TABLE 2-10.   CONTRIBUTIONS OF ACIDS TO FREE  ACIDITY  (%)
                   (ADAPTED  FROM GALLOWAY  ET AL.  1982)

H2S04
HN03
HXa
Amsterdam
Island
< 73
< 14
> 13
Katherine,
Australia
< 33
< 26
> 41
San Carlos,
Venezuela
< 18
< 17
> 65
aHX could be HC1, organic acids, or HaP04;  Galloway et al.  (1982)
 believe it was an organic acid.
                                     2-51

-------
2.2.6  Summary

This discussion of natural emission  sources  has  examined  a  number of factors
related to  precipitation  pH  with reference to the  situation  in  northeastern
United  States  and southeastern  Canada.   In most  cases  it was  necessary  to
draw analogies between  global  conditions  and the situation in the  northeast
region,  so considerable  discussion  was  centered  on  global  background  air
chemistry.   With  specific  regard  to  precipitation  pH,  it  was  shown  by
theoretical chemistry and measurements in remote locations  of the world  that
a  pH value  of  near 5.0 may  occur  as a result of  the acidic compounds  that
occur naturally in the atmosphere.

In  the  eastern  part of the  United States,  it was  shown  that natural  sulfur
compounds emissions  are  relatively minor contributors to the total mass  of
sulfur  emissions  in  the area.   This is shown by  a comparison  of  emissions
from the  United States  east of  the Mississippi  River,  where  the  natural
sources were estimated  to  total  about 0.07 Tg S yr-1  and 1978 anthroppgenic
sources totaled about  11  Tg  S yr"1  (see   Figure 2-6).   For the  contiguous
United  States,  a  total  natural  source emissions  rate   of about  0.5  Tg  S
yr"1 can  be  compared  with  a total  1978  anthropogenic  emissions  rate  of
about 13  Tg S yr-1  (see  Figure  2-4).  Thus,  even considering  the  numerous
probable  errors  that can  be associated  with natural  emissions  estimates,
natural   sulfur  emissions do  not  appear  to be  as  significant  as  pollutant
emissions in establishing the regional  atmospheric  sulfur  cycle.

For nitrogen  compounds, both  acidic NOX  emissions and  basic  NHa  emission
sources must be  considered.   In precipitation  pH,  acidic  NOX compounds may
play an important  role.   In  this  discussion  the  emissions  of NOX  compounds
from natural sources  in the area  east  of the Mississippi  were  estimated  to
range between 0.04  and 0.13  Tg  N yr-1.   This  value  is  significantly  less
than the  estimated 1978 anthropogenic  emissions of 8.9  Tg N yr-1  for  this
same area.   Natural  biogenic emissions of  NH3,  which lead to  NH4+ ions  in
precipitation,  have  been  estimated to be about  1.1 Tg N  yr-1 for  the whole
United    States.      Anthropogenic   sources   of   NH3   include   significant
contributions from  domestic  animal  waste and other sources and have  been
estimated to be about 3 Tg N  yr"1 over the contiguous United Sates.

Chlorides may contribute to precipitation pH, although present evidence  from
areas such as Hubbard Brook,  New Hampshire,  indicates that their  contribution
is perhaps only 10 to 15 percent of the total anion content.  The  source  for
naturally generated Cl" is almost  exclusively sea  salt swept from  the ocean
by  marine  air  masses.    Deposition of  Cl"  on   land  areas  east  of   the
Mississippi  is  estimated  at  about  0.4  g  Cl  nr2  yr"1.    Air  pollutant
sources of  Cl"  are believed to  be relatively small and  are primarily  from
the combustion of fossil fuel  containing trace amounts  of  chlorine.

Fugitive  dust  may  contribute  to  precipitation  pH by contributing  soluble
ions.   For  the most  part these are expected  to be  calcium  and magnesium  and
they would  be expected  to  raise  pH  values.  Estimates  of background  dust
loading in  the northeastern  region of  the United States  show relatively  low
mass loadings  and thus  atmospheric  contributions  of  calcium and  magnesium
would be relatively low.
                                   2-52

-------
2.3  ANTHROPOGENIC EMISSIONS (J. B.  Homolya)

2.3.1  Origins of Anthropogenically  Emitted  Compounds  and  Related  Issues

Large quantities of  sulfur  and  nitrogen oxides are discharged annually  into
the  atmosphere  from  the combustion of  fossil  fuels such as  coal,  oil,  and
gas.   Through chemical  reaction in the atmosphere, these pollutants can be
transformed into  acids,  which may  return  to ground  level  as components of
either rain or snow.   The deposition of these acids by precipitation  has  been
associated  with terrestrial, aquatic,  and  materials  effects (see  Chapters
E-3, E-5, and E-7).

In  addition  to  $03  and  NO,  other   fossil  fuel  combustion  products  are
emitted  that  may  influence  acid  precipitation  formation.    These   include
H2S04, HCl,  and  particulate  matter.    Sulfuric acid  represents  a  variable
fraction (0.01 to  about 0.05) of the $03 emissions and exists as a vapor in
combustion emissions.   Upon mixing and cooling in the atmosphere, the  acid
condenses as  fine  particles.   Field measurements have shown  that a larger
fraction  of S02  is  emitted  as ^$04  from  oil  combustion  than from  coal
burning.    Hydrochloric  acid  emissions   have   been   identified  with   coal
combustion.   Little  information is  available on the rate of  fossil  fuel  HCl
emissions  to  the  atmosphere.  Figure  2-4  illustrates  trends  in   total
anthropogenic  emissions  of  particulate  matter,  SOgj  and  NOX   for   the
United States  from 1940  to 1978.   Sulfur  dioxide emissions  were  about 29
percent higher in  1978 than  in  1940.   Although  the generation of  electricity
has  increased  many-fold,  a  switch   in   fuel   from   coal   to oil   in   the
northeastern  United  States  during  the late  1960's   and early  1970's  has
lowered both  total  S02  and  particulate matter  emissions.  As  noted by  the
marked  reduction   in  particulates  to  about  31 percent  of  the  1940  total
emissions, both fuel  switching  and  incorporating  electrostatic  precipitators
onto  coal-fired units have  dramatically  changed  the pollutant  atmospheric
composition.   The NOX component  has  increased mainly because of increases
in electric power generation and vehicular  traffic.

Reporting  emissions  on  a  nationwide  basis, although  useful as  a   general
indicator of pollutant levels,  has definite  limitations.  National totals or
averages  are  not the   best guide  for  estimating  trends  for   particular
localities.   They  are only  an  indication  of the  extent  of total installed
control technologies and economic growth or  decline.   They are not useful as
an indicator of air quality.  With the concern  for  the  increasing acidity of
precipitation over the eastern United States, it is important  to evaluate the
effects of changing emissions characteristics on the  historical trends  noted
for  the  geographical  distribution of  acidity.    Issues  of  prime  importance
that must be addressed in such an assessment include:

     (1)  Historical   changes of  emissions  with  variations  in   fuel   use
          patterns.  What changes are projected  in future  years?

     (2)  Current  emissions  for  SOX   and  NOX  from  stationary  and   mobile
          source  categories as  a  function  of  geographical  region, urban
          compared to rural, and height of emissions.
                                  2-53

-------
  CQ
   C
   ro
    i
>• —I
CL O
CL 
-S
o
                                     TOTAL EMISSIONS  (ml x  106  yr'1)
   o
   3
00
   o
   -h
   -$
^ * c~t*
 i_i — '•
 UD Cl
 -^1 C.
 00 — •
*— . " Q)
   r-t-
   fD
   oo
   o
  ro
   O
   -S
   0>

   cr
   3
   (D
   Q.

   CO
   r+
   O)
   r+
   fD
   -h

   O
   O

   c+
   O
   CO

-------
     (3)  Current emissions of primary sulfate  and  HC1.   How  significant  are
          these primary emissions by  geographical  region and season of  the
          year?

     (4)  Primary acid emissions in terms of short-range  impact downwind of
          individual  large emission  sources  or  clusters  of sources.

     (5)  Emission  sources of  neutralizing substances  including  NH3   and
          alkaline particles  from combustion sources.   How do such  sources
          vary geographically and by season  of  the  year?   How significant is
          atmospheric neutralization by fly  ash materials?

Examining  these  issues  requires  a  degree   of  geographic   resolution   in
emissions  trends  beyond  that given  in  Figure 2-4.    It is  difficult  to
perceive  the  possibilities  of the  roles of  primary acidic  emissions  and
regional  changes  in emission  levels  on  measured  changes  in precipitation
acidity  without  further   subclassifying   historical  emissions  estimates.
However,  subclassification  to the  single-source level,  if  not  impossible,
would  seem  inappropriate  relative  to the  degree  of spatial  resolution  to
which changes in acidity are noted and discussed.  Therefore,  an  attempt has
been made  to  examine estimates of anthropogenic emissions specifically  from
the eastern United States over the past 30 years and to  present a discussion
of the  trends  of  both  emission quantities and   characteristics in degrees  of
spatial and temporal resolution  that translate to  correlation with  observed
acidity patterns over the same period.

The work  of  Gschwandtner et al.  (1981) was used as  the basis for  examining
historical trends  in the  emissions  of acids,   acid  precursors,   and certain
heavy metals between 1950  and  1978.   Gschwandtner  was able to compile a  data
file of estimates of historical  emissions of  oxides  of sulfur and  nitrogen
for  the eastern  United States.   The  estimates were  calculated from  fuel
consumption data available for each  state, emissions  factors,  and  in  the  case
of sulfur  oxides,  sulfur content of the  fuel.   So that these data  could  be
used for a detailed analysis of emissions  trends, the  files were assembled  in
a  microcomputer and operated  with  additional   emissions factors  for sulfur
dioxide,   nitrogen   oxides,   primary   sulfate  (H2S04),   chloride  (HC1),
volatile metals (As, Hg),  and certain key metals  indigenous  to residual oil
combustion (V, Ni).   There have been no estimates of the  uncertainty of  this
data set.  However,  it is reasonable to assume  that the  earlier records prior
to development of the National Emissions Data System within the EPA  are  less
accurate than those compiled from about 1970.

The calculated  annual  emissions were  then  normalized with  respect to  land
area of  each  state  and reported as  annual  emissions  densities  (kg km~2).
This procedure was chosen to provide a perspective  of  the regional-scale  flux
in emissions to the atmosphere.  Obviously,  one cannot compare fluxes between
states  whose  land  areas  are  quite  different  (e.g., Texas  and  Delaware).
However, emissions density  calculations are useful  to the study  of  relative
contributions of  a  state within a  region (e.g., Indiana in  the  Midwest and
Massachusetts  in  New  England).   Calculations  were  performed on  all   data
between  1950   and  1975  in  5-year   increments   and  for  1978.    The source
categories for sulfur and nitrogen oxides  emissions are  listed in  Table 2-11.


                                  2-55

-------
      TABLE 2-11.  MAJOR SOURCE CATEGORIES AND  SUBCATEGORIES FOR
             EMISSIONS INVENTORY (GSCHWANTDNER  ET AL.  1981)
Electric Utilities

Industrial Sources of Fuel  Combustion

Commercial/Residential Sources of Fuel  Combustion

Pipelines

Highway Vehicles:

     Gasoline Powered
     Diesel Powered

Miscellaneous Sources:

     Railroads
     Vessels
     Miscellaneous Off-Highway Mobile Sources
     Chemical Manufacturers
     Primary Metal Fabricators
     Mineral Products Manufacturers
     Petroleum Refineries
     Other Sources
                                  2-56

-------
A  map  of  the  study  area  for emissions  estimates is  shown  in Figure  2-5.
Since  emissions  estimates  are based  upon fuel  composition  and consumption
data, their validity depends on the detail  with which  fuel  usage records  have
been maintained over the past 30  years.

In  each  state,  Gschwandtner  et  al.  (1981)  compiled  information  on  fuel
consumption by stationary sources  over the years from 1950 to  1978  in  5-year
intervals.   However, data  on statewide  consumption  of  bituminous coal  by
industries and commercial/residential  sources were  not available for 1950.

2.3.2  Historical Trends and Current Emissions of Sulfur Compounds

2.3.2.1  Sulfur  Oxides—Historical trends of  total  sulfur  oxide emissions  by
source category are shown in Figure 2-6.   In  recent years,  electric  utilities
appear  to have  contributed to  more  than  half of the  total  sulfur oxide
emissions.  Sulfur  oxide contributions  from  industrial  sources  increased  up
to  1965  and then  significantly  decreased.   The  marked  increase   in  sulfur
oxide  emissions  from  the  commercial/residential and  industrial  sectors
between  1950  and 1955  may be  somewhat  misleading because  bituminous  coal
combustion data  were  not available for the 1950 input.   During the 1950's,
there  was  a  marked shift  in  residential  fuel from coal  to  oil and natural
gas.  After 1965, industrial sources switched  from coal and high-sulfur  oils
to  natural gas  and  low-sulfur  oils.   Fuel  switches  within  these  source
categories have resulted in their decreasing  contribution  to the total  sulfur
oxides emissions.

Since electric utilities are estimated to contribute  an increasingly greater
proportion of sulfur oxides to the atmosphere, then regions of  rapid utility
power  generation  growth  should have experienced a  proportionate increase  in
sulfur oxide emissions.   Table 2-12 presents a ranking of  the 10 states  that
exhibited  the largest increases  in sulfur oxides emissions densities between
1950 and  1978.   Also given  are  the contributions  (percent)  of utility and
industrial fuel combustion sources to the total  sulfur  oxides emitted  within
each state.  The numerical  ranking indicates  that both Tennessee and Kentucky
have  exhibited  order of  magnitude  increases  in sulfur  oxides   emissions
densities  over  the past 28 years.   In  general,  the  largest  increases  in
emissions  density have been  estimated  for the area bound by 80°W 30°N,  80°W
42°N and 90°W 30°N,  90°W 42°N.  Wisconsin  is  the  only  state that does not lie
within  these bounds.   Within  the region  in  1978,  sulfur oxides emitted  by
electric  utilities  and  industrial fuel  combustion  sources   dominated  the
anthropogenic burden to  the atmosphere.

Along with the  increases in sulfur oxides emissions densities,  the  areas  of
the eastern United States exhibiting the highest  emissions  densities would  be
expected to influence strongly the sulfuric acid component of  precipitation,
whether  through  long-range  transport  and/or  transformation  or by primary
emissions.   Table 2-13  lists annual   sulfur oxides  emissions  densities  by
state  for each  decade   from  1950  through  1970  along  with  1978   and,  in
parentheses,  the numerical  rankings  of  the  10  highest emissions  densities
excluding the District of Columbia. The areas of highest  emissions  densities
have shifted  from the North  Atlantic Coastal  region  in   the  1950's  to the
Midwest in the  1970's.   Connecticut,  New  York,  and Rhode  Island  have  been
                                 2-57

-------
Figure 2-5.   Map showing the study area included for emissions density
             calculations.   Adapted from Gschwandtner et al.  (1981).
                                    2-58

-------
                       LEGEND
                       MISCELLANEOUS
                       HIGHWAY  VEHICLES
                       COMMERCIAL/RESIDENTIAL
                       INDUSTRIAL
                       ELECTRIC UTILITIES
        1950
1955
1960
1965
YEAR
1970
1975   1978
Figure 2-6.   Historical  trends of sulfur oxide emissions by source
             category for the study area.   Adapted from Gschwandtner
             et al.  (1981).
                                   2-59

-------
      TABLE 2-12.  TEN LARGEST INCREASES IN SULFUR OXIDES  EMISSION
                     DENSITIES BETWEEN 1950 and 1978
                                                     SOx
                                           Percentage of total  sulfur
                                             oxides attributable to
                                      electric utilities and industrial
                                             fuel  combustion sources

     State               Increase           1950             1978
                           (*)
Tennessee                  1096              90               93
Kentucky                   1076              76               96
South Carolina              558              61               88
Georgia                     489              48               89
Mississippi                 483              21               83
Alabama                     477              34               84
West Virginia               331              80               96
Ohio                        248              80               93
Indiana                     247              83               93
Wisconsin                   206              67               87
                                  2-60

-------
        TABLE  2-13.
ANNUAL EMISSIONS DENSITIES OF SULFUR OXIDES
      (kg km'2 yr-1)
                         1950
                        1960    1970
1978
A1 abama
Arkansas
Connecticut
Delaware
District of Columbia
Florida
Georgia
Illinois
Indiana
Iowa
Kentucky
Louisiana
Maine
Maryl and
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
New Hampshire
New Jersey
New York
North Carolina
Ohio
Pennsylvania
Rhode Island
South Carolina
Tennessee
Texas
Vermont
Virginia
West Virginia
Wisconsin
944
276
9834 (6)
17960 (4)
169070
1344
714
5403 (10)
5148
1081
981
1680
398
13220 (5)
38436 (2)
3114
1689
344
3596
2615
58539 (1)
6011 (9)
2043
7609 (7)
7500 (8)
19486 (3)
502
807
1199
149
1353
3532
1353
4168
173
16907 (7)
33414 (1)
200904
2043
1180
15245 (9)
17797 (6)
2270
5475
1589
577
13502 (10)
15935 (8)
6538
1634
301
2934
1099
22273 (4)
10088
1553
24943 (3)
18269 (5)
25288 (2)
1308
6065
1180
313
1471
7682
3768
6646
262
22201 (5)
38063 (1)
407038
5757
2443
15672 (9)
18750 (6)
2306
11114
2297
865
15500 (10)
24425 (4)
9153
1880
586
5566
3614
26396 (3)
10288
3550
26568 (2)
17906 (7)
17352 (8)
2088
8199
1516
470
4076
14192
2016
5446
829
7836
32061 (1)
91344
4104
4204
10860 (10)
17851 (3)
2397
11541 (9)
2597
696
11840 (8)
17070 (4)
6728,
1580
2007
6574
2551
14483 (7)
7427
3750
26486 (2)
14691 (6)
6174
3305
9652
1671
317
3087
15209 (5)
4140
Note:   Numbers in parentheses  indicate  numerical ranking of 10 highest
       emissions densities  (D.C.  excluded).
                                 2-61

-------
displaced from the  ranking  by  Indiana,  Kentucky, and West Virginia.   During
1950, the 10 ranked states emitted a total  of  5.9  x  109 kg of sulfur  oxides
compared  with  1.11 x  1010  kg of  sulfur oxides  for the  ranked  states  in
1978, an increase of 88 percent.   Although Delaware remains a  region  of dense
SOX  emissions  because  of  its  large chemical  complexes, notable  reductions
have occurred  in  Connecticut,  Rhode Island,  Maryland,  and New  Jersey as  a
result  of  changes  in   fuel   type  and  fuel  sulfur  content.     If  the
transformation of S02   in  the atmosphere  results in the deposition of  acidic
sulfur  compounds,  then  the  increase  in  midwestern S02  emissions  should
result in an enlarged  geographical  domain in which  acidity  is  measured.

Table 2-14  presents the  estimates of the  annual  emissions of sulfur  oxides
for  each  of  the  31 states  for  the period  from 1950  through  1980.   Total
emissions  from  this   region  declined  slightly after  1970.    The   largest
quantities  of  emissions  can be attributed  to  the  midwestern United  States.
Significant increases  in emissions have occurred in the  southern part  of  the
country,  notably  in Kentucky,  Tennessee, Mississippi, Alabama, Georgia,  and
Florida.   Emissions of sulfur dioxide  in the  Northeast show a substantial
reduction after 1970.

With  establishment  of  sulfur  dioxide  and  particulate   matter  emissions
standards,  most   sources   in   the  northeastern  United  States  found   it
advantageous to  switch  to  fuels  of lower sulfur content rather  than  install
S02  scrubbers, which  were relatively  unproven at  the time.    Also, many
coal-fired  sources  were   design-limited   with respect  to  the  potential
installations  of  high  efficiency  particulate   removal  devices   such   as
electrostatic  precipitators.   Cost  considerations also precluded upgrading
sources that were approaching their design operating  lifetime. Therefore,  as
a  means  of  complying  with   both  sulfur  dioxide  and particulate  matter
emissions  standards,  many  source  operators  switched from burning  coal   to
burning residual  oils, which were lower  in sulfur content,  produced  little
ash, eliminated the need for electrostatic precipitators, and  were  economical
and  readily available  along the East Coast.

2.3.2.2   Primary  Sulfate Emissions—Results over the  past  7 years  have shown
that primary sulfate emissions from  oil  combustion are  5 to  10 times  higher
than those  from  coal  of a similar sulfur  content  (Homolya  and Cheney  1978).
Primary  sulfate   is  that  emitted  as sulfate.   Secondary sulfate  is that
produced  by   atmospheric  reactions  involving  other  chemical  substances.
Sulfuric  acid has  been  identified  as  the major  constituent of  the  total
water-soluble  sulfate  emissions from both  oil and coal  firing  (Cheney  and
Homolya 1978).  Ambient air measurements taken in the  vicinity of an  isolated
oil-fired power plant have demonstrated a correlation  between  primary sulfate
emissions and  an  increase  of up  to twofold in ambient sulfate levels  - 6  km
downwind  from  the source (Boldt et al. 1980).

Shannon  (1979) and Shannon et  al.  (1980),  using  the  Advanced Statistical
Trajectory  Regional   Air  Pollution  model   (ASTRAP),   have  studied   the
relationship  between  primary  and  secondary sulfate  at the  regional  scale.
Using the emissions inventory  compiled  as part of the SURE study  (Klemm  and
Brennan 1979), the model  simulations showed that  primary  sulfate  has  a less
uniform   distribution   than  does  secondary   sulfate,  but   that  in   the


                                  2-62

-------
TABLE 2-14.  ESTIMATES OF ANNUAL  EMISSIONS  OF SULFUR OXIDES
                   (106 kg yr-1)

Alabama
Arkansas
Connecticut
Del aware
District of Columbia
Florida
Georgia
Illinois
Indiana
Iowa
Kentucky
Louisiana
Maine
Maryl and
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
New Hampshire
New Jersey
New York
North Carolina
Ohio
Pennsylvania
Rhode Island
South Carolina
Tennessee
Texas
Vermont
Virginia
West Virginia
Wisconsin

1950
126.2
38.0
127.6
95.7
29.4
203.9
108.9
789.5
484.0
157.6
102.7
211.2
34.3
362.3
822.2
469.7
367.9
42.5
649.2
63.0
1188.3
772.0
278.3
812.6
880.8
61.3
40.4
88.3
830.4
3.7
143.1
321.3
196.8
10784.1
1960
557.3
23.8
219.4
178.1
35.0
309.9
180.0
2227.5
1673.2
331.0
573.0
199.8
49.7
370.0
340.9
986.1
355.9
37.2
529.7
26.5
452.1
1295.7
211.6
2663.7
2145.4
79.5
105.3
663.8
817.3
7.8
155.6
481.2
548.2
18852.6
1970
888.6
36.0
288.1
202.8
70.9
873.4
372.6
2290.0
1762.8
336.3
1163.1
288.8
74.4
424.7
522.5
1380.5
409.5
72.4
1004.9
87.1
535.8
1021.3
483.6
2837.3
2102.8
54.6
168.0
897.4
1050.0
11.7
431.0
889.1
293.3
23492.1
1978
728.2
114.1
101.7
170.9
15.9
622.6
641.2
1586.8
1678.3
349.6
1207.8
326.5
59.9
324.4
365.1
1014.7
344.1
248.1
1186.8
61.5
294.0
953.9
510.9
2828.5
1725.2
19.4
265.3
1056.4
1157.3
7.9
326.4
952.8
602.3
21741.6
1980
821.2
92.1
65.2
99.2
13.4
993.3
761.7
1334.1
1821.5
298.2
1016.7
276.0
86.0
306.6
312.5
822.7
236.2
250.5
1180.4
84.3
253.3
856.7
546.4
2401.1
1834.5
13.8
295.8
976.6
1158.2
6.2
327.5
986.8
566.7
20960.4
                          2-63

-------
acid-sensitive areas  of the northeastern United  States  and eastern  Canada,
primary  sulfate  concentrations are  25  to  50  percent of  secondary  sulfate
during the winter.                                7

To estimate  long-term trends  in  primary  sulfate emissions characteristics,
historical sulfur  oxides  emissions  estimates summarized in Figure  2-6 were
adjusted  by  appropriate primary  sulfate  emissions factors for each  source
category  and fuel  type,   to yield  a  mass emission  of  sulfate  for each
category.  The aggregate mass  emissions for each state were then normalized
with  respect to  state area  and  reported  as  a  primary  sulfate  emissions
density.  Table 2-15  lists the sulfate  emissions  factors used  as multipliers
of the sulfur oxide emissions.  The  factors  are comparable  with those  used by
Shannon et al. (1980)  in ASTRAP simulations  with  the exception of the mobile
and miscellaneous  source  categories.   A conservative  emissions factor of 3
percent was assumed for the mobile source category and a factor of  5  percent
was assumed  to represent  an  average  of the miscellaneous  source categories,
which consist of fossil fuel  combustion, petroleum refining, and chemical and
mineral products manufacturing.

The annual sulfate emissions densities for each state are  presented in Table
2-16  along with  the  ranking of the  10  highest emissions densities for each
period.  The data  indicate that the Northeast has been historically the area
of  highest  primary  sulfate  emissions density   within  the  eastern   United
States.   The  estimates  demonstrate that   primary  sulfate  emissions have
decreased in  the  northeastern  United States, except  for Delaware,  over the
past  28  years,   along  with  the  corresponding   decrease  in   sulfur   oxides
emissions densities given  in Table 2-13.  However, the  Northeast continues to
exhibit the highest primary sulfate  emissions density.

Table 2-17 presents estimates of annual emissions of primary sulfate  for the
31-state  region between 1950 and  1980.   Total  emissions in this  region have
declined  since  1970  in  a  trend  similar  to the  decline   in  S02  emissions
given  in Table 2-14.  However,  the states estimated to  emit the  highest
amounts of primary sulfates are not  the  same states estimated  to be  the major
sources  of SO? emissions.   For example, Pennsylvania, New  York,  and  Florida
are  estimated to  be  the  top  three  states with  highest primary  sulfate
emissions  in 1980.    By  comparison,  Ohio, Pennsylvania,  and  Indiana  are
estimated to be the top three  states with highest  sulfur oxide emissions for
the  same period.   These  differences in  rankings can be  attributed   to the
differences in the types of fuels being burned.   Midwestern states  burn coal
predominantly whereas northeastern  states  consume significant  quantities of
residual  fuel  oils.   The higher primary   sulfate  emission factor  for oil
compared  to  coal   accounts  for the  disproportionate  quantities of  sulfates
estimated to  be emitted from those  states  that burn  the largest volumes of
residual  fuel oils for utility, industrial,  commercial,  and residential use.


The  influence  of  primary   sulfate  emissions  on  acidic   precipitation  is
unclear.   During  the winter season  when  photochemical  activity is minimal,
primary   acid  emissions  should   exert  the greatest  contribution  through
long-range transport  to the northeastern United States  and/or  local  low-level
                                  2-64

-------
TABLE 2-15.  SULFATE EMISSIONS FACTORS FOR SOURCE  CATEGORIES
               AND FUELS (SHANNON ET AL.  1980)
    Source category                Sulfate emissions  factor
 Coal point sources                           1.5

 Residual oil—utility and                    7.0
   industrial

 Residual oil--commercial and                13.4
   residential

 Distillate oil                               3.0

 Mobile sources                               3.0

 Miscellaneous                                5.0
                          2-65

-------
TABLE 2-16.
                      ANNUAL EMISSIONS DENSITIES OF PRIMARY SULFATE
                             (kg km~2 yr~*)
                         1950
                               1960
1970
1978
A1 abama
Arkansas
Connecticut
Del aware
District of Columbia
Florida
Georgia
Illinois
Indiana
Iowa
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
New Hampshire
New Jersey
New York
North Carolina
Ohio
Pennsylvania
Rhode Island
South Carolina
Tennessee
Texas
Vermont
Virginia
West Virginia
Wisconsin
39
15
595
952
6419
80
32
164
154
28
28
94
25
982
2924
115
45
17
158
163
3260
292
50
210
302
1589
28
35
65
8
77
83
38


(6)
(5)



(9)





(4)
(2)




(10)
(1)
(8)


(7)
(3)







89
8
649
1371
10079
111
39
332
333
41
91
79
35
428
962
158
67
15
69
56
1008
380
44
451
431
1090
44
118
57
14
53
143
82


(5)
(1)



(10)
(9)




(8)
(4)





(3)


(6)
(7)
(2)







130
14
1307
1544
32608
230
73
354
344
47
179
118
62
551
1952
193
82
24
133
141
1507
555
84
470
482
1535
62
156
75
23
190
319
51


(5)
(2)



(10)





(7)
(1)





(4)
(6)

(9)
(8)
(3)







116
45
590
1584
6366
214
104
255
367
50
193
151
50
494
1298
168
76
106
133
110
878
459
98
485
387
494
111
182
72
23
151
261
83


(4)
(1)




(10)




(5)
(2)





(3)
(8)

(7)
(9)
(6)







Note:   Numbers in parentheses  indicate  numerical ranking of 10 highest
       emissions densities (D.C.  excluded).
                                  2-66

-------
TABLE 2-17.   ESTIMATES OF ANNUAL EMISSIONS OF PRIMARY SULFATE
                      (106 kg yr'1)

Al abama
Arkansas
Connecticut
Del aware
District of Columbia
Florida
Georgia
Illinois
Indiana
Iowa
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
New Hampshire
New Jersey
New York
North Carolina
Ohio
Pennsylvania
Rhode Island
South Carolina
Tennessee
Texas
Vermont
Virginia
West Virginia
Wisconsin
1950
5.2
2.1
7.7
5.1
1.1
12.1
4.9
24.0
14.5
4.1
2.9
11.8
2.2
26.9
62.6
17.3
9.8
2.1
28.5
4.0
66.2
37.5
6.8
22.4
35.5
5.0
2.3
3.8
45.0
0.2
8.1
5.2
5.5
1960
11.9
1.1
8.4
7.3
1.8
16.8
4.0
48.5
31.3
6.0
9.5
9.9
3.0
11.7
20.6
23.8
14.6
1.9
12.5
1.4
20.5
48.8
6.0
48.2
50.6
3.4
3.5
12.9
39.5
0.4
5.6
9.0
11.9
1970
17.4
1.9
17.0
8.2
5.7
34.9
11.1
51.7
32.3
6.9
18.8
14.8
5.3
15.1
41.8
29.1
17.9
3.0
24.0
3.5
30.6
71.3
11.4
50.2
56.6
4.8
5.0
17.1
51.9
0.6
20.1
20.0
7.4
1978
15.5
6.2
7.7
8.4
1.1
32.5
15.9
37.3
34.6
7.3
20.2
19.0
4.3
13.5
27.8
25.3
16.6
13.1
24.0
2.7
17.8
59.0
13.4
51.8
45.5
1.6
8.9
19.9
49.9
0.6
16.0
16.4
12.1
1980
21.1
4.1
4.7
3.4
1.0
38.0
15.0
23.4
31.8
4.8
13.6
18.3
9.0
9.2
18.6
17.3
5.7
9.9
15.8
4.1
10.7
39.9
13.7
36.5
41.5
1.0
7.9
14.5
31.3
0.5
8.3
14.5
10.6
                   491.5      506.4     704.6      643.0    496.1
                            2-67

-------
emissions sources.   Similarly,  the  Tow-level  emissions source influence may
be exacerbated by space-heating  needs  during winter months.

The differences in the release  height of  point source emissions will affect
the  relative  local  deposition   of  emissions  compared to  those  which may
be  carried  aloft  to  undergo  a variety  of  transport  and  transformation
processes for  extended periods   in  the atmosphere.   As  a comparison,  Table
2-18 was constructed to illustrate  the regional  differences  in the  quantities
of sulfur oxides emitted as a function of  stack  height.   Emissions and  stack
data were taken from the  EPA 1980 National  Emissions  Data  System  (NEDS)  files
for Ohio, Pennsylvania, Florida, and New Jersey.   The number of point sources
and their cumulative emissions of sulfur oxides  were  aggregated  according to
four increments of stack  heights.  The aggregated data indicate that  for Ohio
and Pennsylvania, the bulk of the  sulfur  oxides  emissions in  each state are
emitted  at  stack  heights of from 152 to  305 m.   Emissions in  this  release
height increment represent in excess of 60  percent of the  total sulfur oxides
emitted  and  serve   as  the  basis   of the  hypothesis  involving  long-range
transport/transformation  of sulfur  oxides  with  deposition  in the  northeastern
United States.

Of  the  four states  compared  in Table 2-18, neither Florida  nor  New Jersey
emitted  sulfur oxides at release heights above 152 m during 1980.   In  fact,
60 percent  of  the  point  source  emissions  of sulfur  oxides in New  Jersey are
estimated to  be emitted at  heights  between 31  and  76 m.    In  Florida, 55
percent  of  the  sulfur oxide emissions  from  point  sources  are emitted at
heights between 77  and 151 m.  Therefore,  the deposition  of both primary and
secondary sulfates  and/or acidic  materials  from  point  source  emissions in
these states  may occur  at  shorter downwind distances  than from  midwestern
sources.  In fact the amount of  sulfur oxides emitted from stack  heights less
than 30  meters  in Florida is nearly eight  times  that emitted from a  similar
height in either Ohio or Pennsylvania.

Table 2-19  compares  the  estimated utility-generated  sulfur  oxide and primary
sul fate  emissions  for 1980 from two  states  that  differ  in the  predominate
release  height  of  emissions.   For  both  Ohio and Florida,  utility emissions
account  for all of  the  sulfur   oxides and primary  sulfate estimated  to be
emitted  from  the highest stack  height intervals.  Although the  sulfur  oxide
emissions in  Ohio  are about 3.5 times those emissions from  the  sources in
Florida, the primary sulfate emissions in  Florida  are  about 5  percent higher
than those  from the  sources in Ohio.   These differences can be attributed to
the use  of  residual  fuel oils by the  utility  industry in  Florida.  The  total
emission of primary sulfates by industry  in Florida  is  greater  than  those
emissions generated by  the  coal-fired utilities  in Ohio.   Therefore, one
might expect a  greater deposition  of  primary sulfates from local  sources in
Florida  compared with Ohio.

2.3.3  Historical Trends and Current Emissions  of Nitrogen Oxides

Table 2-20  summarizes the  annual emissions densities of  nitrogen  oxides for
each state  over the  interval from 1950 to  1978.  The  table also  indicates the
numerical ranking  of  the  10 highest emission  densities  for the period of
calculation.    Ohio,  Pennsylvania  and  the  northeastern  Atlantic   coastal


                                   2-68

-------
                TABLE 2-18.  ESTIMATED POINT SOURCE S02 EMISSIONS AS A FUNCTION OF STACK HEIGHT
                                          FOR SELECTED STATES IN 1980
                                                   (106 kg yr)
                                                            Stack Height


                   0-30 meters          31 - 70 m          71 - 152 m           153 - 305 m           Total
2                  No.                 No.                 No.                 No.                  No.
    State        Sources  Emissions  Sources  Emissions  Sources  Emissions  Sources  Emissions  Sources  Emissions
                   14        24.0       70      183.1       47      580.0      48      1,722.9     185     2,510.0

    Pennsylvania    9        24.0      102      412.9       50      238.8      33      1,084.4     194     1,760.1

    Florida         61       184.2       74      205.6       30      469.6       0         0        165       859.4

    New Jersey      16        60.3       18      111.0        4       14.0       0         0         38       185.3

-------
TABLE 2-19.  ESTIMATED S02 AND PRIMARY SULFATE EMISSIONS FOR 1980
             FROM UTILITY SOURCES IN FLORIDA AND OHIO
                         (106 kg yr-1)
         Stack Height      No. of             S02          Sulfate
              m         point  sources     emissions      emissions
Florida



Ohio



0-30
31-76
77-152
153-305
0-30
31-76
77-152
153-305
12
23
30
0
3
17
35
48
99.9
107.9
469.6
0
4.5
48.5
441.4
1722.8
3.7
4.0
17.5
0
0.1
0.5
4.8
18.6
                                  2-70

-------
          TABLE  2-20.
ANNUAL EMISSIONS DENSITIES OF NITROGEN  OXIDES
       (kg km-2 yr~l)
                          1950
                  1960
1970
1978
Al abama
Arkansas
Connecticut
Del aware
District of Columbia
Florida
Georgia
Illinois
Indiana
Iowa
Kentucky
Louisiana
Maine
Maryl and
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
New Hampshire
New Jersey
New York
North Carolina
Ohio
Pennsylvania
Rhode Island
South Carolina
Tennessee
Texas
Vermont
Virginia
West Virginia
Wisconsin
1171
756
6311 (4)
3378 (10)
165891
1235
1017
3723 (6)
2860
1044
1262
2324
449
3604 (8)
6973 (3)
1916
690
672
999
690
12639 (1)
3487 (9)
1280
4240 (5)
3705 (7)
9670 (2)
990
1371
1135
409
1580
1725
1226
2088
763
9851 (4)
8726 (5)
182598
1925
1353
5566 (10)
5648 (9)
1344
2424
3868
518
7382 (8)
10823 (3)
3532
999
1108
1480
1171
16226 (1)
5430
1934
8163 (6)
7890 (7)
13048 (2)
1698
2788
2170
499
2542
3260
1852
2824
1271
14137 (3)
12249 (5)
304180
3305
2370
7019
5566
1925
4313
7346 (9)
799
9897 (7)
15272 (2)
5094
1389
1334
2134
1397
24080 (1)
7073 (10)
3641
9906 (6)
8426 (8)
13910 (4)
2679
3877
3341
1162
3723
5030
2842
3214
1435
12803 (3)
12031 (5)
174790
4649
3269
7019 (10)
5802
1998
4885
11513 (6)
808
10397 (8)
15490 (2)
5076
1662
1998
2833
2515
22110 (1)
6429
3941
10860 (7)
8662 (9)
12240 (4)
3387
4921
4349
944
3741
6701
2951
Note:   Numbers in parentheses  indicate numerical ranking of 10 highest
       emissions densities  (D.C.  excluded).
                                  2-71

-------
states consistently have  been  the  areas of highest emissions density.    The
emissions  densities  have increased  by a  factor of  two  or  three  over  the
28-year  interval  of record.   In New  England,  there  is  a contrast  between
changes in sulfur oxides and nitrogen oxides emissions. Comparing Table  2-13
with  Table 2-20  shows  that,  although  sulfur  oxides emissions  have  been
decreasing substantially  in  the  northeastern  United States,  nitrogen  oxides
emissions have not decreased comparably.

Table 2-21 provides estimates of the annual emissions  of  nitrogen oxides for
the 31-state region during  the  period from 1950  through  1980.   Total emis-
sions  have increased  from  1950 and  show  little  change  over  the  last  ten
years.  During 1980, highest emissions occurred  in Texas,  Ohio,  Pennsylvania,
and Illinois.  With few exceptions,  emissions appear to have increased in all
states from 1960 to 1980.   This contrasts  the apparent regional differences
in S02 and primary sulfate emissions discussed earlier.

The high emissions densities of nitrogen oxides  in the Northeast appear to be
strongly influenced by  mobile  sources.  Table 2-22  gives the  percentage  of
nitrogen oxides emitted by mobile sources  for six  northeastern  states chosen
from the 10  highest nitrogen oxides emissions density  areas  in 1978.   With
the exception of Delaware, this  region  exhibits  a  mobile  source contribution
in excess  of 40  percent  of the  total NOX emitted.    By comparison, areas
such as Ohio and Illinois exhibit a  25 percent contribution by mobile  sources
to nitrogen oxides emissions.  Figure  2-7  summarizes  the  composite  of source
category contributions to total  nitrogen oxide emitted between 1950  and 1978.
Within the last decade, mobile  sources and electric utilities  have been  the
predominant  contributors.   Comparison  with  Figure  2-6,  a  similar  repre-
sentation  of sulfur  oxide  emissions,  indicates  a  marked  and consistent
increase in  nitrogen  oxide emissions  during  a  period (1955-78) when  sulfur
oxide emissions have shown little variation.  Chemical  analyses  (Likens 1976)
of  precipitation  samples suggest  that nitric  acid  is comprising  a  larger
fraction  of  total  acidity  relative  to  sulfuric acid  in  the Northeast.
Because  of the  importance of the low-level mobile source  contribution,  the
argument could be made  that  correlation with the changes  in  emissions could
indicate a substantial  influence of local  and  subregional sources on rain-
water  acidity  through  both  primary  emissions   and  atmospheric transforma-
tions.

2.3.4  Historical  Trends and Current Emissions of Hydrochloric Acid
       THCTT

Hydrochloric acid is an emission component that has received little  attention
with  respect to  its  potential   for  acidic precipitation  formation.  Burning
coal  is  one  of  the major sources of  HC1  emissions to the  atmosphere (StahT
1969).   Chlorine  is present  in  coals in the form of inorganic chloride salts
which  are  soluble  in  water.   During combustion,  most of the chlorine salts
are converted to hydrogen chloride and emitted into the atmosphere.

Chlorine is found in relatively high concentration in coals from the Illinois
Basin and  the eastern United States  (Gluskoter et  al.  1977),  but only in low
concentrations in coals from the western United States.  The chlorine  content
ranges  from  0.01  to 0.50 percent.   Coals  from  western Pennsylvania  through


                                   2-72

-------
TABLE 2-21.
ESTIMATES OF ANNUAL  EMISSIONS OF NITROGEN OXIDES
        (106 kg  yr1)

Alabama
Arkansas
Connecticut
Del aware
District of Columbia
Florida
Georgia
Illinois
Indiana
Iowa
Kentucky
Louisiana
Maine
Maryl and
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
New Hampshire
New Jersey
New York
North Carolina
Ohio
Pennsylvania
Rhode Island
South Carolina
Tennessee
Texas
Vermont
Virginia
West Virginia
Wisconsin
1950
156.6
104.0
91.9
18.0
28.9
187.4
155.1
544.0
268.9
132.3
132.1
292.1
38.6
98.8
149.2
289.0
150.3
83.1
180.4
16.6
256.6
447.9
174.4
452.8
435.1
30.4
79.7
150.1
786.1
10.2
167.1
108.1
178.4
1960
279.2
105.0
127.8
46.5
31.8
292.0
206.4
813.3
531.0
196.0
253.7
486.2
44.6
202.3
231.5
532.7
217.6
137.0
267.2
28.2
329.4
697.4
263.5
871.8
926.6
41.0
136.6
305.1
1302.9
12.4
268.8
204.2
269.4
1970
377.6
174.9
183.4
65.3
52.9
501.4
361.5
1025.6
523.3
280.7
451.4
923.4
68.8
271.2
326.7
768.3
302.5
164.9
385.3
57.8
488.8
908.4
496.0
1057.9
989.5
43.8
215.6
424.3
2313.9
29.9
398.7
315.1
413.5
1978
429.7
197.4
166.1
64.1
30.4
705.3
498.6
1036.1
545.5
291.4
511.2
1447.3
69.5
284.9
331.4
765.6
362.0
247.0
311.4
60.6
448.8
825.7
536.9
1159.8
1017.2
39.5
272.5
533.6
3012.1
23.5
395.6
419.8
429.3
1980
480.5
197.2
121.6
47.1
19.9
588.0
448.3
912.0
701.3
290.9
482.0
842.2
53.9
225.1
230.0
625.9
338.8
258.8
314.9
75.5
368.3
616.5
586.5
1038.4
941.2
33.1
236.1
469.2
2307.7
22.4
367.1
410.3
381.4
                   6386.0   10817.2    15299.6     17609.4   15059.7
                             2-73

-------
      TABLE 2-22.  MOBILE SOURCE CONTRIBUTION TO NITROGEN OXIDES
            EMISSIONS DENSITIES IN NORTHEAST UNITED  STATES
                           Percentage of total  NOX emissions density
                         	attributable  to mobile sources	
State                    1950                 1960                1975

New Jersey                27                    34                  47
Massachusetts             36                    35                  43
Connecticut               23                    34                  46
Rhode Island              30                    34                  64
Delaware                  29                    21                  28
Maryland                  29                    25                  41
                                   2-74

-------
        2.5 r
     01
     _i^


    o
    i—i
     o
     t—I



     CO

     o
     KH
     CO
     oo
                     LEGEND



                     MISCELLANEOUS
HIGHWAY VEHICLES


PIPELINES


COMMERCIAL/RESIDENTIAL


INDUSTRIAL


ELECTRIC UTILITIES
           1950
 1955
1960
1965
1970
1975  1978
                                    YEAR
Figure 2-7.  Historical trends of nitrogen oxide emissions by source

             category for the study area.  Adapted from Gschwandtner

             et al. (1981).
                                    2-75

-------
southern  Illinois  (a high  S02  emission  density  region) contain  about 0.2
percent chlorine.  Estimated emissions of hydrochloric acid from this region
in 1974 amount to over 450,000  tons.   Furthermore,  the  amount of hydrochloric
acid pollution  by  coal  burning may  be  increased  when  calcium  chloride  is
added to the coal as an  antifreeze or dust-proofing agent (Stahl 1969).

Cogbill and  Likens  (1974)  have estimated that  the acidity of precipitation
has  a  5  percent contribution  from  HC1.   However,  the data  set  used  to
apportion the  stoichiometric  balance of  hydrogen  ion and  anions  was  taken
from measurements in  New York  and  New England.   Pack  (1980) noted in his
analysis  of EPRI  and  MAP3S  precipitation  data  that,  excluding  sea   salt
contributions,  the  two networks  were within 6  percent  agreement  on  molar
concentrations of all anions except  chloride,  which differed by 47  percent.
Although no reason could be given for this  discrepancy,  the differences may
be due  to either sampling hardware  and  analytical  errors or a poor  distri-
bution of monitoring  sites  with respect to  major anthropogenic HC1  emission
sources.   The latter possibility could be studied  by examining  individual
precipitation event data.   The high solubility of HC1  in water suggests that
emissions  would be  assimilated  rapidly into  cloud  processes  involved  in
precipitation formation.  Also, during a  precipitation event, washout of HC1
and NH4C1  should occur in the  lower atmosphere.

An estimate  of HC1  emissions densities  as  chloride is given in Table  2-23.
These  values  do not  include  additional  chloride  emissions  due  to  chemical
de-icers  added  to   fuel  prior to  combustion.    The  10  highest  emissions
densities are  also  ranked  for  each  calculation  period.   Consistently,  West
Virginia,  Ohio,  Pennsylvania,  and   Illinois   have  remained  the   greatest
chloride emissions areas.  Significant increases  have been  noted  for Kentucky
and Tennessee because of their increased coal  consumption.

2.3.5  Historical Trends and Current  Emissions of Heavy Metals Emitted
       from Fuel Combustion1

As with calculated  emissions  densities  for  sulfur  and nitrogen oxides, fuel
composition  data can be  used  to estimate  emissions  densities  for  certain
metals  that  might  be  of  use  as  tracers  to   evaluate  the   transport,
transformation, and deposition of acidic components.  Arsenic  and mercury are
emitted as  volatiles from  coal  combustion  but  are present only   in minute
quantities in fuel oils.  In contrast, vanadium  is  the major metal  associated
with residual fuel burning but is only a minor component  of coal.

Table 2-24 is a compilation of arsenic,  mercury,  and vanadium  levels found  in
coals  burned in each state in  the  eastern  United  States.  Gluskoter et al.
(1977)  presented   the  ranges   of   concentrations   and   mean   values   of
concentrations  for  these metals.   The  range of  arsenic concentrations  in
^-Editor's Note:  Although several public reviewers questioned the relevancy
 of this section, it has been included based on the importance of some metals
 to tracer studies and effects studies.  The decision that this section
 remain in the chapter was also made at the November 1982 Technical  Review
 Meeting.


                                   2-76

-------
           TABLE 2-23.  ANNUAL EMISSIONS DENSITIES OF CHLORIDE
                          (kg km'2 yr"1)
                         1950         1960         1970         1978
Alabama
Arkansas
Connecticut
Del aware
District of Columbia
Florida
Georgia
Illinois
Indiana
Iowa
Kentucky
Louisiana
Maine
Maryl and
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
New Hampshire
New Jersey
New York
North Carolina
Ohio
Pennsylvania
Rhode Island
South Carolina
Tennessee
Texas
Vermont
Virginia
West Virginia
Wisconsin
2.1
0.0
71.7 (7)
1.5 (4)
4106.0
0.0
8.6
232.4 (2)
54.6 (10)
2.1
8.3
0.0
0.0
45.2
0.4
35.0
0.5
0.0
3.3
1.1
91.7 (6)
64.1 (9)
12.9
175.9 (3)
99.0 (5)
71.3 (8)
1.0
6.5
0.0
0.0
113.5 (4)
262.9 (1)
0.4
30.0
0.0
254.2 (9)
315.1 (7)
5374.4
2.7
34.1
816.3 (2)
264.2
6.2
63.7
0.1
2.7
252.0 (10)
178.9
139.8
2.6
0.1
22.0
11.7
306.0 (8)
190.8
34.8
697.1 (3)
444.9 (5)
374.1 (6)
69.4
210.3
0.3
2.6
459.0 (4)
829.9 (1)
15.4
37.2
0.0
128.9
298.0 (6)
5843 .9
9.4
80.1
769.9 (2)
258.8 (7)
7.7
114.9
0.0
0.5
240.6 (9)
32.4
171.6
3.4
6.1
39.5
47.0
226.0 (10)
126.8
86.7
746.2 (3)
316.7 (5)
2.0
117.1
255.1 (8)
0.0
2.8
436.4 (4)
1287.5 (1)
21.2
34.5
4.1
4.7
153.5 (9)
437.7
12.5
173.9
728.2 (3)
285.1 (6)
13.3
134.4 (10)
0.4
0.2
172.5 (8)
3.2
133.9
5.3
22.5
66.7
29.8
91.7
65.1
85.5
770.9 (2)
305.1 (5)
3.2
146.9
331.2 (4)
7.4
0.2
261.5 (7)
1905.9 (1)
17.7
Note:   Numbers in parentheses  indicate  numerical ranking of 10 highest
       emissions densities (D.C.  excluded).
                                   2-77

-------
     TABLE 2-24.
ARSENIC, MERCURY, AND VANADIUM CONTENT OF
      BITUMINOUS COAL
State
Al abama
Arkansas
Connecticut
Delaware
District of Columbia
Florida
Georgia
Illinois
Indiana
Iowa
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
New Hampshire
New Jersey
New York
North Carolina
Ohio
Pennsylvania
Rhode Island
South Carolina
Tennessee
Texas
Vermont
Virginia
West Virginia
Wisconsin
Arsenic
(ppm)
53
53
17
17
17
53
26
15
10
22
11
53
17
17
17
10
2
26
9
17
17
17
11
19
17
17
26
26
5
17
6
6
22
Mercury
(ppm)
0.30
0.30
0.18
0.18
0.18
0.30
0.13
0.19
0.30
0.22
0.17
0.30
0.18
0.18
0.18
0.30
0.10
0.13
0.18
0.18
0.18
0.18
0.17
0.23
0.18
0.18
0.13
0.13
0.09
0.18
0.12
0.12
0.22
Vanadium
(ppm)
52
52
40
40
40
52
33
32
26
27
34
52
40
40
40
26
10
33
40
40
40
40
34
38
40
40
33
33
7
40
23
23
27
Source:  Values assigned from Gloshoter et al.  1977.
                                   2-78

-------
eastern U.S. coals  is  1.8  to  100  ppm,  for mercury,  0.05 to 0.47  ppm,  and for
vanadium, 14 to 73 ppm.  The metal concentrations presented in Table 2-24 for
each  state  were obtained  by  assuming that the  fuel  consumed in each  state
forcombustion was obtained from coal producing areas  located  near the state.
For example, an average  arsenic concentration of 53  ppm in coal  was assigned
to Alabama,  Arkansas,  Florida,  and Louisiana with the  assumption that  these
states  would be  receiving  coal   from  about  the   same  producing  region.   Of
course there would  be  a  range of  concentrations  expected  for each  state but
such data are not readily available.

The  fuel  consumption  data computed  by Gschwandnter  et  al.  (1981)  can  be
multiplied by the concentration of  arsenic and mercury in coal to  arrive  at
the normalized annual emissions densities given  in Tables 2-25 and 2-26.  For
1978,  Ohio  exhibited  the highest  emissions density  for  both  arsenic  and
mercury.  These data can  be  used with the corresponding  estimates for S02,
NOX,  and  primary  sulfate to  evaluate  the  transport  and  deposition  of
emissions.    As  tracers,   the  S0x/metals  or  N0x/metals  ratios   could  be
useful in identifying origins  of specific precipitation event  samples.

The ratios of atmospheric sulfate to vanadium, arsenic,  and mercury might be
used  to apportion  that  quantity of  sulfate that  is formed by progressive
oxidation of atmospheric  S02«   The  presence  of  vanadium  in atmospheric
aerosols  could  be  used  in conjunction  with  meteorological measurements  to
estimate the regional origins  of  the air mass containing  such  aerosols.   For
example,  air masses of midwestern  U.S.  origin would be  expected to  contain
less vanadium than  an air mass  being  transported  along  the  eastern United
States  because  of  the  predominant use  of fuel   oil  along the  East Coast.
Estimates of vanadium  in  atmospheric aerosols  as  opposed  to arsenic  or
mercury could be used.

Vanadium  is  not emitted as a volatile element  from  fuel combustion.  It  is
present as  porphyrin compounds in  the  fuels  and is  converted  to  the  oxide
form  in  the  combustion  zone.    The  oxides,  mainly  V20s, are  incorporated
into the  fly ash.   Residual  oil-fired  sources for  utility,  industrial,  and
commercial  categories usually  do  not employ particulate removal  devices.
Therefore, one can  calculate  vanadium  emissions  from oil  burning,  given  the
fuel  consumption, the  particulate emission factors  (U.S. EPA 1981),  and  the
vanadium content of oil  ash.

The vanadium content  of  oil fly ash will vary with  the vanadium content  of
the oil  and  with  certain  combustion  operating  parameters  such  as  excess
boiler  oxygen  and  emissions  controls.    Vanadium   in  fuel  oil  will   vary
according to the  regional  production  source of  the crude and the  degree  of
hydrodesulfurization.  It is  assumed that  most of the residual fuels burned
in the eastern United States are derived from Venezuelen crudes.  These  fuels
are noted for their elevated vanadium levels.  However,  only approximate  fuel
vanadium values  can be applied to  the  fuel  consumption inventories.

For these calculations,  it is assumed  that  the  average vanadium content  of
residual oil consumed by electric  utilities  and industrial  sources  is 200
ppm.    Commercial/residential  sources  are  assumed to burn  hydrodesulfurized
oils  containing 15  ppm vanadium.    Experimental  measurements  of particulate


                                   2-79

-------
TABLE 2-25.  ANNUAL EMISSIONS DENSITIES  OF  ARSENIC
                (kg km~2 yr"1)

Al abama
Arkansas
Connecticut
Del aware
District of Columbia
Florida
Georgia
Illinois
Indiana
Iowa
Kentucky
Louisiana
Maine
Mary! and
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
New Hampshire
New Jersey
New York
North Carolina
Ohio
Pennsylvania
Rhode Island
South Carol ina
Tennessee
Texas
Vermont
Virginia
West Virginia
Wisconsin
1950
0.18
0.01
0.60
0.01
34.70
0.00
0.07
0.56
0.29
0.12
0.08
0.00
0.00
0.38
0.33
0.19
0.06
0.00
0 .03
0.01
0.78
0.54
0.12
1.03
0.84
0.60
0.08
0.05
0.00
0.00
0.08
0.18
0.22
1960
2.62
0.01
2.13
2.66
45.40
0.24
0.26
1.95
1.42
0.34
0.59
0.01
0.02
2.12
1.51
0.76
0.25
0.00
0.17
0.10
2.59
1.62
0.32
4.07
3.72
3.49
0.52
1.59
0.00
0.02
0.32
0.58
0.86
1970
1.96
0.00
0.65
1.51
29.63
0.50
0.36
1.10
0.84
0.26
0.63
0.00
0.00
1.22
0.14
0.56
0.20
0.03
0.18
0.24
1.15
0.64
0.48
2.62
1.61
0.01
0.53
1.15
0.00
0.01
0.18
0.54
0.70
1978
0.61
0.07
0.01
0.26
0.74
0.17
0.26
0.35
0.35
0.15
0.25
0.01
0.00
0.29
0.01
0.14
0.10
0.03
0.10
0.05
0.16
0.13
0.16
0.90
0.51
0.01
0.22
0.50
0.02
0.00
0.04
0.27
0.19
                          2-80

-------
TABLE 2-26.  ANNUAL EMISSIONS DENSITIES OF MERCURY
                 (kg km" ^ yr~*)

Alabama
Arkansas
Connecticut
Del aware
District of Columbia
Florida
Georgia
Illinois
Indiana
Iowa
Kentucky
Louisiana
Maine
Maryl and
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
New Hampshire
New Jersey
New York
North Carolina
Ohio
Pennsylvania
Rhode Island
South Carolina
Tennessee
Texas
Vermont
Virginia
West Virginia
Wisconsin
1950
0.002
0.000
0.013
0.000
0.739
0.000
0.001
0.014
0.018
0.002
0.002
0.000
0.000
0.008
0.007
0.012
0.001
0.000
0 .001
0.000
0.017
0.012
0.004
0.025
0.018
0.013
0.001
0.001
0.000
0.003
0.003
0.007
0.004
1960
0.030
0.000
0.045
0.057
0.968
0.003
0.003
0.049
0.088
0.007
0.018
0.000
0.001
0.045
0.032
0.047
0.003
0.000
0.007
0.002
0.055
0.034
0.010
0.100
0.080
0.067
0.005
0.016
0.001
0.012
0.012
0.022
0.017
1970
0.037
0.000
0.023
0.054
1.052
0.010
0.006
0.046
0.086
0.008
0.033
0.000
0.000
0.043
0.005
0.057
0.003
0.001
0.012
0.009
0.041
0.023
0.025
0.165
0.057
0.000
0.009
0.020
0.001
0.011
0.011
0.034
0.009
1978
0.035
0.004
0.001
0.028
0.079
0.009
0.012
0.043
0.091
0.015
0.038
0.000
0.000
0.031
0.001
0.045
0.005
0.002
0.020
0.005
0.017
0.012
0.024
0.111
0.055
0.001
0.011
0.020
0.000
0.007
0.007
0.050
0.020
                        2-81

-------
emissions from  such  sources under these conditions  have  shown fuel oil  ash
vanadium concentrations  of  5.3 percent  for utility  and  industrial sources
(Boldt et al.  1980)  and 3.4  percent  for residential and commercial sources
(Homolya and Lambert 1981).   Therefore, simply multiplying total  particulate
emissions factors  by vanadium  fly ash  contents  will result  in  a  vanadium
emissions factor for residual  oils.

Estimates of  vanadium  emissions   from  coal  combustion  pose  an   additional
problem in that various levels of  particulate emissions controls were enacted
in each state between 1950 and 1978.  For calculation purposes, an  emissions
control scenario has been assumed to  have  been uniformly implemented  in  the
eastern  United  States   over  this   period.   Between  1950  and  1965,  we have
assumed  that  50  percent  of  the  particulate   matter   generated  by coal
combustion is emitted to the  atmosphere.  This emission  level  is  reduced  to
15 percent  in  1970 and  finally to 10  percent in  1978.   Therefore,  vanadium
emissions were estimated by multiplying the particulate emissions  factor  for
uncontrolled  bituminous coal-fired  sources  by  the fuel  vanadium content
(given  in  Table 2-24)  and  the appropriate particulate  control  factor  for
1950, 1960,  1970, and 1978.

Vanadium  emissions  from both  coal   and  oil  were  summed,  and  the  totals
reported as emissions densities for each state.   The calculations,  shown  in
Table  2-27,  indicate highest vanadium  emissions  densities  in the  northeast
due to  residual  oil  burning.   However, the  values  have  decreased  somewhat
since  1970,  reflecting  a  switch   to  hydrodesulfurized  residuals  containing
less vanadium.  The greatest change in vanadium emissions  has occurred  in  the
Gulf  Coast,  where  utilities  switched  from  gas to  oil  along  with  increased
coal  combustion.

A major  application  of atmospheric trace metal measurements  is  identifying
specific  sources  of  air pollution  at  particular  times  and places.  If  a
particular emitted  quantity can  be  identified with  some  single  source  (or
group  of  sources), then measurements  of its  concentrations  can  be used  to
identify occasions  when air  quality is affected by that  specific source.   The
philosophy is  like that of atmospheric  tracer studies,  except that tracers
"of  opportunity"  are   employed.     In  practice,  however,   it   is usually
impossible to  find  a  single tracer that is unique to some particular  source
or set of sources.   Instead, groups of trace metals  can be chosen  to provide
statistically identifiable  "fingerprints" or  "signatures" of different kinds
of emission sources.  Cooper  and  Watson (1980) identify five  distinct kinds
of statistical analysis  that can  be used, and  they  illustrate  the utility of
the  methods  by  assessing   the  contribution  to  air  pollution in  Portland,
Oregon, of emissions from categories  of sources  such as  automotive exhaust,
kraft mills, home  heating,  asphalt production, coal  burning,  and  road dust.
Kowalczyk et  al.  (1982) used  trace  metal  concentration data  obtained  in
Washington,  DC,  to search for effects  associated with  refuse  incineration,
automotive exhaust, and coal- and  oil-fired  power  plants.

These  statistical techniques (also known as  receptor models) are  designed to
relate observed characteristics of air pollution to corresponding  features of
emissions.   The  statistical  treatments assume   that  the trace  metals  (or
similar  materials)  used in  the  analysis  are transported  in the   same  way


                                   2-82

-------
TABLE 2-27.   ANNUAL EMISSIONS DENSITIES  OF  VANADIUM
                 (kg km'2 yr-1)

A1 abama
Arkansas
Connecticut
Del aware
District of Columbia
Florida
Georgia
Illinois
Indiana
Iowa
Kentucky
Louisiana
Maine
Mary! and
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
New Hampshire
New Jersey
New York
North Carolina
Ohio
Pennsylvania
Rhode Island
South Carolina
Tennessee
Texas
Vermont
Virginia
West Virginia
Wisconsin
1950
0.66
0.48
35.20
12.09
212.69
3.20
0.98
4.31
5.39
0.68
0.57
2.54
1.09
13.91
35.88
2.71
0.33
0.08
0 .08
2.20
63.18
14.24
0.56
6.85
11.26
85.69
1.33
0.41
1.97
0.40
3.14
1.36
0.54
1960
2.95
0.10
33.65
27.39
365.91
4.82
1.22
8.04
7.48
0.55
1.86
0.17
1.56
16.71
46.70
3.86
0.91
0.08
1.11
2.88
56.98
14.19
1.73
11.17
17.29
51.93
1.74
2.08
0.18
0.57
2.41
2.68
1.68
1970
2.25
0.32
77.65
32.77
1,422.65
10.31
2.25
6.76
4.89
0.35
2.16
0.49
3.12
20.44
98.60
3.27
0.73
0.24
1.25
7.89
100.39
27.01
2.69
6.78
16.31
71.30
2.11
1.64
0.11
0.91
7.66
5.21
1.36
1978
2.35
3.54
75.57
56.02
280.44
16.60
2.38
6.17
5.82
0.26
1.39
7.85
3.36
25.81
88.07
4.85
0.52
4.94
0.92
6.16
71.21
28.89
3.04
4.94
14.31
31.23
4.89
1.21
1.00
1.55
9.48
2.18
0.58
                         2-83

-------
between sources and sampling sites, and that they are sampled with precisely
the  same   efficiency.     Although   this   is  undoubtedly   true   in  many
circumstances,  the  accuracy  of  the  assumption  becomes  less obvious  as
distances and time scales increase  or  whenever meteorological  factors  such as
rainfall intervene.

The statistical methods  of  receptor modeling  have recently been extended to
address visibility (Friedlander  1981,  Barone et  al .   1981).  Some attempts to
apply receptor modeling methods  to  investigate long-range  transport have been
conducted,  but the  results  obtained are  contentious.    Applying   methods
involved  in receptor  modeling  to  questions  of  precipitation  chemistry is
difficult   because   of  the  complexity   of   the   processes  involved  in
precipitation  scavenging and the need to assume identical  pollutant pathways
and scavenging rates for source  apportionment methods to work  properly.

2.3.6  Historical  Emissions Trends  in  Canada

Historical  emissions data  have been  developed  for S02  and  NOx  for the
years 1955, 1965, and 1976 as a contribution to the  effort undertaken by the
U.S. /Canada Work  Group  3B  (Engineering,  Costs,  and Emissions)  in accordance
with  the  Memorandum of Intent  on  Transboundary  Air  Pollution  concluded
between  Canada and the United  States  on  August  5,  1980.   Information
regarding  production  and fuel consumption was  obtained from internal  files
and, for  other source  categories,  U.S.  or Canadian  emissions  factors were
applied  to  the  basic   data.    Actual   emissions data  were  available for
copper-nickel smelters and some power plants.  For 1976,  emissions data were
taken from  a  nationwide  inventory prepared by SNC/GECO Canada,  Inc.,  and the
Ontario Research Foundation (1975).
Total  Canadian  emissions  of S02 and  NOX  *°r each of  the years 1955,  1965,
and  1976  are  given  in  Table  2-28.    Total SO?  emissions  in  Canada  were
approximately 5.3 million  metric  tons for 1976, 6.6  million metric tons  in
1965,  and 4.5 million metric  tons in 1955.   The  fluctuations  in  emissions
levels  were  due  to  changes  in  production  by the  copper-nickel   smelting
industry, which is centered in eastern Canada.  Sulfur dioxide emissions from
power  plants were 0.05 million metric  tons  in 1955 and rose  to  0.55 million
metric  tons in 1976, with  over  90 percent  of  the total  emitted in eastern
Canada.   Sulfur dioxide  emissions  from nonutility fuel combustion  decreased
slightly  between  1955 and  1965  as a result  of fuel  switching  from coal  to
oil.     Industrial  fuel   combustion  represents  the  major  contributor  to
nonutility  combustion emissions.

Iron  ore processing emissions of  S02  increased by about 75  percent between
1955  and  1976,  along  with  increases in natural  gas  processing  and  petroleum
refining.   The increases  in  these categories account for 78 percent  of the
"other" S02 emissions for the country.

Tables  2-29 and 2-30  contain estimates  of  emission  densities  for S02  and
primary  sulfate (Vena 1982).   Sulfur dioxide  emission  densities  have been
calculated  for the years  1955,  1965, and  1976.   Primary  sulfate  emission
densities are  available  for 1978.   The highest emissions densities occur in
the Maritime Provinces as  compared  to  western Canada  and  can be explained by


                                   2-84

-------
ro
i
O3
cn
                              TABLE 2-28.  HISTORICAL EMISSIONS OF S02 AND NOX - CANADA
                                 (U.S./CANADA WORK GROUP 3B DRAFT REPORT 1982)
                                               (103 kg yr-1)
Sector
Cu-Ni smeltersb
Power plants
Other combustion0
Transportation
Iron ore processing
Others
TOTAL
1955
S02 N0xa
2,887,420
56,246 10,335
1,210,108 227,837
83,474 323,785
109,732
189,876 68,065
4,536,856 630,022
1965
SOa N0xa
3,901,950
261,837 57,402
1,129,548 247,323
48,669 511,868
155,832
1,095,341 33,778
6,593,177 850,371
1976
S02
2,604,637
614,323
884,867
77,793
175,829
954,215
5,311,664
NO*'
-
206,454
445,315
1,017,936
-
190,327
1,860,032
       aNOx expressed as N02.

       ^Includes emissions from pyrrhotite roasting operations.

       Clnc1udes residential, commercial, industrial, and fuelwood combustion.  Industrial fuel
        combustion also includes fuel combustion emissions from petroleum refining and natural gas
        processing.

-------
      TABLE 2-29.   ESTIMATES OF ANNUAL EMISSIONS DENSITIES OF
                     SULFUR OXIDES (VENA 1982)
                        (kg knr2 yr'1)
                                              Year
Province                       1955           1965           1976
Newfoundland                     52             71            158
Prince Edward Island            675            690          1,557
Nova Scotia                   1,943          1,761          3,180
New Brunswick                 1,894          2,230          2,181
Quebec                          697            949            822
Ontario                       3,136          3,829          2,532
Manitoba                       457          1,047          1,112
Saskatchewan                    108            339             74
Alberta                          98            506            811
British Columbia                125            565            417
Yukon-N.W.T.                    < 1            < 1            < 1
                                   2-86

-------
TABLE 2-30.  ESTIMATED OF ANNUAL EMISSIONS DENSITIES OF PRIMARY
                 SULFATES FOR 1978 (VENA 1982)
                      (kg km-2 yr-1)
Province
Newfoundland
Prince Edward Island
Nova Scotia
New Brunswick
Quebec
Ontario
Manitoba
Saskatchewan
Alberta
British Columbia
Yukon & N.W.T
Total $04
(103 kg)
4,081
435
12,320
12,582
53,452
45,714
13,217
3,742
7,321
33,380
213
Density
11
77
233
176
39
50
24
7
12
37
< 1
                               2-87

-------
the  significant  difference  in   the   size   of   the  provinces.    With  few
exceptions, emissions in Ontario  are  concentrated  near the southern part of
the province.

Total NOX  emissions  for Canada have  increased  significantly  due to changes
in the transportation sector and  power plants.   Automobile  and diesel-powered
engine  emissions  of NOX  have  increased  by   factors  of  three  and  five,
respectively, from 1955  to 1976.   Eastern  Canadian  provinces still contribute
the major  portion  of NOx  emissions,  although a  shift  in industrial activity
and population  to  the west has changed the  contribution from  71 percent in
1955 to 61 percent in 1976.

Table  2-31  contains estimates   of  NOX  emissions  densities  for Canadian
provinces for 1955, 1965, and 1976.  The  highest emission  densities occur in
the maritime  provinces  of Prince Edward  Island and  Nova Scotia.  Over  this
period, NOX emission  densities in Canada  were increasing  similarly to those
estimated for the eastern United  States as shown in Table 2-20.

Qualitative assessments of the geographical  distribution of emissions  in the
United  States and Canada  can be made by  graphically displaying  emissions
aggregated  on a state  or  province level.   Figures  2-8,  2-9,  and 2-10 are
displays  of  annual  emissions of SOg,  primary  sulfate,  and  NOX  for  the
United States and  Canada.   Emissions  data for  the  United States was obtained
from  the  EPA 1980 National  Emissions Data  System  (NEDS)   files.   Canadian
SOg  and NOX  data  are from  Environment  Canada  1980 files and  the Canadian
primary  sulfate data represents  1978  emissions calculated by  Vena  (1982).
The  area  of  highest  SOg  emissions in the   United  States  is  bound by
Pennsylvania  on  the east and  Missouri   on the  west.    Highest Canadian
provincial  S0£  emissions  summaries are  comparable  to  state-level  emissions
tn the southeastern United States.

The  U.S.   region  of  highest  primary  sulfate  emissions extends beyond the
highest  SOg emission region  shown  in Figure  2-8.  Much  of  New England  is
estimated  to  have  total  primary   sulfate emissions comparable to  the  Midwest
because  of the extensive  use of residual  fuel oils  in  the  Northeast.   As
mentioned  earlier, the  combustion emissions from  residual  oils  contain  more
primary  sulfate  than  combustion emissions from  coal   of  similar  sulfur
content.  The use of such fuels in the eastern provinces of Canada results  in
the  estimation  shown  in  Figure 2-9 that  primary sulfate emissions in  eastern
Canada  are  comparable   to total  emission  levels  for the  midwestern and
northeastern  United States.

The  summary of NOX  emissions  shown  in Figure  2-10  illustrates the regional
differences in  the cumulative effect of both stationary and mobile combustion
sources.   The  regions  of  highest  NOX emissions  are   in  the  Midwest,  Gulf
Coast,  and California.   Total  Canadian NOX emissions are  much  lower  than  in
the  United  States with  the  highest  Canadian   NOX  emission  area  occurring
along the  Great Lakes region.
                                   2-88

-------
       TABLE 2-31.  ESTIMATES OF ANNUAL  EMISSION  DENSITIES  OF
                     NITROGEN OXIDES (VENA 1982)
                         (kg km-2 yr-l)
                                              Year
Province                       1955            1965           1976
Newfoundland                     25             37             123
Prince Edward Island            451            767           1,461
Nova Scotia                     529            581           1,483
New Brunswick                   251            364             820
Quebec                           94            130             242
Ontario                         246            294             600
Manitoba                         82             82             156
Saskatchewan                     87            102             231
Alberta                         104            204             515
British Columbia                 86            113             221
Yukon-N.W.T.                      2              1              18
                                   2-89

-------
        n
        D
        D
£  48   IO6
>  48 < 250   10b  kg/yr,
>  250 < 1015  IO6 kg/yr.
>  1015  IO6  kg/yr.
Figure 2-8.
 Annual emissions of SOo by state.
 Emissions Data System 1980.
Data are from National
                                  2-90

-------
               LEGEND
       D £2  xlO6 kg/yr.
       D >2   < 8X106 kg/yr.
       O >8   < 30xl06 kg/yr.
       E3 >30 xlO6 kg/yr.
Figure 2-9.  Annual emissions of $04 by state.
            Emissions Data System 1980.
Data are from National
                                2-91

-------
       D £100 x 106 kg/yr.   v^
       D >100 £370 x TO6 kg/yr.
       ^ > 370^780 x 106 kg/yr
          >780x 106 kg/yr.
Figure 2-10.   Annual  emissions  of NO  by state.
              Emissions  Data  System 1980.
Data are from National
                                 2-92

-------
2.3.7  Future Trends In Emissions

2.3.7.1   United States—Electric  utility  plants fired  by  fossil  fuels  are
projected to continue  to  contribute  the  greatest amount of S02  emissions  as
well as  significant  amounts  of NOX.   One  estimate  of  the  electricity  demand
growth rate is  1.5 percent per year  from 1981  to 1985 and about 2.7  percent
per  year from  1985 to  2000  (U.S./Canada 1982).   These  growth  rates  are
assumed  to  vary slightly by  region,  with  higher growth rates  in  the  West,
West South  Central,  and  Mountain areas,  and lower  than  average  rates  in  the
East.

Within the  nonutility  sectors, industrial  combustors contribute  the  greatest
amount  of S02,  followed  by  nonferrous  smelters and  residential/commercial
furnaces  and   boilers.     Table  2-32   summarizes  current  SOx  and  NOX
emissions  for  1980  and  projected  emissions  to 2000  as  estimated  by  the
U.S./Canada  Work Group  3B  (1982).   The  estimates  are  based  on  numerous
assumptions  incorporated  into simulation  growth models.    The  forecasting
ability and sensitivity of such models are based on  the assumptions made upon
critical  input parameters such as:

     0  Fuel price, boiler capital cost,  operating and maintenance  costs;

     0  Regulatory  assumptions involving  New  Source Performance  Standards
        and State Implementation  Plans,  including nonattainment  policy; and

     0  The technological and physical constraints regarding the use of coal
        or  natural gas.

These  economic  and  regulatory   factors  influence other  source  emissions
categories  of  sulfur and nitrogen oxides  such  as nonferrous  smelting,  where
emissions are  proportional  to the production estimates  of copper,  lead,  and
zinc.

2.3.7.2   Canada—Canada's  electrical  generating  capacity  is  expected  to
increase  substantially by 1980,  exceeding  1977 capacity by over 60 percent.
This  expansion  will  be noticeable in all   three  major  types of generation:
hydroelectric,  nuclear,  and  conventional  fossil  fuels.   Hydroelectric  power
will maintain its leading role in the utility sector, nuclear power will grow
by  a  factor  of  three,  and  thermal  generation  will  increase  by about  50
percent  from  1977  to 1990.   All   projected  fossil-fired  steam  unit additions
will  use coal,  which  will  result in a  12  percent increase in annual coal
consumption over this period.

Natural  gas processing may  be a  significant  source  of S02 emissions over
the  coming  15 years because  approximately half of  the  gas found  to  date in
Canada   contains  significant  quantities   of   hydrogen  sulfide,   which   is
converted to sulfur  during processing.  Residuals, approximately 3  percent of
the  hydrogen  sulfide,  are incinerated and emitted  to the  atmosphere  as S02-
Alberta  and British  Columbia are the major  gas-processing  provinces.   Table
2-33  summarizes Canadian S02  and NOX emissions  projected to  2000.    These
estimates were  compiled  from  the U.S./Canada Work Group  3B forecasts  (1982),
                                   2-93

-------
       TABLE 2-32.  NATIONAL U
0. U.S. CURRENT AND PROJECTED SOa AND NOX
EMISSIONS (Tg yr'1)
Current
1980
Source category
1.
2.
3.
4.
5.
Electric utilities
Industrial boilers and
process heaters
Nonferrous smelters
Re si den ti al /commerc i al
Other industrial
processes
6. Transportation
TOTALS
S02
15
2
1
0
2
0
24
.0
.4
.4
.8
.9
.8
.1
NOX
5.6
3.5

0.7
0.7
8.5
19.0
Projected
1990
SO
15
3
0
1
1
0
22
2 NOX
.9 7.2
.4 3.0
.5
.0 0.7
.2 0.8
.8 7.8
.8 19.5
Projected
2000

16
6
0
0
1
1
26
S02
.2
.5
.5
.9
.5
.0
.6
NOX
8
4

0
1
9
24
.7
.0

.6
.1
.7
.1
Summarized from:   U.S./Canada Work Group 3B  Draft Report (1982).
                                  2-94

-------
     TABLE  2-33.   NATIONAL  CANADIAN CURRENT AND PROJECTED S02 AND NOx
                            EMISSIONS  (Tg yr-1)
Source category
1.
2.

3.
4.
5.
6.
7.
Electric utilities
Industrial boilers and
process heaters
Nonferrous smelters
Residential /commercial
Transportation
Petroleum refining
Natural gas processing
8. Tar sands
TOTALS
Current
1980
S02 NOX
0.7 0.2
0.6 0.3

2.1
0.2 0.1
1.1
0.1
0.4
0.1
4.2 1.7
Projected
1990
S02 NOX
0.7 0.2
0.3 0.3

2.3
0.08 0.07
1.3
0.1
0.5
0.3
4.3 1.9
Projected
2000
S02 NOX
0.7
0.2

2.3
0.03

0.0
0.4
0.3
4.0
0.3
0.3


0.07
1.7



2.4
Summarized from:   U.S./Canada Work Group  3B Draft  Report  (1982).
                                   2-95

-------
which again are based on assumptions  concerning  costs  and  regulatory controls
similar to those used to prepare  the  U.S.  estimates.

2.3.8  Emissions Inventories

Numerous source emission inventories  have  been  used  by EPA and  the  Department
of Energy.   Historically, most of these inventories  start with  the National
Emissions Data System (NEDS) data base to  modify,  correct, or update specific
source categories such as electric power plants.  With different  assumptions,
time frames,  and  emissions  factors,  these various  inventories have yielded
differing   results   in   terms   of   emissions  totals   and    geographical
distributions.   Inventories have  been developed  that range  from national
trends summaries to  annual  and seasonal point and area source-specific  data
at the county and metropolitan level.   The diversity  of inventories reflects
the  differences in  the  objectives  for  which  they  were produced.   These
include:

     1.  Historical Trends  Analysis.    An  example is  the Emissions History
         Information  System5ytFe   Office  of  Air  Quality  Programs   and
         Standards.   The inventory  contains  national  emissions  levels  of
         particulate matter, sulfur oxides, hydrocarbons,  and carbon monoxide
         for  1940,  1950,  1960,   and   all  years  from  1970   to  1980.   The
         Historical  Trends  inventory  (which was  used extensively  for  the
         emission  density  calculations in  this contribution)  is a  set  of
         S02  and  NOX  state-level  emissions  for  33  states in  the eastern
         United States for 1950 to 1978.

     2.  Air  Quality  Simulation  Models. The  SURE  inventory was  sponsored  by
         the  Electric Power Research  Institute as  a  point  and area  source
         SOX  inventory for  the eastern United  States  for 1977-78.  The  data
         were  compiled  to  reflect  spatial,  seasonal, and  temporal   source
         variabilities.  Similarly, Brookhaven  National  Laboratory compiled a
         national  inventory  of  criteria  pollutants  from 1978  to include
         selected Canadian emitters.

The EPA  and Environment Canada sponsored  a collaborative  effort  through the
Emissions,  Costs,  and  Engineering Assessment  Subgroup  (Work  Group  3B)  in
response  to the needs  identified  in  the  Memorandum of  Intent between  the
United  States  and Canada  on acidic  deposition.   The  inventory for  1980
presents  state-level  and provincial  summaries  of  SO? and  NOx f°r a^  area
and  point  source  categories.   The  inventory  will   oe  used   in  comparative
Lagrangian  transport  and transformation model   studies  by the United  States
and Canada.

The  Northeast  Corridor  Regional  Modeling  Program  (NECRMP)  inventory  is
perhaps  the  most sophisticated inventory  to  have  been developed  for modeling
purposes.   NECRMP  contains 1980  area  and point source emissions  of NOX and
hydrocarbons  for  a  13-state  area in   the northeastern  United States.  Area
sources  have  been  gridded to 20 x  20  km  resolution,  and  a  complex  data
handling  system  applies   seasonal  and  temporal  distribution  factors  to
                                   2-96

-------
emissions.   The inventory  is  to be used  as  Input to an  oxidant simulation
model for control strategy assessment.

Because  the  research  community is using many  of these inventories  to  study
acidic  deposition  from  various  perspectives,   it   is  essential  that  the
inventories  be consistent  and accurate.   The  National  Acid  Precipitation
Assessment Program (NAPAP) has established a Task Group on Man-made Emissions
(Task  Group  B).    The  primary   function  of  Task  Group  B  is  to  provide
quantitative  information  on  the  emissions  of  pollutants from  significant
manmade  sources in relevant  areas of  the United States  for  selected  time
periods.  Task Group B is responsible for four major objectives:

     1.  Quantify  emissions  of pollutants of  interest from  various  sources
         and regions at various times.

     2.  Provide economic, energy, and emissions information to  support NAPAP
         research areas.

     3.  Provide  data and  tools  to  assist policy  analysts in  other  task
         groups to identify and  assess cost-effective strategies to  control
         acidic precipitation.

     4.  Ensure  that  the  information  and  analytic  tools  used to  evaluate
         possible control strategies are accurate and available.

In response to the latter  objective, Task  Group  B  has undertaken  development
of a  coordinated emissions inventory  plan, which  embodies an  assessment  of
the  current  emissions   data  needs  for  transport/transformation  modeling,
source-receptor  modeling,  historical  studies  relating  to materials  damage
effects,  and  the  disaggregation  of manmade  sources  from  natural  sources.
Through  this  activity,  the 1980  U.S./Canada  inventory and the NECRMP  1980
inventory will be cross-checked  and  augmented  to provide a common  basis  for
acidic deposition modeling efforts.  A uniform historical  emissions  data base
will   also be  established for  use  in  supporting  retrospective  studies  of
materials damage.

2.3.9  The Potential  for Neutralization of  Atmospheric Acidity by
       Suspended Fly Ash'

Likens and  Bormann  (1974)  have  suggested  that increases  in  the  acidity  of
precipitation  in  the northeastern  United  States  have been  associated  with
augmented  use of  natural  gas  and  with  installation  of  particle-removal
devices  in  tall smoke  stacks.    They  have maintained that where the  major
source of anthropogenic  sulfur to the atmosphere  was  coal  combustion,  much  of
the  sulfur was  precipitated  to  the   land  near  the combustion  source  in
particulate form as neutralized salts.

The speculative conclusion by  Likens  and Bormann  is based  on their assumption
that fly  ash  is  a  highly reactive alkaline material.  Table  2-34  summarizes
approximate limits of  ash composition for various coals in  the United  States,
England,  and  Germany.   Examining Table  2-34  reveals that  the  potential  for
alkalinity of eastern  U.S.  bituminous coals is  associated  with their calcium,


                                   2-97

-------
i
VD
CO
                        TABLE 2-34.  APPROXIMATE  LIMITS OF  FLY  ASH COMPOSITION FOR VARIOUS COALS
                                               (GLOSKOTER ET  AL.  1977)
                                        Chemical  analysis, weight-percent of ash
                                    Fe2°3
                                          CaO
         MgO
            Na20
     British coal
 S03
American coals
Anthracite
B i tun i nous
Subbituminous
Lignite

48-68
7-68
17-58
6-40

25-44
4-39
4-35
4-26

2-10
2-44
3-19
1-34

1.0-2
0.5-4
0.6-2
0.0-0.8

0.1-4
0.0-3
0.0-3
0.0-1

0.2-4
0.7-36
2.2-52
12.4-52

0.2-1
0.1-4
0.5-8
2.8-14


0.2-3.0
_
0.2-28


0.2-4
_
0.1-1.3

0.1-1
0.1-32
3.0-16
8.3-32
      Britimrinous
25-50  20-40   0-30   0.0-3.0
1.0-10  0.5-5
           1.0-6
1.0-12
     German coal s
      Bituminous
      Brown
25-45  15-21  20-45
 7-46   6-29  17.26
2.0-4
4.0-43
0.5-1
0.9-4
4.0-10
2.0-22

-------
magnesium,  sodium,  and  potassium  content.   However,  it is also reported that
these  elements are  found  in  ash  samples  in  the  sulfate  form.    Aqueous
solutions  of  these salts  are neutral  and,  therefore,  should  exhibit  no
appreciable  scavenging  of  $03.   Newman  (1975)  has  also  pointed  out  the
inability of coal fly ash to neutralize SOg further in the atmosphere.

Therefore,  from available  data,  we  could conclude  that  the roles  of S02,
NOX,  and mineral   acid  emissions  from eastern  and  midwestern  coal-fired
sources  in  producing acidic precipitation are  not changed  significantly  by
incorporating  particulate  emissions controls such  as electrostatic  precip-
itators.  Even  if one could demonstrate a  minimal  effect of further reaction
of  combustion  particles with  $02  a*  atmospheric concentrations,  asserting
that eliminating all particulate controls would enhance neutralization of the
atmosphere  is misleading.   The absence of   controls  would result in  a con-
tinual massive  fallout  of  large particles  from each  combustion  source.  The
short  residence  time of these particles  in  the atmosphere  would exert  no
positive benefit on  air quality because their deposition  velocity  would  not
permit appreciable reaction with ambient S02-

The composition of oil  ashes differs significantly from  that  of coal.   Table
2-35 is a summary of the analysis  of a typical  residual oil-fired  power plant
fly  ash.    Water-soluble  sulfate,  carbon,  and  vanadium  are the  principal
components.  Vanadium is  a  characteristic  element present as a  porphyrin  in
Venezuelan  crude  oil.    This  particular  type  of  crude  serves  as the main
source  of  heavy  residual  and  base-hydrode sulfurized  residual  oils  for
fuel-firing in the  Northeast and  Gulf Coast areas.   Recent  studies  {Homolya
and Fortune  1978)  have  shown  that  ash emitted  from the combustion of these
oils is highly  acidic due  to  the absorption of  sulfuric acid on  the  carbo-
naceous oil ash  particles.   Table 2-36 compares total water-soluble  sulfate
and free sulfuric acid content of  particulate matter  collected from coal-  and
oil-fired boilers.  Oil  ash  samples are  found to  contain about 20 times more
water-soluble sulfate and about 10  times more free  sulfuric acid than does
ash from coal  combustion.

The implication of sulfate and sulfuric acid aerosols  as direct emissions  to
the acidification of  precipitation  is complex.   Coal  typically  contains  10
percent ash, but major combustion  sources employ  particulate  controls  such  as
electrostatic  precipitators   with  collection   efficiencies  exceeding   95
percent.  Residual  oils contain 0.05 percent ash;  therefore,  sources  burning
residuals generally have no  particulate controls  other than  perhaps  mechan-
ical collectors if the  power plant was of the type  converted  from  coal  to  oil
in  the mid-1960's.   The mean  aerodynamic  particle diameter  of  oil  ash  has
been measured  as 3 urn,  with  30  percent  weight of  the  ash  sized less than
0.5 urn (Boldt et al.  1980).   This  suggests  that mechanical   cyclones  remove
little material  and that material  emitted  to the atmosphere  is transportable
in  the same  air parcels  wherein  atmospheric  transformations   of  S02  and
NO-J occur.  Therefore,  it  is conceivable that the sulfuric acid  fraction  of
acidic  precipitation  consists of  a   mixture  of  primary   (particles and
condensed  ^$04  aerosols)  and   secondary  (atmospheric  oxidation of $02)
components of  varying  properties,  depending upon  the  origin,  season, and
transport time  of an air parcel and  the magnitude of  a precipitation event.
                                   2-99

-------
TABLE 2-35.  ANALYSIS OF A TYPICAL RESIDUAL OIL ASH
                (BOLDT ET AL. 1980)
Oil Ash Constituents
Water-soluble components
so^-
Cl-
NH4+
N03.,
Metal s
V
Na
Mg
Ni
Fe
K
Mn
Carbon
C
Mean
deviation
(wt. %)

47.5
1.1
0.7
0.1

5.4
3.7
3.2
1.3
0.3
0.1
0.02

38.1
101.5
Standard

9.1
1.5
0.5
0.03

1.2
1.5
1.1
0.3
0.2
0.1
0.01

6.3
                     2-100

-------
          TABLE  2-36.
SULFURIC ACID AND SULFATE CONTENT IN PARTICULATE MATTER COLLECTED FROM COAL- AND
          OIL-FIRED BOILERS  (HOMOLYA AND  FORTUNE 1978)
ro
Source of ash
Collection
site
Sulfur content
Wt %
Ash composition (dry basis)
Wt % H2S04
Wt %
total S04
A. Coal -fired boilers:
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
Wilmington, N.C.
Chapel Hill, N.C.
Moncure, N.C.
Kentucky, CR No. 4
Kentucky, CR No. 6
Kentucky, MC No. 1
Kentucky, MC No. 2
Ohio, PC
Kansas City, Mo.
Arizona, NFL
ESP
Stack
ESP
ESP
ESP
ESP
ESP
ESP
ESP
ESP
1.7
1.7
2.0
3.9
3.9
3.9
3.9
3.9
1.7
0.5
0.06
0.08
0.02
0.04
0.07
0.03
0.01
0.02
0.02
0.01
0.41
0.97
0.20
1.06
4.96
1.31
1.44
0.79
0.90
0.42
B. Oil-fired boilers:
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
Raleigh, N.C. —2nd week
Raleigh, N.C.— 4th week
Raleigh, N.C. —6th week
Raleigh, N.C.— 8th week
Anclote, Fla.
Nassau Co., N.Y.
Albany, N.Y., No. 1, 4/77
Albany, N.Y., No. 2, 4/77
Albany, N.Y., No. 1, 7/77
Albany, N.Y., No. 2, 7/77
Long Island, N.Y., No. 2
Long Island, N.Y., No. 3
Long Island, N.Y., No. 3
Stack
Stack
Stack
Stack
Stack
Cycl one
Cyclone
Cyclone
Cycl one
Cyclone
Air heater
Air heater
ESP
1.5
1.5
1.5
1.5
2.6
0.3
1.8
1.8
1.8
1.8
2.4
2.4
2.4
0.45
1.25
1.46
5.66
0.20
0.03
0.34
0.26
0.35
0.34
0.03
0.02
0.26
15.31
23.35
30.33
43.89
22.24
21.62
30.62
34.35
35.56
33.40
29.01
25.75
32.45
    ESP = Electrostatic  precipitator.

-------
Other  types  of  alkaline particulate  emissions may  have  an  effect on  the
deposition  of  acids.    For  example,   participate   emissions  from cement
manufacturing processes could  act  as  a neutralizing  sink in the  atmosphere.
However, no  assessments  have been performed  to examine the distribution  of
such sources and their emissions relative to historical  deposition patterns.

2.4  CONCLUSIONS (E. Robinson and J.  B.  Homolya)

The  review  of  natural  sources of  sulfur,  nitrogen  oxides,  ammonia,  and
chlorine compounds has been directed toward natural emissions  and  background
concentrations  of   those  compounds  that   may   have  direct   impacts   on
precipitation pH, more popularly known as  acid  rain.   The emphasis  has  been
on conditions that  relate to  the  northeastern  region of the  United States.
Within  the  definition  of  "natural"   sources  are  the  emissions  from  the
biosphere, which  include  biological  processes  on  land  and  in  the water,
volcanos,  oceanic  or   marine  sources,  atmospheric  processes   including
lightning, and,  in some cases, combustion of a nonindustrial  nature.

The  most  important  conclusions  for  this  assessment  appear   to   be  the
following:

    o   Present  evidence does  not  show  that  natural  sources  of sulfur
        compounds   are   significant    contributors    to    excessively    low
        precipitation pH  when compared  to anthropogenic  sources  (Sections
        2.2.1 and 2.3.1).

    0   On a quantitative basis and for the area of the United  States east of
        the  Mississippi   River,  soil-generated  natural  sources  of sulfur
        compounds are estimated  to  total about 0.07  Tg S yr-1.   Thus,  less
        than 1 percent of the sulfur compound  emissions in this regional  area
        seem to be  due to natural sources,  even though this  natural source
        estimate might vary by a factor of 2  or  3  (Section 2.2.1.3).

    o   Natural  emissions of  nitrogen  oxides  (NOX)   a/e  primarily  due  to
        processes in  the biosphere,  although these  emissions  are much  less
        well  known than the natural  sulfur compounds  (Section 2.2.2.1).

    o   NOX  from natural  sources in  the area east of the Mississippi  River
        have been  estimated  to  be in  the  range of  0.04  to  1.5  Tg N  yr-1
        with values from  the  lower  part of the  range being the more recent
        and  more  likely correct ones.   These  estimates  should  be  compared
        with estimated  anthropogenic  NOx emissions in  1978  of about 8.9  Tg
        N  yr-1  from this  same area.   Thus,  perhaps only  a few percent  of
        the  NOX  contribution  to  acid precipitation  may  be due to  natural
        NOX sources (Sections 2.2.2.6, 2.2.2.13, and  2.2.6).

    0   Ammonia,  when  incorporated  into  precipitation,  tends  to  counter-
        balance  the effects of acidic  compounds such as sulfates,  nitrates,
        and chlorides.  Most  of  the ammonium  compounds  in the  atmosphere and
        thus  in  precipitation  are due  to  nonindustrial  sources  (Section
        2.2.2.7).
                                   2-102

-------
        Biogenic  sources of  ammonium  compounds in  the  area east  of
        Mississippi  River  are  estimated to  be about  0.3  Tg N  yr~l,
                                                                  the
                                                                  but
certainly a  factor of  2  or more  must be induced  in this  estimate
(Sections 2.2.2.9 and  2.2.2.13).
    0   Chloride  compounds  may   also  contribute   to   acidic  values   of
        precipitation  pH.    Anthropogenic  sources  of chlorine or  chloride
        compounds  are believed  to  be  small  relative  to  natural   sources
        (Section 2.2.3.1).

    0   Natural  chlorine sources  affecting the  eastern  United  States  are
        almost total ly--99  percent  or more—due  to  oceanic area  processes.
        These  mainly  involve the  generation of  sea  salt  aerosol  particles
        (Section 2.2.3.2).

    o   The total natural chlorine  compound deposition affecting the  United
        States east of  the  Mississippi  River  is about  0.9  (0.4  x 2.34  x
        1012)  Tg Cl  yr'1, mostly sea salt (Sections  2.2.3.5 and  2.2.6).

    o   Fugitive dust  concentrations  in  rural  and more  remote  locations  in
        the northeastern region  are relatively  low (Section 2.2.6).

Thus, in  areas where  the acidity  of precipitation occurs  outside the  normal
range  of  variations   and   where  ecological  impacts  are  suspected  to   be
occurring,  it  seems very unlikely  that  the products  of natural  sources  of
acidic material are significant  factors (Section  2.2.5).

A review of the  historical  anthropogenic  emissions  in the  United States  and
Canada from 1950 to about 1980  identified the following trends:

(1)  Sulfur Dioxide (Section 2.3.2.1)

    0   Total  emissions  in  the  eastern  United States doubled  from 1950  to
        1980 with a peak in 1970.   Emissions  in 1980 were about 9  percent
        less than those in  1970.

    0   Electric  utility contributions  tripled  over  this  period.
        Highest SOe  emissions occur  in  the  Midwest.   Within  the 31-state
        region, the five highest  levels  of  estimated SO?  emissions for 1980
        occurred  in   Ohio,   Indiana,  Pennsylvania,  Illinois,  and Missouri
        (Table 2-14).

        The largest increases in  S02  emissions  over this  period occurred  in
        the Southeast,  where nearly 90  percent of  the total  sulfur oxides
        emitted are  attributed to  electric  utilities  and  industrial  fuel
        combustion sources.

        Changes in fuels from coal  to  oil  reduced  emissions  in New  England  by
        20 percent.   These   reductions in  SOg  emissions  ocurred  during the
        mid-  to late-1960s.
                                   2-103

-------
    0    Estimates  of Canadian SOg emissions  indicate  a 20 percent  increase
        from 1955  to 1976 (Section 2.3.6).  There  was  a marked increase in
        S02  emissions in Canada between 1955  and 1965  of about 44  percent.

    0    Copper and nickel smelters  represent  the major Canadian S02  source
        category,  with  most  point  sources  located in eastern Canada.

(2)   Primary Sulfate (Section  2.3.2.2)

    o    Sulfate  emission   factors   were   significantly  larger   for  oil
        combustion than  for  coal.   Primary  sulfate  emission  factors for
        industrial  and  residential   oil   combustion  were  larger  than for
        utility oil  combustion.

    0    The  highest primary sulfate emission densities occur in New England
        and   the  Atlantic  seaboard.    Emissions  from  nonutility  sources
        concentrated in metropolitan areas may be  significant during  winter
        months because  of space-heating.

    o    Primary sulfate emissions  increased in the Midwest in proportion to
        increases  in coal  consumption.

(3)   Nitrogen Oxides (Section  2.3.3)

    0    Total emissions in the  eastern  United States  increased by  a  factor
        of 2.4 from 1950 to  1980 with a  peak  in  1978.

    o    Electric utilities and highway  vehicles  are the largest  contributors
        to NOX.

    0    Highest NOX  emissions  densities  occur   in  the northeastern  United
        States and are  influenced  by highway  vehicles.

    °    Coal-fired utilities  significantly affect  the  NOX emissions  in the
        Midwest.

    0    Canadian  NOX  emissions  tripled  between  1955  and  1976   (Section
        2.3.6).


(4)   Hydrochloric Acid  (Section 2.3.4)

    0    Coal combustion represents the  major  HC1  emitter.

    0    Midwestern coals contain the highest  chloride levels.

    0    Mass emissions of HC1  from major  coal-consuming states  are  equal to
        or  greater  than corresponding  primary  sulfate  emissions.   Because
        chloride is  emitted as  free HC1  and primary sulfate may consist of
        free  H2S04  and  sulfated   ash,  their  relative  contribution  to
        acidity patterns  is unclear.  A detailed  analysis of  precipitation
                                   2-104

-------
        chemistry  data is  needed  to  discern  local   deposition  of  HC1   in
        precipitation samples.

(5)  Arsenic, Mercury, and Vanadium (Section 2.3.5)

    0   Arsenic and  mercury  are emitted  from coal  combustion.   Mercury  is
        emitted  in  the  vapor  phase  and  is  not  collected  efficiently   by
        particulate emissions controls.

    o   Implementing particulate controls  reduced  arsenic  emissions  in  the
        eastern United States, but mercury  emissions increased  in proportion
        to coal  consumption.

    o   Vanadium is emitted from residual  oil  combustion  in  varying amounts.

    o   Highest vanadium emissions  occur  in  the  northeastern United States.

(6)  Acid Neutralization in the  Atmosphere by  Fly Ash or  Alkaline Particles
     (Section 2.3.9)

    0   Available data on  the chemical analysis of  fly ash from coal or  oil
        combustion indicate these  materials are either  neutral  or slightly
        acidic.   The capacity of fly ash for  neutralizing acidic aerosols in
        the atmosphere is  not apparent.

    o   Data is lacking on neutralization capacity  of other particles (e.g.,
        cement dust) which should be alkaline.
                                  2-105

-------
2.5  REFERENCES

Adams, D. F.,  S. 0.  Farwell,  E.  Robinson,  and M. R. Pack.   1980.   Biogenic
sulfur  emissions  in  the SURE  region.    Final   Report  by Washington  State
University for Electric  Power  Research Institute, EPRI Report No. EA-1516.

Adams, D. F., S.  0.  Farwell,  E.  Robinson,  M.  R.  Pack, and W.  L. Bamesberger.
1981a.   Biogenic sulfur  source  strengths.  Presented at 74th Annual  Meeting
Air Pollution Control Assoc.,  Philadelphia, PA,  June 21-26,  1981.   Paper No.
81-153.

Adams, D. F.,  S. 0.  Farwell,  M.  R.  Pack,  and E. Robinson.  1981b.   Biogenic
sulfur gas emissions  from  soils in eastern  and  southeastern United  States.
J. Air Pollut. Contr. Assoc. 31:1083-1089.

Adams, D. F.,  S. 0.  Farwell, E. Robinson, M. R.  Pack, and W. L. Bamesberger.
1981c.    Biogenic   sulfur   source   strengths.      Environ.   Sci.   Technol.
15:1493-1498.

Altshuller,  A. P.   1958.    Natural  sources of  gaseous pollutants  in the
atmosphere.   Tell us 10:479-492.

Altshuller,  A. P.    1979.    Model  predictions  of  the  rates  of homogeneous
oxidation of  sulfur  dioxide to  sulfate  in  the troposphere.   Atmos. Environ.
13:1653-1661.

Aneja, V. P., J. H. Overton, and  A.  P. Aneja.   1981.    Emission  survey  of
biogenic sulfur flux from terrestrial  surfaces.   J.  Air  Pollut. Contr. Assoc.
31:256-258.

Ayers, G.  P. and J. L.  Gras.   1980.   Ammonia  gas  concentrations  over the
Southern Ocean.  Nature 284:539-540.

Bartels,  0.  G.     1972.    An  estimate of  volcanic  contributions   to the
atmosphere and volcanic  gases and  sublimates as the source of radioisotopes
10Be, 35S, 32P, and 22Na.  Health Phys.  22:387-392.

Barone,  J.  B., L.  L. Ashbaugh,  B.  H. Kusko, and T.  A. Cahill.   1981.  The
effect of Owens Dry Lake on air  quality  in  the Owens Valley with implications
for  the Mono  Lake  area, pp.  327-346.   In  Atmospheric Aerosol, Source/Air
Quality  Relationships.   E.  S. Macias and~~P. K. Hopke,  eds.   ACS Symposium
Series Number  167.   American Chemical  Society, Washington, D.C.  359 pp.

Bates,  D.  R.  and  P. B. Hays.   1967.  Atmospheric  nitrous  oxide.    Planet.
Space Sci. 15:189-197.

Boldt, K.  R., C. P. Chany, E.  J.  Kaplin, J. M.  Stansfield, and B.  Webber.
1980.  Impact  of a primary emission source on air quality. EPA-600/2-80-109,
U.S.  Environmental  Protection Agency,  Research Triangle  Park,  NC.
                                   2-106

-------
Braman, R. S. and T. J. Shelley.  1981.  Gaseous and particulate ammonia and
nitric acid concentrations:   Columbus,  Ohio area - Summer 1980.   Report No.
EPA-600/57-80-179,  Environmental  Protection Agency, Research  Triangle  Park,
NC, February 1981.   (Available as PB 81-125007 from NTIS).

Bulla, L. A., Jr.,  C.  M.  Gilmour,  and  W. B. Bollen.  1970.   Non-biological
reduction of nitrite in soil.  Nature 225:664.

Butcher, S.  S.  and R. J. Charlson.   1972.  An  Introduction  to Atmospheric
Chemistry.  Academic Press,  New York.

Cadle, R. D.  1975.  Volcanic emissions of chlorides and sulfur compounds to
the troposphere and stratosphere.  J. Geophys. Res. 80:1650-1652.

Cadle R. D.  1980.   A comparison of volcanic with other fluxes of atmospheric
trace gas constituents.  Rev. Geophys.  Space Phys. 18:746-752.

Cadle,  R.  D.,  W.  H.  Fisher,  E.  R.   Frank,  and J.  P.  Lodge,  Jr.    1968.
Particles in the Antarctic atmosphere.  J.  Atmos. Sci. 25:100-103.

Charlson, R.  J.,  and H.  Rodhe.   1982.  Factors controlling  the  acidity of
natural rainwater.   Nature 295:683-685.

Cheney, J.  L. and J.  B.  Homolya.   1978.   Workshop  proceeding  on  primary
sulfate  emissions  from the  combustion of  fossil  fuels.   EPA-600/9-78-020a.
U.  S.   Environmental   Protection   Agency,  Research  Triangle   Park,   NC.
pp. 53-63.

Cicerone, R.  J.    1981.   Halogens in  the atmosphere.    Rev.  Geophys.  Space
Phys. 19:123-139.

Cogbill,  C.  V.  and  G.  E.  Likens.    1974.     Acid  precipitation  in  the
northeastern United States.   Water Resources Res. 10:1133-1137.

Conway, E. J.  1942.  Mean geochemical  data in relation to oceanic evolution.
Royal Irish Acad.  Proc. A48:119-159.

Cooper, J. A. and J. G. Watson, Jr.   1980.  Receptor oriented methods of air
particulate   source   apportionment.      J.   Air   Pollut.    Contr.   Assoc.
30:1116-1125.

Cox, R. A. and F. J. Sandalls.  1974.   The  photooxidation of hydrogen sulfide
and dimethyl sulfide in air.  Atmos. Environ. 8:1269-1281.

Cronn,  D.   R.,  R.   A.  Rasmussen,  E.   Robinson,  and D.  E.  Harsch.    1977.
Halogenated compound identification and measurement in  the  trosposphere and
lower stratosphere.  J. Geophys. Res. 82:5935-5944.

Crutzen, P. J.  1974.   Photochemical  reactions  initiated by and influencing
ozone in unpolluted tropospheric air.   Tellus 26:47-57.
                                  2-107

-------
Dawson, G. A.  1977.   Atmospheric  ammonia from undisturbed land.  J. Geophys.
Res. 82:3125-3133.

Dawson,  G.  A.   1980.   Nitrogen  fixation  by  lightning.    J.  Atmos.  Sci.
37:174-178.

Delmas,  R.,  J.  Baudet, J.  Servant,  and Y.  Baziard.   1980.   Emissions and
concentrations of hydrogen  sulfide in the air of the  tropical  forest  of the
Ivory  Coast  and  of  temperate  regions  in  France.    J.  Geophys.  Res.
85:4468-4474.

Denmead, 0. T., J. R.  Simpson,  and J.  R. Freney.  1974.   Ammonia  flux into
the atmosphere from a grazed sheep pasture.   Science 185:609-610.

Denmead, 0.  T.,  J. R. Freney,  and J.  R. Simpson.   1976.   A closed ammonia
cycle within a plant  canopy.   Soil  Biol.  Biochem. 8:161-164.

Drapcho, D.  L.,  D. Sisterson,  and  R.  Kumar.   1983.   Nitrogen  fixation  by
lightning activity in a thunderstorm.  Atmos. Environ. 17:729-734.

Duce, R. A.  1969.   The source of gaseous chlorine  in the marine atmosphere.
J. Geophys. Res.  74:4597-4599.

Eriksson, E.   1952.    Composition of atmospheric  precipitation  I.  Nitrogen
compounds.  Tell us 6:261-267.

Eriksson, E.  1959.  The yearly  circulation of chloride and sulfur in nature;
meteorological,  geochemical, and pedological  implications.   Part I.   Tellus
11:375-403.

Eriksson, E.  1960.  The yearly  circulation of chloride and sulfur in nature;
meteorological,  geochemical, and pedological  implications.   Part II.   Tellus
12:63-109.

Eriksson, E.  1963.  The yearly  circulation of sulfur  in nature.  J. Geophys.
Res. 68:4001-4008.

Farwell, S.  0.,  S. J. Gluck, W.  L.  Bamesberger, T.  M.  Schutte, and  D.  F.
Adams.   1979.   Determination of  sulfur-containing  gases by  a deactivated
cryogenic enrichment and capillary gas  chromatographic system.   Anal. Chem.
51:609-615.

Fishman, J.   1981.   The  distribution   of NOX  and  the production  of  ozone:
comments on "origin of tropospheric ozone"  by S. C. Liu et  al.   J.  Geophys.
Res. 86:12161-12164.

Friedlander, S.  K.  1981.  New  developments  in receptor modeling theory, pp.
1-20.    In  atmospheric Aerosol,  Source/Air  Quality  Relationships.    E.  S.
Macias aTuTP. K.  Hopke, eds.  ACS  Symposium Series No. 167, American Chemical
Society, Washington,  D.C. 359  p.
                                   2-108

-------
Friend, J. F. 1973.   The  global  sulfur  cycle,  pp.  177-201.   In  Chemistry of
the Lower Atmosphere.   S.  I. Rasool, ed.  Plenum Press, New Yo"?k".

Galbally,  I.  E.  1975.   Emission  of  oxides of  nitrogen  (NOX)  and  ammonia
from the earth's  surface.   Tell us 27:67-70.

Galbally, I. E. and C.  R.  Roy.   1978.   Loss of fixed nitrogen  from soils by
nitric oxide exhalation.  Nature 275:734-735.

Galloway, J. N.  and D. M.  Whelpdale.  1980.  An atmospheric sulfur budget for
eastern North America.  Atmos.  Environ.  14:409-417.

Galloway, J. N.,  G.  E. Likens, W.  C. Keene,  and J. M. Miller.   1982.   The
composition of precipitation  in remote areas of the world.   J.  Geophys.  Res.
87:8771-8786.

Gandrud, B. W.  and A.  L. Lazrus.  1981.  Filter measurements of stratospheric
sulfate  and  chloride   in  the   eruption  plume  of  Mt. St.  Helens.   Science
211:826-827.

Garland, J. A. and J. R. Branson.  1976.  The mixing height and mass balance
of S02 in the atmosphere above  Great Britain.  Atmos. Environ. 10:353-362.

Georgii, H. W.  1963.   Oxides of nitrogen and  ammonia  in  the atmosphere. J.
Geophys. Res. 68:3963-3970.

Georgii, H.  W.  and G.  Gravenhorst.  1977.   The ocean as  source  or  sink of
reactive trace-gases.   Pur. and Appl. Geophys. (PAGEOPH) 115:503-511.

Gluskoter, H. J.,  R. R. Ruch, W.  G. Miller,  R.  A. Chahill,  G. B. Dreher, and
J.  K.  Kuhn.   1977.   Trace elements  in coal, occurrence  and  distribution.
EPA-600/7-77-064.   U. S.  Environmental  Protection  Agency,  Research Triangle
Park, NC.

Graedel, T. E.   1977.   The homogeneous chemistry of atmospheric sulfur.  Rev.
Geophys. Space Phys. 15:421-428.

Graedel, T. E.   1978.   Chemical  Compounds in the Atmosphere.  Academic Press,
New York.  440 pp.

Graedel, T.  E.   1979.   The kinetic photochemistry of the marine atmosphere.
J. Geophys. Res.  84:273-286.

Granat,  L.,  R. 0.  Hallberg, and H. Rodhe.   1976.  The global  sulfur cycle,
pp.  89-134.   In  Nitrogen, Phosphorus  and  Sulfur -  Global  Cycles.   B. H.
Svensson and R7~Soderlund, eds.   Ecological  Bulletins No. 22, SCOPE Report 7,
Royal Swedish Academy of Sciences,  Stockholm.
                                   2-109

-------
Gschwandtner,  G.,  C.  0.  Mann,  B.  C.  Jordan,  and  J.  C.  Bosch.    1981.
Historical emissions  of sulfur  and  nitrogen oxides  in  the  Eastern  United
States  by state and  county.   Presented  at the 74th  Annual Meeting,  Air
Pollution Control  Association,  Philadelphia,  Pennsylvania,  June 21-26,  1981.
Paper 81-30.1.

Harriss,  R.  C.  and J. T. Michaels.  1982.   Sources  of atmospheric ammonia.
Proceedings Second Symposium,  Composition  of the Nonurban  Troposphere,  pp.
33-35, American Meteorological  Society, Boston, Mass.

Hidy, G.  M.   1982.   Bridging  the  gap  between  air quality  and precipitation
chemistry.  Water,  Air,  Soil  Pollut. 18:191-198.

Hill, R.  D.,  R.  G. Rinker, and  H. D.  Wilson.   1980.   Atmospheric nitrogen
fixation by lightning.  J.  Atmos. Sci. 37:179-192.

Hitchcock, D. R.  1975.   Dimethyl sulfide emissions to the global  atmosphere.
Chemosphere 4:137-138.

Hitchcock, D. R.   1976.   Atmospheric  sulfates  from biological sources.   J.
Air Pollut. Contr.  Assoc. 26:210-215.

Hobbs,  P.  V., J.  P.  Tuell,  L. F. Radke,  D. A.  Hegg,  and M. W.  Eltgroth.
1982.    Particles  and  gases  in the  emissions  from  the  1980-81  volcanic
eruptions of Mt. St.  Helens.  J.  Geophys. Res. 87:11062-11086.

Hoell,  J.  M., C.  N.  Harward,  and  B.  S.  Williams.   1980.   Remote infrared
heterodyne radiometer  measurements  of atmospheric ammonia profiles.  Geophys.
Res. Lett. 7:313-316.

Homolya, J.  B.  and C. R.  Fortune.   1978.    The measurement  of  the sulfuric
acid and  sulfate content of particulate matter  resulting from the combustion
of coal  and oil.  Atmos. Environ. 12:2511-2514.

Homolya,  J.  B.  and J.  L.  Cheney.    1978.    Workshop  proceedings  on  primary
sulfate emissions  from  the combustion of fossil  fuels.   EPA-600/9- 78-020b,
U.S. Environmental  Protection Agency, Research Triangle Park, NC.   pp. 3-13.

Homolya, J. B. and S. Lambert.   1981.  Characterization of sulfate emissions
from nonutility boilers  firing  low-S  residual oils  in  New York City.  J.  Air
Pollut. Contr. Assoc.  31:139-143.

Husar, R. B., D. E. Patterson,  J. D. Husar,  N. V. Gillani, and W. E.  Wilson,
Jr.  1978.  Sulfur  budget of  a  power plant.   Atmos. Environ. 12:549-568.

Hutchinson, G. L., R. J. Millington, and D.  B. Peters.  1972.   Atmospheric
ammonia:  Absorption by  plant leaves.  Science 175:771-772.

Jaeschke,  W.,  H-W.     Georgii,  H.  Claude,  and   H.  Malewski.     1978.
Contributions  of   H2$  to  the  atmospheric  sulfur  cycle.    Pure  and  Appl.
Geophys. 116:465-475.
                                  2-110

-------
Johannes,   A.   H.,   E.   R.   Altwicker,   and  N.   L.   Clesceri.     1981.
Characterization of  acidic  precipitation  in the  Adirondack  region.   Final
EPRI  Research  Project 1155-1,  EPRI Report  EA-1826,  Electric  Power  Research
Institute, Palo Alto, California.

Johnston,  D.  A.     1980.     Volcanic  contribution  of   chlorine   to  the
stratosphere:  More significant to ozone than previously estimated?   Science
209:491-493.

Johnston, H. S.,  0.  Serang, and J. Podolski.   1979.   Instantaneous  global
nitrous oxide photochemical  rates.  J. Geophys. Res. 84:5077-5082.

Junge,  C.  E.   1956.    Recent  investigations of  air  chemistry.    Tellus
8:127-130.

Junge, C. E.  1958.   The  distribution of  ammonia  and nitrate in  rain  water
over the United States. Trans  Am.  Geophys.  Union 39:241-248.

Junge, C. E.  1960.  Sulfur  in  the  atmosphere.  J. Geophys. Res. 65:227-237.

Junge,  C.  E.   1963.   Atmospheric Chemistry  and Radioactivity.   Academic
Press, New York, 382  pp.

Junge,  C.  E.,  and P.  E.  Gustafson.   1956.    Precipitation sampling  for
chemical analysis.  Bull.  Am. Meteorol.  Soc., 37:244-245.

Junge, C. E. and R. T. Werby.  1958.  The concentration of chloride, sodium,
potassium, calcium and sulfate in rain water over  the  United  States.   J.
Meteorol.  15:417-425.

Junge,  C.  E.,  C. W. Chagnon,  and  J.  W.  Manson.    1961.    Stratospheric
aerosols.  J. Meteorol. 18:81-108.

Keeney, D. R., I. R.  Filbery, and G. P. Marx.  1979.   Effect of temperature
on the gaseous nitrogen products of dentrification in a silt loam soil.  Soil
Sci. Soc. Am. J. 43:1124-1128.

Kellogg, W.  W., R. D. Cadle, E. R.  Allen,  A. L. Lazrus,  E.  A.  Martel.   1972.
The sulfur cycle.  Science 175:587-596.

Kelly,  T.  J.,  D.  H. Stedman,  J. A.  Ritter,  and R.  B.  Harvey.    1980.
Measurements of oxides of  nitrogen  and nitric acid in clean air.  J.  Geophys.
Res. 85:7417-7425.

Kerr, R. A.   1982.  El Chichon  forebodes climate change.  Science 217:1023.

Kim, C. M.  1973.  Influence of vegetation  types on the intensity of ammonia
and nitrogen dioxide  liberation from soil.   Soil Biol. Biochem. 5:163-166.
                                  2-111

-------
Klemm, H.  A.  and R. J.  Brennan.    1979.   Emissions inventory for the  SURE
region.  EPRI Report EA-1913, Electric Power Research  Institute,  Palo Alto,
California.

Kley, D., J. W.  Drummond, M.  McFarland, and S.  C.  Liu.   1981.   Tropospheric
profiles of NOX.  J.  Geophys.  Res. 86:3153-3161.

Kowalczyk,  6.  S., G.  E. Gordon, and  S. W. Rheingrover.  1982.  Identification
of  atmospheric   particulate   sources  in  Washington,  D.C.,  using chemical
element balances. Environ. Sci. Technol. 16:79-90.

Lawson, D.  R. and J. W. Winchester.   1979.  A standard crystal aerosol  as  a
reference for elemental enrichment factors.  Atmos. Environ.  13:925-930.

Lazrus, A.  L.,  H. W.  Boynton,  and J. P. Lodge, Jr.  1970.   Trace constituents
in oceanic  cloud  water  and their origin.  Tellus 22:106-114.

Lazrus, A.  L., R. D. Cadle,  B.  W.  Gandrud,  J.  P. Greenberg, B. J. Heubert,
and W. I. Rose, Jr.   1979.   Sulfur  and halogen chemistry of the stratosphere
and of volcanic eruption  plumes.  J. Geophys. Res. 84:7869-7875.

Levine, J.  S., R. S. Rogoruski,  G.  L. Gregory,  W. E. Howell  and J. Fishman.
1981.    Simultaneous measurements  of  NOX,  NO,   and  03  production  in  a
laboratory   discharge:    Atmospheric  implications.    Geophys.  Res. Lett.
8:357-360.

Likens, G.  E.   1976.  Acid precipitation.  Chem. Eng. News 54:29-44.

Likens, G.  E.  and  F.  H.  Bormann.    1974.   Acid  rain:   A  serious regional
environmental  problem.  Science 184:1176-1179.

Likens, G.  E.  and T.  J. Butler.  1981.  Recent acidification of precipitation
in North America. Atmos. Environ. 15:1103-1109.

Likens, G.  E., F. H. Bormann, J. S. Eaton, R.  S.  Pierce,  and N.  M. Johnson.
1976.   Hydrogen  ion input  to the  Hubbard  Brook  Experimental  Forest,   New
Hampshire during  the  last decade.  Water, Air, Soil Pollut. 6:435-445.

Lodge, J. P. Jr. and J.  B.  Pate.   1966.   Atmospheric gases and particulates
in Panama.   Science  153:408-410.

Lodge, J. P., Jr.,  A. J. MacDonald, Jr.,  and E. Vihman.   1960.   A study of
the composition of marine atmosphere.  Tellus 12:184-187.

Logan, J. A.  1983.   Nitrogen  oxides in the troposphere:  Global and regional
budgets. J. Geophys. Res.:   In press.

Lovelock, J.  E., R. J.  Maggs,  and R.  A.  Rasmussen.    1972.    Atmospheric
dimethyl sulfide  and  the  natural sulfur cycle.  Nature 237:452-453.
                                  2-112

-------
Makarov, B. N.  1969.   Liberation of nitrogen  dioxide from soil (in Russian).
Pochvovedeniye,  49-53.     (Translation  from  Russian   available  in:    Soil
Chemistry).

Maroulis, P.  J.,  and  A. R.  Bandy.   1977.   Estimate of the contribution of
biologically produced dimethyl sulfide to  the global sulfur cycle.   Science
196:647-648.

McConnel, J. C.  1973.  Atmospheric  ammonia.   J. Geophys. Res. 78:7812-7821.

Naughton, J.  J., V.  Lewis, D.  Thomas,  and J. B.  Finlayson.   1975.   Fume
compositions found at various stages of activity at Kilavea volcano, Hawaii.
J. Geophys. Res. 80:2963-2966.

Nelson,  D.  W.  and J.  M.  Bremner.   1970.   Gaseous  products  of  nitrite
decomposition in soils.   Soil  Biol.  Biochem. 2:203-215.

Newman,  L.     1975.    Correspondence:    Alkalinity of  fly ash.    Science
185:957-959.

Noxon, J.  F.   1976.  Atmospheric nitrogen fixation by lightning.   Geophys.
Res. Lett.  3:463-465.

Noxon, J.F.  1978.  Tropospheric  N02-  J- Geophys. Res. 83:3051-3057.

Ontario Research Foundation.  1975.   A nationwide inventory of anthropogenic
sources  and  emissions  of  primary   fine   particulate  matter.    Unpublished
Document.

Pack, D. H.   1980.   Precipitation  chemistry  patterns:   A  two-network data
set.  Science 108:1143-1145.

Palmer,  T.  Y.   1976.   Combustion sources of atmospheric chlorine.   Nature
263:44-46.

Perhac,  R.  M.   1978.   Sulfate  regional  experiment in  northeastern  United
States:  The "SURE"  program.  Atmos.  Environ.  12:641-647.

Pierson, W.,  W. Brachaczek, T.  Truex,  J.  Butler,  and T.   Korniski.   1980.
Ambient  sulfate measurements  on Allegheny  mountain  and   the  question  of
atmospheric   sulfate   in   Northeastern   United  States.     In   Aerosols:
Anthropogenic and  Natural,  Sources  and Transport.   T. J.  Kneip  and  P.  J.
Lioy, eds.   Ann. N.Y.  Acad. Sci.  338-145-173.

Pollack, J.  B.   1981.   Measurements  of  the volcanic plumes of Mount St.
Helens  in  the  stratosphere  and  troposphere:     Introduction.     Science
211:815-816.

Porter, L.  K., F. G. Viets, Jr.,  and  G. L.  Hutchinson.  1972.  Air containing
nitrogen-15  ammonia:     Foliar  absorption  by  corn   seedlings.     Science
175:759-761.
                                  2-113

-------
Rasmussen, R. A.,  L.  E.  Rasmussen,  M. A. Khali!, and R.  W.  Dalluge.   1980.
Concentration distribution  of methyl chloride in the atmosphere.  J. Geophys.
Res. 85:7350-7356.

Rasmussen, R.  A.,  M.  A.  K. Khalil,  R.  W. Dalluge,  S.  A. Penkett,  and B.
Jones.  1982.   Carbonyl  sulfide  and carbon disulfice from the  eruptions of
Mount St.  Helens.   Science  215:665-667.

Ratner, B.   1957.   Upper-air  climatology of the United States.  Tech. Paper
No. 32, U.S.  Weather Bureau, U.S. Dept. of Commerce, Washington, D.C.

Ravishankara, R. A., N. M.  Kreutter, R. C. Shah, and P. H. Wine.  1980.  Rate
of reaction of OH with COS. Geophys.  Res. Lett. 7:861-864.

Reiter, R. and  M.  Reiter.   1958.   Relations between the contents of nitrate
and  nitrite  ions  in  precipitation  and  simultaneous atmospheric  electric
processes, pp.  175.   j£ Recent Advances in Atmospheric  Electricity.   L. G.
Smith, ed.  Pergamon Press, London.

Rice,  H., D.   H.  Nochumson,   and  G.  M.  Hidy.    1981.    Contribution  of
anthropogenic and  natural   sources  to atmospheric  sulfur in  parts  of  the
United States.   Atmos. Environ. 15:1-9.

Richardson,  C.  J.  and G.  E.   Merva.   1976.    The  chemical  composition  of
atmospheric  precipitation  from  selected  stations in  Michigan.   Water, Air,
Soil Pollut.  6:385-393.

Robbins, R.  C., R.  D.  Cadle,  and  D. L. Eckhardt.   1959.   The  conversion of
sodium  chloride to  hydrogen   chloride  in  the  atmosphere.    J.  Meteorol.
16:53-56.

Robinson,  E.  and R. C. Robbins.  1970a.  Gaseous sulfur pollutants from urban
and natural sources.  J.  Air Pollut. Contr. Assoc. 20:233-235.

Robinson,  E.  and R. C. Robbins.  1970b.  Gaseous nitrogen compound pollutants
from urban and  natural  sources.  J.  Air Pollut. Contr. Assoc.  20:303-306.

Rodhe, H.  and I. Isaksen.   1980.   Global  distribution of sulfur compounds in
the troposphere estimated in a height/latitude  transport model.  J. Geophys.
Res. 85:7401-7409.

Ryan, J.  A.  and N. R. Mukherjee.   1975.  Sources  of stratospheric gaseous
chlorine.   Rev.  Geophys.  Space  Phys. 13:650-658.

Shannon,  J.  D.   1979.   The  advanced  statistical  trajectory  regional  air
pollution  model.  Argonne National  Laboratory  Radiological  and Environmental
Research Division  Topical Report ANL/RER-79-1.   pp. 1-34.

Shannon,  J.  D., J.  B.  Homolya,  and  J.  L. Cheney.    1980.    The relative
importance   of  primary   vs.   secondary  sulfate.     EPA   Project  Report
IAG-AD-89-F-1-116-0.
                                  2-114

-------
Smith,  C.  J.  and  P. M.  Chalk.   1980.    Gaseous  nitrogen evolution during
nitrification of ammonia fertilizer and nitrite  transformation  in soils. Soil
Sci. Soc. Am. J. 44:277-282.

Soderlund, R.  and  B. H.  Svensson.   1976.   The global  nitrogen  cycle, pp.
22-73.   In Nitrogen,  Phosphorous  and  Sulfur-Global  Cycles.  SCOPE Report 7,
EcologicaT  Bulletins No.  22.    Swedish  Natural  Science  Research Council,
Stockholm.

Spirtas,  R.,  and  H.  J.  Levin.    1970.    Characteristics  of  particulate
patterns.  National Air Pollution  Control  Admin. Publ  AP-61, Raleigh, NC.

Stahl,  Q.  R.  1969.   Preliminary air pollution survey of hydrochloric acid
APTD-69-36.  U. S.  Department of  Health, Education and Welfare, National Air
Pollution Control  Administration,  Cincinnati, Ohio.

Stensland, G. J.,  and R. G. Semonin.  1982.   Another  interpretation of the pH
trend in the United States.  Bull. Am. Meteorol. Soc.  63:1277-1284.

Stith, J. L.,  P. V.  Hobbs, and L.  F.  Radke.   1978.   Airborne  particles and
gas  measurements   in  the  emissions  from  six volcanoes.   J.  Geophys.  Res.
83:4009-4017.

Stoiber, R.  E.  and A.  Jepsen.   1973.  Sulfur  dioxide  contributions  to the
atmosphere by volcanoes.  Science  182:577-578.

Sze,  N. D.  and M.  K.   W.  Ko.   1980.   Photochemistry  of  COS,  CS2,  CHs,
SCH3,  and  HgS:    Implications  for the  atmospheric   sulfur  cycle.   Atmos.
Environ. 14:1223-1239.

U.S. Department of Commerce.   1968.   Climatic Atlas of  the  United States.
Environmental Data  Service, Washington, D.C.

U.S. Environmental   Protection Agency.  1973.   The  National  Air Monitoring
Program:   Air Quality  and  Emissions Trends,  Annual   Report,  Volume  II,
EPA-450/l-73-001-b,  Office  of Air  and   Water  Programs,  USEPA,  Research
Triangle Park, NC.

U.S. Environmental  Protection Agency.   1978.  National air  pollution emission
estimates, 1940-1976.  EPA-450/1-78-003.

U.S. Environmental  Protection  Agency.  1981.   Compilation of  air pollution
emission factors.   2nd Edition.  Publication  No. AP-42.

U.S./Canada  Memorandum  of  Intent  on Transboundary  Air  Pollution.   1982.
Emissions, costs,   and engineering  assessment.   Work  Group 3B.  Final Report.
June 15, 1982.

Valach,  R.   1967.    The origin of the gaseous form  of natural  atmospheric
chlorine.  Tellus 19:509-516.
                                   2-115

-------
Vena, F.   1982.   Environment  Canada,  Personal  Communication.   August 19,
1982.

Viemeister, P.  E.   1960.   Lightning and  the origin  of nitrates  found  in
precipitation.  J.  Meterol.  17:681-683.

Visser,  S.  1961.   Chemical  composition of rainwater in Kampala, Uganda, in
its  relation  to meteorological  and  topographical  conditions.    J.  Geophys.
Res. 66:3759-37.

von  Liebig, J.   1827.   Une  note sur la nitrification.   Annales de Chemie et
de Physique 35:329-333.

Weast, R. C. (ed).  1973.  Handbook  of  Chemistry  and Physics,  54th  Ed., pg.
F-188.  CRC Press,  Cleveland,  Ohio.

Whitby,  K. T., and B. Cantrell.  1976.  Atmospheric  aerosols-measurement and
characteristics, Paper 29-1.   In International  Conference  on  Environmental
Sensing  and Assessment, Inst. oT"Electrical and Electronics Engineers, Inc.,
New York.  (IEEE Catalog  #75-CH  1004-1 ICESA).

Yue,  G.  K.,  V. A.  Mohnen,  and C.  S.  Kiang.    1976.    A  mechanism  for
hydrochloric acid production in  cloud.   Water,  Air, Soil Pollut. 6:277-294.

Zafiriou,  0.   C.   and  M.  McFarland.   1981.    Nitric  oxide  from  nitrate
photolysis  in   the  central   equatorial   Pacific.      J.  Geophys.   Res.
86:3173-3182.

Zafiriou,  O.C.,  M. McFarland, and  R.  H.   Bromund.   1980.  Nitric  oxide  in
seawater.  Science  207:637-639.
                                   2-116

-------
              THE ACIDIC DEPOSITION PHENOMENON AND ITS EFFECTS

                          A-3.  TRANSPORT PROCESSES


3.1  INTRODUCTION (N. V. GiHani)

Atmospheric contributions  to  the acidification of a  sensitive  receptor  site
can  best be  assessed if  the contributing sources  can be  identified  unam-
biguously,  and  if   the  atmospheric  transport  of  their  emissions  can  be
determined  quantitatively.   For a  number  of  reasons, it is  not  possible to
make such  a assessment precisely.   Atmospheric  depositions  include  not  only
primary  emissions,  but also  chemically-transformed secondary products.   Some
pollutants  such  as  sulfate are  both  primary  and  secondary  in  origin.   The
transport  age of  the  deposited materials  cannot  generally be  determined
accurately  from  their physical-chemical form.   Furthermore, the  complexity
and  considerable  variability  of   the   transport  winds  and  the  practical
limitations  on  the  detail  and  resolution  with which  we do,  or even  can,
measure  them  in  a routine manner,  make the tasks of source  recognition and
uncertainty assessment extremely difficult.

For several years now researchers in North America as well as in  Europe  have
recognized  that  the  regional  distribution of  secondary  pollutants such  as
sulfates is a consequence of  long-range transport and chemical  transforma-
tions  of  pollutant  emissions  into the atmosphere  (Altshuller  1977,  OECD
1977).  Transboundary exchanges of acidic pollutants no  doubt occur among the
nations  of Europe  as well  as  between  the United States and  Canada.   The
extent  to   which  pollutants   are  dispersed and  deposited far  beyond  their
sources  is highly  variable  and depends significantly  on the  processes  of
atmospheric transport  and  dispersion.   Atmospheric transport processes  also
play an  important,  sometimes critical  role  in  the chemical  transformations
and  deposition  of  pollutants  during   plume  transport.   For  example,  the
gas-to-particle  conversion of  sulfur  in  power plant plumes  depends  upon
atmospheric mixing,  which  facilitates interaction  between primary  species  in
the plume and reactive species  from the  polluted background  air (Gillani and
Wilson 1980).  Also, turbulent vertical  dispersion  is the  principal  mechanism
for delivering elevated  emissions  to the  ground for dry deposition.  Thus,
indirectly, transport processes play an important role  in   determining  the
overall atmospheric  residence time  of pollutants in the  atmosphere.

Deposition  of a  pollutant marks the end of its atmospheric   residence.   The
concept  of  atmospheric residence  time   (T) is  of critical   concern  in  any
assessment  of relative  locations  of  source  areas  of  acid  precursors  and
impacted areas of acidic depositions.  The other critical factor  influencing
such an  assessment  is the  spread of material  trajectories during  the  atmos-
pheric  residence  time.  Transport processes  exert a major, or possibly  even a
controlling, influence on T and the  trajectories.


                                     3-1

-------
The main objective of this chapter is to  identify  and  describe  the  principal
mechanisms of pollutant  transport,  specifically  in terms of their  influence
on the  atmospheric  residence time of the  pollutant.   To depict the role  of
transport, an attempt has  been  made to estimate T  of  sulfur emissions  from
different types of major sources  and during different seasons.  Atmospheric
processes influencing pollutant trajectories  and  spread  over regional  areas
are  described,  but  methods  of  trajectory calculations  and a  quantitative
assessment  of uncertainties  associated  with them are  not  covered  here.
Chapter A-9 discusses transport models and their  status as operational  tools.

3.1.1  The Concept of Atmospheric  Residence Time

The atmospheric residence time of a  given  pollutant emission is defined here
as the  characteristic  time  during  which   the  emission mass is depleted  by
removal processes (transformation and deposition)  to 1/e  or  about 37  percent
of its  initial  value.    If  the  depletion   were due to  first-order  processes
only,  such  a definition of  T  would make   it  the  effective time constant  of
exponential  decay  of the  pollutant  from   the  atmosphere.   In  general,  the
value  of  T  depends  on the  kinetics  and mechanisms  of the  processes  of
transport, transformation, and deposition.  Because transformation  and depo-
sition  rates are  specific  to chemical  species, T  is different  for  different
species   (for  example,   SOX  versus  NOX,   or   even   S02  versus   aerosol
sulfates).

Transport  processes  are, however, essentially  independent of  chemical  spe-
ciation.   In  this  chapter,  the  nature   and  significance  of  the  role  of
transport  processes  are explored  specifically  for   S02  emissions,  partly
because  SOg  is an important precursor of acidification  and partly  because
we have a better quantitative understanding  of  the rates of  transformation
and  deposition  of  S02   than  for  other  precursor species.     This  role  of
transport  processes  may  also vary depending on the type  of  emission source.
Consequently, we explore the  difference for the  two most  important  types  of
acid  precursor sources:  large, tall-stack power  plants  and  urban-industrial
complexes.

Acidification  of  an  ecological   system  is a  long-term  process.    Seasonal
averages  of  T and of  the influencing  transport  parameters are,  therefore,
more  pertinent in the present context than short-term  variations and effects.
Accordingly,  this chapter  reflects  such  a  bias  in  favor of monthly-  or
seasonally-averaged  data and interpretations.   Seasonal   averages,  however,
are  merely  integrations of shorter-term events.    In particular,  atmospheric
transmission  processes  (transport,  transformations,   and  deposition)  are
characterized by strong diurnal  variations, and proper  resolution of these is
necessary.   Therefore,  we  have  also tried  to  describe the  diurnal  cycle  of
transport layer structure and dynamics in  some detail.

Four meteorological variables are of particular significance  in  the transport
and  dispersion  of air  pollution:   the  height  of the   pollutant  transport
layer,  and the wind, temperature,  and moisture fields  within  this layer.  The
Earth's  atmosphere  is  about  100  km  deep.   Anthropogenic pollutants  are
typically confined and  transported within  the  lowest 2 km of the atmosphere.
The  flow field  within this  boundary layer is driven  by the planetary  flow


                                     3-2

-------
above and at  the  same time is subject to influences of  interaction with  the
Earth's surface below.   This flow  field  governs the mean  transport  of  the
pollutants.   The  spread of the  pollutants  during transport is largely gov-
erned  by  spatial  and  temporal  inhomogeneities  in the flow  field.    The
dispersive capacity of the transport layer is also influenced strongly by  the
temperature  distribution   within  it,  which  is  determined  principally   by
insolation and the nature  of the ground surface.   The moisture  field  governs
cloudiness and precipitation and also influences  atmospheric chemistry.   The
local moisture  field  depends  on transport  from  upwind,  as  well as on  local
evaporation of surface water.

General features of the planetary and the boundary layer  flows  are  described
in Section 3.2.   The  structure and dynamics of  the transport layer,  as well
as more detailed features  of the boundary  layer  flow and  dispersive  capacity,
are presented in Section 3.3.   The remainder of  the chapter  describes  how  the
transport  of pollutant  emissions  takes  place   by  atmospheric  motions   of
various scales.

3.2  METEOROLOGICAL SCALES AND ATMOSPHERIC MOTIONS (N. V. Gillani)

3.2.1  Meteorological  Scales

Atmospheric motions and transport phenomena  vary over a wide range of  spatial
scales.   In  general,  as a  pollutant  plume  spreads during  transport,  atmos-
pheric motions  of progressively larger scales influence  its  further  disper-
sion.  The  relationship between plume dynamics and atmospheric motions must
therefore  be considered  in the context  of their relative  spatial-temporal
scales.

Meteorological  scales  are typically classified  into  micro, meso,  synoptic,
and global regimes.  The meteorological  microscale is defined by the vertical
dimension of  the  planetary boundary layer (PBL), within  which  anthropogenic
pollutants are typically emitted and distributed.  This  dimension is  about a
kilometer,  and  its associated  time scale  is measured  in   tens  of  minutes
(approximately  the  time  required  for a  plume  to  spread over the vertical
extent of the mixing layer  under daytime  convective conditions).  The micro-
oscale phenomena  include  atmospheric turbulence.1  The  meteorological  meso-
scale extends up  to  about 500 km,  and its  associated time  scale is  about a
day, approximately the time needed for a mean horizontal  transport of  500  km.
Mesoscale effects include  plume  dynamics  and the diurnal variability  of  the
PBL.   They  are  strongly influenced by surface inhomogeneities  of terrain as
well as  heat and moisture fluxes.    Within the  range  of  the  mesoscale,  a
specific  plume  from a  power  plant or urban  complex  will commonly lose  its
identity  by  mixing  with other  plumes  or  by diluting indistinguishably into
1Atmospheric  turbulence  is sometimes  interpreted  broadly to  include  vortex
 motions over all meteorological  scales.  Our use of  the  term  is  more  speci-
 fic, and refers only  to  random microscale  eddy  motions  ranging in size from
 a few millimeters  to  a few hundred meters.  Thus,  we use the terms  turbu-
 lence and microscale turbulence synonymously.
                                     3-3

-------
the background.   Transport over  the microscale and  mesoscale is  sometimes
also referred  to  as short- and  intermediate-range transport,  respectively.
Beyond the mesoscale  is the synoptic  scale,  the scale of the  weather maps,
with characteristic horizontal dimensions  of about 1000  km  and a  transport
time of about 1 to 5 days (the approximate  range of residence  times  of sulfur
in the air in eastern North America).   Finally, the hemispherical   or  global
scale  is  about a  week   and  includes intercontinental transport.   The  dis-
cussion of pollutant transport processes is divided into mesoscale  transport
(Section 3.4)  and  continental  (synoptic) and hemispheric transport (Section
3.5).   The term "long-range transport" commonly  refers to transport over the
synoptic and hemispherical  scales.

3.2.2   Atmospheric Motions

The energy  that drives  the atmosphere  comes from the  sun  in the form  of
radiation.   However,  solar radiation is not  uniformly distributed  over  the
surface of the Earth.   Because the Earth's  pole  is  tilted, a given  horizontal
area in high latitudes receives far  less solar  radiation than  an equal  area
closer  to  the  equator.    If  there were  no transfer  of  heat  poleward,  the
equatorial regions would heat up.  In  a  fluid as mobile as air, temperature
differences  will  immediately  give rise  to  currents that tend  to reduce the
thermal gradient.   Unequal  heating  of  the  Earth's  surface   thus  leads  to
horizontal pressure gradients that provide  the driving force of the  winds.

Wind,  of  course,  is  air  in  motion  and  although  it  is a motion   in  three
directions,  usually  only the horizontal  component is reported in  terms  of
direction  and   speed.    In  the free  atmosphere (above  the  effects of  the
Earth's friction)  two forces are  important  in describing fluid motion  in the
moving  reference  frame  of an observer  on  the Earth's surface.  One  is the
pressure gradient force, which tends  to move the air in a direction  from  high
to  low pressure.   The second  force  is  called  the  Coriolis  force.    The
Coriolis/force is« a consequence of the rotation  of  the Earth,  and is directly
proportional to the speed  of  this rotation.   It  increases  at  higher  lati-
tudes.  The Coriolis force also increases with wind speed, and its  effect is
to deflect the wind to the right (in  the northern hemisphere)  relative to the
pressure gradient force.   In  the  free atmosphere where the Earth's friction
is  not felt  significantly,  the  horizontal  flow becomes established  nearly
normal  to  the  pressure   gradient  force  (hence,   parallel   to  the iso-  bars).
The pressure gradient force and  the  Coriolis force act  equally and opposite
to  each  other.    This   condition is  called geostrophic  balance,  and  the
corresponding flow is the geostrophic flow.

Friction  between   the  flow  and   the  surface is felt significantly in  the
so-called Ekman layer which typically extends one to  three kilometers above
the  surface.   Ordinarily  the wind  speed and  wind  deflection (veer)  are
maximum at  the  top of the Ekman  layer.   Within the Ekman layer, wind speed
decreases as the  surface is approached.  Correspondingly, the Coriolis force
decreases and  so  also  does the amount of  wind  deflection.    Wind  deflection
under  the idealized Ekman  layer conditions  decreases  from 90°  at geostrophic
level   to 0°  at  the  surface.   Thus, the surface  flow  is  nearly perpendicular
to  the pressure  isobars while geostrophic  flow  is  nearly  parallel  to the
isobars.   The  condition of wind  speed  shear  and wind directional   veer  with


                                     3-4

-------
height  in  the idealized  Ekman  layer is  called the  Ekman  spiral  (see,  for
example, Brown  1974 and  Figure  3-1).    In  actuality, the  surface  is  never
completely  homogeneous,   and  the  Ekman  layer  is  characterized  by  varying
degrees of  vertical  stratification  (i.e.,  lack  of homogeneity of turbulence
structure), and the idealized Ekman spiral  is only approximately  realized.

On the  global  scale, the general  circulation outside the boundary layer  is
driven  by  the  global pressure gradients  due to the  unequal  heating  of  the
Earth's surface between the equator and  the  poles,  and  it is modified by the
CorioHs force.   This planetary flow  is approximately geostrophic horizon-
tally.  Vertically,  a  weak  pressure gradient force  (pressure  decreases  with
height) is  nearly balanced by the  gravitational  force (hydrostatic  balance).
Hence, on the global scale,  vertical motions are relatively weak,  except over
the  high  and  low  pressure   zones  of  the  Earth.    Hot  air  rises  over  the
equatorial   low pressure belt and sinks at the tropics (25°  to 30° latitude).
Aloft,  the  wind  blows  horizontally from the  equator to the  tropics  (south-
westerlies  in the northern hemisphere);  near the surface,  the flow is  towards
the equator (northeasterlies).   Poleward of the tropics, the  Coriolis  force
is stronger, and the flow pattern is more complicated, being characterized by
synoptic-scale  cyclones   and  anticyclones,  which  are  rotating horizontal
flows,  rather  than  simple  straight flows  (see, for example,  Chapter 4  in
Anthes et al. 1975).

Cyclones are low pressure cells  with  rising motion near the  center and  a
counterclockwise  flow  spiral ing  towards the  eye  near  the  ground.    Anti-
cyclones are large high pressure cells with  slowly  sinking  air at the center
and  weaker  outward  and   clockwise  spiral ing surface flow  in  the northern
hemisphere.   Cyclones  and anticyclones  rotate   about their  own centers  but
also move downstream, generally eastward, in the broad-scale westerly  general
circulation  in which they are embedded.   Anticyclones are  characterized  not
only  by weak rotating flow  within the  cell, particularly  in the core,  but
frequently  they are  also  characterized by  weak  or stagnant motion.    When  an
anticyclone  stagnates for multiday periods over  pollutant source  regions such
as the  Ohio River  Valley, considerable pollutant accumulation and  aging  can
occur over  a synoptic  scale,  and episodes of regional haziness  occur.   Such
hazy air masses become richly loaded with acidic material.   A summary of the
climatology  of  synoptic-scale  "air stagnations" (covering area  greater  than
200,000 km2 for  more  than  36 hours)  in the eastern  United States  is  pre-
sented  in  Figure  3-2.   The  greatest likelihood of  such  stagnations  is  over
the dense  source  regions  of the TVA and  the Ohio River Valley.  For a  dis-
cussion of  the  relationship  between haziness  and  concentrations of acidic
substances  see Chapter A-5.

Another important large-scale  flow feature  is the jet stream.   Temperatures
do not  vary gradually  from the  tropics toward the poles.   Sometimes,  regions
of relatively  weak  thermal   gradients  are interrupted  by regions of strong
gradients,  called "frontal zones."   These frontal zones  are  associated  with
localized  regions of strong  winds located above these  zones.   Such  frontal
zones exist at interfaces of air masses  of different origins  and physical
properties.   In  the interior of the North  American continent, there  are  no
significant  geophysical obstructions  to  air  movements,  particularly  between
the  north  and the  south.  Southward  intrusions of  the  dry, cold Canadian


                                     3-5

-------
   500 - 1000 m a,
                       GEOSTROPHIC WIND
Figure 3-1.   The Ekman spiral  of wind with height in the northern
             hemisphere.   Adapted from Barry and Chorley (1977).
                                 3-6

-------
Figure 3-2.   Climatology of air stagnation advisories issued over a ten-
             year period.  Adapted from Lyons (1975).
                                 3-7

-------
continental  polar  air mass  and northward  intrusions  of the moisture-laden
maritime air mass from the Gulf of  Mexico  often give rise to frontal  zones,
with  the  associated  jet stream  and its  strong, generally  westerly  flow.
Associated with such  frontal  zones  is also strong horizontal convergence  of
flow at lower  levels,  and  upward  motion aloft; clouds and precipitation are
concentrated at frontal zones.  (For  detailed  descriptions of North  American
air masses, frontal zones, and the  jet  stream,  see Chapters  4 and  5  of Barry
and Chorley 1977.)

Mesoscale systems are  perturbations  of  the synoptic  flow on scales  that are
too small to be resolved on  weather maps but  are  larger  than the microscale.
They  are  particularly important  in  producing  local  weather,  which can  be
quite variable spatially within the  same synoptic system.  Except  in frontal
zones and near cyclone centers, synoptic  and  global  flows are  largely  domi-
nated by  horizontal  winds,  with  very weak vertical  components.   Mesoscale
systems, in contrast,  are characterized by  significant vertical flows,  hence
are  often  termed  complex  flows.    Whereas  average  vertical  velocities  in
large-scale systems are typically on  the order of 1  cm  s"1, vertical  speeds
in  local mesoscale  systems are typically  on  the  order  of  1 m s"1,  and may
even  exceed  10  m  s~l  in   strong   updrafts,  especially  in  thunderstorms
(Panofsky 1982).

Mesoscale complex  flows may  be  terrain-induced or synoptically-induced (see,
for  example,  Pielke  1981).   Terrain-induced  effects  include  land   and sea
breezes  and other  effects  related  to  shoreline  environments, as  well  as
forced  air  flow over rough  terrain, mountain valley  winds resulting  from
natural  convection  phenomena,  and  urban   and  other  circulations  related  to
specific land  use  patterns.   Synoptically-induced vertical  motions,  such  as
at  frontal  zones,  may be  complicated by  interactions  with local mesoscale
disturbances  such  as  squall  lines, which  are  narrow  lines of thunderstorm
cells that  may extend for  several  hundred kilometers.   Later  sections will
show  that  substantial  depositions  of sulfur emissions  occur  within the
mesoscale  range,   particularly  in  summer,  in the  eastern  United  States.
Mesoscale flow  systems are  therefore of  considerable  importance  in source-
receptor  relationships.    A  more  detailed discussion  of mesoscale complex
flows is given in Section 3.3.4.

Turbulence  is the  most  important  microscale  motion.    Unlike  large-scale
motions (synoptic and global), it is essentially random and three-dimensional
motion.  The vertical component of the motion  is comparable to the horizontal
component.   Microscale  turbulent  eddies   may  be  generated  in  two  ways,  by
thermal convection or by mechanical  shear.   Water  boiling in a pan is full  of
thermal turbulence.   In  the  atmosphere, heating from the ground below in the
daytime sets  up convection  currents with  turbulent eddies often as  large  as
100  m or more  in  size.   On the other hand,  the  interaction  of wind with
surface roughness also generates turbulent eddies  that are characteristically
smaller than  thermal  eddies.  Friction between the  ground and  the air gives
rise to strong wind shear in  the surface layer of air (lowest few meters) and
gives rise  to  intense small-scale mechanical  turbulence.  Patches  of mechan-
ical  turbulence  may  sometimes  also  occur  high in  the  upper  atmosphere  in
locally  strong wind  shear  zones  associated  with  frontal  zones  (see,  for
                                     3-8

-------
example, Panofsky  1982).   This type of clear-air turbulence  (CAT)  sometimes
causes discomfort to aircraft passengers even at cruising altitudes.

Turbulence  is an  important  mechanism  for  mixing or  spreading a  pollutant
emission horizontally  but, more  importantly,  it is often the only  mechanism
for vertical  mixing.   It is principally responsible for  delivering  elevated
emissions  to the  ground.    It is  also  an important  agent  for dilution  of
concentrated  pollutant  releases  from  point sources.   Turbulence is  also  the
mechanism  for vertical  spreading  of moisture  evaporating  from the  ground.
This, of  course,  is the  stuff of which clouds  and  precipitation are  made.
The significance of turbulence as a dispersion mechanism,  particularly in  the
vertical, is  not restricted to mass only (i.e., pollutants and moisture).   It
disperses momentum and energy just as effectively. Turbulent eddies distrib-
ute surface  drag  (friction)  over the Ekman  layer.   Vertical turbulence,  in
fact, is the  principal means for  communication  of mass,  momentum,  and energy
between  the   Earth's  surface  and  the  large  scale  upper  air flow,  thereby
gradually changing large-scale conditions.   This is an  example of interaction
between the extreme scales of atmospheric motions.

Interactions  occur  between  all scales  of  atmospheric  motions.   Such  inter-
actions  play an important role  in pollutant  transport  and  dispersion.   In
fact, such interactions pose a major difficulty in the  modeling  of  long-range
transport, in which a  rather coarse spatial-temporal  resolution of  the mean
flow  field   is  commonly used.    Mesoscale  and  microscale  effects  are  not
resolved adequately in an explicit manner  in  such a coarse "grid"  structure.
The net  effects  of such  "sub-grid"  phenomena are often  most important  and
must be included by means of parameterizations or bulk  representations.

As  an important  example, consider the  question of  long-range  trajectory
calculations.    It  is  still  common  practice  to  calculate  an   "average"
long-range trajectory  of a polluted  air parcel, based  on the  average wind
speed and  direction in  the  entire vertical  domain  of  the  transport  layer
(see, for example, Heffter 1980).   Such  an  average trajectory hides  the fact
that, as  a  result  of  the  spatial-temporal   variation  of wind  speed, wind
direction, and  turbulence characteristics  within the  transport layer,  the
ensemble  of   pollutant  particles  in  the  air  parcel  of  interest  actually
follows  an   ensemble  of  noncoincident  trajectories.    The  spread  of this
ensemble of trajectories is, in fact, the measure of pollutant  spread  during
transport.   In  long-range  transport, such  spread can  amount to hundreds  of
kilometers.    For  proper modeling  of  pollutant  transport and  spread,  the
average calculated trajectory  must  be accompanied by  a measure of  pollutant
spread based  on an appropriate  parameterization of the  wind variations  within
the transport layer.

A  considerable  amount of  micrometeorological  field  data  and research have
yielded more  or less  acceptable  approximate parameterizations of  dispersion
due to microscale  wind fluctuations.  Dispersion due to shear  and veer  in  the
mean  wind field  is  only  now beginning  to  be  modeled  realistically  and
explicitly,   and  has  not progressed  to the  point  of  formulating  reliable
parameterizations.    Field  data  pertinent  to  mesoscale  motions  are very
limited.   Routine monitoring  of upper  air winds is  confined  to  a  sparse
                                     3-9

-------
spatial network (stations being  separated,  on  the average, by well over  300
km), and the  temporal  resolution of the measurements  is also coarse  (typi-
cally  at  12-hourly  intervals).    Such monitoring  is  adequate  for  the  re-
construction of the  synoptic  flow field (as  seen on  the  weather maps)  but
inadequate to resolve mesoscale  effects.   Possibly the major uncertainty  in
the  assessment  of regional  impacts of  emissions  is  due  to this  lack  of
resolution of mesoscale  and diurnal  variations  of the flow field, particu-
larly under short-term episodic conditions.

The extremely important  role  of  microscale  turbulence  in vertical mixing  is
characterized by strong spatial-temporal  variabilities  in vertical  turbulence
structure.  Turbulent eddies  range over  a wide  spectrum  of  size as well  as
turbulent kinetic energy distribution.  The large thermally-generated  eddies
contain the most turbulent energy, and thus are  capable  of the most vigorous
mixing up  to  a  scale of several  hundred meters.  They exist in  the central
part of the PBL, which is generally quite well-mixed.  Because the source  of
their  energy  is  surface  heat  flux  which,  in  turn, depends  directly  on
insolation, their existence exhibits a strong diurnal  cycle.   Close  to  the
surface, small-scale  mechanically-generated  eddies predominate.   They contain
much less energy  and have  more  limited mixing  capacity.  Consequently,  the
near-surface layer presents the most resistance  to  the downward  transport  of
momentum and elevated emissions, or to upward transport  of heat  and moisture
fluxes.  Small-scale  turbulence exists  also  in  the well-mixed  bulk of the  PBL
because individual large eddies  are very  transient in  nature (as  indeed  are
all eddies), and are continuously  being generated on the  one  hand by surface
heating, and  degenerated on the  other hand to  small  eddies  by  a rapid  and
continuous  transfer  of  energy from larger  to  smaller  eddies.   At the lower
end of this "spectral energy cascade"  (Tennekes 1974),  viscous dissipation of
the  smaller eddies ultimately  removes  turbulent  kinetic  energy by converting
it  to  heat.   This process of  kinetic  energy  dissipation is  responsible  for
dissipation of  as much  as half of the  kinetic  energy  of  the  large-scale
atmospheric flow patterns (Tennekes 1974).

The  role  of these spatially-temporally  varying  microscale  motions  must  be
included  in  transport   models  by  appropriate  parameterizations.    Because
vertical stratification of the transport  layer occurs in  terms of wind  speed,
wind  direction,  and wind  shear  as  well  as  turbulence,  it is  increasingly
evident  that  realistic  transport models  must  adopt  a  degree  of  vertical
layering.    In  the  next  section, we  explore  the characteristics  of  the
transport layer in somewhat greater detail.

3.3  POLLUTANT TRANSPORT LAYER:  ITS STRUCTURE AND DYNAMICS (N.  V.
     Gillani)

3.3.1  The Planetary Boundary Layer (Mixing  Layer)

The  troposphere  is  the  lowest  portion  of the  Earth's  atmosphere in which
temperature,  on  the  average,  decreases  with height.    In  the  tropics,  its
depth  is  about 10 km.   The bulk  of  anthropogenic pollutant emissions,  in-
cluding precursors of acidic depositions, is  released  and transported  in  the
lowest  2  km or  so  of the  troposphere.   This  is  also  the  layer where  the
primary meteorological variables [i.e., the thermal  field  (temperature),  the


                                     3-10

-------
momentum field  (winds),  and the moisture field] are perturbed  significantly
as a direct consequence of  the  Earth's  surface.   In air pollution meteorol-
ogy, pollutant  concentrations in  the  air represent a fourth type of primary
variable.  For  each variable, the  layer  perturbed  by surface effects is its
boundary layer.  The surface sources of  disturbances of  the  primary variables
may be  different for  the  different variables,  and for each  variable, the
distribution of such  sources may  be  spatially inhomogeneous and temporally
variable also.   However,  all types of  disturbances are communicated verti-
cally by the same physical  mechanism,  turbulence.   Consequently, the boundary
layer of most  practical  significance is  the  so-called mixing  layer  (also
called the planetary boundary layer, PBL).  The  principal  characteristic of
this layer  is  the continuous presence of  significant  microscale turbulence
within it.

The definition  of  the mixing  layer  as the  vertical   domain  of microscale
turbulence  must be  qualified.   In certain  complex  flow  situations,  this
definition may  be  inappropriate.   For  example,  in the  presence  of strong
convective instability associated with towering  cumulus clouds and thunder-
storms,   vigorous  turbulent mixing  within  clouds  may extend  into  the  upper
troposphere.  In such  cases, the base  of the clouds may be considered as the
PBL  height.    When  strong  orographic,  shoreline,  or  other  topographical
effects   are present,  the  PBL needs special  consideration.    Perhaps  a  more
appropriate definition of the top of the  mixing layer is "the lowest level in
the atmosphere at which the ground  surface no  longer directly influences the
dependent variables through turbulent  mixing" (Pielke 1981).

The mixing  layer  is  so called  because, within it, atmospheric turbulence
effectively  and  quickly  manages  to  mix  up,  spread  out,  or dilute any
concentrated release of mass, momentum,  or heat.    In all other parts of the
atmosphere, the dilution of pollutants is very  slow.  The mixing layer  grows
during  the  daytime, typically  to  heights of  1 to 2   km,  due  to increased
thermal  convection, and subsides at night to heights typically ranging  up to
about 200 m.

While the  deep  daytime mixing  layer is  dominated by  large-scale  thermal
turbulence, the shallow nighttime  mixing layer  contains  only small-scale
mechanical  turbulence.   The daytime mixing layer  is extremely efficient in
quickly   delivering  any  elevated pollutant  releases  within  it  to  its entire
vertical extent, including  the ground.  On  the  other hand, elevated nighttime
releases from tall  stacks  are typically outside the  shallow  mixing layer and,
in the absence  of any mechanism to  bring them  down  to the ground, are trans-
ported over long distances  while remaining  decoupled from the ground.  Night-
time urban releases within  the shallow mixing layer, on  the  other hand,  often
remain  trapped  at  relatively  high concentration  and,  being  in  constant
contact  with the ground sink, may become  substantially depleted  of pollutants
during relatively short-range transport.   Pollutants that become well-mixed
in  the   deep  daytime  mixing  layer are  transported at  night  in  this  deep
transport layer, decoupled from the ground except  for  the  lowest portion in
the shallow nocturnal  mixing layer.

The depth  of  the  mixing  layer  is a  critical parameter  with  respect  to
pollutant  transport.    The   top  of  the  mixing  layer usually  distinctly


                                    3-11

-------
delineates the  turbulent,  polluted air  below  from the  calmer,  cleaner air
above.  This is particularly the case  during  midday, convective periods.  The
height  of the mixing  layer can  be  measured most  accurately  by turbulence
monitors  in  instrumented  research aircraft flying a  vertical  spiral,  or by
remote soundings of the turbulent fluctuations of temperature and atmospheric
refractive index  using sodars  and lidars.    In  daytime,  the  mixing  height
commonly coincides with the lowest temperature inversion.   Accordingly,  it is
most commonly estimated from  vertical  temperature and humidity soundings by
standard  radiosonde  releases.   The daytime  mixing  height may even be  esti-
mated from the height  of  the  cloud base in  fair-weather  cumulus conditions,
or often from the height of the visible polluted  layer.

A number  of  excellent  review articles  describe the  structure and dynamics of
the PBL.   Tennekes  (1974)  presents a  useful  qualitative description of the
PBL.  Arya (1982) presents a more detailed review of the  PBL over homogeneous
smooth  terrain,  including  a  section   summarizing  techniques  of parameteri-
zation  of the  PBL.   PBL  parameterization  and attempts at  simulation of
observed  PBL structure and dynamics are  thoroughly  reviewed  also  by  Pielke
(1981).  The features of the PBL over non-homogeneous  terrain,  and  simulation
of these,  are  described  in detail by  Hunt and Simpson  (1982).   Also,   a WHO
Technical  Note  devoted to  the PBL (McBean et al. 1979)  contains a number of
excellent  chapters summarizing PBL features,  observed  and modeled,  for  simple
and complex  terrain.

The sections that follow are substantially based on the above  references.   In
addition,  however,  the author  has  chosen to  present illustrative  examples
from  previously  unpublished data of  very recent, very  sophisticated,  major
EPA-sponsored mesoscale field  programs,  particularly  Projects  MISTT  (Midwest
Interstate  Sulfur  Transport  and  Transformations),   RAPS  (the  St.   Louis
Regional  Air Pollution Study),  and TPS (Tennessee Plume  Study).  Collective-
ly,  these data  bases reflect  state-of-the art  technology, seasonal  coverage,
and some  of  the most detailed measurements of mesoscale plume  transport. The
results of  earlier  well-known PBL field studies such  as the Great  Plains
Experiment at O'Neill,  Nebraska  (Lettau and  Davidson  1957),  the Wangara
Experiment in Australia (Clarke et al. 1971,  Deardorff 1980),  the 1968  Kansas
Field  Program (Izumi  1971, Haugen et al. 1971,  Businger et  al. 1971),  the
1973  Minnesota study (Kaimal et al. 1976, Caughey et al. 1979), and the 1975,
1976  Sangamon Field  Program   (Hicks  et al.  1981)  are  well  covered  in  the
original  references  and are also included in the PBL  review articles identi-
fied  earlier.  These earlier studies were focused more on micrometeorological
measurements and  analyses.

3.3.2   Structure  of  the Transport Layer  (TL)

For  a given  day,  the transport layer may be defined as the layer between the
surface and  the peak mixing height of  the  day.   For  any  given  instant, it  is
therefore made up of  the  current mixing layer  below and a relatively  quies-
cent  layer above.  This minimum stratification  of the TL  into two  layers  is
essential  in any  transport model.  The daytime mixing  layer itself  may  be
further subdivided  into  a  surface layer (extending typically to 50  m  or so)
and  a "mixed"  layer  above.
                                     3-12

-------
The surface layer is principally characterized by strong  gradients  in  all  the
primary variables, the  influence  of  surface effects being most concentrated
there.  The wind speed increases from zero at the surface to  near-geostrophic
in the mixed layer.  The land surface has  a  relatively smaller heat capacity
than  the air  above,  and therefore undergoes more rapid  and  greater tempera-
ture changes than the air  during  the diurnal cycle.  The transition  between
the surface temperature and the mixed layer  temperature  distribution  is also
most  concentrated  in the  surface layer.    Owing  to  the dry  deposition of
pollutants at the surface, a significant  increase  in  pollutant concentration
occurs as  height in  the  surface layer  increases.   Also pronounced  in  the
surface layer is  the  frictional  force.   Thus, the average wind speed is  low
here,  and  consequently  the Coriolis  effect is  relatively  unimportant.   In
turn, the wind direction remains  relatively  constant and more nearly  aligned
with the pressure gradient.

The large wind  speed  shear in  the surface layer  leads  to the generation of
intense small-scale  mechanical  turbulence.   While thermal   buoyancy  effects
are also intense here in the daytime, the proximity of the surface  limits  the
size of turbulent  eddies.   As  a  result,  surface layer turbulence  is  charac-
terized chiefly  by small  eddies.   Consequently,  the dispersion  within  the
surface layer  is relatively much slower than in  the mixed  layer, and dis-
sipation of turbulent kinetic  energy is locally  high relative  to the total
amount  of  turbulent  energy  present.   Also,  the  relatively  slow vertical
transfer of  the  pollutants  in  this layer  is at  a  nearly constant rate.
Hence,  it  is often  also called  the  "constant flux  layer."   Shear  effects
generally predominate over buoyancy effects  in the lower part of the  surface
layer  (forced convection  layer), but  under  midday  convective conditions,
buoyancy effects may predominate in the  upper part of  the surface layer (free
convection layer).

The surface layer is by  far the most studied part of the  PBL. The  parameter-
ization of  the  mean  flow  as  well  as   its  turbulent components  are well-
established and,  at  least over  smooth  terrain  under relatively  stationary
conditions, fairly reliable.  Turbulent dispersion is parameterized in terms
of  an "eddy  diffusivity,"  by  analogy  with the  concepts  of molecular dif-
fusion.  Eddy "diffusion"  is on  a relatively larger scale,  however,  because
the scale of the transporting medium, the eddies,  is considerably larger than
the mean free path (mean distance between collisions) of the molecules.  In
the  surface  layer,  the vertical  eddy  diffusivity,  Kz,  increases linearly
with height as larger eddies can exist farther from the  surface.   Higher  up,
in the mixed  layer,  the distribution of eddy  scales and turbulent energy is
more nonlinearly distributed with height,  and the concept of eddy diffusion
becomes less reliable.

In  the  mixed layer,  as  the   name   suggests,  the  variables  (wind  speed,
"potential" temperature, moisture, and pollutant concentrations)  are more or
less  homogeneously  distributed vertically,  owing to the more  thorough  and
rapid  mixing  by the  large-scale, thermally-generated eddies  or  convection
currents.  Buoyant effects predominate,  and the turbulent dispersive capacity
of  the  atmosphere  is  more  commonly  expressed  in  terms   of  atmospheric
stability.    The  potential temperature  (e)   is  a closely  related concept.
Both concepts are defined below.


                                     3-13

-------
A hot (buoyant) puff  of  gas  released into the atmosphere will rise, expand,
and  cool   nearly   adiabatically  (i.e.,  without  exchanging  heat  with  its
surroundings) at the rate of about 1 C per 100 m  in dry  air (a dry  adiabatic
lapse rate,  r^y), and  more  slowly in  moist  air  (a  wet adiabatic  lapse
rate, r).  The  puff will  continue  to rise and expand  as long as it remains
buoyant,  i.e.,  warmer  than  the ambient air.    Whether its  buoyancy will
increase, decrease, or  remain  unaltered as  it  rises  depends on whether the
ambient  atmospheric  lapse   rate  (dT/dz)  is  superadiabatic  (dT/dz  <  r),
subadiabatic  (dT/dz > r ),   or adiabatic  (dT/dz  =r,  which  is negative).
The  potential  temperature  is defined by  de/dz  =  dT/dz  -r .   The  potential
temperature  decreases with height  in a  superadiabatic atmosphere,  increases
with height in a subadiabatic atmosphere, and remains constant with  height  in
an adiabatic atmosphere.   A superadiabatic layer is  unstable because the puff
will become continuously more buoyant in it  and will rise and  dilute faster.
A subadiabatic  layer  is  stable because  it  tends  to slow down and  terminate
puff  rise.   An adiabatic  layer is  neutral  because it  does  not alter the
initial  puff  buoyancy.   The puff will  thus  continue to  rise in  neutral and
unstable surroundings until  it reaches a  stable thermal  environment.   In the
daytime, the surface layer is typically very  unstable,  and  the mixed layer  is
in near-neutral condition.   Any surface perturbations  of mass, momentum,  or
energy  in  the  daytime mixing  layer will  thus  be convected  upwards by the
turbulent eddies.    Surface heating will  continually release "thermal plumes"
or  convective  updrafts,  some  of which  may  rise to  the top  of the  mixing
layer,  carrying along with  them any  evaporated  moisture.    Some  of these
updrafts  will  also rise  into   the  quiescent layers aloft, thus causing  an
upward growth of the mixing layer by penetrative convection.

The rise of buoyant updrafts in the  unstable daytime convective mixing layer
is  frequently  obstructed by  a thin  temperature  "inversion" layer  (stable)
capping the mixing layer.  The  climatology of daytime  mixing  layers over the
continental  United  States has  been  documented (Holzworth 1972).  Figure 3-3
illustrates  the vertical  structure  of  temperature,  small-scale  turbulence,
and  S02 in  a  rather well-mixed power  plant  plume within  the  bulk  of the
peak daytime mixing layer on a  cloudless  summer day in the  midwestern  United
States.   The  turbulence  clearly decays  rapidly  at the elevated  inversion
base.   Unlike the  rather uniform  distribution  of small-scale turbulence  in
the mixing layer,  the vertical  distribution  of large-scale  turbulence  in the
mixing  layer (that most responsible for rapid mixing) is  quite inhomogeneous,
peaking  in  the middle  of  the  mixing  layer  (where  tall-stack  plumes are
released) and  decaying  rapidly at  the  top  and  bottom  boundaries (much  like
the  S02  profile).   Typically,  no  physical  or  stable  boundaries  exist
horizontally,  and  the turbulence structure  is more  homogeneous.   Turbulent
eddies  are  horizontally  larger, and turbulent plume dispersion  is  generally
faster  horizontally than it is vertically.

A number of major factors influence the  structure of the PBL.  The  mean  flow
field is  principally  driven  by the  planetary  flow, and modified by surface
friction and the local thermal  wind  due  to  horizontal  temperature gradients.
The modifications can be locally dominant as over extremely complex terrain,
in  shoreline  environments, over urban heat  islands, and in the  vicinity  of
mesoscale  convective  precipitation  systems.   The  turbulence  structure  is
principally governed by surface heating and cooling  and by wind shear,  either


                                     3-14

-------
     2000
  J   1500
  CD


  £
      1000
       500
           ^-S02
             \
    TEMPERATURE
          TURBULENCE
                   10
20       30
  TEMP  (°C)
40
50
0
1
0
5
i
2
10
S02
_l
4
TURB (i
(ppb)
i
cm2/3 6S-1)
20
i
8
25
i
10
Figure 3-3.  Vertical  profiles of temperature,  small  scale  turbulence,
             and SOg concentration in  a  diluted power  plant plume
             within the daytime mixed  layer  near St.  Louis,  MO.
             Observe the temperature inversion  and  sharp turbulence
             decay between 1700 and 1900 m (Gillani 1978).
                                 3-15

-------
due  to  surface  roughness  or other  causes.   Wind  shears  and turbulence
intensities also depend  strongly  on mixing  layer  height,  which essentially
fixes the dimensions of the largest eddies.  This  height depends principally
on the sensible heat flux from the ground, which in turn depends strongly on
insolation, local  land  use,  and surface  condition.   The  heat flux not only
has strong diurnal  variability, but also substantial  spatial  variability in
urban as well  as  rural  areas  on  the scale of a few kilometers (Ching et al.
1983).   The mixing height can also  be  influenced  significantly by synoptic
influences on mixed layer growth,  such  as cold  air  subsidence  and large-scale
lifting as in frontal  zones (Ching  et al. 1983).

3.3.3  Dynamics of the Transport  Layer

Strong diurnal and seasonal  variations  occur  in  the mean  thermal  and flow
fields, as well as in  the turbulent fields,  within  the PBL.  Good qualitative
descriptions of the diurnal effects have been  given  by  Plate (1971)  and by
Smith and Hunt (1978).

Diurnal  and seasonal variations of the thermal  stratification of the trans-
port layer are shown in  Figure 3-4,  and  the average diurnal  profiles of the
mixing height  during  the  different seasons are shown  in Figure  3-5.   The
temperature data are based  on RAPS radiosonde measurements  at  a rural  site
near St.  Louis, and each profile  is based on 31 daily  soundings  in 1976.  The
mixing height  data  are  deduced from a  composite  of 6-hourly  temperature and
wind  soundings as  well  as turbulence  measurements  during  a  large number of
aircraft spirals.

At night, the  ground is  cooler than  the  air layers above.   Hence, a  surface
based inversion (very  stable)  extends upward to about  300 m  in the  summer and
to nearly 600  m  in the winter near  St.  Louis.   A  shallow mechanical mixing
layer exists within the  inversion  layer.  As the sun  comes  up in the  morning
and  heats  up  the  ground,  surface   temperature rises  above  that of  the air
layers immediately above.   Consequently, an  upward  sensible  heat  flux by
conduction and convection  is  established, and  a  continuous warming trend of
the surface layer  air occurs.  With  increasing insolation and warming of the
air,  the nocturnal  inversion  layer  is eroded  from the surface  up. "AS the
heating  continues  into  the mid-  and late-morning  hours,  an unstable  layer
develops near  the  ground,  while  convective  eddies  aid in  the growth of the
mixing layer by penetrative convection  into  the quiescent layers  aloft.  On  a
clear day, this growth proceeds quite rapidly in the  morning  and more slowly
in the early afternoon,  until the  transport  layer  is  fully  established, with
the mixing height at its peak value  typically by midafternoon.   This  daytime
mixing layer  is  typically capped  by an  elevated  inversion layer, which is
very stable and quite  thick in the winter (700 to 1200 m, on  the average, in
January  in St. Louis;  Figure  3-4)  and  quite high and narrow  in  summer  (1800
to 2000 m, on the average, in  July  in St. Louis).   The peak  mixing  height, or
the  full  transport layer height,  is thus much deeper in the summer  than in
the winter.   This  fact, above all   else,  is  likely to lead to a  substantial
difference in  the  atmospheric residence  times  of emissions from tall stacks
during summer and winter.  Within  this daytime mixing layer  are  embedded the
surface  layer  with high  gradients of  the  primary variables, and the  mixed
layer with nearly uniform vertical  distribution of  the variables.


                                     3-16

-------
     2500
     2000
   1
     1500
co
   2 1000
      500
        0
        -10
              if-
JANUARY
      DAYTIME
      ELEVATED
      INVERSION
                                         UNSTABLE
                                                                  ELEVATED
                                                                  INVERSION  LAYER
                                                                                     NEAR NEUTRAL
                                                                                                  UNSTABLE
                                                                NOCTURNAL
                                                             SURFACED-BASED
                                                             INVERSION LAYER
10                           20

 AMBIENT TEMPERATURE (°C)
                                                                                30
       Figure 3-4.   Monthly-average  diurnal and seasonal variations of the  vertical  thermal  structure  of the
                    PBL for a  rural  site near St. Louis, MO based on  1976 data.

-------
2000
1500
LU


CD

I—I
X
•—I
s:

LU
CD

DC
LU
1000
 500
       MIXING HEIGHT


      ST. LOUIS 1976
                                                  JULY
                                   i   I   i   i   i   i   i   I   i
                 06
                                         12

                               HOUR OF DAY
18
 Figure 3-5.   Monthly-average  diurnal  and seasonal  variations of mixing
              height  near  St.  Louis, MO,  based on 1976 data (Gillani et
              al.  1981).
                                  3-18

-------
Late  in  the  afternoon,  when  ground  level  insolation  has  diminished
siderably,  the  ground  begins  to  cool  gradually.   For  a  brief  period,  it
attains  nearly  the same  temperature  as  the air  immediately  above,  there is
negligible  heat  flux  at  the  interface,  and  the  potential   temperature  is
nearly  constant throughout  the  PBL (neutral).   Thereafter,   no  upward  heat
flux  occurs,  and  no energy  supply sustains the  convective  eddies.   Conse-
quently,  the  intensity of the  turbulence  diminishes quite rapidly  from the
top of the PBL downwards  (Caughey and Kaimal 1977, Ching et al. 1983) and the
mixed layer collapses.  After sunset, the ground cools off rapidly by release
of  its  stored thermal   energy  in the form  of  long-wave radiation.   Thermal
relaxation  of  the  air  above  is  much  slower.   Hence,  the  ground becomes
increasingly colder than  the air above,  and a  deepening surface-based inver-
sion  slowly develops.

The change in the lowest  portion of the transport layer from very unstable in
the day  to  very stable at night is especially dramatic in  summer.   Particu-
larly on evenings with clear skies and light-to-moderate winds,  the surface
inversion  layer becomes  extremely stable  and strongly  suppresses  vertical
transport of mass, momentum, or  energy.   The heat flux is now downward owing
to the inverted temperature  profile.  Turbulence  is  inhibited except for the
small-scale turbulence in the  shallow  surface layer  (also  the  only  mixing
layer, since there is  no  nocturnal  mixed layer).  The  height  of this surface
mixing layer  is typically 100  to 200  m  (Garrett 1982).  Above  the inversion
layer, remnant  small-scale turbulence from  the daytime gradually dissipates.
In the absence of any  effective vertical  transfer mechanism, the layers above
the stable layer become decoupled from the mixing layer and the ground.

Because  turbulent interaction is limited,  the nocturnal boundary layer reacts
slowly to change.   The surface  inversion continues  to grow very slowly  long
after surface cooling has ceased.  This growth may be by a process of gradual
entrainment of  air  from  above, made  possible by  local  generation of  weak
turbulence by wind shear  (Blackadar 1957).   The existence of very strong wind
shear  in the  inversion  layer  will  be discussed  in  the  next  paragraphs.
Because  the  nocturnal   inversion  layer  continues to  grow  for a  long  time,
steady-state  assumptions  concerning nocturnal  dynamics may not be warranted
in  some  problems  (Businger and  Arya  1974).   For a  fine  review  of  the  noc-
turnal boundary layer  dynamics, the reader is referred to Shipman (1979).

The stable inversion layer not  only decouples  trapped  as  well as new release
of pollutants in  the elevated  daytime mixed layer from the ground sink, but
also  prevents  communication  of  surface  friction to  these  layers  above the
nocturnal inversion layer.  The winds in these upper layers are thus released
from  the retarding effect of  friction, and thus begin  to accelerate.    In
contrast, layers  further  aloft  where friction  is   weak  at  all   times,  are
relatively unaffected.   The  surface layer winds, however,  now are subjected
to a more concentrated effect of friction in the absence of momentum transfer
from above, and are decelerated.   There  is  thus an  opposite diurnal  oscilla-
tion of  winds in  the middle  layers as compared to that in  the surface layer
(Goualt 1938,  Wagner 1939, Farquaharson  1939).

The behavior of the flow above  the nocturnal  inversion layer was described by
Blackadar (1957).    The inertial oscillation there  is quite  pronounced,  and


                                     3-19

-------
wind speeds  frequently  become supergeostrophic  in  these layers.   The  phe-
nomenon  has  become widely  known as  the "nocturnal  jet."    Perhaps  a  more
appropriate description of  it is "low-level  nocturnal wind maxima" (Frenzen
1980), because  these  accelerated layers are  not restricted horizontally as
jets are, in the usual sense.   Rather, they  are broad sheets of  faster moving
air.

The nocturnal jet is a very frequent occurrence in St. Louis,  particularly in
summer,  as  shown  in the upper air  St. Louis  wind  data  of January and  July
1976  (Figure  3-6).   The  figure  shows  monthly-average  vertical profiles of
wind  speed  near midday  and midnight  for January and July near St.  Louis,
based  on RAPS  data.   The  following major  observations may be made  about
diurnal  and seasonal  variations  in  transport  layer  wind  speeds, based  on the
average  St. Louis wind data:

   o   There is a  nearly three-fold increase  in the  free stream wind  speed
       (at  2  km,  say)   from  summer  ( ~  6   m   s~l)  to  winter  ( ~  18 m
       s~l).   Wind  speeds are  correspondingly  greater in  winter  in  the
       boundary layer below.

   0   In summer as well as in winter,  the  wind speeds  are  greater  at  night
       than during  the day in the  layers between 100 and 1000  m.   In  par-
       ticular, the wind speed  is supergeostrophic  in much of  these layers
       in the  middle  of  summer nights  and, on the  average, peaks at  about
       500 m.   The peak  value is about 10 m  s'1, on the average.   However,
       values  as   high  as   20 m  s'1  (72  km  hr"1)  have been  observed  on
       occasions.

   o   Based  on  the average  mixing  height  data of  St.  Louis (Figure  3-5),
       the maximum  transport  layer  depth  (peak mixing height  of the  day) is
       about 700 m in January and  about 1700  m in July.   During the daytime
       in both seasons,  relatively  little wind  shear with  height occurs in
       the  transport  layers  above  the surface  layer (  ~ 100 m).   In  con-
       trast,  considerable  wind  shear occurs  at night on the lower  side of
       the nocturnal jet (below 500 m) in both seasons.

   o   In  the  mean pollutant transport layers,  the  average 24-hr  transport
       range based  on St. Louis winds and mixing heights is estimated at 500
       to 600  km  in the summer, and  about 800 to 900 km in winter.   These,
       however,  are  transport  distances along wind  trajectories  and  not
       along  straight lines.   They  thus represent upper  bounds  on  the  aver-
       age  seasonal transport ranges.  The  actual straightline  displacement
       of point emissions  during 24 hr of transport may, on the average, be
       closer  to half of these upper  bounds.   It  is  quite possible, however,
       for  an  individual  elevated pollutant  release to start its  journey
       lodged  in  a strong  nocturnal  jet  and  be transported 500 km  or more
       within  a single night.  On the other hand, it is also quite possible
       for  pollutant  trajectories to  be quite  stagnant or highly meandering,
       thus  resulting in  very  short net displacement  from  the source in
       several  hours.
                                     3-20

-------
o
UJ

o
CO
   2500
   2000
   1500
   1000
    500
           ST.  LOUIS 1976
                     JULY
       0
10
                           WIND SPEED  (m s'1)
20
   Figure  3-6.   Monthly-average  diurnal and  seasonal  variations of the
                vertical  profiles  of  wind  speed  near  St.  Louis, MO,  based
                on  1976 data.
                                   3-21

-------
The inertial oscillation is  not  restricted  to wind speed only.  As the wind
speed increases from the surface wind  to  the  peak jet wind, a corresponding
increase occurs in the strength of  the  Coriolis  force,  and  hence  in wind veer
with height.  Thus, a strong wind speed shear on the underside of the jet is
also associated with a strong wind  directional  shear.   This is  evident in the
St. Louis  data (Figure 3-7),  which show  average vertical  profiles  of the
absolute difference  in  local  wind  direction  at  any  height relative to the
direction of the surface wind.   On  summer nights,  on  the average, the 500 m
winds (at peak jet level) blow at a 60° angle compared to  surface winds, and
this difference is about 100° for layers near the  top  of the transport layer
(about 1700 m).  In  other words, a  daytime  summer pollutant release that has
become well-mixed over the  entire afternoon  transport  layer, may  be subjected
at night to a layered transport in  which the uppermost layers may move nearly
perpendicular  to  the surface  layers.   Clearly,  this  phenomenon will cause
highly distorted and extensive lateral  dispersion of  the  pollutant plume at
night.   The combined effect of  nocturnal  amplification  of wind  speed and
directional  shear,  followed  next day by  vertical  homogenization of all the
separated  layers  into a deep  mixing layer,  will result  in vastly  greater
lateral dispersion over the  time scale of a day  than  that due to horizontal
turbulence.  The role of vertical turbulence  in mixing  all  individual layers
throughout the next  daytime mixing layer, however, is  of critical importance
in such large-scale  pollutant  dilution  and dispersion.   Only  such  a large-
scale  dispersive  mechanism  can  explain  the  rather  rapid incorporation of
strong pollutant plumes indistinguishably  into  the regional background.  In
special plume  studies based  on  aircraft  sampling designed to  track large
power plant  or urban plumes  over long  distances, our  success  in identifying
daytime well-mixed plumes  during  subsequent night-time  transport  has been
rather limited.   Only on  rare occasions has  it  been  possible  to track such
plumes for over 300 km (Gillani et al.  1978).

Blackadar (1957)  attributed the cause of the nocturnal  jet to be  the  shift of
the lower-level thermal  structure from unstable and convective in the day to
stable and  inhibitive of turbulence at night.   This  is consistent with the
St. Louis  observation that the  jet  is  most pronounced  in summer,  when the
lower-level  thermal  oscillation  is  also most pronounced.   This  explanation,
however, may not  be  complete, particularly since  the occurrence of the jet
shows some geographical  preference also, as well  as some extreme  behavior not
fully consistent with Blackadar1s explanation  (Paegel   1969).   Other  possible
influencing  factors  that  have been  implicated are horizontal  variations of
surface heat flux  (Hoiton  1967)  and variations of surface elevation  (Lettau
1967, Mahrt  and Schwerdtfeger 1969).

While nocturnal low-level  wind maxima have been observed in many  parts of the
world (for a comparison of Wangara, Australia  and O'Neill, Nebraska  data see
Mahrt 1980), they are especially remarkable in the Great Plains region of the
United States.   It  is  there  also  that the  phenomenon  has been most  fully
documented.

Strong,  southerly jets  over  the  Great  Plains  have  been  observed  in all
seasons, but especially  in  summer  (Bonner et al.  1968,  Bonner  1968).  They
are most  frequent  and generally  better developed  at  night. The jet  becomes
most  pronounced  sometime  between midnight  and sunrise.    The observed wind


                                     3-22

-------
           ST. LOUIS 1976
    1000
     500-
 o
 a:
 UJ

 o
 CO
 
-------
speeds in the jets are frequently supergeostrophlc.   In  their  analysis  of  10
selected cases over the Great Plains, Bonner et al.  (1968)  observed  the peak
speed to be, on the average, 1.7 times  the  apparent  geostrophic  speed,  which
ranged  from 10  to 26  m  s'1,  and  the ratio  was  as  high as  2.8  on one
occasion.   Measurements in  Australia  showed  speeds at 300 m reaching 1.5
times the  magnitude  of  geostrophic  wind (Clarke  1970).   Perhaps  the  most
remarkable documented jet (on the night of  March  18,  1918  at Drexel,  NB) was
characterized by speeds  of  up  to 36 m  s"1  (130  km  tr1) at a  height of 238
m, while  surface  winds were at  3  m s"1, and  the geostrophic  wind at  about
10 m s-1 (Blackadar 1957).

Spatially, the diurnal inertial  oscillation  is believed  to be  a function  of
latitude  (Thompson  et al.  1976),  being  stronger at lower latitudes.   The
amplitude of  the  oscillation about  the mean  speed  was just  detectable  in
Minnesota, significant in Kansas (amplitude =  2 m s'1), and more pronounced
in Texas  (2  to 3 m  s'1).   Hering  and   Borden  (1962)  observed  the  average
amplitude  based  on   6-hourly  data  of  July  1958  in   Fort  Worth,  TX and
Shreveport, LA  to  be about 3.5  m  s"1. The  average  St. Louis  data of  July
1976 show the amplitude to  be about 2 to 3 m s-1.

Wind  field  measurements  are routinely  made in  the  United States at hourly
intervals at  several  hundred ground stations.   Rawinsonde measurements  of
upper air  winds are  made  over  a  much sparser  network, typically  at  12-hr
intervals, at noon and midnight GMT, or approximately early morning  and  early
evening in  the  eastern  United  States.  At  some  stations,  6-hourly  soundings
are made.  Bonner (1968)  studied the climatology of the  lower-level  jet  based
on the 6-hourly (if available)  or 12-hourly  data of 47 rawinsonde stations  in
the United  States  over  a period of  2 years.   The most   relaxed criterion  he
used  for  the  definition  of a low-level jet was the  occurrence of wind  speed
of  at least  12  m s'1  in   the  boundary layer,  and decreasing  above  by  at
least 6  m s"1 below  a  height  of 3  km.   His plots of the  frequency  distri-
butions of  low-level  nocturnal  jet occurrence in  the United  States east  of
the Rockies for the periods October  through  March  (winter)  and April  through
September (summer) are  reproduced  in Figure 3-8.  Bonner1s findings  confirm
the prominence  of  the Great Plains  as  the  most likely  region  of these  jets
and  that,  in  this region  at  least,  nocturnal  jets  are more  common and
stronger  in  summer than in  winter.   He  also  found  that  the  early  morning
period was preferred over daytime.   From Kansas southwards the jets  tend  to
be more  southerly and  in  the  northern plains  more  northerly.   Between the
Mississippi River and  the  Appalachian Mountains,  the frequency of  low-level
jet  observations  drops  off sharply.   There  is  a  second but  much weaker
maximum in frequency along  the East Coast.

Presumably, St.  Louis represents a  borderline  location  as far as  frequency
and strength of nocturnal jets are  concerned.   Nocturnal jets  are apparently
much stronger west of St. Louis, and somewhat weaker to  the east.

Bonner's  plots  also   show  a generally  westerly  flow in the  states  between
Missouri and  the  Appalachians.   The St. Louis wind  direction data  are  shown
in Figure 3-9  in the  form  of wind roses (wind direction frequency  distribu-
tions)  in  22  1/2°  sectors  for  the winds at 500  m MSL   (about 1000 ft  above
ground).   By  and  large,  the  transport winds are  southwesterly  in summer


                                     3-24

-------
Figure 3-8.   Frequency  distribution  of low-level jet observations
             within 30°  class  intervals  of wind  direction  at the level
             of maximum wind.   Distributions are for (A) winter months;
             October through March,  and  (B) summer months; April
             through September.   Total number of jets observed during
             each season (over the two years) are given  for each site.
             Black circles  in  (A)  indicate stations with greater
             frequency  of jets in summer than in winter.   Adapted from
             Bonner (1968).
                                 3-25

-------
                                 ST. LOUIS
                                JANUARY 1976
    DAY
o
o
un
                                        NIGHT
                                 ST. LOUIS
                                 JULY 1976
    DAY
o
o
LO
                                        NIGHT
  Figure 3-9.  Monthly-average diurnal and seasonal variations of the
               frequency distribution of wind direction (wind rose)  based
               on 500 m (MSL) wind data near St. Louis, MO, for 1976.
                                   3-26

-------
and  westerly  In  winter,  with  northwesterly  as  well   as   southwesterly
components.

The emphasis on St. Louis data  in this  chapter  is  not intended to claim  its
representativeness for  eastern U.S.  conditions.   Primarily  the  choice  was
based on  data availability and their broad  diurnal  and seasonal coverage.
For  a  comparison of  average  seasonal St.  Louis  winds  with  those in  other
parts of the continent, the reader  is referred  to  Figure 3-28  (Section 3.5)
which shows a regional distribution of wind vectors at  four levels over many
U.S. rawinsonde  sites.    The   seasonal  averages in  the  regional  wind plot
include twice  daily  soundings  at  each site.   For a  comparison  of  average
winds in Missouri and Ohio, the annual average (1960-64)  wind roses at 1000 m
MSL  for the  Columbia,  MO  and the  Dayton,  OH  rawinsonde  sites were also
examined.   Those  wind roses (not presented here)  indicate  little difference
in the evening soundings, and  about 10 percent higher  wind speeds  at  Columbia
in the morning soundings.   On  the  average, the wind direction  over  Columbia
had  a somewhat greater northwesterly  component  and somewhat smaller  westerly
component  than over  Dayton.   The regional  and  seasonal  distribution  of  the
peak afternoon mixing heights  are shown  in  Figure 3-29 and may  be  compared
with the St. Louis data presented in Figure 3-5.

3.3.4  Effect of  Mesoscale  Complex  Systems on Transport Layer  Structure  and
       Dynamics (N. V. Gillani)                                            ~~

Mesoscale  complex systems  are  subdivided  here   into  mesoscale   convective
precipitation systems and complex terrain-induced systems.

3.3.4.1  Effect  of Mesoscale  Convective Precipitation Systems  (MCPS)--Among
the mesoscale storm systems are air mass thundershower cells,  frontal  storms,
squall  lines, and  mesoscale convective  complexes.   Such  systems  are  charac-
terized by significant vertical as  well  as  horizontal  motions.   Lyons  and
Calby  (1983)   have  recently  summarized  the effects  of  MCPS  on  polluted
boundary layers.

In  frontal  zones where cold and  warm air masses  meet,  warm air  rises over
cold air,  and  if sufficient  moisture  is  present in  the  rising air,  the
formation of clouds and precipitation may occur. An  advancing cold front  may
cause cold air to move under warmer air (Figure  3-10a),  while in an advancing
warm front, warm air will ride over colder air (Figure 3-10b).   In each case,
a frontal  inversion forms atop the cold air layer.   Horizontal  convergence of
surface  flow  into the  frontal zone  is also associated with  such  vertical
motions.   A  pollutant  plume   reaching  a  frontal  zone  may  be subjected  to
complex vertical motions, encounters with the liquid  phase,  and sharp changes
of  transport direction  if  it  traverses  into  the   other   air mass.    The
situation  is further complicated by the dynamic  nature of fronts and  by local
interactions with terrain inhomogeneities.  For example, squall  lines form in
frontal zones  and are undoubtedly influenced by  geographic features.  They
are  also highly variable in space and time (Pielke  1981).  Fritsch and Maddox
(1980)  have shown that  the occurrence  of these  squall lines causes major
alteration  in  the synoptic flow field.   These areas  of  intensive  cumulus
convection  can  be tracked for  days  across the  United States.  Squall lines
                                     3-27

-------
          (a)
                         (b)
                          WARM
                            WARM
                                                    COLD
          (c)
                                            INVERSION
                                             — TOP
                                             — BASE
                          ' "  ' '
                                    '''LAND'7' '
min»
RURAL
                 URBAN
                                              RURAL
Figure 3-10.
   Inversions  due  to  advection  and  internal boundary  layer  growth.
   (a)  Frontal  inversion  caused by  cold  air wedging under warmer
   air  (advancing  cold  front);(b) Frontal  inversion caused  by
   warm air overriding  colder  air (advancing warm  front);  (c)
   Modification of an unstable  overland  mixing  layer  within a
   growing stable  internal  boundary layer  (dashed) over water
   during offshore daytime  advection  on  a  warm  day (temperature
   profiles are shown);  (d)  Modification of a stable  over-water
   inversion layer within a  growing unstable internal  boundary
   layer (dashed)  over  land  during  onshore daytime advection on
   a warm day;  (e) The  growth  of an internal mixed layer  (between
   dashed lines) due  to  urban  heat  flux  into an otherwise  stable
   nocturnal boundary layer.   Adapted from Oke  (1978).
                       3-28

-------
that become stagnant over one area can  produce  devastating  floods  such as the
one in Johnstown, PA in July 1977  (Hoxit et al. 1978).

Cloud  processes  in MCPS  have a  strong influence  on  PBL  height,  mean and
turbulent flow and thermal  structure, and pollutant  distribution.  The forma-
tion of cumulus  clouds,  like PBL growth,  is  related to vertical convection
(e.g.,  see  Manton  1982).    The  top of  the mixing  layer  is an  uneven and
undulating interface, characterized by  patches of mixed layer air extending
into the quiescent layers above.   The mixing layer is deepened by  penetrative
convection; i.e.,  individual  thermals  or updrafts that  rise  to  the tops of
these  patches penetrate  further  into  the upper  layer  (e.g.,  Mahrt and
Lenschow  1976).    Cumulus  clouds  form  when  rising moisture-laden  air in
updrafts  finds  its condensation  level  at  or  below the elevated inversion
base.   The  latent heat  released  by the condensation  of moisture generates
strong  convective  currents  within the  clouds  and causes  them to expand
upwards.  Large  storm  clouds can  grow to  heights of several  kilometers and
can  thus  provide  an  avenue for  boundary  layer material  to  ascend  to  such
heights.

Convective mesosystems  ranging  in size   from large  isolated  cumulonimbus
clouds  to massive mesoscale convective complexes (MCCs) (Fritsch and Maddox
1981)  profoundly  alter the  structure  of the  PBL out  of  which  they evolve
(Lyons and Calby  1983).   The upward transport of PBL material in relatively
compact supercell thunderstorm systems  has been estimated  to be on the  order
of 10  million metric  tons  per second  (Mack and Wylie 1982).  MCC storms are
larger,  with  greater  associated  upward  transport.  Associated  with   such
updrafts  are  compensating  downdrafts  around the  clouds, large  infusions of
mid- and  upper-tropospheric cold and  clear air  into  the  PBL,  and surface
mesoscale high-pressure regions.   Such mesoscale vertical  circulations  were
detailed  by  Byers  and Braham  (1949),  and  the  production  of  the surface
mesohighs were reported  by  Fujita  (1959).   The divergent surface mesohighs
associated with  the larger MCC storms occupy multi-state  areas (Maddox 1980).
Such mesoscale systems are also common  over much of  the  eastern United States
during the warm  seasons.

Cloud venting of  PBL pollutants has been  discussed  by  Lamb (1981),   Ching et
al.  (1983), and  Lyons and  Calby  (1983).   With satellite  imagery, Lyons and
Calby observed the development of  a mesoscale "hole"  of  clean  air  in  the PBL,
embedded  within  a polluted  air mass.   They performed a case  study of  this
event, and attributed its cause to several  types of  MCPS.   The "hole" covered
Virginia, Maryland, Delaware, northern North Carolina and  extended more  than
500  km  out  to sea.   Within  the  "hole",  daytime surface  ozone  levels  were
considerably   depressed  and  visibility considerably enhanced.    The "hole"
existed for  at  least 36  hours.  The authors used visibility data  and assumed
typical sulfate/visibility relations to estimate the  total  removal of sulfate
in the development of the hole.  This  estimate ranged from 16 to  32  thousand
metric tons of total  sulfate removal  in the MCPS area.  Based on precipita-
tion amount and  "typical"   precipitation sulfate  concentrations  reported in
the  literature  for the  area, the  authors established  an estimate  for the
likely  fraction  of  sulfate removal attributable to  wet  deposition.   The
remainder was  assumed to  have been transported  vertically by the clouds.  The
conclusion was  that massive  quantities of  sulfate, perhaps  two-thirds of


                                    3-29

-------
the total  removal, may  have  been  transported  in  thunderstorm  updrafts to
heights of 10 km or more.

Cloud venting of pollutants out of  the  PBL subsequently results in floating
elevated debris  when  moisture  supply to the  cloud  system terminates in the
evening and  the clouds finally dissipate.    Such  floating  debris manifests
itself  as  elevated haze layers,  which  have  been  observed  frequently (over
large areas  of  eastern United States,  according  to lidar measurements made
during EPA's Project PEPE field study in summer 1980).   Such  floating debris
is likely to have a long residence time in the atmosphere and may be brought
down by downdrafts of future mesoscale systems.  Cloud  venting processes, and
many  other  vertical  motions,  are   largely   ignored  in  current long-range
transport  process  models.   A highly sophisticated  regional  model, currently
under development by EPA (Lamb  1981), aims  to  incorporate many such processes
in  the  formulation.   However,  considerable further quantitative research is
needed  before   adequate  information  is  available  to  parameterize   such
processes.

Even nonprecipitating fair-weather cumuli  play an  important  role  in pollutant
budgets.   Cloud  droplets provide  the medium for rapid  liquid-phase chemistry
resulting  in the  transformation  of  precursor emissions to acidic products.
Once formed, the aerosol products  may have  longer  atmospheric  residence time,
hence farther range of  impact.  Gillani and Wilson  (1983) have observed  that
when an elevated power  plant plume  is entrained  into a  growing  late morning
mixing  layer capped by  clouds, it passes en masse  through the clouds, giving
a  rapid burst of aerosol formation.   In the  afternoon,  such  a plume becomes
well-mixed in the  mixing layer, and  if  scattered  clouds still  prevail at the
elevated inversion base, the  plume  material  is cycled  into  and out of  such
clouds,  giving   rise  to additional   aerosol  formation.   The  period  of  such
cycling may  typically be  about 30 to 50 minutes,  with  perhaps  one-tenth of
the time being spent in the cloud stage (Lamb  1981).

Cloud  processes  also  influence  PBL  growth.    By  reducing  ground  level
insolation  and  heating, clouds cause a decrease  in  surface  heat  flux and
hence in PBL growth by penetrative convection.  The downdraft of colder upper
level  air  around  clouds,  injected  into the  sub-cloud  layer,  leads  to the
stabilization  of  the  cloud base level layer  in  the  region between cloud
patches, thus tending  to inhibit  further  cloud formation as  well as  further
mixing  layer growth  in  the   cloud  free  areas  (Garstang  and  Betts 1974).
Reduction  of insolation  by  clouds  also  inhibits  photochemical  reactions
involved  in  the processes  of  chemical  transformation  of precursor emissions
to acidic  products.

Mesoscale  convective  systems cannot  be  adequately resolved  spatially or  tem-
porally  by the  existing upper air  weather monitoring  network.  Nor can the
denser monitoring  network of surface winds adequately  fill  the gap,  particu-
larly with respect to vertical motions.  Errors once introduced in  long-range
trajectory calculations as  a result  of  inadequate  treatment  of the mesoscale
flow  will,  of   course,  be simply   amplified  during  subsequent simulation.
Uncertainties in such trajectory calculations  must be recognized and  assessed
through special  field measurements aimed at characterizing and parameterizing
mesoscale  flow  systems.  A number  of mesoscale observational  programs  have


                                     3-30

-------
probed  into  such  mesoscale  phenomena  (e.g.,  Project SESAME,  Lilly  1975,
Alberty et al. 1979; Project  GATE,  Zipser  and Gautier 1978, Frank 1978;  and
Project VIMHEX, Betts et al.  1976), while  a Prototype  Regional Observing  and
Forecast Service (PROFS, Beran 1978) has proposed development of a mesoscale
forecast service, initially for the Denver  area.

3.3.4.2  Complex  Terrain  Effects—Surface  inhomogeneities in terrain rough-
ness, height, and heat and moisture fluxes  can perturb  the downwind condition
of the existing atmospheric boundary layer. The  perturbed layer, originating
at  the surface  source of the  disturbance,   grows  upward  with increasing
downwind distance  and  constitutes  an  internal growing boundary layer within
the  outer  existing boundary  layer.   Such mesoscale  perturbations  are most
commonly encountered in shoreline  environments,  downwind of urban complexes
or other heterogeneous land use  sites,  and in hilly or mountainous regions.
The  internal boundary layer may be characterized by altered mean flow field,
mechanical  turbulence, stability,  or a  combination  of any of these changes.
Examples of inner boundary layer growth are shown in Figure 3-10 (c,d,e)  for
offshore and onshore flows at land/sea  interfaces, and  for flow past an  urban
complex.   These  examples   are for  relatively  strong  upwind flow (i.e.,  the
undisturbed synoptic flow).   In  such cases,  the effects  of the  disturbances
are  transported along in a growing internal boundary layer  until they weaken
and  become  indistinguishable  within the outer boundary  layer.   Under weak
synoptic  flow conditions,  the  effects  of  the disturbances  are  not  thus
stretched out  far  downwind,  but are trapped  in localized  recirculating flow
patterns dominated  by  the  nature  of  the  disturbance.   In such  cases, pol-
lutant accumulation is likely.

3.3.4.2.1    Shoreline environment  effects.   The continental  United  States
(excluding Alaska) has about  16,000 miles  of  coastline (including the  Great
Lakes).  The  Great Lakes  cover 95,000  square miles and  have  a  shoreline of
nearly 3600 miles.  About  15 percent of  the United States  population,  over 60
percent of  the Canadian  population, and even larger  fractions  of  U.S.  and
Canadian national  industrial  activities  are concentrated  in the Great  Lakes
Basin (Lyons 1975).  A large  number of  power  plants  and  several  major  urban
complexes dot the shoreline of the  Great Lakes.   Large  bodies of water under-
go  far  fewer  diurnal  and  seasonal  variations in  temperature  than  do  the
surrounding lands.  Also,  the  water surface is relatively  smooth.  Turbulence
and mixing depths over water  are thus considerably different from those over
land.  Because of these sharp  differences in thermal and mechanical features,
the  potential  exists for  extreme mesoscale air mass modifications in shore-
line  environments.   Only   a  brief  outline  of some  of the  major effects of
coastal flows  on pollutant transport  is  given  here.   For a  more  detailed
review of this subject, see Lyons (1975), Hunt and Simpson (1982), and Pielke
(1981).

During the "warm" season,  as warm and well-mixed air flows offshore over  the
cooler water  surface,  intense  stabilization  occurs,  giving rise to  a low-
level inversion that decouples  the warmer  air aloft  from the water  surface
(Figure 3-10c).    Pollutants  from elevated  sources  in  such  cases  may  be
transported  over  water  for  long  distances   without  any  deposition.    In
contrast,   during  periods   of  cold  air  advection over warmer water  in  the
"cold" season, a stable air  mass  can  be  rapidly  transformed to  a  growing


                                     3-31

-------
boundary  layer  of  neutral  or  slightly  superadiabatic  lapse  rate.    As a
result, the mixing  depth  and  diffusion may  increase,  and  also snow  squalls
frequently develop.  Shoreline  plume  releases  may be fumigated to the water
surface more quickly than  inland plumes are  fumigated to  the  land  surface.

Of  greater interest  is  the  behavior  of  shoreline plume  releases during
onshore flow conditions (Figure 3-10d).  During the  warm season,  the  land  is
warmer than the water  during  the day.   Even in July, it  is common   to  find
pools of  cold water (4 C)  at the center of  the Great Lakes.   Sharp  tempera-
ture  gradients  exist  in  a  narrow band of  warmer  near-shore  water.     An
airstream  blowing  toward  land  and  already   stabilized by  long passage  over
water is  subjected  to  internal  boundary-layer growth as it  passes over  the
warmer  surfaces  during the  daytime.    Within  this  boundary  layer,  the  air
becomes unstable and conducive to rapid mixing.   Above, the air is relatively
stable. Emissions  released from  short-stack sources at the  shoreline  will
become  trapped  within  this internal  boundary  layer  and rapidly  brought  to
ground.   Emissions  from tall  stacks,  however,  may be transported inland  in
the stable  layer  aloft for many  kilometers until  the boundary-layer growth
reaches the plume height.   Subsequently, the elevated plume will be fumigated
to the ground.  Because the internal  boundary  layer may  be present  for  many
hours in the daytime, continuous elevated source emissions may continue to  be
fumigated  for several  hours,  thus creating  potentially  high doses of local
pollutants.  Similar elevated emissions farther inland would  be  released  in
the convective  daytime mixing  layer  and would be rapidly mixed  vertically
within a short distance from the source.

Analyzing  onshore   flows  under  weak  synoptic   flow  conditions is  far  more
complex in  the  presence  of recirculating land,  sea, or  lake  breezes, which
are caused by the  thermal  gradients  between land  and water.  An excellent
qualitative description  of the diurnal variations  of coastal circulations
during  weak synoptic  flows is  given  by Defant  (1951).   In the daytime,  the
land surface is warmer and causes the  air above  to rise.  Colder air  from the
sea flows  onshore  to  fill  the void.  The risen air  over the  land then flows
offshore and sinks over water.  A vertical  circulation with a sea  breeze  near
the surface is  thus established  if the prevailing  synoptic winds are weak.
At night, the air over the sea is warmer, and the  situation  is reversed,  with
an  offshore  land   breeze.    An  example  of  the   lake  breeze  recirculation
observed  by means   of  the  trajectory  of  a  balloon  launched  at  the  Chicago
shoreline  is shown  in  Figure 3-11.  In  the  case of a coastal  urban area  with
a high emission density, pollution levels can become quite elevated  during a
lake  breeze due to the recirculation  effect.   During  the  lake  breeze,  an
elevated emission can be released in the upper offshore air  flow and  be blown
back  in  the  lower level  onshore flow of  the  circulation.    Land  and  sea
breezes play a particularly important role in local air pollution  climatology
in  locations  such  as  the Los Angeles  basin, where  significant  blocking
effects of complex  terrain are also present.

Numerous observational  studies of coastal circulations and  precipitation  have
been made.  A sampling of these includes Day (1953), Gentry  and Moore (1954),
Plank (1966), and Burpee (1979)  for the  Florida coast; Lyons  (1975)  and  Keen
and  Lyons  (1978)  for   the  Lake  Michigan  coast;   Hsu (1969)  for the Texas
coasts; Neumann (1951)  and Skibin and Hod (1979)  for Israel;  and  Johnson and


                                     3-32

-------
        1200

        1000

         800

         600

         400

         200

           0
                                    1030
                                  1045
                      1100

                        0900
                    ,    . RELEASE
                    54321
                      INLAND
                            DISTANCE  (km)
                                -1   -2  -3
                                OFFSHORE
Figure 3-11.
Side view of the trajectory of a balloon launched at
0900 hr on 12 August 1967 at the Chicago shoreline of
Lake Michigan.  Positions of the balloon are plotted
every 5 min.   Also shown are the positions of the lake
breeze front at 0945 hr and of prevailing clouds.
Adapted from Lyons and Olsson (1973).
                                  3-33

-------
O'Brien (1973)  for  the  Oregon coast.   These  studies have demonstrated  that
transport, and diffusion and precipitation patterns are  significantly  altered
in the coastal zone, and that such mesoscale circulations  are  poorly resolved
in conventional  weather-observing  network  systems,  thus  creating a  serious
problem in  developing  routine operational  forecasts of mesoscale  phenomena.
Analytical and numerical models of mesoscale systems, based on  field  data  of
special studies, are thus   particularly important.   Early model  studies  were
based  on  linearized analytical simulations (e.g., Defant  1950,  Kimura and
Eguchi  1978).   Nonlinear  numerical  models  were  at  first  two-dimensional
(e.g., Estoque 1961, 1962; Pielke  1974a; Estoque et al.  1976).  With extended
computer  capabilities in  the last decade or so,  three-dimensional numerical
models are now possible and provide valuable new insight (e.g.,  Pielke 1974b,
Warner  et al.  1978,  Carpenter 1979).   For a  complete  review of mesoscale
numerical modeling, the reader is  referred to  Pielke (1981).

3.3.4.2.2   Urban effects.   As in the case of coastal  circulations,  urban-
induced circulations are  primarily due  to  the  differential  heating and cool-
ing  between  urban  and  rural  areas.    Indeed,  this phenomenon is commonly
referred  to  as  the  urban  heat island  effect.   The urban area  also represents
rougher  terrain  and a  source  of  enhanced mechanical turbulence  (automobile
traffic  also  contributes to  this effect).    Moisture  fluxes may  also  be
greater in the urban area.

The  most  direct evidence  of the heat island  concept is the  observed higher
air  temperatures in  the  urban areas,  on the  average,  than  in  rural  areas
(Chandler 1970,  Clarke  and  McElroy 1970,  Landsberg 1956, Oke 1974).   Matson
et  al .  (1978)  used  satellite  imagery  to  illustrate  maximum  urban-rural
differences  ranging up  to  6.5  C  in  the  midwestern and  northeastern United
States  on  a particular  summer  day.   Price   (1979),  using  high  resolution
statellite  imagery, estimated  this difference  to  be as  high as 17 C  for New
York  City--a value  considerably  higher  than  those estimated  from  surface-
based  air temperature measurements.  His explanation  for the  apparent dis-
crepancy  is that the satellite sensing includes  industrial  areas,  rooftops,
as well as  the  trapping of  energy  within  urban canyons  (Nunez and Oke 1977),
which  are not sensed  by surface observations.   Numerous other studies of the
urban  heat   island  have been  based  on  satellite  and  surface-based  observa-
tions,  as well as on numerical calculations.   Many of  these  are reviewed by
Pielke  (1981)  and by McBean  et al. (1979, Chapter 6).  In particular,  the St.
Louis  area   has  been  studied  extensively  as  part  of  the RAPS  and  METROMEX
programs  (a  series of articles  in   the  May  1978  issue of  the Journal  of
Applied Meteorology was devoted to results of Project METROMEX).

The  urban heat island effect is most pronounced at night under weak synoptic
flow conditions.  The  rise  of heated air over the city  is compensated  by a
radial  and  horizontal  convergence  of  flow  into  the  urban  area near the
surface.    A  vertical  circulation is  completed  when  the  risen air  flows
outwards, then subsides over  the rural  areas,  and recirculates to the  urban
source near the surface.   Such  a recirculation  traps  urban  pollution  emis-
sions  when  the  larger-scale  flow  is weak.   When the outer flow  is strong, the
urban  boundary  layer  is  stretched out  downwind  (Figure 3-10e)  rather than
closed and  recirculating.    The  inflow  velocity  in the  recirculating heat
                                     3-34

-------
island  flow is typically  about  1.0 m  s-1  in New  York City  (Bornstein  and
Johnson 1977)  and about 0.4 m  s'1  in  St.  Louis (Schreffler 1978).   There  is
also apparently a tendency for anticyclonic  turning  in this convergent inflow
(Bornstein  and Johnson  1977,  Lee  1977).   The heating  within the  nocturnal
urban heat island also produces a local  unstable  mixing  layer  deeper than  the
rural mechanical  mixing  layer.  Oke  (1973)  concluded  that the heat  island
effect  of a  city on  the surroundings  under cloudless  skies is  inversely
proportional  to the  large-scale  wind  speeds,  and  directly  related to  the
logarithm of the urban populations.

Quite apart from the local stability and circulation changes due  to  the  urban
area, the emission of  primary  fine aerosols and the  secondary generation  of
aerosols  during downwind  transport of  urban  plumes can produce  significant
haziness  and  reduction  of  incoming  solar  radiation  (White et  al.  1976,
Viskanta et  al.  1977).   There is also evidence of  the  effect of  large  urban
areas  on  climate  and weather.    Project  METROMEX   (1976)  results  indicate
preferred regions of thunderstorm development  downwind of urban areas.

3.3.4.2.3   Hilly terrain  effects.   Hills and mountains alter local atmos-
pheric flows in two ways—by physically blocking or channeling the  flow,  and
by  adding  a  secondary thermally-induced  flow  resulting  from  differential
heating of the  surface and the  free atmosphere at  the  same elevation  (above
mean  sea  level).   Complex terrain  effects  are  particularly important  for
urban and  industrial  complexes in  river  valleys and in coastal  and  inland
plains  backed  by  mountains.    Denver  and  Los  Angeles are  good  examples.
Emissions  from tall  stacks  in  mountainous  terrain may   impinge  upon  the
elevated  ground after  only  short-range  transport.   Stagnation  in  blocked
flows (e.g.,  Los  Angeles) can lead  to  high  levels of  secondary pollution.
Also, mesoscale modifications  of pollutant flow trajectories past mountainous
terrain (e.g., the Appalachians)  cannot be ignored  in an assessment of  long-
range transport when  the  source  and the impacted regions are  separated by  a
mountain chain.

In the discussion below,  certain important features of  complex  terrain  flows
are  highlighted.   More  detailed reviews are given  by Egan  (1975), Pielke
(1981), and Hunt and Simpson  (1982).

The  principal  features  of the  primary  flow  in  and  immediately  upwind  and
downwind of  the complex terrain  will  be determined  largely by the  shape  and
size  of  the  obstruction,  the  strength  and  direction  (relative  to  the
orientation  of the obstruction)  of the  oncoming  flow, and  by the  strati-
fication (stability)  of  the  undisturbed upwind  boundary layer.   There will
naturally be preferential and accelerated flow through mountain  gaps  and
passes.  When the flow can neither go over or around  the obstruction  because
it is  too  slow or stable, blocking will  occur,  with propagation of  effects
upwind.   Such  damming effect  of  the  Southern Appalachians is discussed  by
Richwien (1978).

The flow of a neutrally stratified  atmosphere  with an  elevated inversion atop
(the typical daytime mixing layer) past a two-dimensional obstruction (i.e.,
perpendicular  to  the  flow) of height  H  less than the mixing  height  h  is
illustrated  in  Figure 3-12  for low (a)  and  high (b) wind  speeds.   In each


                                     3-35

-------
                                                    HYDRAULIC JUMP
Figure 3-12.
Air flow over a two-dimensional  ridge with an elevated
inversion upwind.
(A) Case of low wind speed;  separation can occur downwind.
(B) Case of high wind speed;  mixed  layer  flows down lee
side; no separation; hydraulic jump downwind.  Adapted
from Hunt and Simpson (1982).
                                  3-36

-------
case,  as  the  flow  ascends  the windward  slope,  it accelerates,  and  the
elevated inversion drops somewhat.   If  the  upwind slope is  steep,  a  captive
recirculating  eddy  may  form  at  the  base of  the slope.   The leeward  flow
pattern  is  generally more  complicated.   Depending on  the  speed of the  flow
and  the  leeward slope  of  the  hill,  flow separation may occur  downwind,  and
separate the  main  flow above  from  a captive  recirculating  eddy below  (a).
The wavy nature of the  main flow field can persist for a  significant distance
downwind and  can  even  generate  additional  secondary  eddy motions  downwind.
For  increasing  oncoming  wind  speeds,   the  downward  displacement  of  the
elevated inversion base increases until,  under  an  appropriate  combination  of
the  flow speed,  atmospheric   stability,  and obstruction  height,  the  whole
mixed  layer may flow  down the  lee  side  of the  hill,  producing  a  highly
turbulent and  sometimes recirculating flow (b).   Such  a  wind is known as the
Chinook or foehn.  Lilly and Zipser (1972) observed wind gusts  of about  50 m
s~l  associated with a  Chinook  immediately  downwind  of  the Rockies.   With
the downwind displacement of the  warmer  inversion  layer  air, such a  flow  is
often  also  associated  with some warming of the  lower elevation air  on  the
leeward  side.   At  some point downwind,  the  mixing  layer will   return  to  its
prevailing  larger-scale  condition by rapid  dissipation  of  the mean  kinetic
energy through a phenomenon known as the hydraulic jump.   Considerable mixing
and  dilution  is associated with  the  hydraulic  jump,  while  captive recircu-
lating eddies  represent  localized stagnant  flow.   The atmospheric  residence
time,  dilution, and  overall  trajectory  of  pollutants  in such  flows  is  sig-
nificantly influenced by these mesoscale  features.  Also,  the  forced  lifting
of moist air  on  the  upwind  slopes  causes  condensation and  precipitation,
while comparatively dry air flows on  the  lee side.  Such orographic rainfall
can be responsible for  significant localized acidification  (OECD 1977).

When the  flow is three-dimensional  around  isolated or clustered hills,  the
flow may also go around the obstructions.  The  flow field on the lee  side  is
generally even  more  complex in  such  cases.   The  relative split between  the
flow around and over the obstruction will depend  not  only  on   the height  of
the obstruction and the free flow speed,  but also  significantly on  free  flow
stability.   The greater the stability,  the  less  will be  the  likelihood  of
flow going over the hill.

Thermal or mountain-valley winds  result  from the unequal heating and  cooling
of the  terrain surface at different  heights.   Consequently,  such  secondary
flows exhibit a strong  diurnal  variation. During  the  day,  the  higher  terrain
becomes an elevated heat source, while at night it is  an elevated heat sink.
In the  day, heated air  rises  from  the  higher  terrain  drawing compensating
upslope flow.   A vertical circulation may be completed by  sinking air motion
to the  valley floor.   At  night, the reverse  situation  prevails,  with  noc-
turnal  drainage down the slope.  These daytime  upslope and  nocturnal drainage
flows  are  also called anabatic  and katabatic winds, respectively.    In a
closed  valley, a  recirculating  flow pattern  may be  established  by  such
mountain-valley winds,  and  if  a pollutant  source  emits   into  this  flow,
considerable accumulation can occur.  A  number  of  observational  and modeling
studies of complex  terrain  flows have been reviewed by Pielke (1981).
                                     3-37

-------
3.4  MESOSCALE PLUME TRANSPORT AND  DILUTION (N.  V.  Gillani)

Mesoscale plume transport and dilution are  influenced by  the  height  of plume
release and the configuration  of the source, as well  as by  transport  layer
structure  and dynamics.    Two principal  types of  source releases  are  of
special concern:  stationary elevated point-source releases,  and  near-ground
releases  from an  aggregate  of  sources  in a  broad  area  such  as an  urban-
industrial complex.   In  the eastern United States,  about 92  percent  of  the
anthropogenic  S02  emissions are due  to  fossil  fuel  combustion,  with  about
70 percent from power plants, many with  tall stacks.   Automobiles  emit little
sulfur.   In contrast,  NOX  emissions in  the United States  are almost equally
due to automobiles, electric utility sources, and  industrial  fuel  combustion
(Husar and  Patterson  1980;  see also Chapter  A-2).   Thus, while most  SOe  is
emitted  from  elevated sources,  NOX emissions  are  more  evenly  distributed
between elevated and low sources.   On the  average, elevated  releases spend a
substantial fraction  of  their mesoscale  transport  time  decoupled from  the
ground sink,  while  near-ground releases maintain continuous  ground  contact.
Important  diurnal   and  seasonal   patterns   of dry  deposition,   attributable
directly  to variations  in  the transport phenomena,  exist for both  types  of
sources.

3.4.1  Elevated Point-Source Emissions (Power Plant Plumes)

The proliferation of tall stacks in the  eastern United States  in  the  past two
decades was motivated  primarily  by the regulatory requirement for abatement
of  ground-level concentrations of  SOg  from  large emission   sources   such  as
central  power-generating  stations  (Thomas  et al.  1963).   That  tall  stacks
were  largely   successful  in  this   objective  is  quite  evident  (Pooler  and
Niemeyer 1970).  At the same time,  however, taller stacks and  greater thermal
effluxes from them may have resulted in  increased atmospheric  residence times
for pollutant emissions.  In  turn,  farther distribution  of the  emissions and
increased formation of secondary products  may be occurring.  Tall stacks  no
doubt  result  in substantial  reductions  in ground losses  during  short-range
transport.   But source  height is  unimportant  once   the  plume  becomes  well
mixed  vertically  in the mixed layer.   The extent to which  tall  stacks  in-
crease  pollutant  residence  time during  long-range  transport and result  in
increased secondary formation and deposition has not  yet been  fully resolved.
Results  of  some new and previously unpublished analyses  pertaining  to this
question are presented in this chapter.

The Ohio  River Valley  (ORV) region is well known  to have a  large concentra-
tion  of  central  electrical   power-generating  stations burning fossil  fuels,
particularly coal.  In a recent  study of trends related to power plant stack
heights  and  SOg emissions  in  this region, Koerber  (1982)  focused attention
on  power plants with a generating capacity greater than 50 MW, and located in
a two county  row on both sides of  the Ohio River in  Illinois, Indiana, Ohio,
Kentucky, West  Virginia, and Pennsylvania.   A  total  of 62 such  power plants
were  operational  there between 1950 and 1980.   Figure  3-13  (top) shows the
trend  of  total  SOg  emissions from the study  plants  during the 30 year study
period.    Nearly  a ten-fold  increase  in   generating  capacity  was  realized
during  this  period.  Figure  3-13  (bottom) shows the corresponding  trend of
S02 emissions broken  down  by stack heights.   In 1950,  more  than  75 percent


                                     3-38

-------
          O
          •—«
          I/O
          CVI
          O
          oo
                  1950
                                                          200 m
                                                    100 - 200 m
                                                      0 - 100 m
                  1960
1970
1980
                                      YEAR
Figure 3-13.
Trend in emissions of S02 from 62 study power plants in
the Ohio River Valley:
(A) Total tonnage;
(B) Tonnage breakdown according to specified physical
stack height intervals.
Adapted from Koerber (1982).

                      3-39

-------
of the  S02 emissions  were from  stacks  lower than  100  m, most  of the re-
mainder being from stacks between 100 and 200 m  tall.   By 1980, less than 5
percent  of the  S02  emissions  were  from  stacks  lower  than  100  m,  while
nearly 60  percent  of  the  emissions  were from stacks taller  than  200  m.   Of
the 62  stacks in  1980,  32 were  taller  than 244  m  (800  ft.), and 11 were
superstacks of 305 m (1000 ft.)  height or taller.  The average  stack height,
based on  weighting with respect  to  S02  emissions, increased from  under 100
m in 1950 to about 225 m in 1980.   The ORV study  area is  quite  representative
of the  corresponding  picture  for the United States  and Canada, as a whole.
In the  latter case, more  than  90 percent  of the SOX  emissions  from major
point sources  during  1977-78 were  from  stacks higher than  100 m, about 63
percent from stacks taller than 200 m, and  about 38  percent  from  superstacks
taller than 300 m (Benkovitz 1982).   It is interesting  to note,  however, that
relatively little of this national increase  in stack heights occurred in the
northeast  coastal  states, where  the average  height of  major point source
stacks remained close to 100 m (Benkovitz 1982).

The range  over  which  an elevated emission  maintains its  identity  is  highly
variable.   Tall-stack  emissions  may  be brought  down  to  ground  and mixed
rather  uniformly  throughout  a deep  daytime mixing layer  within  just a few
kilometers  of the source  (Figure 3-14,  top), or  they may  remain  elevated,
coherent,  and decoupled from  the  ground  for hundreds of  kilometers at  night
and in winter (Figure 3-14, bottom).   Such diverse  plume  dispersion is due to
the  pronounced  vertical  stratification  in  the   transport  layer  structure
(unstable mixing layer versus stable  layers aloft),  and  the  enormous  diurnal
and seasonal  variations in PBL dynamics.    Vertical plume spread  is  caused
predominantly by atmospheric turbulence;  turbulence continues to play  a  vital
role in plume dilution long after the plume fills  up the  peak  daytime mixing
layer,  and loses  its  source identity.   Horizontal plume  spread by turbulent
diffusion,  on  the other  hand,  is  mostly   significant  only during  initial
transport,  i.e.,  until the  plume is a  few kilometers wide.   Increasingly,
wind shear and veer effects, and wind shifts, become the principal mechanisms
of horizontal spread.   As  a  well-mixed  daytime plume journeys  into night,  it
may become sheared into multiple  layers moving off  in  different  directions.
The next  day, as  the  mixing  layer  grows, each higher layer is entrained  in
turn  and  diluted  over  the entire height of the  mixing  layer by  turbulent
vertical  diffusion.   This process of nocturnal  horizontal  shearing followed
by  daytime vertical  dilution may  be  repeated  through   successive  diurnal
cycles  and is most probably the mechanism whereby individual  large plumes  are
homogenized rather quickly into the regional background.

The  vertical  and  temporal  features  of  the  transport  and  dispersion  of  a
tall-stack  plume  during a typical hot  and  humid  midwestern U.S.  summer  day
are illustrated in Figure  3-15.   The emissions represent the 0700 hr release
on 23 August  1978  from the two identical 305-m stacks of the Tennessee Valley
Authority's  (TVA)  Cumberland Steam  Plant  (2600 MW  generating capacity)  in
rural northwestern Tennessee.  Such multiple stack emissions typically become
mixed and indistinguishable  rather  quickly.  The  buoyancy of  the efflux  led
to  a  plume rise  that resulted  in an effective  stack  height (physical  stack
height  plus plume  rise) of about  500  m  and  an initial  plume  spread in excess
of  100  m  vertically.   The bent-over plume  was then transported  in a stable
environment  at  this   height  in  relatively  coherent  form  until   the  rapid


                                      3-40

-------
Figure 3-14.   (TOP)   Rapid  vertical  dispersion  of  a  tall-stack  plume within
              a midday unstable  mixing  layer  in the  summer.  Such a plume  is
              typically brought  down to ground  within a  short distance from
              the source.

              (BOTTOM)  Transport  of a  coherent tall-stack plume in an ele-
              vated  stable  layer during winter.  Such a  plume has a signi-
              ficant likelihood  of remaining  aloft over  long-range transport.
                                    3-41

-------
3-43

-------
                    1500
                    1000 -
CO
i
                           CUMBERLAND PLUME
                           AUGUST 23,1978
                          (BNL and EMI  DATA)
  500 -
                              0800
                       1000
1200

TIME OF DAY
1400
                                                                             1600
                                                                      1800
                                                  80 km  110 km
                                           DOWNWIND DISTANCE
                                            AT SAMPLING
                                                                                 160 km
    Figure 3-15.
The physical  behavior of a tall-stack plume on  a  rather typical  summer day.  The  plume
shown is the  reconstruction of  the  Lagrangian transport of the 0700 release on  23 August
1978 from  the 305 m tall stacks  of  the 2600 MWe Cumberland Stream  Plant in northwestern
Tennessee.  The reconstruction  is based on aircraft sampling, ground-based lidar  returns.
and tetroon transport data (Gillani  and Wilson  1983).

-------
midmorning rise of  the  unstable  mixing  layer reached and exceeded the plume
height.  Entrainment into the mixing layer followed, subjecting the plume  to
vigorous  mixing  and  rapid  spread.   Within  about  1  hour,  plume touchdown
occurred on  the  ground, and ground  removal  of the  pollutants  by dry depo-
sition began.   The  plume  quickly filled the  entire mixing layer following
entrainment, becoming  rather  uniformly spread  out  in  the  vertical   domain.
Thereafter,  pollutant   concentration,  and hence  the  rate  of  ground loss,
varied inversely with the mixing  height.   The plume  continued  to dilute until
the mixing height reached its peak value  in  the midafternoon.  Subsequently,
as the mixing  intensity diminished and  the  mixed layer collapsed, the plume
remained  diluted,  with  its  top  at  the   height of  the  peak  daytime mixing
height.   If  any  further upward dilution  occurred,  it  must  have been small.
In the evening, with the formation of the nocturnal, surface-based inversion
layer, the bulk of this daytime plume (except the bottom part in  the  shallow
nocturnal,  mechanical   mixing  layer)  presumably  became decoupled  from  the
ground sink  (no  data was taken  after  1800   hr).   During the  night,  if the
nocturnal jet  developed (as  it  frequently does),  this bulk probably  experi-
enced relatively rapid transport, as  well  as  considerable  shearing  spread and
distortion.

In  the  example  described  above,  convective clouds  also developed  at the
elevated  inversion   base  during  midday.     Direct  evidence  of  substantial
plume-cloud  interaction,  particularly  during  plume  entrainment  into  the
mixing layer,  was observed;  this interaction was accompanied by  significant
in-cloud chemistry (Gillani  and Wilson  1983,  Gillani et al.  1983).  Such fair
weather  cumulus formation is  fairly  common   in  the  eastern  United States  on
summer days,  being  more common  in the southern  half  of the eastern United
States than  it is  in the north.   Elevated  nocturnal  plume  releases  that  do
not rise  sufficiently  high  and  become  entrained before such cloud formation
begins may experience no interaction  with clouds during entrainment.

The reconstruction of  the physical evolution of the example plume was based
on  aircraft  data   and  on  ground-based  lidar data.    It  illustrated  the
"Lagrangian" transport  of a particular plume  release (the 0700  hr  Cumberland
plume  release  of  23 August 1978)  in  terms  of variations in the  time-height
plane.  The lidar data  (Figure 3-16) were collected by  the  Stanford  Research
Institute (SRI) lidar  (Uthe et al. 1980)--a  laser-radar  system operated from
a mobile  van.   In  this  system,  a  laser  beam is fired at equal intervals  of
travel  distance (horizontally   under  the  plume  section  in  the crosswind
direction, in  the samples  shown),  and  the lidar returns  (backscatter of the
beam by  atmospheric  aerosols) are  processed  into  these  visual  images.  Dense
aerosol  layers (e.g.,  the  plume and clouds)  appear whiter  than the back-
ground,  as  does  the more polluted mixing layer,  in contrast to  the  cleaner
stable air  farther  aloft.   As  the laser  beam penetrates a cloud,  it  becomes
attenuated; black bands thus appear above the point  of total  beam  extraction.
In the pictures the  letter  C  identifies  a cloud,  P  refers to a plume, and T
denotes  the  top  of  the mixing layer.  The time frame  of  the measurements  is
marked atop  each picture.   In  the example shown,  the lidar was in  operation
about 30 km downwind of the power plant.

In  Figure  3-17,  "Eulerian"  views  of  the  plume vertical cross sections at a
fixed downwind distance  (35 km) from the Cumberland stacks are illustrated  at


                                     3-44

-------
Figure 3-16.   SRI  lidar photographs  showing  the  structure  and  dynamics  of
              the  boundary layer and the Cumberland  power  plant  plume,  30
              km downwind of the source, on  23 August  1978.   (P=plume,  C=
              cloud,  T=top of mixing layer.)  Adapted  from Gillani  and
              Wilson  (1983).
                                    3-45

-------
    0940
                                09 BO
                                                                                1020
  79 Wett    U         79 East
      Indian  Mound Rd.
          U
       Cook Rd.
79 West      U       79 East       U
       Indian Mound Rd.         Cooper Creek
     1030
                                   1040
                                                       1130
                                                                                    1140
Watt          U
       Indian Mound Rd.
79 East           U
           Wood lawn Rd.
                                                              79 East

                   U
              Lylewood Rd.
                                                                                            79 West
1250CDT
                                1300
                                                                                   1640
             79 East
                                    X      79 West
                                  Co. Line
                       X   U    X 79 East
                     79/120      79/120
                                                  3-46

-------
Figure 3-17.   Lidar photographs depicting the diurnal  variation of the
              vertical  cross-sectional  structure of the Cumberland plume
              on 18 August 1978.  All  data were collected  at the same
              distance (about 35 km)  downwind of the source (Uthe et al.
              1980).
                                    3-47

-------
                                                                                                ALTITUDE - km
                                              Or
       Or
GO
!
-n>
oo
       ro
       O
                                       CD
                                       ro
                                       O
                                      o>
m
l
o
                                              ro
                                              o
                                     •2
                                      o
                                             (Ji
                                             o
                     CJI
                                                                                     Or
                                                                             O)
                                                                             01
                                                                             O
                                                                                    ro
                                                                                    O
m
I

o
o
                                                                            •s
                                                                            o
                                                                            g>
                                                                            o
                                                                                    o



                                                                                    3

-------
different times  of another day  (18  August 1978)  under different stability
conditions.   At 0540 hr, the elevated Cumberland  plume  is  in stable air and
has  a  curious >-shaped  vertical  cross  section,  which is  anything  but the
horizontal,   elliptical,   Gaussian  shape  commonly   assumed   in  many  plume
diffusion models.  The distorted shape  is a consequence  of wind shear both of
speed and direction with  height.   At  1000 hr,  the  plume  section is vertically
very thin (100 to  200 m) but is  fanned  out (about 10  km or more wide) in the
crosswind direction, and  is tilted.   Such plume fanning is typical in stable
air.  The plume  is  still elevated and decoupled from  the ground sink, but an
unstable daytime mixed layer has formed and risen  to a height of about 400 m
(P = plume,  T  =  top of mixed layer).  Upon further rise of  the mixing-layer
top, this elevated plume  would  become entrained and  mixed down to the ground.
Subsequent plume  releases  within this  layer might  fail  to penetrate out of
the inversion lid at the  top of the mixing layers.

By 1600  hr,  the  mixed  layer has grown to  1500 m,  and the  plume is entirely
within it, well  mixed  throughout, and  subject to ground removal.  Also, the
plume has a  large  cross  section,  with  lateral spread exceeding 25  km  (at a
distance of  35  km downwind from  the source).   The  plume  is diluted by the
background air, and the conditions within  it are conducive to photochemical-
ly-driven  formation of  sulfates  and   nitrates   (assuming  the  presence  of
reactive radical  and  organic  species in  the  background).   By 1830  hr,  the
mixing layer has  collapsed  (the  daytime mixed layer of aerosols, of course,
cannot reconcentrate).   The boundary layer has a neutral-to-stable strati-
fication.  Two plumes are evident:   (1)  a fresher  (about 1.5  hr old) elevated
plume (middle right), released at about  1700  hr,  which  has risen quite high
(1500 m or five  times the physical stack  height) and  is coherent, and (2) an
older well-mixed plume (lower left),  within the daytime mixed layer.  During
the night, the lower plume  has a greater  likelihood of  getting a ride in the
nocturnal jet, with  expected wind maxima  in  the  300  to 900  m  layers.   The
upper plume would be expected  to remain  concentrated and transported at about
1500 m  throughout the night and much  of the next day until (and  if)  the
mixing layer on  the next day  rises  high  enough to  entrain  it.   If the next
day's mixing layer  does  not rise to 1500 m,  the plume will  travel  on,  de-
coupled from the ground,  until  it is brought down in  the future, either by a
deep enough mixing layer, by sinking  air, or by rain.   That  particular plume
release is likely  to  have  a longer  atmospheric residence time than does the
average  summer  plume and,  accordingly,  its  impact range  is likely  to  be
farther afield.  Rise of  coherent plumes to heights  of 1500 m  is probably not
very common except possibly in  the case of emissions from superstacks (> 300
m).

An  important feature of  tall-stack  emissions is  that they  can  remain  de-
coupled from the ground for a long time.   An  example  of such elevated plume
transport  in the  stable  layers  appears  in  Figure  3-18,  which  shows  the
nocturnal transport of the Labadie power  plant plume  near  St. Louis,  MO,  on
14-15 July 1976.   The  Labadie  stacks are 214 m high.   Lidar data (Uthe and
Wilson 1979)  show  a  side  view (time-height  plane)  of longitudinal  plume
transport over  85 km  and  a vertical cross-sectional  view of the  plume  at
nearly 100 km  downwind distance.  During  much  of the  night, the  plume  was
transported in a thin  layer at  a  height of 400  to 500 m and  had the fanning
spread characteristic  of stable  plumes (see the  cross section  at 100  km


                                     3-49

-------
Figure 3-18.   The longitudinal  and cross-sectional  structure of the  Labadie
              power plant (2400 MW)  plume during  nocturnal  transport on  14-
              15 July 1976 (Uthe and Wilson 1979).
                                    3-50

-------
                                                                  Top
                                                                  View
      Route  of mobile  lidar observations of  the Labadie plume
                          on 14-15 July  1976
        ?320
                                 LOCAl. TIMF .COT}  - lionts

                           2340      J350       ?WO
     24
           28
     i       i

 Missouri f*- Missouri -i—	
  100    340
   43             59
       DISTA"»CF l.hom Labdt

East on US 40 		~f~
                                                                 0020
                                                                              Side
                                                                              View
                                                                       85
                                                      East on US 70
                       0040
                   0 0,75 U
                                          G,,i Si,,i

                                0050       U100 P120
                                  'Cross-sectional  View
                                  at  90-100 km  Downwind
Downwind Labadie plume  structure observed on  14-15 July 1976 using the
                    SRI mobile Mark IX  lidar system
                                  3-51

-------
downwind, with  a  lateral  width of 13 km and  a  vertical  thickness < 100 m).
The plume was also horizontally tilted  at  this  cross section.  The  apparent
looping  of  the  plume  during  early transport (over  rather flat terrain)  is
most probably not what it seems to be; rather,  in  its  zig-zag course  under
the plume, the lidar may simply have been  sequentially looking up  at  parts  of
a  tilted  or a >-shaped plume  that had highly  variable  local  heights.  The
nocturnal  plume transport  shown   had  a  speed   of  about  10  m  s~l  (35  km
hr~M.   Trapped in  such  a high-speed  layer,  the plume  can be  transported
well over 500 km from 1800 hr to 1000 hr the next day without any  deposition.

Because tall-stack emissions of acid precursors  represent a large  fraction  of
the  total,   the  following  question is  of  considerable  importance  to the
subject  of  chemical   transformations,  atmospheric residence  time,  range  of
transport,  and  deposition:   How much time  does a given  tall-stack  emission
spend  aloft and decoupled from the  ground  sink?  This  question  pertains  to
interactions  of  the  plume  and  the mixing  layer.    Because mixing-layer
dynamics are out of  our control,  the height of  the  plume is the controllable
variable of  interest.  This  height depends  on the physical  stack height and
the  plume  rise  (Figure  3-15),  which  at  times can  be  several  times the
physical stack height.

The emissions from  a  tall  stack  are accompanied by an  efflux  of heat and
momentum.   Consequently,  the plume  initially is a  rising buoyant jet.  Its
interaction  with  the prevailing wind and the ambient atmospheric  turbulence
results  in  plume  bending  and plume spread  by the entrainment of  ambient air
(Briggs  1969,   1975).    In  a  stable  atmosphere, the   plume  rapidly  loses
buoyancy and attains  its  final plume rise.   It  remains  vertically quite thin
while  fanning out horizontally by  shearing effects.   In  a neutral  or  unstable
atmosphere,  the plume  maintains buoyancy for  longer  times  as it loops  up and
down in  the  convective up-and-down drafts.   Plume dilution  counters its net
buoyant  rise, and the prevailing  wind  causes it to bend over.   In  general,
plume  rise  increases  with  increasing  stack heat  flux  and decreases  with
increasing  wind speed and  atmospheric  stability.   For  the  same  stability,
wind  speed, and  exit conditions,  plume rise   is  also   greater  with  lower
ambient  temperature.   At  night   and  in winter, the effects of  increased
stability   and   wind  speed  are   partially  countered   by   lower   ambient
temperature.

Local  wind  speed,   stability,  and  ambient  temperature  in the  vertically
stratified  atmosphere are  in  turn  related to   physical  stack height.    An
example  of  the  effect of physical  stack  height on plume  rise  is  shown  in
Figure  3-19.   The   Johnsonville  stacks (all  shorter than 125  m)  and the
Cumberland   stacks  (305  m  tall)  are  only  60  km  apart  (in  northwestern
Tennessee).  The  plume releases shown  are  rather close  in  time and  are both
in a  nocturnal-type  regime.    The lower Johnsonville  release,  however,  is
within  the  very  stable  nocturnal  inversion  layer,  while  the  Cumberland
release  is  in  near-neutral  layers  aloft.    Even with   somewhat  higher  wind
speeds  acting on  the Cumberland plume, this  plume  rose  up to 1000 m  in  the
example  shown and remained  decoupled from the ground throughout  the  morning.
In stark contrast,   the  Johnsonville plume  remained  trapped in  the  surface
inversion  layer and was  "fumigated" to  ground  before 0800  hr, when  the  sun
caused  the  erosion  of the  surface inversion.   At least  during  short-range


                                     3-52

-------
                 1500 -
                • 1000 -
CO
i
en
CO
               o
               cc.
               03
O
CO
               a:
               CD
          AUGUST 27, 1978
           (EMI DATA)
                  500 -
                                                     10     11     12    13
                                                      TIME OF DAY
                                       DOWNWIND DISTANCE AT SAMPLING: 100 km  150 km
                                                  EMISSION SOURCE:   CUM    JHV
                                                              14
15
16
   Figure 3-19.
   The  physical behavior of the emissions fronrthe Johnsonville (ten stacks, all  less  than
   125  m tall) and Cumberland (two stacks, both  305 m tall)  power plants.   Reconstruction
   is based on aircraft  and tetroon  data.  Adapted from Gillani and Wilson (1983).

-------
transport  (<  100  km),  the  Johnsonville  plume  probably  experienced con-
siderable ground removal, while the Cumberland plume was  spared  such  losses.
The Johnsonville  plume was also  exposed to morning  fog  and its chemistry,
while the Cumberland plume was not.   On  this  day  (27  August  1978), no  cumulus
formation occurred  before  1400 hr at the  top  of the  mixing layer.  If such
clouds had  formed,  the Cumberland plume would  have experienced substantial
interaction with  them  during  entrainment  into  the mixing  layer,  while  the
Johnsonville plume would not have.  Evidently, plume rise can have  important
influence on  plume  sulfur and  nitrogen budgets,  but the  relationship   is
complex.

To  investigate  the  diurnal  and  seasonal   dynamics  of  plume   mixing-layer
interactions, one  must  resort to a  time-varying,  plume-transport  and dif-
fusion model  that  explicitly considers  the distinction  between   diffusion
characteristics in  the  mixing layer and  aloft.    Such  a two-layer  (mixing
layer below and a decoupled "reservoir"  layer aloft) model was used by Husar
et al. (1978) to study the  sulfur budget of  a  power plant plume.  That model
did not  include temporally variable  plume  rise  or atmospheric  stability  in
the two layers.  We have refined that earlier model  to  include plume rise  and
spread more  explicitly  in  terms  of local meteorological  parameters.  (De-
tailed description  of  the  model  will  be included in another paper  now under
preparation by  Gillani.)  The  meteorological data used in the model calcula-
tions are  from  ground-level and upper-air measurements  made as part of  the
St. Louis Regional Air  Pollution Study  (RAPS).   All  plume calculations refer
to the case  of  emissions from the largest of the three stacks (height =  214
m) of the  Labadie  power plant near St.  Louis.  A steady   thermal  output from
this  stack  corresponding   to  electrical  power  generation  of  1000  MW   is
assumed.  (This assumption  is  quite realistic.)   In  the model, plume  rise is
calculated based  on the well-known Briggs empirical formulas (Briggs 1969).
The model  results  for  such  an emission are believed  to  be quite  represen-
tative also  for the average  current  tall-stack  emissions in the Ohio River
Valley source region.

The model  results  are presented  in  Figures 3-20 through 3-22.   The  upper
graphs of  Figure  3-20  show the diurnal   patterns  of monthly  median  values of
mixing-layer  height and effective stack  height  for January and July.   The
reader is reminded of  the substantial difference in daytime  mixing  heights in
summer and  winter—peak mixing heights  averaging about  1800 m  in July  and
only  about  700  m in January.   The greater  stability and  wind  speeds  typical
in January tend to  keep plume  rise lower,  but  the lower  ambient  temperatures
tend  to  offset this  tendency significantly.  The  result is that  the 24-hr
average values  of median plume  rise are  about 525 m in January  and  about  625
in  July, but  a somewhat  greater  day-to-day variability exists about this
average  in  July.  On  the   median  basis, the  July  plume generally  remains
confined within the mixing  layer for releases between 0900 and 1700  hr, while
the  January  plume release  even  during  midday has  nearly a 50-50  chance of
rising out of the mixing layer.

The lower graphs of Figure 3-20 show plots of the probability,  for two plume
releases at  12-hr  intervals in the diurnal  cycle, that the  plume will remain
decoupled  from the  ground during and  up to  24 hr of  transport.    The  two
releases  chosen for each  month represent  nearly the extreme conditions of


                                     3-54

-------
                  JANUARY
 E

O
z

o
O
CO
2000


1600


1200


 800


 400


   0
                                                        JULY
            I     i     i     i     I
        (MONTHLY MEDIAN HEIGHTS)
          EFFECTIVE
          STACK HEIGHT
                     MIXING\
                     HEIGHT
           04   08    12
                         16   20   24    0    04

                                  TIME OF DAY
                                                      08
                                          12
16   20    24
   1.0
                          \
                          \
         ioeoo
           PLUME
           RELEASE
   Figure 3-20.
                           \1800
                           V
                            \
                            \
                            \
                             V
                                                  10400
                                                   PLUME
                                                   RELEASE
                                                       i     I     I
                                                        1600
8    12   16    20    24   0
                                                      8
                                                         12   16
                                                                      20   24
                                PLUME TRANSPORT TIME
                            (Hours after Plume Release)
              A summary of the expected diurnal  and seasonal  variation
              of the interaction of the Labadie  power  plant plume  with
              the mixing layer.   The upper graphs show comparisons of
              the monthly-median diurnal  profiles of the  measured  mixing
              heights and calculated effective stack heights (based on
              Briggs formula  for plume  rise and  1976 upper air
              meteorological  data from  a site near the source).  The
              lower graphs show the distributions,  for two extreme plume
              release conditions, of the probability that the plume will
              remain aloft and decoupled from the ground  up to 24  hr
              after release.
                                     3-55

-------
plume rise.   The  probability distribution functions  for  all  other  releases
fall  more or  less  within these two  extremes.   The July  data  show  that  the
0400  hr  release  will always  start out decoupled  but that within  12 hr  of
transport it  will  almost certainly  experience  entrainment into  the mixing
layer.    The  late  afternoon  release (1600  hr)  has  a low  probability  (12
percent)  of penetrating  out of the mixing layer and,  except for some outlier
cases,  this release is also  almost certain to have  experienced ground contact
within  24  hr  of transport.   Thus,  the  probability is almost  zero  for  any
release  from  such  a  large  emission at  about 200  m  to  remain continuously
decoupled from the ground for a full 24  hr  during  summer.  The situation  is
significantly different  in January.  For  almost all  January releases, a 20  to
30 percent chance exists that, even after 24  hr of  plume transport,  the plume
is likely  not to  have experienced  any  interaction with  the  mixing  layer  or
the ground.  Plume  measurements in summer are plentiful and fully  support  the
above  summer  results.    Winter  plume measurements are  indeed  rare.    The
limited observations of  the  recent Cold  Weather Plume  Study jointly  conducted
by the  U.S.  Environmental  Protection  Agency (EPA) and  the  Electric Power
Research Institute  (EPRI) in February 1981 at the  Kincaid power plant (183 m
high stacks)  near  Springfield,   IL, do  indeed  bear  out the  above  winter
results.    In  that  field study, measurements were  made on 5  different days.
Of these  5  days,  3 were  typified  by very cold  winter conditions (Tmax <  -5
C), while the other  2 days  were  not typical  of winter (Tmax  >  15 C).  On 2
out of the 3  cold  days,  the  plume  releases,  even  those at midday, rose above
the mixing layer and remained decoupled from the ground.   In winter, then, a
significant fraction of  the  plume  releases  may   remain  decoupled   from  the
ground for  well  over 24 hr, and  even  over 36  hr.   In  the  meantime, this
fraction may  be transported to well  beyond 500 km  without any ground removal
at all.

To investigate  the  implications  of this important  seasonal  difference  in
plume-mixing-layer interaction on  seasonal plume  sulfur budgets,  transforma-
tion and ground removal   modules are  added to the  above plume model.   Trans-
formations of S02  to sulfates by  the  gas-phase  and  liquid-phase mechanisms
are included  in accordance with their empirical parameter!zations by Gillani
et al. (1981, 1983).   All transformation and removal rates are based on  St.
Louis,  MO,  data  for 1976,  are  assumed to  be  pseudo-first-order rates,  and
include  diurnal  and  seasonal  variabilities.  The transformation  rates  are
assumed to have seasonal  and diurnal variations  such that the  24-hr average
rates  are  about 1.3  percent hr-1  in  July  (about  0.8  percent hr"1  average
by gas-phase  mechanism  and  about  0.5  percent hr"1  by  liquid-phase mecha-
nism)  and  about  0.4  percent  hr~l  in  January  (mostly  by   liquid-phase
mechanism).   Ground  removal  of S0£ by  dry deposition is based  on a diurnal -
ly varying deposition  velocity,  being  0.3  cm  s"1 at night  and peaking  at
1.9  cm  s~^ at noon  in  July, with  corresponding  values  of 0.15  cm  s"1  and
0.95 cm  s"1  in January.  Deposition  velocity  of  sul fate  is assumed to  be
constant (0.1 cm s"1) at all  times.  These values  are consistent with those
most commonly used in current regional  models.  The model  calculations assume
that  no  precipitation   scavenging  occurs  during  the simulated  48 hr  of
transport.

The results of the model  calculations are shown in Figures 3-21  (January)  and
3-22  (July).   The  figures  illustrate plume dynamics (top)  and the sulfur


                                     3-56

-------
   2000
o
Q£
CD
O
CO
CD
   1000
           PLUME DYNAMICS
        '(Power  Plant Plume)


              JANUARY
      0
       00
          PLUME RELEASE TIME
              1200
                                                            '    '
                                             ' •'>.-.;'.. MIXING-
                                                  HEIGHT.
                                                            .' . • -v • • .- • '•'. •

                                                            •  •''.. t •'••••
06      12       18       24      06

                    TIME OF DAY
                 12
                 18
                24
    100
 UJ
 oo
 u.
 o
 C_5
 ce.
     50
           PLUME  SULFUR  BUDGET
           (Power Plant  Plume)
 JANUARY
                                   DAY 1
                                                         % AEROSOL
                                                     ISULFUR FORMED!
    PLUME
   RELEASE
     TIME
                                      % GASEOUS SULFUR
                                     REMAINING AIRBORNFJ
                                                         % SULFUR
                                                      DRY DEPOSITED
                                                                 j_
                                                                     DAY  2
                          0

                          10

                          20

                          30

                          40

                          50

                          40

                          30

                          20

                          10
                        12
                 18
24
30
36
42
48
                             HOURS AFTER PLUME RELEASE
 Figure 3-21.
(TOP)  Calculated Labadie plume  dynamics,  on  a
monthly-average basis, for plume releases  at  000, 0600,
1200, and 1800 hr in January 1976.
(BOTTOM)  Calculated monthly-average  sulfur budget  of the
Labadie plume in January during  48 hr of transport,  in the
absence of wet deposition.  Results are shown for the 0600
1800 hr plume releases.
                                      3-57

-------
   2000
o
o;
LU
o
CO
1000
       PLUME DYNAMICS     	
     (Power Plant Plume)^^
                                    1000
              JULY
           0040
                                      1600
                                     PLUME
                                    RELEASE
                                     TIME
                                     2ZDQ—
0       06       12       18      24      06
                           TIME OF DAY
                                                       12
                                                             18
                                                                24
    100
GO
GO
to
u_
o
o
  50
. PLUME SULFUR BUDGET
        JULY
          PLUME  RELEASE  TIME
                   04
 Figure 3-22.
                  	r
                         -16-
                               DAY 1
                                                    %  AEROSOL
                                                  SULFUR  FORMED
                                               % GASEOUS SULFUR
                                              REMAINING AIRBORNE
                                                        % SULFUR
                                                     DRY DEPOSITED
                       12
                            18
                                                    _L
                                                           DAY 2
                  0
                  10
                  20
                  30
                  40
                  50
                  40
                  30
                  20
                  10
                               24
                                        30
36
42
48
                         HOURS  AFTER  PLUME  RELEASE
            (TOP)   Calculated Labadie plume dynamics,  on  a
            monthly-average basis,  for plume releases  at  2200, 0400,
            1000,  and 1600  hr in July 1976.
            (BOTTOM)   Calculated monthly-average  sulfur budget of the
            Labadie plume in July during 48 hr of transport,  in  the
            absence of wet  deposition.   Results are  shown for the 0400
            and 1600 hr plume releases.
                                3-58

-------
budget (bottom) for different plume release times during 48 hr of transport.
The median  plume-rise  (at  the  time of  release)  and mixing-height  (diurnal
profile)  values  are used  in these model  calculations.   Ground removal  is
about 16 percent on each day  in  January.   In  July,  the ground loss  is  about
25 to 30 percent  on the  first day and an additional 10 to 12 percent on  the
second day.  In the absence of wet  deposition, the  1/e  atmospheric  residence
time of  S02 in such  a plume is  about 30 hours  in  summer and about double
that in winter.  With wet deposition,  this time will be shorter.  Of greater
importance, however, is the  residence  time of total sulfur.  In July,  about
40 percent  of  the  sulfur  emission is  dry  deposited  in  48  hours.  While  the
wet deposition is highly  variable and  discrete in  nature, it is reasonable to
assume that, on the average,  another  20 to 40 percent  of  the sulfur may  be
wet deposited during this period.   It  would appear  reasonable then  to assume
that about two-thirds of the sulfur emission from  a  typical  tall  stack in the
Midwest may be deposited  (wet and dry) within two  days during summer,  i.e.,
the 1/e residence time of total  sulfur emission from tall  stacks is  probably
about 2  days  during summer in the  Midwest.   During this  time, the  plume  is
likely to  have been transported about 1000  km   along  the particle trajec-
tories, and probably half  that distance  along the  straight line joining  the
source and  the plume  center of  mass,  on the average.  After two days,  the
plume is likely to  be  so spread out that  it  is probably not  even meaningful
to speculate about  the transport of the plume center of mass.  Parts of  the
plume may  even be  moving closer to the source as  other  parts move farther
away.  In any case, it would appear that perhaps more than  half of the sulfur
released in St. Louis from a 200 m stack  may become  deposited  within  a 500 km
radius of St. Louis.   In the Ohio River Valley, with less frequent and weaker
nocturnal jets and  generally somewhat lighter winds  than  in St. Louis,  the
effective  transport range of  the  emissions  is likely  to  be  shorter.    The
presence of the mountainous terrain of the Appalachian, and vertical motions
due  to other  mesoscale   influences,  may  further slow down  net horizontal
transport  and  reduce  the  sphere of  influence  of  the  source region.  Cloud
venting of  pollutants, however,  could increase  the atmospheric  residence  of
pollutants considerably.   Emissions from shorter  stacks (less  than 215 m)  may
be expected to have shorter  atmospheric  residence, while  those from super-
stacks may remain  airborne  for longer  periods.    Emissions  in  the coastal
areas of the  northeast,  may experience significant local   shoreline  recircu-
lations, thereby reducing their impact range over  the land  mass.

In winter,  the atmospheric  residence  of  sulfur  is expected  to be  signifi-
cantly  longer, and the  potential  for  long-range  transport  significantly
greater.    Cloud  venting  is  expected to  be of  less  significance  than  in
summer.  The tall-stack effect,  that  is  a  significant  increase in long-range
transport as a direct result of increasing the average  stack  height  from less
than 100 m  in 1950  to more than 200 m  by 1975, for example, is also  likely to
be much more important in winter than  in  summer.

The sulfur  budgets  described above depend on the particular  choices of con-
version and removal parameters.  While the reliability  of  the  absolute values
of the results may be questioned,  important  and  consistent information lies
in the  relative  values corresponding  to different release  times.   In both
seasons, ground loss is highest  for the  early morning  releases (0400 or 0600
hr)  because plume  rise  is  lowest  at  these  times  due  to  maximum  stability


                                     3-59

-------
and wind speeds.   Consequently,  these  releases  are entrained early in the day
and fumigated  to  ground at relatively  high  concentrations,  leading to sub-
stantial ground removal  within  the first 12 hr.  The higher ground loss of
S02 from these early morning  releases leads  to lower net  sulfate formation.
At the other extreme, ground loss  is  minimum for the  late  afternoon releases
(1600 or  1800  hr), which  have  the highest plume rise  and,  consequently, a
late  entrainment  the next day.   In  the  case of  the  1800 hr  releases in
January, a significant  portion  do  not get  at all  entrained into the average
peak mixing layer  and are  transported over long distances  without any  deple-
tion.   In winter,  the plume spends more time decoupled from the ground than
it does  in  summer, mainly because of the  substantially lower  daytime  mixing
height.   When  the winter  plume is entrained,  however, ground-level concen-
trations  will  be  higher  for  the  same reason.    In  terms  of ground  removal,
these two effects have partially offsetting results.

3.4.2  Broad Area! Emissions Near Ground (Urban Plumes)

Urban plumes result from urban emissions from low sources  such  as automobiles
and  short stacks.   Emissions  from  such  multiple  point  sources  in  urban-
industrial  complexes  are  generally treated  as broad  areal  emissions.   The
effective plume release height  of such an urban plume  is  typically  close  to
the ground.

From  the point of view of secondary  product  formation and deposition,  two
principal  differences   exist  between   the  power plant  plume  and  the  urban
plume.   The  first difference is  in plume  release   height  (elevated  vs low);
the  second  is in  the   chemical  composition  of emissions  from  precursors  of
acidic  products.    Compared to  urban emissions, power plant  emissions  are
relatively  richer  in  SOX  than   they  are  in  NOX.    Urban  emissions  are
substantially  richer in reactive  hydrocarbon species, which play an important
role  in  the  chemistry  not only  of  urban  plumes  but also  of power plant
plumes.   The role of transport and  turbulent  mixing in the  physical  inter-
action  of  power   plant plumes  with   polluted air  originating from  urban-
industrial  complexes is  thus  important in  determining the contribution  of
power  plant  emissions to  secondary  product  formation  during  long-range
transport.

The  difference in the  characteristic  release  heights of  the  two plume types
 is important  only during mesoscale  transport.   Once  the  two plumes  become
vertically  well  mixed  throughout the  mixing  layer, they  are  physically
 indistinguishable.  The  principal difference  during mesoscale transport is
 that elevated  releases spend their early transport period decoupled from the
 ground and  in a  relatively  stable  environment, while near-ground releases
 continuously experience  ground removal,  and  at least in the  daytime,  are
 subjected immediately  to rapid  dilution.

The  principal  difference  between elevated  and  low-level  plume  transport
 concerns nocturnal transport.   While  an elevated nocturnal  plume release is
 decoupled from the ground, a plume released near  the ground will be  trapped
 within the ground-based  shallow,  stable,  mechanical  mixing layer  unless it
 has  sufficient  buoyancy  to  escape  this  mixing layer.    If  trapped, plume
 concentrations of the  primary  emissions  in  contact  with  the  ground will be


                                      3-60

-------
high, and,  accordingly,  even  with  the reduced  nocturnal  ground absorption
capacity, substantial  ground losses  can occur.   Husar  et  al.  (1978)  presented
convincing evidence (Figure 3-23)  that the  central-city plume of  St.  Louis  is
at least partially trapped  in  the  nocturnal mixing  layer  in  summer.   The
figure  shows  Sg (gaseous  sulfur)  and  NOX concentration  data  averaged for
five ground-level monitoring stations of the St.  Louis Regional  Air  Monitor-
ing  network  for the  month  of  July 1976.   The  Sg  data  are segregated  by
sectors  pointing to three  major local  sources:   tne  central-city area; the
Alton-Wood River  petroleum refinery complex, which  includes a  power  plant;
and the  tall-stack Portage des  Sioux  Power Plant.  The  diurnal  patterns for
the Sg  data  show that while the Alton-Wood River and Sioux  contributions  to
ground-level  sulfur concentration peak in the  daytime  (when their  elevated
source plumes are entrained into the mixing layer and  brought to  ground), the
central-city concentration peaks at night  (presumably due  to trapping  in the
shallow  nocturnal  mixing  layer)  and  is  minimal  during the  day,  when the
emissions are effectively  diluted in  the  deeper, daytime mixing layer.  The
drop in  contribution  of  the elevated source plumes at night indicates  their
nocturnal decoupling from the ground.

The NOX  data shown are averaged not only  for all five stations  but  also for
all  sectors.   The  sector-segregated  NOX  data  (not  sh9wn here) support the
conclusions  drawn  below.    The diurnal NOX  pattern  is  indicative  of the
predominance  of local,  low-level   sources  of  NOX,  particularly  automobile
emissions.   During the  day, NOx is  dilute, both at  gound-level  and  aloft
(except  in  a fresh plume).   During  the  evening traffic rush hour,  ground-
level  NOX increases  sharply and remains  high  throughout  the  night,  indi-
cating  that  it  is  trapped in the shallow mixing layer.   This observation  is
consistent  with the  fact that  automobile  exhaust is  rich  in  NOX but not
SOX.

The  diurnal  and  seasonal  variations  of  urban  plume  dynamics  in  the  time-
height  plane and of plume  sulfur budget (not including  precipitation scaven-
ging)  based  on model  calculations  using  St.  Louis  meteorological   data  for
1976 are shown  in  Figures  3-24  and  3-25 (January and  July, respectively).  In
the  urban plume model,  the gas-phase  oxidation rate of  S02 is assumed  to
depend  only  on  sunlight  (linearly),  such  that  its  peak daytime values  are
typically  5.5  percent  hr-1  in July and  3.5 percent  hr-1  in  January.
Li quid-phase oxidation  of SOg  is  calculated in the  same way  as  it  is  for
power  plant plumes.   The  resulting  estimates  of  sulfate  formation  in  the
urban plume may be considered as reasonable but  unsubstantiated (particularly
for  winter).   However,  sulfate formation  only  weakly influences  the  sulfur
ground-loss  estimates.   The model  calculations  of  the  ground losses  may  be
considered valid  at least for  comparing diurnal  and  seasonal  variations  for
the  urban plume and differences between urban  and  power plant  plumes.   For
the  daytime  urban  releases (for example,  the 1200 hr releases in January and
the  1000 hr releases  in July)  during  both seasons,  the  plume is  brought  to
ground  close to the  source area at high concentration  and is  subsequently
rapidly diluted  throughout the  mixing layer.  Consequently,  ground removal  is
more  rapid initially  and  much slower  as  the  plume  dilutes  and the ground-
level  concentration of  the pollutants diminishes.  As a result  of  the rapid
daytime plume-spread  throughout the  mixing  layer, the  transport  range over
which   source   characteristics  are   still   physically   distinguishable   is


                                      3-61

-------
   50
   40
.n
 Q.
 Q.
  X
o
CO
 I
   30
 CD
 cn
I 20
o
o
CJ>
oo
CD
   10-
                          ALTON and WOOD RIVER
                                                                       \
                                                                         \
                NOX
            ALL SECTORS
                                DES SOUIX

       CENTRAL CITY
       /\
                             I
                 1
                        I
                        I
                            8
                10
12
14
16    18
20
22
24
    Figure 3-23.
                TIME OF DAY
The diurnal behavior of sulfur and nitrogen concentrations
in St. Louis, MO, based on monthly average data of the
RAPS ground network for July 1975.  The data are averaged
for five stations.  For gaseous sulfur, Sg, they are
segregated by wind-direction sectors which pointed to
three major sources:  the central  city area; the Wood
River refinery complex (including  a 650 MW power plant);
and the tall-stack Portage des Sioux power plant (1000 MW)
(Husar et al. 1978).
                                      3-62

-------
  2000
         PLUME DYNAMICS
       •  (Urban  Plume)


             JANUARY
§
O
CD
   1000
              0600

                              1200
                         &£
                •>;•   PLUME RELEASE
                .-.V       TIME

                ;,.-;.'•. \ isocL      OOOQ,
                      ft>T
                      zk
               06      12       18      24       06

                                   TIME OF DAY
                                                             ''\V MIXING.
                                                             /Vr HEIGHT
                                                              .•' A;- ' •  '•••"
                                                             ' . '  \ -• .' ••.•••
                                                                '
                                                                 •• ' '
                                              12
 18
24
    100
 GO
 GO
 UJ

 oc
 oo
 u_
 o
     50
PLUME SULFUR BUDGET
  (Urban Plume)
                JANUARY
                                                                 AEROSOL  .
                                                       -1800
                                                                0

                                                                10
                                  PLUME RELEASE
                                         TIME

                                 	1800'
.GROUND
 LOSS
                                                                           30
         H40   m

           50 _
                                                                              330
                                                                            30
                                                                              •-•o
                                                                              ^m
                                                                              00
    Figure 3-24.
                 HOURS AFTER PLUME RELEASE
       (TOP)   Calculated dynamics of the St. Louis  plume  (low-
       level  emissions only), on a monthly-average  basis,  for
       plume  releases at 000, 0600, 1200, and  1800  hr  in  January
       1976.
       (BOTTOM)  Calculated monthly-average sulfur  budget of the
       St.  Louis plume in January during 48 hr of transport,  in
       the  absence of wet deposition.  Results are  shown  for the
       1200 and 1800 hr.
                            3-63

-------
  2000
        PLUME  DYNAMICS
      - (Urban Plume)


            JULY
  1000
o
CO
«c
       - 0400
                       1000

                       PLUME  RELEASE TIME
                       '  1600
                             2200
              06       12       18       24

                                   TIME OF  DAY
                                   06
12
                                                                I.
                                                                I
                                                                I
                                                       MIXING
                                                       'HEIGHT
                                                                 I
                                                                 I
                                                                  I
                                                                  \
                                                                  x
18
24
   lOOi -^^—
(St
ts>
       • PLUME SULFUR BUDGET
           (Urban Plume)  ___
oo
LU
CJ
ce.
     50
JULY  /
     0
     Figure 3-25.
                                                ~ -  2200

                                             	2200
                                             AEROSOL
                                             FORMATION
                                                      1000
                  0

                  10



                  30

                  40

                  50
                 HOURS  AFTER  PLUME  RELEASE
        (TOP)   Calculated  dynamics  of the St.  Louis  city plume
        (low-level  emissions  only),  on a monthly-average basis,
        for plume releases at 0400,  1000, 1600,  and  2200 hr
        January to July 1976.
        (BOTTOM)   Calculated  monthly-average sulfur  budget  of  the
        St.  Louis city  plume  in January to July  during  48 hr of
        transport,  in the  absence of wet deposition.
        Results are shown  for the 1200 and 2200  hr plume releases.
              70

              60 oo

              50
                                                                   jino-H
                                                                   tum
                                                                     •}o;jo
                                                                    „ om

                                                                     SI
                                                                   20 |g

                                                                   10

                                                                   0
                                        3-64

-------
short.   Hence, the  difference between  ground  losses from  urban  and power
plant plumes is smallest for the daytime releases.  An exception is  apparent
in the  daytime power plant  releases  in winter,  which  penetrate  out of  the
mixing layer and remain  detached from  the ground for long  distances.

In stark contrast to the daytime urban plume  releases,  the nocturnal  releases
(1800  hr  in  January and  2200 hr  in July)   remain trapped  in  the  shallow
mechanical  mixing  layer throughout  the night.   Being concentrated and  in
continuous ground contact,  nocturnal  releases experience heavy ground losses.
After 12 hr of such  nighttime  transport, the  urban  plume  ground losses range
between about 40 and 60 percent  of  the  emissions,  compared to  almost  no
ground loss in 12 hr for the elevated nocturnal releases  from power plants.
Thus,  for  the nocturnal releases,  the  effect  of source  height difference,
though short-lived  in terms  of multiday, long-range transport, can  be quite
substantial.   The  loss  of about  half of the precursor emissions during  the
nighttime transport  of  the  urban  plume in  July before  the chemistry  even
begins  (assuming  the absence  of the  liquid phase  at night) substantially
limits the  amount  of secondary  formation  during further  transport.  Actual
nighttime measurements  of  ground loss  from trapped  urban  plumes  are  not
available in  the published literature.   Nor  does any  documentation  exist for
the fraction  of all  urban releases  (from either low or intermediate and  tall
stacks)  that  remains  trapped  within  the  shallow  nocturnal  mixing layer.
Analyses of  field  data  of  pollutant transformation and removal during urban
plume transport have lagged behind such analyses for power plant plumes.

In summary,  dry  deposition during  the  first 12 hr  of transport  appears  to
play a dominant role in  urban plume sulfur budget.  This is particularly  true
for nocturnal  releases.  After the  first 12  hr, most further  loss  of sulfur
and  nitrogen  compounds  may  be significant only  for daytime  releases under
convective  conditions.   While long-range  transport  of urban plumes  is  more
likely in winter, seasonal  differences in sulfur budget are not as  pronounced
as they are  in the  case of  power plant  plumes.   The bulk  of  the urban emis-
sions  of acid precursors,  particularly  NOX,  are  likely  to  be   deposited
within 500 km  of the source.

3.5  CONTINENTAL AND HEMISPHERIC TRANSPORT (J. D. Shannon  and D. E.
     Patterson)

Pollutants  transported  over  continental  and  larger scales may be  subject to
repeated  "breathing"  of the  planetary boundary  layer  (PBL)  over land,  i.e.,
the diurnal  cycle of daytime growth of the mixing layer and vertical  coupling
between upper  layers and the surface,  followed by the nocturnal decoupling of
flow  and pollutants aloft  from surface  removal  processes (Sisterson  and
Frenzen 1978).  In addition,  transport  over  long ranges may  be sufficient in
duration  that vertical  motions associated with  large-scale  weather systems,
such  as  subsidence   in  a region of high pressure  or  ascent over  a frontal
surface (Davis and Wendell 1976), become  significant and  result in  a greater
depth  of  the  troposphere affecting long-range  transport  than  is typical  for
mesoscale  transport.    This  leads  to  more   uncertainty  in  defining   the
transport  layer,  particularly  in  simulation  models  that  use   a  single
horizontal  transport layer.   Decoupled layers of haze  and sulfate on  the
regional  scale above the  mixing  layer have  been  noted  in  the  literature


                                     3-65

-------
(Slsterson et al.  1979,  McNaughton  and  Orgill  1980)  and  during  the  recent  EPA
Project PEPE/NEROS.

In addition,  transport  over continental  and  larger  scales may involve flow
over oceanic  areas, such  as anticyclonic  flow from the Midwest or Northeast
around an offshore high pressure center into  the  South (Lyons et  al.  1978).
The structure  and dynamics of  the PBL over  water differ considerably from
that over  land.   Oceanic  (or  Great Lake)  surface  temperatures show  little
diurnal variation because  of mixing processes.   As a result,  the  marine  PBL
is relatively constant.   In addition, the  ocean  is  a homogeneous surface over
large  areas,  while  the  continent varies  from forest  to field to  city, etc.
Broad  stretches of  strong atmospheric inversions  overlie  cold water, while
well-mixed regions  overlie relatively warm water.   While pollutants  within
the PBL  are  subject  to  dry deposition  processes  and will  eventually   be
removed, pollutants above the PBL,  perhaps  transported there by  convective
processes  over  land,  will  remain  above  the  PBL  until  transported down  by
precipitation processes  or by large-scale  subsidence.

Any single  trajectory  is a stochastic process  from an ensemble of possible
trajectories  for  a  given  set of meteorological conditions.   There  are some
occasions, such as a stationary pattern of  well-defined flow,  in which there
is considerable  accuracy  (i.e., little  ensemble   spread)  for an  individual
trajectory calculated for  daytime well-mixed  flow.  However,  if the meteoro-
logical  systems  are  moving,  a small  initial error  produced  in  temporal
interpolation can lead  to  a large  eventual  error, and  if the flow is ill-
defined or rapidly  changing, a  small initial  error in  calculations can lead
to a large change in downstream  position.   Currently,  the  network  of routine
upper  air  wind measurements is  sparser than  the network for  measurements of
precipitation chemistry  over  eastern North America.  Considering  the  normal
12-hr  spacing of  the  upper air  measurements, it  is  optimistic  to  hope  for
knowledge  of  the  prevailing wind at an arbitrary  location in  space and  time
to better  than 5  degrees  about  the  "actual" advecting  wind;  this  alone leads
to  an  uncertainty  in  the  crosswind direction of  15  to  20  percent  of  the
trajectory length for every timestep in  the  simulation.   The  statistics of
multiple  trajectories   contain  much  less  uncertainty  than  individual  tra-
jectories,  because  the  sample  size is  much larger,  and can  be  extended
further  downstream  in   time.   In  addition,  the problem of estimating hori-
zontal  diffusion becomes  easier  because  over long-term regional  scales,
horizontal dispersion is  due  primarily to the  spread of  plume or  trajectory
center!ines,  rather than  to the spread about  some  individual  plume centerline
(Durst et  al. 1959, Sheih  1980).

Calculation  of  transport  distances  for pollutants  subject  to  chemical  trans-
formation  and deposition  requires  simulation  modeling  (as  is  done  earlier in
this chapter  when wet removal  processes are not considered),  but  the results
are a  function of the modeling  parameter!zations,  such  as  the dry deposition
velocities or the transport layer  height, and  the  source  location and mete-
orological  conditions.    Therefore, the  transport distance  associated  with
sulfur oxides will  differ  from  the  corresponding scale of influence  for
nitrogen oxides,  even when  both are  emitted in one plume.  The regional-scale
transport  field  experiments currently  planned,  such as the Cross-Appalachian
Transport  Experiment  (CAPTEX)  sponsored  by  the   Department  of Energy,  use


                                     3-66

-------
inert, non-depositing  tracers.   The  CAPTEX  experiment is  intended  to be a
diagnostic study  of the transport  and diffusion  processes associated with
flow  over  large-scale mountainous  terrain  and,  as  such,  could  be  said  to
examine,  for  the  situations studied,  the upper  limit of transport  distance
scales associated  with depositing pollutants.   More definitive  experiments
must await development of suitable reactive and depositing tracers.

Another transport issue requiring simulation  models is the importance of tall
stacks.  Qualitatively,  use  of  tall  stacks  must  increase transport  distance
scales because upper-level  emissions are often decoupled  from  surface removal
processes,  thus  decreasing  near-source  dry  deposition,   and  because wind
speeds generally  increase  with  height.   A model  comparison of  hypothetical
surface-layer and  upper-level  emissions  from  a  source in  southern  Ohio  by
Shannon (1981) indicates that net transport past  the  Atlantic coast  could  be
one third higher  for  the elevated source.  The difference  between mid-level
and upper-level  sources,  somewhat  more  realistic  for  examination  of the
effect of the introduction of tall  stacks, would  be less.

It may prove instructive to examine  a few examples of key  "forcing functions"
which  determine  the  transmission  of  pollutant   emissions over  the   North
American continent.   For elucidation  of  the  meteorological nature  of  long-
range transport, two excellent reviews are those  of Munn  and Bolin (1971) and
Pack et al. (1978).  For a more  thorough exposition of climatological factors
influencing  long-range  deposition,   the  reader is  referred to  a series  of
studies by Niemann et al. (e.g., Niemann 1982).

That  long-range transport  of  acidifying pollutants  actually  occurs can  be
inferred or modeled in  a number of  ways.   The simplest demonstration  may  be
seen  in  observations  of the motion  of  polluted  air masses from satellite
images or from surface reports of aerosol  sulfate or reduced visibility (Tong
et al. 1976, Chung 1978, Wolff et al. 1981).   The  episode  during  23 June to 7
July 1975 shown in Figure 3-26  indicates  the  apparent motion of  a large hazy
air mass over  a  two-week  period;  this particular episode  of long-range
transport in  a  stagnating anticyclonic system was documented through  visi-
bility,  sulfate,  and  ozone measurements  (Husar et al.  1976), as well  as  by
satellite imagery (Lyons and Husar 1976).

It is evident that the day-to-day transport  of air pollutants on  the  regional
scale  is  controlled  by  the  synoptic  passages   of  fronts,  cyclonic, and
anticyclonic systems.   Smith  and Hunt (1978)   have pointed  out that  receptor
regions  remote  from  major  sources  may  receive  a disproportionately  large
fraction of  deposition  during  a few  events,  and  thus the  average transport
conditions may be  irrelevant because  the  episodes have their  own distinctive
meteorology.  In  particular,  precipitation  along  a frontal  zone  on  the edge
of  an anticyclone can  contribute  a  large deposition of acidifying species
which are built up over the prolonged continental  residence.  Vukovich  et  al.
(1977) illustrated that  the  air with  the  longest  residence  time  (and highest
mass  loading  of pollutants) within  an anticyclonic  system is  found  on  the
periphery, where frontal activity is most likely.

On the regional scale,  the  spreading  of emissions is dominated  by the  action
of vertical wind  shear  and  wind direction changes acting  in combination with


                                     3-67

-------
          JUNE  25,  1975
                                 JUNE  27,  1975
          JUNE 29, 1975
                                  JULY 1,  1975
           JULY 3, 1975
                              1000 km
                                 JULY 5, 1975
Figure 3-26.
Sequential  contour maps of noon  visibility for June
25-July 5,  1975 illustrate the evolution and transport of
a large-scale hazy air mass.   Contours  correspond  to
visual range 6.5-10 km (light shade), 5-65 km (medium
shade) and  <5 km (black).   (Husar et al.  1976).
                                  3-68

-------
the diurnal cycle of daytime mixing and nighttime layering of the atmosphere
(e.g., Draxler and Taylor 1982).  A graphical  example  of the dispersal of a
puff released in St.  Louis,  during four  days  of transport, by interactions of
vertical  wind  shear  and synoptic motion  is  given  in Figure 3-27.   Here an
ensemble of  100 trajectories begun at  midday  are  represented  by the lines
shown; the  mean trajectory  is   indicated  by  the  heavier line  with  dotted
nodes, and ellipses  at  12-hr intervals  indicate the spread of end points of
the ensemble relative to the mean position.   During daylight hours, lateral
puff  spread  is minimal  due to  lack  of wind  shear.   By early evening, as
mixing greatly  diminishes,  vertical  layers  (here  simulated by  four 300-m
layers) begin to diverge, and continue independent paths until midmorning of
the next day.   At  that time, the clusters in  each layer act as  a new  puff
beginning  a  well-mixed  day until  the  next  evening,  when  each  puff again
divides into  layers, and so on.   Within  one  day  of  such dispersion, shear
spreads the puff out over a  scale of  the  width of  Michigan.  After four  days
(trajectory  endpoints), the  puff  is  smeared  across  all   of  the  eastern
Canadian border.    Edinger  and  Press (1982)  expressed  the effect  of  such
spreading and mixing in terms of a regional dilution volume  over 1  to 3 days.
They show that episodes of haze  occur  when the dilution volumes  from  sites in
the northeastern U.S. overlap;  the overlap produces sufficient homogeneity to
explain large regions of haze emanating from just  four representative  source
cities.  The mixing and spreading are due more  to  shear  in  the  vertical  than
to horizontal nonuniformity in  the flows field.

Rodhe  (1974)  illustrated that  the  assumptions made about  the  intensity of
turbulent mixing in  the vertical  may  dramatically  alter the output of model
transport  computations.   Other  vertical  motions are important  in  long-range
transport  in  the  troposphere,   although  difficult  to  simulate properly.
Transmission  of pollutants  across major  topographical  obstacles (e.g., the
Rocky  Mountains),  along  warm   and  cold  fronts,  and  near  convective cells
involves vertical  transport that is  problematic for  the modeler.   Unfortu-
nately, these are also  the  situations which  are crucial  in  simulating  events
of wet deposition.   The motion  of low pressure systems  and,  more importantly,
the significant accumulation of  pollutants during  the  passage of  slow-moving
anticyclonic  systems are also  major  factors  in determining  the extent and
severity of  source  impacts.   Korshover  (1967)  has  shown that  the Smoky
Mountain area  is particularly subject to  stagnating  anticyclones,  leading to
a lower overall ventilation of its emissions  on a regional  scale.

Although the shorter temporal and spatial  scales of transport are known to be
important,  the  characterization of episodes has been limited  for the  most
part either to case  studies or to simple term tabulations of occurrence.  The
understanding  of  such  events in the detail   required  for policy  decisions,
including the development of models, is incomplete  at present (see Bass 1979,
for  review).    The  estimation  of long-term  transmission coefficients  from
sources  to receptors is  inextricably tied  to transformation  chemistry and
deposition mechanisms,  and  is beyond  the  scope of  this section  (see  Chapters
A-4 and A-7).   Similarly, consideration of "pure transport"  without  kinetics
involves model simulations which are not described  here.   It may be mentioned
that  very  recent  computations  at  Washington   University indicate  that the
seasonal and  annual  mean trajectories within  eastern North  America give  mean
displacement  rate  on the order of 3 m s"1   over  the first  few days,  with


                                     3-69

-------
Figure 3-27,
Dispersion of a plume emitted at St.  Louis,  on August 26,
1977, assuming 1-layer daytime transport and 4-layer
nighttime transport.   The  spread occurs as a result of the
interactions of vertical wind shear with synoptic wind
fields over a 4-day period.
                                  3-70

-------
root mean square deviation  from  the  mean path being large enough to include
the source.   Comparable computations by  several  models in  the  MOI studies
yielded roughly comparable  results.   It is perhaps more direct, however, to
examine  climatological   examples  of  key  meteorological   parameters:  wind
fields, mixing height, and precipitation.

The most obvious determinant of transport is,  of  course,  the  wind field.  For
the years  1975-77,  the  available rawinsonde  upper air data (Figure 3-28)
yield some clear patterns:  (1) the general  flow  is west to  east, with  also  a
significant  flow  upward  from  the Gulf  of Mexico  to  the  Great  Lakes;  (2)
winter  and   fall  exhibit the  highest  speeds;  (3)  the  southeastern  United
States  lies  within  a region  of low mean  velocity during  late  spring and
summer; (4)  the midwestern United States  exhibits very strong shear  during
summer  and spring, with  southerly  surface  flow and westerlies at the  top of
the PBL.  Mean winds include artifacts of averaging and should be interpreted
with caution; for example,  alternating NW and  SW  flows  will  produce  a  mean  W
flow.   It is also important to note that these are local  mean winds;  not only
are the existence  and interactions of synoptic-scale circulations  not  shown,
but  as mentioned  earlier,  the  flow associated  with  wet  deposition may be
quite  different from the mean.  Wendland and  Bryson (1981)  have used  clima-
tological  near-surface  wind fields to  identify  airstream source regions and
mean frontal  locations;  the Ohio Valley is  identified  as an airstream  source
region  during summer and fall.

An  important notion  in  both mesoscale and continental-scale transport  is  the
existence of a  top to the layer  in which pollutants  are found.  The height of
such a layer will  vary during the day as well as geographically and from day
to  day. There  is  also  an unknown but likely important loss of material  from
the mixed layer  to  upper  layers  by convective  motion (Ching et  al.  1983).
Well-mixed aged pollutants  in  nocturnal  stable layers aloft may sometimes not
be reentrained  into  the mixing layer the next morning.  As noted earlier the
maximum afternoon mixing depths  at  several  locations in  the  United  States
 have been  determined by Holzworth  (1972).   Similar  studies were conducted for
Canadian  sites  by Portelli  (1977).   Contours of  these  literature  values  of
 representative  mixed depths (Figure  3-29) provide  some insight into the gross
 interactions of advecting winds and  the  depth of the  mixing layer, although
 synoptic  temporal  and  spatial  scales of interaction may be  at  least  as im-
 portant as  the seasonal  averages  in determining the  net  transport of emis-
 sions.  It  is  seen  that the northern regions generally have lower inversion
 heights,  with  the  deepest  layers occurring  in the  desert regions  of  the
 United States.  Most important is the considerable  uniformity, separately, in
 the eastern  United States and in the western United States.  On the average,
 some of  the well-mixed,  aged  pollutants  will  ride over  the  daytime mixed
 layer   when  moving either from  south to  north or from west  to  east,  due  to
 decreasing  mixed  depths along  the  trajectory.    Thus,  an  appropriate para-
 meterization of the spatial-temporal variation of the mixing layer height  is
 required for simulation  of continental-scale  transport over  several days and
 thousands of kilometers.

 Another  "forcing   function,"   precipitation,  is  critical  in   long-range
 transport,  not only in determining the  local  impact of  wet deposition  of
 pollutants,  but  also  as  a  mechanism  for removal of pollutants  from the


                                      3-71

-------
Figure 3-28.
Averages for 1975-77 of winds  in  the  layers 0-500,
50-1000, 1000-2000,  and 2000-3000 m ag  1  for  the 0000 and
1200 GMT soundings.   Lower-level  winds  generally lie to
the left and are of  lower speed,  (a) January through
March; (b)  April through June;  (c) July through September;
and (d) October through December.
                                   3-72

-------
Figure 3-29.
Contour plots of maximum afternoon mixing depths by
season, indicating qualitative patterns  only.   Note
change of contour scales,   (a)  January  through  March;  (b)
April  through June;  (c)  July through  September;  and (d)
October through December.

                     3-73

-------
atmosphere, thus preventing  further  transport.  Prevalent trajectories from a
source to a receptor region will not  indicate  actual  impact if the air mass
is very likely  to  experience  precipitation  along the way.  The exact nature
of wet removal  is  still  a matter of  debate;  presumably  some combination of
the amount of precipitation, the type and intensity of precipitation events,
and the  frequency  of precipitation  may be  an appropriate  measure  of this
"forcing function"  on a  regional  scale.   As  illustrated in  Figure 3-30, these
three alternative  measures  can  lead  to very  different  conclusions.  A pol-
lutant emitted  in  northeast Canada is  more  likely,  less likely, or equally
likely than a pollutant in  the  southeastern  United States to be locally  wet
deposited,  depending on  whether frequency,  intensity,  or  total  amount of
rainfall  is the determining wet deposition  factor during  the summer months.

To examine the  average sulfur deposition pattern  produced  by a  single  source
as  a  function  of  time  after emission, the  ASTRAP model  (Shannon 1981)  has
been  exercised  with  summer meteorological   data  for a  single  hypothetical
elevated  source located  near  Kansas  City.   The  wet  and  dry deposition
patterns for  the first, second,  and  third days after  emission,  respectively,
are shown  (Figures 3-31 through 3-33).   Note that these  are season average
patterns, and not the patterns produced by emissions on  a particular  day;  the
latter patterns likely would be  much  more plume-shaped.   If flow during both
wet and  dry patterns were  random,  with no  prevailing  direction, the  depo-
sition patterns would be centered on the  source location.  Here, the  depo-
sition maxima progress  to  the  northeast with time,  but  since  flow is  not
always in the prevailing direction, some deposition occurs in all  quandrants,
particularly  during  the first 24 hr  of transport.  In the  Midwest,  a  region
where  rainfall  is  typically  75   to  100   cm yr-1,  with  frequent  summer
showers, wet  deposition dominates  dry deposition after the  first day.   This
is  because dry deposition  is  a  function of  the  steadily  decreasing  surface
concentration,  while wet removal occurs through the depth of the mixed layer.
The wet  deposition maxima can also be  seen  to progress  faster  with  time; in
the Midwest,  the Gulf of Mexico  is the  usual source of precipitation moisture
and  thus  the  flow  during  precipitation  has a  somewhat  higher degree of
prevalence  than during dry  periods.

A similar exercise  has been  carried out for ten hypothetical  sources   dis-
tributed  across the United States and  southern  Canada  (Figures 3-34 through
3-36).    Even  though the  sources  (indicated by  the   symbols)  are  widely
separated,  the maxima  become difficult  to associate with a  single  source
(other  than  the  western sources)  after the  first 24  hours.    The  greater
relative  importance  of  dry deposition  for  the southern  California  source is
due  both  to  lighter winds and  to less precipitation.    The  wet deposition
contours  over the  ocean have little  meaning because  no  precipitation obser-
vations  beyond coastal  regions  were  available for model  use;  thus,  the wet
deposition maxima  cannot progress beyond  the coast,  although there  is no
 significant bias caused  in  simulations  over  the  land.

An example of both current modeling capabilities  in long-range transport and
deposition and  current  source/receptor spatial  relationships (at  least as
treated  by a particular model)  is  given in the  series  Figures 3-37 through
3-40.  The  ASTRAP  model  (Shannon 1981) was exercised with a current sulfur
                                      3-74

-------
GO
I
en
                    QURRTER 3,1977
                                                                                  QURRTER 3,1977
                                FRRCTION

                                do.01-0.02

                                E3 0.02-0. Oil

                                go.Oil-O.OB

                                • >O.OB
   Fiqure 3-30.
Statistics  of  hourly precipitation  data during July-September of 1977.   (a) Fraction of
hours with  precipitation; (b) intensity of rate of  rainfall  during  precipitation events;
and (c) total  rainfall  during the quarter, which  is  the product of  .(a)  and (b).

-------
                 WET  DEPOSITION
     FIRST  DRY DEPOSITION
            DRY 'DEPOSITION
                                                                   '*•*/          1 °*»
FIRST DRY  DEPOSITION
Figure  3-31.  Cumulative wet  and dry total  sulfur deposition patterns  during the first day of transport,
             for a hypothetical source near Kansas City in summer.

-------
                WET DEPOSITION
    SECOND DRY DEPOSITION
            DRT'DEPOSITION
                                                                    x /
                                                                        "
SECOND DRY  DEPOSITION
Figure  3-32.  Cumulative wet and  dry total sulfur deposition patterns during the  second day of transport.

-------
—I

00
                     WET DEPOSITION
        THIRD  DRY DEPOSITION
             DRY 'DEPOSITION
                                                                            •—*/
                                                                                       •>• •:
                                                                                       •'• •_.£
                                                                                       "!•.«'"
THIRD DRY  DEPOSITION
    Figure 3-33.  Cumulative wet and dry total sulfur deposition patterns during the third day  of  transport.

-------
co
i
IO
                     WET DEPOSITION
         FIRST  DRY DEPOSITION
                                                      DRY- DEPOSITION
                                          FIRST DRY  DEPOSITION
    Figure 3-34.
Cumulative wet and dry  total sulfur deposition patterns during  the first day of transport,

for ten  hypothetical  sources.

-------
co
i
oo
o
                    WET  DEPOSITION
        SECOND  DRY DEPOSITION
                                                      DRY'DEPOSIT ION
                                          SECOND DRY  DEPOSITION
   Figure 3-35.
Cumulative wet and dry total sulfur deposition patterns during the second day of transport,
simulated for ten hypothetical sources.

-------
oo

00
                    WET DEPOSITION
        THIRD DRY  DEPOSITION
                                                      DRY .DEPOSITION
                                          THIRD  DRY DEPOSITION
    Figure 3-36.
Cumulative wet and dry total sulfur deposition  patterns during the  third day of transport,
simulated for ten hypothetical sources.

-------
CO
oo
ro
                      SULFUR DIOXIDE
           SUMMER RVDWGE
           TROM SOURCES HITHIN'500 KM
           pG/CUBIC METER
           MRX - 42.2
                                                             SULFUR DIOXIDE

                                                                           •          •--..   •  .
                                                  SUMMCR
                                                  TROM SOURCES BCTONO'500 KM
                                                  uG/cueic MCTCR
                                                  MflX - 3.36
    Figure 3-37.
Contribution  to average  summer S02  concentrations  resulting from U.S. and  Canadian
anthropogenic sulfur sources within 500 km and from sources beyond  500 km.

-------
CO
I
o-
CJ
                        SULPflTE
            SUMMER rlVCIWGE
            FROM SOURCES HITHIN'500 HI
            uG/OIBlC METER
            M«X -  10.1
                                                                  SULFflTE
                                                     SUMMER flVERflGC COMCENTRflTION '-'
                                                     FROM SOURCES BEYOND"500 KM
                                                     uG/CUeiC METER
                                                     MRX -  4.73
    Figure  3-38.
Contribution  to average sulfate  concentration resulting  from U'.S. and Canadian anthropogenic
sulfur sources within 500 km and  from sources  beyond 500  km.

-------
CO
I
00
                    DRY  DEPOSITION
                            ON)
      SUMMER flCCUMULflTlON
      FROM SOURCES WITHIN'500 KM
      KG SULPUR/HECTRRE
      MflX -  6.83
                                                                             DRY  DEPOSITION
                                                                  SUMMER HCCUMULBTIdN /"
                                                                  fROM SOURCES BETONO'SOO KH
                                                                  KG SULruR/HECTflRE
                                                                  MflX - 1.39
     Figure 3-39.   Contribution to cumulative dry  deposition of total sulfur resulting from U.S.  and
                    Canadian anthropogenic sources  within 500 km and from  sources beyond 500 km.

-------
CO
oo
en
                         WET DEPOSITION
                     •     /   c
            SWtCH flCCUMULHTIdN /
            TROn SOURCES MITHIN'500 W1
            KG SULrUR/HCCTflRE
            MRX - 1.97
                                                                WET DEPOSITION
                                                   SUMMER flCCUMULHTlON /
                                                   FROM SOURCES BErOND'SOO KM
                                                   KG SULrUR/HECTHRE
                                                   MBX -  2.25
     Fiqure  3-40.
Contribution  to cumulative wet deposition of total  sulfur  resulting from U.S.  and
Canadian anthropogenic  sulfur  sources  within 500 km and from sources beyond 500  km.

-------
oxide emission  inventory for  the  United  States  and Canada  and  with mete-
orological  data  for June-August  1980.    The  concentration  and  deposition
patterns were separately calculated for sources within 500 km of each of the
(51 x 37) points in a grid across  North  America,  and for sources beyond 500
km from  each point.   If the  two  source/receptor separation categories are
termed local  and long-range,  respectively,  it  can be seen  that average S02
concentrations from  sources  beyond 500  km are almost  nil,  while  the long-
range contribution  to sulfate is more than half of the  average  concentration
in New England and  much  of eastern  Canada.   The fraction of dry  deposition in
those regions  from long-range transport  is also  significant,  although the
total amounts  are  low.    Wet deposition of  sulfur  has  the most significant
long-range component of  the four fields examined for this single season.  If
one considers the low emission density of most  of  New England, upper New York
State, and the Maritimes, the relatively greater  influence of sources beyond
500 km is not surprising.  In  particular,  there are  few emissions within 200
km of most  of the area.   While other models  might  give somewhat different
results, there is general agreement that sulfate  and wet deposition of total
sulfur have  a larger long-range component than  do  sulfur  dioxide  and dry
deposition of total  sulfur.   Since the input  data have  a minimum resolution
of about 100 km, local  deposition maxima on smaller scales are not simulated.
It should be emphasized  that the  results shown are  from a particular model,
and  that no  model   of  long-range  transport  and  deposition is  as  yet fully
verifiable.

Seasonal ASTRAP  simulations  for all  anthropogenic sulfur emissions from the
United States and  Canada during  1980 gave a budget  of  28 to 32 percent dry
deposition on the  continent  and 13 to 31  percent wet,  with  annual totals of
29 percent dry  and 24  percent wet.   The budget  remainder,  47 percent for
annual  totals,  is  an  upper  estimate  of  coastal  net mass  sulfur  flux, as
deposition,  particularly dry deposition, is likely underestimated within 100
km of sources, the minimum resolution of the ASTRAP simulations. Wet  deposi-
tion  is  more variable  than dry deposition  because  of regional  droughts and
wet periods.  Rigorously determined confidence  limits cannot  be  placed on the
simulation results, because only wet deposition is monitored.

Hemispheric  transport of acidic deposition precursors from sources in  North
America  to   receptor  regions  in the  Northern  Hemisphere  has  been examined
primarily in regard to  two  particular  issues:   the contribution  of  North
American  sources  to acidic  deposition  in  Europe, particularly Scandinavia;
and  the  contribution  of North American sources to  Arctic haze.  The latter
issue  has  been  raised  more  in  reference  to  visibility or modification of
radiation balance.   For long  periods,  the Arctic is a polar  "desert"  with
essentially  no  wet deposition and very  little dry deposition  due to strong
low-level stability.

According  to  Rahn  (1981),  the  two  pathways to  the  Arctic  of greatest
significance are  northward   transport  from  Europe  via  Scandinavia  and  a
cyclonic  pathway from  Europe and the  central U.S.S.R.  into  the  Norwegian
Arctic.   These air masses may be transported over  the pole into the  North
American  Arctic.   The  cyclonic   track  is  less  effective  as  a  transport
mechanism  because  of much greater  wet removal.   North  American  pollutant
sources,  which  lie  mostly  in  the  eastern  or  downwind  portion   of  the


                                     3-86

-------
continent, occassionally contribute  haze  precursors to  the Canadian Arctic
islands via a  track  around  Greenland.   Concentrations  of pollutant aerosols
in the  Arctic  show  a  definite  winter peak when  the  removal  mechanisms are
almost inactive.  Rahn  and  McCaffrey (1980)  indicate winter residence times
of 2 to 3 weeks for  Arctic  aerosol  particles.

The contribution of  North  American  sources to  acidic  deposition in Europe,
particularly Scandinavia,  is not  firmly  established  but  is  thought  to be
relatively small.  Studies  of "clean" Atlantic aerosol  (i.e., not downwind of
European  sources)  indicate  concentrations  of  0.2  yg  m~3  of S02  and  0.8
yg  m"3 of  sulfate   (Prahm  et  al.  1976),  but  in  part  the  concentrations
result from production/destruction activities in the sea, greatly complicat-
ing the analysis of  box-budget  studies.   While the North American contribu-
tion  is  not   the  major  share   in   acidic  deposition   in   Scandinavia,  the
multiplicity of  sovereign  source regions  in Europe  and the resulting frag-
mentation of contributions  to the deposition burden make  quantification of
the North American input desirable.

An  issue receiving  increasing   attention  is  the occurrence  in presumably
pristine areas of precipitation  pH as low  as  4.3 (Miller and Yoshinaga  1981).
While  most  pristine  areas  receive precipitation hydrogen ion  concentrations
an order of magnitude less than  in industrialized regions,  the  pH of  elevated
sites, in particular, can be  considerably lower.  The  relative importance of
natural biogenic sources and  hemispheric transport of  manmade  pollutants has
yet to be determined.   Transport above the PBL over oceanic areas might not
encounter  either wet  or dry removal processes  for  great distances  until
mountainous  islands, which  can  extend above  the marine  PBL, are  reached.
Calculations of back trajectories  from  Hawaii  (Miller  1981)  show a  strong
east-west flow dichotomy.

There  are many uncertainties in diagnostic analysis  and modeling of  transport
of  acidic or acidifying pollutants.   These uncertainties involve both  under-
standing  and  quantifying individual  processes, and development of  tractable
parameter!'zations  for   use  In computer simulation  models  of   transport and
deposition.    An  illustrative,  although not  necessarily complete,  list in-
cludes the  following:

     1)   The transport  layer  or  layers must  be  defined.  Should calculations
          be for constant-level flow, or for isentropic  flow (common  above the
         mixed layer)?

     2)   Synoptic-scale  and mesoscale vertical  motions  redistribute  the  pol-
          lutants and thus complicate the definition  of the transport layer.

     3)  Transport   and  diffusion over  complex  terrain,   such as  mountain
          ranges  or  shorelines,  is more complicated and  less  understood  than
          over  homogeneous terrain.   Current experimental plans such as CAPTEX
          will  help here.

     4)   Three-dimensional  flows  through  precipitation  systems   over  all
          scales  are  not well  understood.
                                     3-87

-------
     5)   The effect of  wet  and dry removal  cannot  be separated from trans-
         port distance  calculations.   For continental  transport, the air mass
         must pass over surfaces  of very different roughness, vegetation, and
         stability characteristics.   Dry deposition  rates  are  still  conten-
         tious matters, and  the  "best estimate" can  vary  widely.  Wet depo-
         sition has been  investigated  in  detail mostly on  the local  scale,
         although the OSCAR experiment of  the  EPA/DOE MAP3S program  in 1981
         was aimed at the regional  scale (Easter 1981).  Wet removal  parame-
         terizations,  developed  for  the local  scale but  then modified for
         continental  scales,  have yet  to be  thoroughly verified.

     6)   Most atmospheric processes  have a  strong diurnal  variation,  such as
         the  pronounced  shear effects associated with  nocturnal  decoupling
         and  the  nocturnal  "jet."    While   in  simulation  modeling of long-
         range transport and deposition one may elect not to apply diurnally
         varying parameters explicitly, the diurnal  variations in  the real
         atmosphere  must  be  considered  in  the  choice  of  any  average
         parameterization values.

     7)   Evaluation of  recurvature of  trajectories back  to  the  North American
         land  mass  has  been  far  more    qualitative  than   quantitative.

3.6  CONCLUSIONS (N. V. Gillani,  J.  D.  Shannon,  and  D. E. Patterson).

The flow field  in  the  PEL,  which is responsible for  pollutant  transport be-
tween a source and the  receptor  sites,  is  characterized by a broad spectrum
of atmospheric  motions ranging  from microscale  turbulent  eddies  to global-
scale circulation.  As  a pollutant cloud is  transported and dispersed, it is
influenced  by a  progressively  larger  range of  atmospheric  motions.   The
horizontal   winds  are  primarily  responsible for pollutant advection, while
turbulent eddies, wind  shear,  and  direction changes with  height,  as well as
sudden wind shifts, cause vertical  and lateral  pollutant dispersion (Section
3.3).

There is no universal  agreement as  to  proper scale divisions in the transport
of acidic or acidifying pollution.   In general, the dominant time scales are
diurnal, synoptic  (2 to 5 days), and annual.   The  diurnal  scale is critical
because so  many  transport and removal  processes (including air mass  convec-
tion showers) are strongly affected  by the  solar heating cycle.  The synoptic
scale is significant both because flow patterns may  "box the compass" during
passage of  a circulation center and  because the precipitation frequency is
largely controlled on this scale.  The  annual  scale  is important because so
many  important atmospheric  variables show  a marked  seasonal   pattern  (e.g.,
synoptic  flow pattern,  PBL   height,  pollutant transformation  rates,  etc.)
(Section 3.2).

We  wish to highlight   the  following aspects  of transport processes  which
appear  to be  of particular  significance at  this stage  in  our  assessment of
acidic deposition.

    0   Mixing height  is an important transport parameter.   It governs not
        only  vertical   dilution  of  the   pollutant,   but  also   horizontal


                                     3-88

-------
dilution by wind shear  effects  in  the vertical  domain of  transport.
Mixing height has a very pronounced diurnal and seasonal variability
but is spatially relatively uniform in the eastern United  States.  It
peaks daily in the  afternoon  and  seasonally  in  summer.  In  particu-
lar, as a result of substantially  lower mixing heights  in winter  than
in summer, a sirMficant portion (perhaps greater  than  20 percent)  of
the elevated emissions  from tall  power plant stacks in northeastern
United States may  remain elevated and relatively  coherent for  more
than 24 hr and 500  km of transport (Section 3.3.1).

Atmospheric dispersive processes also  play critical roles in chemical
transformations of  emissions (by  facilitating  their  dilution  with
chemically different background air) and  in pollutant  removal  by dry
deposition (by governing  the vertical delivery to or  away from the
ground sink).   Elevated emissions  remain mostly  decoupled from the
ground at  night  and reach  it substantially  diluted  during the  day.
In contrast, ground-level  emissions  (for example, from automobiles)
may rerr-'n  trapped  within  a  shallow  mixing layer  at night,  exper-
iencing  substantial  dry  deposition  within  short-range   transport.
Tall-stack emissions of  sulfur  and  nitrogen  oxides thus have  longer
atmospheric residence  times than do  the general  urban emissions  of
these compounds (Sections 3.4 and 3.5).

The PBL  flow  field is  characterized  by  strong  diurnal and  seasonal
variations.   In  the dense  source  region in  the northeastern  United
States, prevailing  winds  are, on  the  average,  from  the  southwestern
quadrant  in  summer  and more westerly  in winter.   The vertical  pol-
lutant transport layer for long-range  transport  varies  typically  from
the ground up  to  1  or 2 km in  summer and about half that  in  winter.
Diurnal variability of  the flow field is  particularly pronounced  in
summer, especially  in the  midwestern  states, where a "nocturnal  jet"
with strong associated wind shear is a frequent  occurrence, following
relatively  slower  and  vertically  more  homogeneous  wind   during the
daytime.  The pollutant  plumes  undergo a sequence  of sheared  strati-
fication  and distortion  during  the night followed  by vertical  homog-
enization by day.   This results in a rapid dispersion of emissions
over a regional scale (Sections  3.2, 3.3.2, and  3.4.1).

Prevailing winds are strongly influenced  locally by mesoscale  effects
such as complex terrain  and  storm  fronts.  Alterations of  air parcel
trajectories by local  vertical  flows  remain  inadequately understood,
at least partly due to the  lack of routine vertical  wind data.  Con-
ventional methods of air trajectory analyses in frontal zones,  near
squall  lines  and   other  storm systems,  may  be  quite   inadequate
(Section 3.3.4).

A major  source  of  uncertainty in long range trajectory calculations
is related to the inadequacy of currently available routine upper air
wind   data,  which   represent  relatively  sparse,  two-dimensional,
Eulerian  measurements.     Their   spatial-temporal   coverage   cannot
provide  important  information  concerning mesoscale  flows.    Field
                             3-89

-------
  experiments  to characterize long-range transport  under  a  variety of
  flow conditions are needed  (Sections 3.2.2 and 3.5).

  Individual  trajectory  calculations  can be highly  uncertain,  and the
  use  of the  statistics  of  multiple  trajectories is  to  be  preferred.
  In general,  the uncertainties associated with transport processes are
  known  only in a qualitative sense;  rigorous estimation of uncertain-
  ties is limited to particular models, at best (Sections 3.1 and 3.5).

  Deposition from a pollutant  source  is greatest near  the  source and
  decreases  more or less  exponentially  away from the  source.   In the
  summer,  on the average, well  over half  of the eastern  U.S. sulfur
  emissions  may  be deposited  within  two  days  and 500  km from the
  source.  The transport range  is likely to be considerably greater in
  winter.   Average or cumulative deposition,  particularly dry  deposi-
  tion,  extends in  all  directions from  the  source,  but the deposition
  pattern  is not homogeneous.   The  prevailing flow is  reflected  in a
  shift  of  the  deposition maxima downstream in time;  in  the ecologi-
  cally  sensitive regions of eastern  North America, downstream gener-
  ally means toward the  east or  northeast.  This conclusion  is based
^primarily   on  observations   and  modeling  of  SO/.  The  conclusion
  probably   applies  to  NOX,  but  in  general,  information  related to
  atmospheric residence  times  of nitrogen compounds  is  less complete
  and more tentative than for sulfur compounds  (Sections 3.4 and 3.5).

  Based  on modeling simulations for summer  conditions, one may  identify
  three  approximate regions  in  northeastern  United  States and eastern
  Canada in  which the  relative contributions to acidic depositions due
  to emissions f»»om near  (<  500  km)  versus distant (> 500 km)  sources
  may  be significantly different.   In  the upper Ohio  River Valley (a
  dense  source  region),  sources  within 500 km appear  to dominate the
  maxima of  ambient  SOg and aerosol  sulfate  concentrations,  as   well
  as the  total  wet and dry  depositions  of  sulfur.   At  the  other
  extreme,  in upper New  England  and  parts  of eastern Canada which are
  remote from  major sources  of  sulfur,  long-range transport may be
  responsible for most of the aerosol sulfate and total wet deposition
  of sulfur.   In  the  intermediate regions, including the AdiVondacks,
  contributions to total acidic  depositions  from  near and far  sources
  may be more  comparable, considered on  a  regional  and summer average
  basis.  These simulations  have  a minimum resolution of about 100 km
  and thus  do  not reflect  local  source  "hot spots."   The  relative
  contributions of long-range  transport and  local  circulations to the
  deposition  patterns  in  the  eastern  coastal  region of  the United
  States are  not well  understood.   In general, modeling  uncertainties
  make  the  boundary  between local  and long-range  domination  somewhat
  tentative.   Also, estimates  of  regional   dry  depositions  must be
  viewed as  tentative  since  they  are  based  on indirect,  very local, and
  rather sparse measurements of  dry  deposition parameters  rather  than
  on  direct  regional  monitoring  of  dry  deposition  fluxes  (Section
  3.5).
                               3-90

-------
Acknowledgment:   A  significant  amount  of  the material  presented  in this
                 chapter was  developed  under  cooperative  agreement  between
                 Washington University and the U.S. Environmental Protection
                 Agency (CR-80-9713,  CR-81-0325, and CR-81-0351).
                                     3-91

-------
3.7  REFERENCES
Alberty, R. L., D.  W.  Burgess,  C.  E.  Hand,  and J.  F. Weaver.  1979.  SESAME
1979 Operations Summary,  Technical  Report, NOAA-ERL,  Boulder, CO.
Altshuller, A.  P.   1977.   Formation  and removal  of S02  and  oxidants from
the atmosphere.  Adv.  Environ.  Sci.  Techno!. 8:9.
Anthes, R. A.,  H.  A.  Panofsky,  J.  J.  Cahir  and A.  Rango.  1975.  The Atmos-
phere.  Chas.  E. Merrill  Pub1,.,  Columbus, OH.
Arya, S. P.  1982.   Atmospheric  boundary  layers over  homogeneous  terrain, Ch.
6.  Jjn Engineering  Meteorology.   E.  J.  Plate,  ed.   Elsevier,  Amsterdam.
Barry,  R.  G.  and R. J.  Chorley.   1977.  Atmosphere,  Weather, and Climate.
Third Ed., Methuen  & Co., Ltd.,  London.
Bass, A.   1979.  Modeling  long  range  transport and diffusion.    Proceedings,
Second  Joint  Conference  on  Applications  of Air  Pollution   Meteorology.
American Meteorol.  Society, Boston  MA.
Benkovitz,  C.  M.    1982.    Compilation of  an  inventory  of   anthropogenic
emissions  in the United States and  Canada.   Atmos.  Environ.  16:1551-1564.
Beran,  D.  W.    1978.   Prototype Regional  Observing and  Forecasting Service.
NOAA Executive Summary of a Program Development Plan.   NOAA-ERL,  Boulder, CO.
Betts, A.  K., R. W. Grover and M. W. Monerieff.  1976.   Structure and motion
of  tropical  squall  lines  over  Venezuela.    Quart.   J.   Roy.  Met.  Soc.
102:395-404.
Blackadar, A.  K.   1957.   Boundary  layer wind maxima and their  significance
for  the growth  of nocturnal  inversions.   Bull.  Amer.  Meteorol.  Soc. 38:
283-290.
Bonner, W. D.  1968.  Climatology of the low level  jet.   Monthly Weather Rev.
96:833-850.
Bonner,  W. D.,  S.  Esbensen, and  R.  Greenberg.   1968.   Kinematics  of the
low-level  jet.  J.  Appl.  Meteorol.  7:339-347.
Bernstein,  R.  D.  and D.  S.  Johnson.   1977.   Urban-rural  wind  differences.
Atmos.  Environ. 11:597.
Briggs,  G. A.  1969.   Plume Rise.  USAEC Critical Review Series, TID-25075,
Clearinghouse  for Federal Scientific and Technical  Information.
Briggs,  G. A.   1975.   Plurne rise predictions,  ^n Lectures  on  Air  Pollution
and Environmental Impact Analyses.   Amer. Meteorol. Soc., Boston, MA.
                                     3-92

-------
Brown, R.  A.    1974.   Analytical Methods  in PBL Modelling.   Halsted Press
(John Wiley),  New York.

Burpee, R. W.   1979.   Peninsula-scale convergence in  the  south Florida sea
breeze.  Mon.  Weather Rev.  107:852-860.

Businger, J. A.  and  S.  P.  Arya.   1974.   Height of the mixed layer in stably
stratified boundary layer.   Advances  in  Geophysics, Vol. 18A:73-92.

Businger, J. A., J. C. Wyngaard, Y.  Izumi, and E. F.  Bradley.   1971.  Flux-
profile  relationships  in  the  atmospheric  surface  layer.   J.  Atmos.  Sci.
28:181-189.

Byers, H. R. and R. R. Braham, Jr.   1949.  The Thunderstorm.   U.S.  Printing
Office, Washington, DC.

Carpenter, K.   1979.   An  experimental  forecast  using non-hydrostatic meso-
scale model.  Quart.  J.  Roy.  Met.  Soc. 105:629-655.

Caughey,  S. J.  and J.  C.  Kaimal.   1977.   Vertical  heat  flux in the convective
boundary layer.  Quart.  J.  Roy.  Met.  Soc. 103:811-815.

Caughey,  S. J.,  J. C.  Wyngaard,  and  J.  C.  Kaimal.  1979.   Turbulence in the
evolving  stable boundary layer.   J. Atmos.  Sci.   36:1041-1052.

Chandler, T. J.  1970.   Urban Climatology-Inventory and Prospect.  WMO TN No.
108:2-14.

Ching, J.  K.   S.,  J.  F.  Clark,  J.  S.  Irwin, and J. M.  Godowitch.   1983.
Relevance of mixed layer sealing for daytime  dispersion  based  on  RAPS and
other field programs.   Atmos. Environ. :In press.

Chung, Y.  S.   1978.   The distribution of atmospheric sulfates in Canada and
its  relationship to  long-range  transport  of  pollutants.    Atmos.   Environ.
12:1471-1480.

Clarke,  R.  H.   1970.   Observational  studies  in the  atmospheric  boundary
layer.  Quart.  J. Roy.  Met.  Soc.  96:91-114.

Clarke, T. F.  and  F.  G.  McElroy.  1970.   Experimental  studies  of Nocturnal
Urban Boundary Layer.   WMO  TN No.  108:108-112.

Clarke, R. H.,  A. J. Dyer,  R.  R.  Brook, D. G.  Reid,  and A. J. Troup.  1971.
The  Wangara Experiment:   Boundary Layer  Data,  Tech.  Paper  No.  19, CSIRO.
Div. Meteor. Phys.

Davis, W.  E. and L.  L.  Wendell.  1976.   Some effects of isentropic  vertical
motion simulation in  a  regional-scale quasi-Lagrangian air  quality model.  In
Proc., Third Symposium  on Atmospheric Turbulence,  Diffusion, and  Air  Quality,
Oct. 19-22, Raleigh,  NC.   American Meteorological  Society, 403-406.
                                     3-93

-------
Day, S.   1953.   Horizontal  convergence and  the occurrence  of  summer pre-
cipitation at Miami, Florida.   Won.  Weather  Rev.  81:155-161.

Deardorff, J. W.  1980.  Progress in understanding entrainment at  the  top of
the mixed layer.  Proc. AMS Workshop on the Planetary Boundary Layer.  J. C.
Wyngaard, ed.  American Meteorological  Society,  Boston, MA.

Defant, F.   1950.   Theorie der land-und  seewind.  Arch. Meteorol., Geophys.
Bioklimatol., Ser. A 2:404-425.

Defant, F.   1951.   Local  winds.  In Compendium  of Meteorology, pp. 655-672.
American Meteorological Society, Boston, MA.

Draxler, R. R. and A. D. Taylor.  1982.   Horizontal dispersion parameters for
long-range transport modeling.   J. Appl. Meteorol. 21:367-372.

Durst, C. S., A. F. Crossley  and N.  E.  Davis.   1959.  Horizontal diffusion in
the  atmosphere  as determined  by geostrophic  trajectories.   J.  Fluid Mech.
6:401-422.

Easter, R. C.   1981.   The OSCAR experiment.   In Proc., ACS Symposium on Acid
Rain.  Las Vegas, NV.

Edinger,  J.  G.  and  T.  F.   Press.    1982.    Meteorological   factors  in the
formation of regional haze.   Final report,  EPA ORD  ESRL.  October  1982.

Egan,  B.  A.   1975.   Turbulent  diffusion in complex terrain.  lr±  Lectures on
Air  Pollution  and  Environmental  Impact  Analyses.    Amer.  Meteorol.   Soc.,
Boston, MA.

Estoque, M.  A.   1961.  A  theoretical  investigation of  sea  breeze.   Quart. J.
Roy. Met. Soc. 87:136-146.

Estoque,  M.  A.   1962.  The  sea breeze  as  a function of prevailing  synoptic
situation.   J. Atmos. Sci. 19:244-250.

Estoque,  M.  A.,  J. Gross  and H. W. Lai.  1976.  A lake  breeze  over southern
L.  Ontario.  Mon. Weather Rev.  104:386-396.

Farquaharson,  S. J.   1939.    The diurnal  variations of wind over  tropical
Africa.  Quart.  J. Roy. Met.  Soc. 65:165-183.

Frank,  W.  M.  1978.  The life  cycles of GATE  convective systems.   J.  Atmos.
Sci. 35:1256-1264.

Frenzen, P.  1980.   Discussion  following paper  by L. Mahrt in Proceedings  of
the Workshop on the Planetary  Boundary  Layer, Boulder,  CO, 14-18  August.
Amer.  Meteorol.  Soc.,  Boston, MA.

Fritsch, J.  M.  and  R.  A. Maddox.   1980.  Analysis  of upper tropospheric wind
perturbations  associated  with  mid-altitude mesoscale convective  complexes.
Preprint  volume, AMS  Conf. on  Weather Forecasting Anal., 339-345.


                                     3-94

-------
Fritsch, J.  M.  and  R.  A.  Maddox.   1981.   Convectively  driven  mesoscale
weather systems aloft.   Part  I:  Observations.  J. App. Meterol. 20:1, 9-19.

Fujita, T.  T.    1959.   Precipitation and  cold  air production  in  mesoscale
thunderstorm systems.   J. Meteorol. 16:459-466.

Garrett,  J.  R.    1982.   Observations  in  the  nocturnal  boundary  layer.
Boundary Layer Met.  22:21-48.

Garstang,  M.,  and A.  K. Betts.   1974.   A review of  the  tropical  boundary
layer  and  cumulus  convection:   Structure  parameterization  and  modeling.
Bull. Am.  Meteorol.  Soc.  55:1195-1205.

Gentry, R. C.  and P. L. Moore.    1954.  Relation  of  local  and general  wind
interaction near the sea coast to time and location of air-mass showers.  J.
Meteorol.  11:507-511.

Gillani, N.  V.   1978.   Project  MISTT:    Mesoscale  plume  modeling  of  the
dispersion,  transformation  and  ground  removal  of  SO?.   Atmos.  Environ.
12:569-588.

Gillani, N. V. and W.  E. Wilson.   1980.   Formation  and  transport of ozone and
aerosols in power plant plumes.  Annals N.Y.  Acad.  Sci. 338:276-296.

Gillani, N. V. and W.  E. Wilson.   1983.   Gas-to-particle conversion of sulfur
in power plant plumes:   II.  Observation of liquid-phase conversions.  Atmos.
Environ. 17{9):1739-1752.

Gillani, N.  V.,  J.  A.   Colby, and W.  E.  Wilson.   1983. Gas-to-particle con-
version of  sulfur  in  power plant  plumes:  III.    Parameterization  of plume-
cloud interactions.   Atmos. Environ.  17(9):1753-1764.

Gillani, N.  V.,  S.  Kohli,  and W.  E.  Wilson.  1981.  Gas-to-particle conver-
sion of sulfur in power plant plumes - I.  Parameterization of the conversion
rate  for  dry,  moderately  polluted   ambient conditions.   Atmos.  Environ.
15:2293-2313.

Gillani, N. V., B. R.  Husar,  J. D. Husar, D. E.  Patterson, and W. E. Wilson.
1978  .   Project MISTT:  Kinetics of particulate sulfur formation in a power
plant plume out to 300  km. Atmos.  Environ. 12:589-598.

Goualt, J.   1938.   Vents en  altitude a fort Lamy (Tchad).   Ann.  Phys.  du
Globe de la France d'Outre-Mer 5:70-91.

Haugen, D. A., J. C. Kaimal,  and E. F.  Bradley.   1971.  An experimental study
of Reynolds  stress and  heat flux  in  the  atmospheric surface layer. Quart J.
Roy. Met. Soc. 97:168-180.

Heffter, J.  L.  1980.   Air Resources Laboratories Atmospheric Transport and
Disperion  Model  (ARL-ATAD),   NOAA Tech.  Memo.,   ERL  ARL-81,   Air  Resources
Laboratories,  Silver Spring,  MD.
                                     3-95

-------
Hering, VI. S. and T.  R.  Borden.   1962.   Diurnal variations in the summer wind
field over the central  U.S.   J.  Atmos.  Sci. 19:81-86.

Hicks, B. B., G.  D.  Hess,  M.  L.  Wesely,  T. Yamada, P. Freugen, R. L. Hart, D.
L. Sisterson, P.  E.  Hess,  F. C. Kulhanek, R. C.  Lipschutz,  and G. A. Zerbe.
1981.  The  Sangamon  Field  Experiments:   Observations  of  the diurnal  varia-
tions of  the PBL over land.   Atmospheric Physics Section,  Argonne National
Laboratory, Report ANL-RER-81-1, Argonne,  IL.

Holton, J. R.  1967.   The diurnal boundary layer wind oscillation over slop-
ing terrain.  Tell us  19:199-205.

Holzworth, G. C.   1972.   Mixing  heights, wind speeds, and potential for urban
air pollution throughout the  contiguous United States.  U.S. EPA AP-101.

Hoxit, L. R., R.  A.  Maddox,  C.  F. Chappell, F. L.  Zurkerberg, H. M. Mogil, I.
Jones, D. R. Greene,  R.  E.  Saffle, and  R.  E.  Scofield.  1978.  Meteorological
aspects of  the  Johnstown,  PA flash  flood,  19-20  July 1977.   NOAA   TR ERL
401-APCL-43:1-71.

Hunt, J.  C. R.  and Simpson, J.  E.   1982.  Atmospheric boundary layers over
non-homogeneous terrain.  Ch. 7.   I\± Engineering  Meteorology.   E. J. Plate,
ed.  Elsevier, Amsterdam.

Hsu, S.   1969.   Mesoscale structure  of the Texas  Coast Sea  Breeze.  Rep. No.
16,  Atmospheric   Science  Group,  Univ. of  Texas,  College   of  Engineering,
Austin, TX.

Husar,  R. B. and D. E.  Patterson.    1980.    Regional  scale air pollution:
Sources and effects.   Annals  N.Y. Acad. Sci.  338:399-417.

Husar, R. B., D.  E.  Patterson,  J. D.  Husar,  N. V. Gillani, and W. E. Wilson.
1978.  Sulfur budget of a power plant plume.   Atmos.  Environ.
12:549-568.

Husar, R. B., D.  E.  Patterson,  C. C.  Paley,  and N. V. Gillani.  1976.  Ozone
in hazy air masses.   Proceedings of International  Conference on Photochemical
Oxidant and its Control, Sept.  12-17, Raleigh NC.

Izumi, Y.  1971.   Kansas 1968 Field Program Data  Report.  Air Force Cambridge
Research  Laboratories, Environmental  Research Papers. No. 379.

Johnson,  A.  Jr.  and  O'Brien, J. J.  1973.   A study of an  Oregon sea breeze
event.  J. Appl.  Meteorol.  12:1267-1283.

Kaimal, J.  C.,  J. C. Wyngaard,  D.  A.   Haugen, 0.  R.  Cote,  Y.  Izumi,  J. S.
Caughey,  and  C.  J.  Readings.  1976.    Turbulence  structure  in the convective
boundary  layer.  J.  Atmos.  Sci.  33:2152-2169.

Keen, C.  S.  and  W.  A.  Lyons.   1978.   Lake/land  breeze  circulations on the
western shore of L. Michigan.  J. Appl. Meteorol.  17:1843-1855.
                                     3-96

-------
Kimura R.  and T.  Eguchi.    1978.   On  dynamical  processes of  sea  and land
breeze circulations.   J.  Meteorol.  Soc.  Japan  56:67-85.

Koerber,  W.  M.   1982.    Trends  in  S02  emissions  and  associated release
height for  Ohio  River Valley  power plants.   Paper  No.  82-105, 75th  Annual
Meeting of the Air Pollution Control  Association,  New Orleans, LA.

Korshover,  J.   1967.   Climatology of  stagnating  anticyclones east  of the
Rocky  Mountains,   1936-65.     U.S.  Public  Health  Service  Publication  No.
999-AP-34.

Lamb, R.  G.   1981.    A regional  scale (1000  km)  model  of photochemical air
pollution.   Part I:   Theoretical  formulation.  USEPA  Technical Report.

Landsberg,  H.  E.   1956.   The  climate of  towns.   Proc.  Int. Symp.  on Man's
Role in Changing the  Face of the  Earth.   Univ. of  Chicago  Press, 584-606.

Lee, D. 0.  1977.  Urban influence  on  wind direction over London.  Weather.
32:162.

Lettau,  H.  H.    1967.   Small  to  large  scale  features  of  boundary layer
structure over mountain slopes. J^ Proc.,  Symposium  on Mountain Meteorology.
Atmospheric Sciences  Paper No. 122.  Colorado State  Univ., Fort Collins, CO,
pp. 1-74.

Lettau, H.  H.  and B.  Davidson.   1957.   Exploring the  Atmosphere's  First Mile,
Vol. 1 and 2.   Pergamon Press,  New York, NY.

Lilly, D.  K.   1975.    "Open SESAME," Proceedings  of SESAME  Open  Meeting at
Boulder,  CO, Sept. 4-6, 1974.  NOAA-ERL,  Boulder,  CO.

Lilly, D.  K.  and E.  J.  Zipser.   1972.   The  front range  windstorm  of 11
January 1972.   Weatherwise. 25:56-63.

Lyons, W. A.  1975.   Turbulent diffusion and  pollutant transport in  shoreline
environments.     In  Lectures  on   Air  Pollution   and  Environmental   Impact
Analyses.  Amer. "Meteorol.  Soc.,  Boston, MA.

Lyons, W.  A.  and R.   H.  Calby.  1983.   Impact  of Mesoscale Convective Pre-
cipitation Systems on Regional  Visibility  and Ozone  Distributions.  MESOMET,
Inc.  Final Report of Contract No.  68-02-3740  to EPA.

Lyons, W.  A.   and  R.  B.  Husar.    1976.   SMS/GOES visible  images  detect a
synoptic-scale air pollution episode.   Mon. Weather Rev.  104:1623-1626.

Lyons, W. A. and L.  E. Olsson.   1973.   Detailed mesometeorological studies of
air  pollution  dispersion  in the  Chicago lake breeze.    Mon.  Wea. Rev. 101:
387-403.
                                     3-97

-------
Lyons,  W.  A.,  J.  C.  Dooley,  Jr.,  and  K. T.  Whitby.   1978.   Satellite
detection  of long-range  pollution  transport  and   sulfate  aerosol  hazes.
Atmos. Environ.  12:621-531.

Mack, R. A. and D.  P. Wylie.  1982.  An estimation of the condensation rates
in three severe  storm  systems from  satellite  observation  of the convective
mass flux.   Mon. Wea. Rev.  110:725-744.

Maddox, R.  A.  1980.  Mesoscale convective  complexes.   Bull. Aner. Meteorol.
Soc. 61:1374-1387.

Mahrt, L.  1980.  Boundary layer mean flow  dynamics.   Proc.  AMS Workshop on
Planetary  Boundary   Layer,  J.  C.  Wyngaard,   ed.    American Meteorological
Society, Boston, MA.

Mahrt, L. and W. Schwerdtfeger.   1969.   Ekman  spiral for exponential thermal
wind.  Boundary Layer Met.  1:137-145.

Mahrt,  L.  and D. H. Lenschow.   1976.   Growth dynamics of the convectively
mixed layer.   J. Atmos.  Sci.  33:41-51.

Manton, M.  J.  1982.  A model of  fair-weather cumulus convection.  Boundary
Layer Met.  22:91-107.

Matson, M.,  E.  P. McClain,  D. F.  McGinnis, Jr., and J. A. Pritchard.  1978.
Satellite detection  of urban heat  islands.   Mon.  Wea.  Rev.  106:  1725-1734.

McBean, G.  A., K. Bernhardt,  S. Bodin, Z.  Lytyuska,  A.  P.  Van  Ulden, and J.
C.  Wyngaard.   1979.   The  Planetary Boundary  Layer,  WMO  Technical  Note No.
165, G. A.  McBean, ed.   WMO, Geneva,  Switzerland.

McNaughton, D J.  and M.  M. Orgill.  1980.   Synoptic  case  study of elevated
layers   of  high   airborne   sulfate  concentration.     Mon.  Weather   Rev.
108:655-662.

Miller, J.  M.   1981.  A  five-year  climatology of back  trajectories from the
Mauna Loa Observatory, Hawaii.  Atmos.  Environ. 15:1553-1558.

Miller, J.  M. and A. M.  Yoshinaga.   1981.   The pH of Hawaiian precipitation  -
a preliminary report.  Geophysical  Research Letters  8:779-782.

Munn,  R.  E.   and  B. Bolin.   1971.   Global air  pollution - meteorological
aspects.   Atmos. Environ. 5:363-402.

Neumann, J.   1951.   Land breezes and nocturnal  thunderstorms.  J.  Meteorol.
8:60-67.

Niemann, B.  L.   1982.   Analysis of  wind  and  precipitation data for assess-
ments  of transboundary  transport  and acid deposition between Canada and the
United  States.    Proceedings,  American  Chemical  Society  Symposium  on  Acid
Precipitation March 29,  Las Vegas,  NV.
                                     3-98

-------
Nunez, M. and T.  R.  Oke.   1977.   The energy balance of an urban canyon.  J.
Appl. Meteorol.  16:11-19.

Oke,  T.  R.   1973.   City  size and the  urban  heat  island.   Atmos. Environ.
7:769-779.

Oke,  T. R.  1974.   Review of Urban Climatology,  1968-1973.  WMO TN  No. 134.

Oke,  T.  R.   1978.   Air pollution on  the boundary layer, Ch. 9.  j£ Boundary
Layer Climates.   Methuen & Co.  Ltd.,  pp.  268-301.

Organization for Economic  Cooperation and  Development.   1977.   Final  Report
of  the   OECD   Programme   on  long   range   transport  of  air  pollutants.
Measurements and Findings.

Pack, D. H., 6. J. Ferber, J.  L.  Heffter,  K.  Telegadas,  J.  K.  Angell, W. H.
Hoecker  and L.  Machta.   1978.   Meteorology of long range transport.  Atmos.
Environ. 12:425-444.

Paegel, J.  1969.   Studies of diurnally  periodic  boundary layer winds.  Ph.D.
Thesis, Dept. of Meterology, UCLA.

Panofsky, H. A.   1982.   The Atmosphere, Ch. 1.  Jji Engineering Meteorology.
E. J. Plate (ed.), Elsevier, Amsterdam.

Pielke, R. A.  1974a.  A three-dimensional  numerical model of the sea  breezes
over south Florida.  Mon.  Wea.  Rev. 102:115-139.

Pielke,  R. A.   1974b.   A  comparison  of  three-dimensional  and two-dimensional
numerical predictions of sea breeze.   J.  Atmos.  Sci. 31:1577-1585.

Pielke,  R.  A.   1981.   Mesoscale  numerical modeling.   Jji  Adv.  in Geophys.
23:185-344.

Plank,  V. G.    1966.    Wind  conditions in  situations of  patternform and
non-patternform cumulus convection.  Tell us  18:1-12.

Plate,  E.  J.   1971.   Aerodynamic Characteristics of  Atmospheric Boundary
Layers, U.S. AEC Critical  Review Series.

Pooler,  F.  and L.  E.  Niemeyer.   1970.   Dispersion  from tall  stacks:   An
evaluation.  Paper No. ME-14D presented  at  the Second  International Clean Air
Congress, Washington, D.C., December  6-11,  1970.

Portelli,  R.  V.     1977.    Mixing  heights,  wind  speeds  and  ventilation
coefficients  for  Canada.     Environment  Canada,  Atmospheric  Environment
Service, Climatological  Studies Number 31,  UDC:551.554.

Prahm, L. P., U. Torp, and R. M.  Stern.   1976.   Deposition and  transformation
rates  of sulfur  oxides  during   atmospheric  transport  over  the  Atlantic.
Tell us 28:335-372.
                                     3-99

-------
Price, J. C.   1979.   Assessment of the urban heat island effect through the
use of satellite data.   Mon.  Weather  Rev. 107:1554-1557.
Project METROMEX.   1976.   METROMEX  update.   Bull.  Am. Meteorol.  Soc.  57,
304-308.
Rahn, K.  A.    1981.   Relative  importances  of North America  and Eurasia as
sources of Arctic aerosol.   Atmos.  Environ.  15:1447-1455.
Rahn, K. A. and  R. J. McCaffrey.   1980.  On the  origin  and transport of the
winter Arctic aerosol.   Ann.  N.  Y.  Acad.  Sci.  388:486-503.
Richwien, B. A.  1978.   The  damming  effect  of the  southern Appalachians. Am.
Meteorol. Soc.  Conf. Proc. Weather Forecast.  Anal.;  Aviat.  Meteorol., 1978.
pp. 94-101.
Rodhe,  H.    1974.    Some aspects  of  the  use of  air  trajectories  for the
computation of  large-scale dispersion  and fallout  patterns.   Jji Advances in
Geophysics, Vol. 18B, Academic Press, New York.
Schreffler, J.  W.  1978.  Detection  of centripetal, heat island  circulations
from tower data in St.  Louis.  Boundary Layer Meteorology.  15:229.
Shannon,  J.   D.   1981.    A model  of  regional  long-term   average  sulfur
atmospheric  pollution,  surface  removal,  and  net  horizontal   flux.   Atmos.
Environ. 15:689-701.
Sheih,  C.  M.   1980.   On lateral dispersion coefficients as functions of
averaging time.  J. Appl. Meteorol. 19:557-561.
Shipman,  M.  S.    1979.   Dynamics of the  nocturnal  boundary layer.    M.S.
Thesis, Dept. of Meteorology, NC State University,  Raleigh.
Sisterson, D. L. and P.  Frenzen.  1978.  Nocturnal  boundary-layer wind maxima
and  the problem of wind power assessment.  Environ. Sci. Technol. 12:218-221.
Sisterson, D.  L.,  J.  D. Shannon and J.  M.  Hales.   1979.   An  examination of
regional  pollutant structure in the lower  troposphere—some  results of the
diagnostic   atmospheric   cross-section  experiment  (DASCE-I).    J.   Appl.
Meteorol. 18:1421-1428.
Skibin, D. and  A. Hod.  1979.  Subjective analysis  of  mesoscale flow patterns
in northern  Israel.  J. Appl. Meteorol. 18:329-337.
Smith,  F.  B.  and R.  D.  Hunt.  1978.   Meteorological  aspects of the transport
of pollution over long distances.  Atmos. Environ.  12:461-477.
Tennekes, H.   1974.  The atmospheric  boundary layer.   Phys.  Today 52-63.
Thomas,  F.  W.,  S. B. Carpenter,  and F. E.  Gartrell.   1963.    Stacks - how
high?  JAPCA 13(5):198-204.
                                     3-100

-------
Thompson, D. E., P. A. Arkin, and W. D.  Bonner.   1976.   Diurnal variations of
the  summertime  winds  and force  field  at three mideastern  locations.   Mon.
Weather Rev. 104:1012-1022.

long, E. Y., G. M. Hidy, T.  F.  Lavery,  and  F.  Berlandi.   1976.   Regional and
local aspects  of atmospheric sulfates  in  the northeastern  quadrant of the
U.S.  Proceedings, Third Symposium  on Turbulence, Diffusion and  Air  Quality,
American Meteor. Society, Boston, MA.

Uthe, E. E. and W. E.  Wilson.   1979.   Lidar observations of the density and
behavior of the Labadie power plant plume.   Atmos. Environ.  13:1395-1412.

Uthe, E. E., F.  L. Ludwig,  and F. Pooler.   1980.  Lidar  observations of the
diurnal  behavior of the Cumberland power plant  plume.  JAPCA 30(8)-.889-893.

Viskanta, R.,  R.  W.  Bergstrom,  and R.  0.  Johnson.    1977.    Effects of air
pollution on thermal  structure and  dispersion in an urban planetary  boundary
layer.  Contrib. Atmos. Phys. 50:419-440.

Vukovich, F. M., W. D.  Bach,  Jr., B. W.  Cressman,  and  W. J.  King.  1977.  On
the  relationships  between  high  ozone  in  the rural  boundary  layer and high
pressure systems.  Atmos. Environ. 11:967-983.

Wagner,  A.   1939.   Uber die Tageswinde in der freien  Atmosphere.   Beitr.
Phys. Atmos. 25:145-170.

Warner,  T. J.,  R.  A.  Anthes, and A. L. McNab.  1978.   Numerical simulations
with a three-dimensional mesoscale model.  Mon. Wea. Rev.  106:1079-1099.

Wendland, W.  M.  and  R. A.  Bryson.   1981.    Northern  hemisphere airstream
regions.  Mon.  Weather Rev.  109:255-270.

White, W. H.,  J.  A.  Andersen, D. L. Blumenthal, R.  B. Husar, N. V.  Gillani,
J. D. Husar, and W. E.  Wilson.   1976.   Formation  and  transport of secondary
air  pollutants:   Ozone and aerosols in the  St.  Louis  urban plume.  Science
194:187-189.

Wolff, G. T.,  N. A. Kelly, M. A.  Furman.  1981.  On the  sources of  summertime
haze in  the eastern United States.  Science  211:703-705.

Zipser,  E.  J.  and C.  Gautier.  1978. Mesoscale events within a GATE  tropical
depression.   Mon.  Wea.  Rev. 106:789-805.
                                    3-101

-------
               THE ACIDIC DEPOSITION PHENOMENON AND  ITS  EFFECTS

                        A-4.  TRANSFORMATION PROCESSES
4.1  INTRODUCTION (D. F. Miller)

This  chapter addresses  the atmospheric  processes by  which pollutants  are
transformed chemically into species that ultimately may result  in  deposition
of  acidic   matter.     When  chemical  transformations  are  considered,   a
fundamental  concern  is  for  the  kinetics  of  reactions  that  limit  the
production and consumption  of  acidic  species and  their precursors.   In this
chapter, many individual equations pertaining to  gas-phase  and  aqueous-phase
reactions  have been  written and assigned best estimates for  their kinetics.
However, to assess the relative importance of these reactions with  respect to
acid  deposition  under  various  atmospheric conditions,  one  must  evaluate this
information along with the other facets of this document;  i.e.,  the pollutant
emissions  and distributions  (Chapters  A-2 and  A-5);  transport (Chapter A-3);
and   other  meteorological   processes   (Chapters   A-6   and   A-7),   including
preci pi tation-deposi tion processes.

To  integrate the detailed  aspects  of  atmospheric chemistry with models  of
atmospheric  physics  requires  an  operational   scheme  referred to  in  this
chapter  as transformation  modeling.   The basic  approaches  to transformation
modeling,  the problems encountered  and some exemplary  results  are discussed
at the end of this chapter.

Figure 4-1,  taken from Schwartz  (1982), depicts  in simplified form the types
of transformation processes  by which  common pollutants become more acidic in
the atmosphere.

The diagram  shows areas for interactions between gas-phase and aqueous-phase
processes.  While gas-phase oxidation is conceptualized as a direct route for
producing  acidic products,  the aqueous-phase route is somewhat  more complex.
There is  partitioning  of   the  gaseous  reactants  between  the   two  phases
followed by  oxidation  and possibly neutralization.  Since most of this occurs
in  cloud  droplets  which  evaporate   rather  than precipitate,   the  acidic
products are vented  into  the atmosphere,  primarily  in the  form  of aerosol
particles.    In general,   these  particles  will  have  longer  atmospheric
lifetimes  (and  transport  times)  than their  gaseous  precursors.    In many
respects,  cloud  droplets have the property  of  forcing pollutants  to undergo
reactions  at much  faster rates than  experienced in the gas phase.  Oxidation
of S02 by  03  and H202  are the  two familiar  examples.

In  Sections  4.2  and  4.3  of  this   chapter,   gas-phase   and  aqueous-phase
transformations  are   discussed  separately.    The  section  on   homogeneous
gas-phase  reactions suggests  that the fundamental chemistry is  fairly well


                                    4-1

-------
              GAS PHASE
                  GASEOUS OXIDE
                       S02
                     NO;  N02
ro
   AEROSOL,
CLOUD DROPLET,
 OXIDATION
(HYDRATION^
HO, H02, 03
                          AQUEOUS OXIDE
                           (WEAK ACID)
                                          .
                       S02(aq.)= H+t  HS03
                         N0(aq.),
GASEOUS ACID
    H2S04
    HN03
  OXIDATION^    AQUEOUS     NEUTRALIZATION
            ""   STRONG ACID
  °3; H2°2*.     2H+, S042'   NH3;MO; MC03
 02 + Fe, Pin...   H+, N03~
                                 AQUEOUS or
                                 DRY SALT
                               (NH4)2S04; MS04
                               NH4N03;  M(N03)2
  Figure 4-1.  Schematic representation of pathways for atmospheric formation of sulfate and nitrate.
               Adapted from Schwartz (1982).

-------
established,  although  there are specific areas of  uncertainty  pertaining to
the formation of acidic  species.   A major  problem is that field measurements
have  not  been  adequate  to  definitely  test  the chemical  models  based  on
laboratory studies.

    An   appreciation   of  the   time   scales  that   characterize   gas-phase
transformation  paths   can  be  had   by   direct   measurements,  theoretical
calculations, or budget  calculations based  on  time  and  space averages (Rodhe
1978).    When  a  gas-phase  transformation  process  can  be  described  by  a
first-order  reaction,  the lifetime of the  reacting species  with respect to
the  particular  reaction is equal  to  the reciprocal  of  the rate coefficient
(IT1).   For  a  bimolecular gas-phase  reaction (A+B^-C+D),  a  pseudo
first-order  rate  for  the removal of A may  be  approximated  by K [B]  when  the
concentration of B can be estimated.

    In  contrast to the  situation  for gas-phase  chemistry,  the  fundamental
chemistry  of  aqueous-phase  reactions  leading  to  acid  products  in  the
atmosphere is  not well known.   Thus,  in this  chapter  there is very  little
discussion of the myriad chemical mechanisms likely to be occurring  in cloud,
fog and even dew droplets.  Aqueous-phase chemistry is discussed primarily on
the basis of generalized rate expressions,  and assessments of the atmospheric
significance  of various  chemical  processes in  clouds  are made using  best
available information and necessary assumptions.

    The rate of a gas-liquid reaction (as in aqueous cloud droplets)   depends
upon the physical  solubility of  the reactant gas,  the  rate  of mass  transport
of  the  reactant  and  the  aqueous phase  reaction  rate.    To  estimate  the
lifetime of  a  given  reactant,  one must  further  consider  the  liquid  water
content  of  the  cloud;  other  solutes  which may affect ionic  strength,  pH or
act as  oxidizers;  and the  residence  time  of air  within clouds.  Since  the
liquid  water content  may  vary from  1  x  10~5  g  m~3  for  embryonic  cloud
nuclei to >  1  g m~3 for  dense  clouds, there are  problems  in evaluating  the
lifetimes of species that react under such  conditions.

    References  specifically  to heterogenous  (gas-solid)   reactions  in  the
atmosphere are not included in this chapter.  Although there has been  valuble
research on  this  topic,  it is  not yet possible to assess  the  importance of
these reactions to the acidic  deposition  problem.   The consensus at  this time
seems  to  be that heterogeneous reactions  make significant  contributions  to
acidic deposition but only  under rather  special circumstances which  have  not
been well defined.

4.2  HOMOGENEOUS GAS-PHASE REACTIONS   (D. F. Miller and M. R. Whitbeck)

4.2.1  Fundamental  Reactions

4.2.1.1  Reduced  Sulfur Compounds—Sulfur  (S)  occurs in the troposphere  in
diverse  forms  involving  oxidation  states  from   -2  (H£S)  to  +6   (H2S04).
The chemical  mechanisms and kinetics of reduced S  compounds such  as  hydrogen
sulfide  (HgS)  and  carbonyl   sulfide  (COS)   have  not   been  studied   as
extensively as sulfur dioxide  ($02) and sulfuric acid ^$04)  have.
                                    4-3

-------
The oxidation of  reduced  S  compounds  in the troposphere presumably leads to
S02 formation.   Some possible  reactions  are listed  in Table 4-1.   Except
for the  first reaction,  OH +  H2S,  considerable  uncertainty  surrounds the
products and rate constants  (Baulch et al .  1980).

The  atmospheric   lifetimes  of  these  reduced  S  compounds  with  respect to
gas-phase  reactions  are expected  to  be determined by  their reactions  with
hydroxyl (OH) radicals.   Table  4-2 lists some typical  background  concentra-
tions for the compounds (Sze and Ko 1980)  and estimated  lifetimes  for  removal
by a background OH level of  4 x  10~5  ppb.

Data are  insufficient to assess quantitatively  the  importance  of reduced  S
compounds  on acidic  precipitation;  but,  relative  to  the  strong local S02
emissions   from   anthropogenic   sources,    their   contribution  may  be
insignificant.   They do, however, significantly  contribute  to  the global  S
budget, but further work in  this area is needed  to clarify  reaction pathways.
In particular, rate  constants and  products  for the  reactions of OH with COS,
carbon  disulfide  (C$2),  dimethyl   sulfide   (CHaSCHs),  and  other  biogenic,
reduced S compounds need to  be identified.

4.2.1.2  Sulfur  Dioxide—The  atmospheric  chemistry of  S02  has been  studied
extensively,  yet  some   aspects  are  still   not   well   delineated.    Removal
mechanisms  for  S02 are complex  and  involve aqueous droplet, gas-phase,  and
possibly  particulate reactions.   The gas-phase  reactions  for S02  represent
a  major  oxi dative path  in the troposphere,  although  it has  been  argued that
the aqueous-phase route is dominant (Holler 1980).

Direct   photo-oxidation   reactions   for  S02  play  a  minor  role   in  its
oxidation.    Reactions  4-7a and  4-7b  (Table  4-3)  dominate  the  fate of
S02(3Bi),  while  reactions   4-8,  4-9  or  4-10,   and  4-11  may  account  for
photo-oxidation  of  S02  at  a rate of  ~ 0.02 percent  hr"1  (Calvert et al .
1978) .
 Oxidation of  S02 by  excited  oxygen  (Ug,  1£g+)'  nitrogen dioxide (N02) ,
 nitrogen  trioxide  (NOa),  nitrogen  pentoxide  INpOc) ,  or  ozone  (03),  is
 unimportant  in  the  troposphere (Calvert et al .  19/8).   The reaction  of S02
 with  0(3p)  is not  a  significant route for oxidation  in  the  troposphere but
 should  be  included in  models  for  plume  chemistry,  where  it  may   play  a
 significant  role in early stages of plume dilution (Calvert et al . 1978).

 The  reaction of  S02  with  hydroperoxy (H02)  radicals  is not  well  defined.
 At one  time, it was felt that the  reaction with H02 was  a  significant path
 for   oxidation  in  a  highly  polluted troposphere  with  [H02] ~  0.24  ppb
 (Calvert  et al . 1978).   More recent  evidence,  e.g.,  Graham et  al .   (1979),
 suggests  that  the reaction  of  S02  with  H02  is  much too slow  to  be
 significant  in  the  troposphere.  An  analogous  reaction  is that of  S02 with
 methylperoxy  radicals   (^302).      Although   this  system   has   received
 attention  in  recent  years,  the  tropospheric  role  of  the  CH302  +  S02
 reaction  has not been  interpreted  concretely.    Table 4-4 lists some recent
 rate  constant determinations  for this  reaction.
                                     4-4

-------
                  TABLE 4-1.  REACTIONS OF REDUCED SULFUR
Reaction
Rate constant at 298 K
(cm3 molecule"1 s"1)
                        Reference
                     Reaction
                      number
OH+HS+HS+HO
      2          2
OH + CS2
HS + 0. -»• SO + OH
HS + 02 + S02 + H
SO + 0£ -* S02 + 0
     5.3 x 10-12
OH + OCS -*•  C02  + HS(?)    _< 6 x 10-14
                                    -14
                              1 x 10

                                   ,-16
                            5.8 x 10
     < 2 x 10
        -13

        -13
4.3 x 10

1.5 x lO"15
                                  -13
       < 10
       9 x 10-18
Baulch et al. (1980)  [4-1]


Baulch et al. (1980)  [4-2]

Demore et al. (1981)

Leu and Smith (1981)

Baulch et al. (1980)  [4-3]

Cox and Sheppard (1980)

Wine et al. (1980)


Baulch et al. (1980)  [4-4a]


                      [4-4b]


Baulch et al. (1980)  [4-5]
                 TABLE 4-2.  OCCURRENCE OF REDUCED SULFUR
                    Typical Concentrationa
    Molecule               (ppb)
asze and Ko (1980) .
                                    4-5
                            Lifetime for removal
                             by OH (s x 10-5)
H2S
COS
CS2
0.004
0
0.069
- 0.40
.49
- 0.370
1.9
1,000
6,750

-------
              TABLE 4-3.  PHOTOOXIDATION REACTIONS OF S02
                                                                Reaction
      Reaction                                                   number
S02(X lAi) + hv (340-400 nm) + S02(3Bi)                            [4-6]


S02(3Bi)  + 02(3zg") -> SO (X !A!)  + 02(lzg+)                        [4_7a]


S02(3Bi)  + 02(3£g~) ^ S°2^ IA!^  + ^^Ag)                         [4-7b]


S02 (3Bi) + 02 (3£g~) •* S04 (cyclic)                               [4-8]


S04( cyclic) + 02 •> S03 + 03                                        [4-9]


                 ~) ^ S03 + 0(3P)                                  [4-10]


                  M                                                [4-11]
                                  4-6

-------
        TABLE 4-4.  RATE CONSTANTS FOR CH302 + S02 + PRODUCTS

k (cm3 molecule"1 s"1)                             Reference
      <  5  x  10'17                              Sander and Watson (1981)
      8.2  x  10-15                              Sanhueza et al. (1979)
      5.3  x  lO-1^                              Kan et al. (1979)
      1.4  x  ID"14                              Kan et al. (1981)
                                    4-7

-------
The  rate  constant for  the SC>2  and methoxy  radical  (CH30)  reaction  should
be measured to  assess  its  significance accurately;  a  rough  estimate of 6  x
10-15  cm3  molecule'1   s-1  for  this  reaction   (Calvert  et  al ,  1978)   has
been reported.  Kan et  al .  (1981)  used a  larger  rate  (5.5 x 10-13)  in  their
assessment of this mechanism.

An  important  competitive  fate for  methoxy  radicals  is the reaction  with  Q£
which  has  a  rate  constant  of  5.7   x  10~16  cm3  molecule-1  s"1   (Demore
et  al .  1981).   That rate,  combined with  the ambient  level  of 63, keeps  the
level of  CH30 very low; probably  lower  than that for  OH.   Thus, if [CH30]
«  [OH]  and k(CH30   + S02)  <  k(OH  +  S02) ,   then   oxidation  of  S02  by
     is not important.
The  combined  oxidation of  S02  will  depend  on  the  concentration  of  other
reactive  species  (e.g.,  H02,  CH302,  CHaO,  NO,  N02) ,   as  suggested  in
a  recent  study by Kan  et al .  (1981).   Their  mechanism  and  suggested  rate
constants are  given  in Table  4-5.   Further study is needed  to  evaluate the
significance of this reaction  sequence.   If the Kan et al .  (1981)  mechanism
is  correct,  the  influence  of  atmospheric  levels  of NO on  the rate  of S02
oxidation by CHs02 will need to be assessed.

Ozone-alkene reactions are complex and give rise to diverse reactive radicals
that  may  oxidize  S02-    Some  possible  reactions  are listed in Table  4-6.
Cox  and  Penkett (1972) observed  that water markedly inhibits  S02  oxidation
in  these  systems.   Calvert et al . (1978)  have  evaluated the  data  of Cox and
Penkett (1972) for the cis-2-butene, 03, S02, H20 system in terms of:

    03 + C4Hs -" molozonide -> CHsCHOO + CHsCHO                          [4-23]

    03 + C4Hs -* RCHO, RCOOH, etc.                                      [4-24]

    CH3CHOO + S02 + CH3CHO + $03                                       [4-25]

    CH3CHOO + C4H8 + CH3CHO +  C4H80 + other products                   [4-26]

    CH3CHOO + 03 •* CH3CHO + 202                                        [4-27]

    CH3CHOO + H20 •> CH3COOH +  H20                                      [4-28]

    CH3CHOO +  (CH3COOH)t+CH4  +  C02  ( + CH3OH, CO, etc.)                [4-29]

and have concluded  that reactions  with  the Criegee  intermediate (Criegee
1957)  cannot be neglected  as  a loss mechanism for  S02«   The lack of direct
observation  of these  elementary reactions and  subsequent determinations of
their rate constants  hampers  a quantitative assessment  but S02   conversion
rates by this mechanism are  not  expected to be large.

The predominant gas-phase mechanism  for S02 oxidation is the reaction  with
OH.

          OH +  S02 •* HOS02                                             [4-30]
                                     4-8

-------
TABLE 4-5.  CH302 + SOg  MECHANISM OF  KAN  ET  AL.  (1981)
Reaction Suggested rate constant
CHo09 + S09 ->• CH,0?SO? 1.4 x 10" cm molecules" s"
O £ £ O £ £>
CH302S02 -*• CH302 + S02 < 24 s"
20
CH302S02 + 02 ->• CH302S0202~] K14/k15 = 1<7 x 10
3 . . -1
CH302S0202 -* CH302S02+ 02 cm molecule
CH302S0202 + NO +
N02 + CH302S020 6.2 x 10" cm molecule" s
CH,00S000 + CH00 + 00 3.3 x 1013 cm3 molecule"1 s"1
322 3 2
ru n en n^ru n -4- <;n 	
Reaction
number
[4-12]
[4-13]
[4-14]
[4-15]
[4-16]
[4-17]
rd-ifti
                         4-9

-------
        TABLE 4-6.  POSSIBLE SOg-Oa-ALKENE REACTIONS
 Reaction                               .               Reaction
                                                        number
       o-o-o


R - CH	CHR + S02 -> 2RCHO + S03                 [4-19]
   0.  0-0.
  f   I


RCH - CHR + S02 + 2RCHO + S03                            [4-20]




 •

RCHOO + S02 -> RCHO + S03                                 [4-21]







RCHO- + S02 + RCHO + S03                                 [4-22]
                              4-10

-------
The  recommended  rate   constant   for   this  reaction  is  2  x  10-12   Cm3
molecule"1  s'1  (Baulch  et al .  1980).    Further  improvement  on  this  rate
constant  and  studies  on the  subsequent  fate  of the HOSO? radical have  been
recommended (Seinfeld et al . 1981).  Calvert et al . (1978), Davis  and  Klauber
(1975),  and Davis et  al .  (1979)   have  speculated  on the  fate  of the HOS02
radical in  the troposphere  (Table  4-7).  The  determination of rate constants
and  fate  of  the HOS02  radical   constitute   a pressing  need  for   further
research.   At this  writing,  however,  all  evidence  suggests  that  a final
product  of the HO  +  S02  reaction  is  sulfuric acid and  that  this  initial'
step is rate limiting.

The  fate  of  sulfur  tri oxide  ($03) in  the  atmosphere  is  expected  to  be
dominated  by  its  reaction  with  water (Calvert et  al . 1978), although Baulch
et  al .  (1980) make  no  recommendation  for  this  reaction because  only  one
investigation of  the  process  (Castleman et al .  1975) was conducted  and  the
reaction products were not identified.   The  presumed reaction is:

          $03 + H20 + (S03-H20)  -*  H2$04                                 [4-58]
 4.2.1.3   Nitrogen  Compounds— The chemistry of  N  in the troposphere  rivals
that of S, both in the diversity of compounds present and in their impacts  on
acidity of precipitation.  N  is  found  with oxidation states ranging from  -3
(ammonia  [N^])  to  +5   (pernitric  [HO?^]  acid),  including  both   bases
(ammonia  [NHs]  and  amines)  and  acids  (nitrous  [HOMO],  nitric [HNOsL and
pernitric [H02N02] acids).

NH3  is the most  abundant form  of  reduced N  (after molecular nitrogen and
nitrous  oxide)  in  the  troposphere,  but,  it  is   one   of  the  most  poorly
understood of the trace  atmospheric  gases.    It is the only common gaseous
base  and plays  a key  role  in  neutralizing  acidic gases,  particles, and
droplets.

The  principal loss mechanism  for NH3  is probably heterogeneous  (Seinfeld  et
al .  1981).    Recent  model  calculations were  made   to  fit a  set of ambient
measurements  when  the heterogeneous lifetime  of NH3 was set at 10  days and
its  homogeneous lifetime was  set at  40   days (Levine   et  al .  1980).   The
homogeneous loss mechanism  should be  dominated by  reaction with  OH, but the
fate of  the   product  of  this  reaction, NH2,  is  unknown.   The NH3  reaction
rate   with  gaseous   acids   (HN03,   H2$04)  is  not  well   established but
should be rapid (Seinfeld et al . 1981).

The  most abundant nitrogen oxides  (NOX)   1n  the  troposphere (excluding the
relatively  unreactive  nitrous   oxide  [N20])   are   nitric   oxide  (NO) and
nitrogen  dioxide  (N02).     Chemistry   that  is   rather  complex   and not
completely understood interconverts  these  compounds (which are  also primary
emissions)  to  N03,  N205,   HONO,  HN03,   and  H02N02  (Table  4-8).     NO
is' converted  to   N02 and  HONO  through  reactions  with  02,  03,  HO, and
H20.    Nitric  oxide, as  such,  does  not  contribute  to   the  acidity   of
precipitation.

Nitrous  acid  (HONO)   has  been measured in  urban  areas  at  concentrations  as
high as  1 ppb  (Perner  and  Platt 1979).   Concentrations this  high are not


                                    4-11

-------
TABLE 4-7.  PROPOSED MECHANISMS FOR THE FATE OF HOS02
Reaction ~ A
Mechanism of Calvert et
HO + S02 + (+M) + HOS02 (+M)
HOS02+ 02 + HOS0200
HOS0200 + NO * HOS020 + N02
HOS0200 + N02 J HOS02OON02
HOS02OON02 + HOS020 + N03
HOS0200 + N02 -> HOS020 + N03
HOS0200 + H02 -> HOS0202H + 02
2HOS0200 -> 2HOS020 + 02
HOS020 + NO + HOS02ONO
HOS02ONO + hv -> HOS020 + NO
HOS020 + N02 -> HOS02ON02
HOS020 + H02 -> HOS02OH + 02
HOS000 + C, HQ + HOSO.OH + iso-C^H-,
COO L. 0 /
HOS000 + C0HC -»• HOSO.OCH0CHCH.
c ob e. e. o
H2S04+ aerosol (H20, NH3, CH20, CnH2n ..) +
HOS02ONO + aerosol (HgO) -> aerosol (H2$04,
HOSO,ONO + aerosol (H90) •* aerosol (H9SO/,,
H, kcal mole-1
al. (1978)
-37
-16
-25
?
?
- 2
-43
-22
-26
-22
-57
-10
?
(growing aerosol)
HON02...)
HONO ...)
Reaction
number

[4-31]
[4-32]
[4-33]
[4-34]
[4-35]
[4-36]
[4-37]
[4-38]
[4-39]
[4-40]
[4-41]
[4-42]
[4-43]
[4-44]
[4-45]
[4-46]
[4^47]
                          4-12

-------
                         TABLE 4-7.  CONTINUED
                                                                Reaction

           Reaction                      -  AH, kcal mole-1       number
            Alternative mechanisms of Davis and Klauber (1975)




HOS020 + 02(+M) +  HOS0203(+M)                                   [4-48]



HOS0203 + NO + HOS0202 + N02                                     [4-49]



      2 + NO + HOS020 + N02                                      [4-50]
              Mechanisms of Davis et al. (1979) for HOS02
HOS02 + 02 + M + HOS04 + M



HOS04 + H20 -> HS05-H20



HS05-H20 t HS05-(H20)2



HS05-(H20)X t HS05-(H20)X+1



HSO,-(H00)V + NO * HSO.(H00)VN09
   b  d.  X           <\  i.  X  c.


HS05(H20)X + S02 -> HS04(H20)XS03



HS05(H20)x + H02 -^  H2S05(H20)x
                                   4-13

-------
                               TABLE 4-8.  REACTIONS OF NITROGEN COMPOUNDS
        Reaction
      Rate constant
 k (cm3 molecule-1 s-1)
    Reference
Reaction
 number
NH3 + HO •* NH2 +
NO + N02 + H20 t 2HONO
2NO + 02 •+ 2N02
HO + NO + M •+ M + HONO

NO + 03 + N02 + 02
N02 + 03 + N03 + 02
HONO + hv •+ HO + NO

HO + HNOa -v H20 +
N02 + NOa + N205
N03 + NO   2N02
N20s + N02 + N03
N02 + hv + NO + 0
2.3 x 10-12 exp (-800/T)
    k = 1.56 atm-1
3.3 x 1039 exp {53Q/T)a
       1 x 10-11
       3 x 10-H
      1.8 x 10-14
      3.2 x 10-17

      8.5 x 10-14
      1.3 x 10-13
      8.2 x 10-14
       5 x 10-12
       2 x 10-11
       0.2 s-1
Hampson and Garvin (1977)
Hampson and Garvin (1977)
Hampson and Garvin (1977)
Baulch et al. (1980)
Demore et al. (1981)
Baulch et al. (1980)
Baulch et al. (1980)
Baulch et al. (1980)
Baulch et al. (1980)
Demore et al. (1981)

Baulch et al. (1980)
Baulch et al. (1980)
Baulch et al. (1980)
Baulch et al. (1980)
  [4-59]
  [4-60]
  [4-61]
  [4-62]
  [4-63]
  [4-64]
  [4-65]
  [4-66]

  [4-67]
  [4-68]
  [4-69]
  [4-70]

-------
                                               TABLE 4-8.   CONTINUED
              Reaction
                                           Rate constant
                                      k (cm3 molecule-1 s-1)
                                Reference
                                                            Reaction
                                                             number
en
    + hv •*• N02 + 0
N03 + hv -> NO + 02
HO + N02 + M -> HN03 + M

H02 + N02 + H02N02
H02N02 + H02 + N02
        + N02 + CH3C002N02
CH3C002N02 + C
N205 H20 -> 2HN03
                             N02
   1.6 x 10-11
   2.4 x 10-11
   5.0 x 10-12
0.09 s-1 at 298 K
   1.4 x 10-12
    ,-14
                                Baulch et al. (1980)
                                Baulch et al. (1980)
                                Baulch et al. (1980)
                                Demore et al. (1981)
                                Baulch et al. (1980)
                                Baulch et al. (1980)
                                Cox and Roffey (1977)
7.94 x ID'1* exp (-25000/RT)    Cox and Roffey (1977)
       < 1 x 10-20              Hampson and Garvin (1977)
[4-71]
[4-72]
[4-73]
[4-74]
[4-75]
[4-76]

[4-77]
[4-78]
          molecule
                  "2  "l

-------
readily explained from the known homogeneous reactions that produce MONO and
the photolysis rates that destroy  it.   Additional  homogeneous sources might
exist, and  the  heterogeneous promotions of  the  reaction  of NO +  N02 + H?p
 * 2HONO  are  possibilities.    MONO  is  a  relatively  weak acid(pKa 5.22}
and has its greatest tropospheric  significance  as  a photolytic source of OH
radicals.

N02 has a gas-phase removal  mechanism  dominated  by  reaction with  OH  to form
HN03.   With  an  OH  concentration  of  4  x 10~5  ppb,  NOg   would  have  a
lifetime  of ~  17 hr.   N02  also  reacts with ozone  to form  N03,  which can
photolyze  to give back  N02.
HN03  like  H2$04,  is  a  major  acidic  compound  in  the  troposphere.    It is
likely  removed  from the  atmosphere  by  both heterogeneous  and homogeneous
routes.  The  gas-phase removal  mechanism is  relatively  slow, because it is
dominated  by  reaction  with  OH  to  form  N03.   The  lifetime  of  HNOs   with
respect to the OH reaction,  HO  ~ 4  x  10~5  ppb, is  2  to 3  x  103 hr.

N03 is  a  strong oxidizer in the atmosphere  and  may be  removed by oxidation
of NO to N02, reactions with organic compounds  (Bandow et  al . 1980)  such as
terpenes (Noxon  et al . 1980, Platt et al .  1980), and by  photolysis (Graham
and Johnston  1978).   The  oxidation  of  S02 by  N03 is  not  considered an
important  reaction  (Calvert  et al .  1978).    NOs  also exists in equilibrium
with  N205  which may be  removed by heterogeneous  or homogeneous  hydrolysis
to  HN03-   Because  N03  readily  photolyzes in daylight,  peak concentrations
are expected in the evening hours, and  levels as  high as 0.35 ppb have  been
reported  in  the  Los  Angeles  area,  with  calculated equilibrium  values of
N205  as high as  11 ppb  (Platt et  al .  1980).    Similar  values  have   been
reported for a more remote Colorado mountain  site  (Noxon  et al .  1980).

The  chemistry  of  NO,  N02,  N03,  N205,  OH,   and  03   involves  a  close
interrelationship that should have a profound significance to the  acidity of
precipitation, especially in remote areas where  HN03  may dominate the pH of
acidic  precipitation (Seinfeld et  al .  1981).   Further studies  are warranted
involving  field  measurements   of  N03  and   N205  and   kinetic  studies of
their reactions.

The organic nitrate esters should not hydrolyze  under ordinary conditions and
thus  should  not contribute  to  the acidity  of  precipitation.   Peroxyacetyl
nitrate  (PAN),  found  in  urban  smog,  hydrolyzes  to  give  nitrate  in basic
solutions  (as would the other organic  nitrates),  but  its behavior  in  neutral
or slightly acidic solution is  unknown.

The dominant gas-phase  loss  mechanism  for PAN is thermal  decomposition,  k ~
7.94   x  1014   exp  (-25000/RT)  (Cox   and   Roffey  1977).      Its   thermal
decomposition  rate is  considerably  slower  than  that  for  pernitric acid,
H02N02,   k  -   1.4   x   1014   exp   (-20700/RT)    (Graham   et  al .  1977).
Pernitric  acid  has a thermal decomposition  lifetime  of  only 12  s at 298  K
(Graham  et al .  1977).   Both PAN and  H02N02  are essentially in  equilibrium
with  their decomposition  products,  and  although  an assessment  to acidic
deposition cannot  be made  at this time, any  of these species  is  potentially
important.


                                    4-16

-------
4.2.1.4   Halogens--Table  4-9  lists some  halogenated  compounds found in  the
troposphere"!  The compounds characterized as predominantly natural  emissions
are thought to be oceanic  in origin (Seinfeld et al .  1981).
Methyl chloride  (CHsCl)  and  methyl bromide  (CH3Br)  have  tropospheric  life-
times probably dominated by aqueous-phase processes  that produce  and   consume
hydrochloric acid  (HC1).   HC1 is  also  produced by gas-phase  reactions  fol-
lowing the  reaction  of  OH with a  halocarbon as  the  rate-limiting step.   It
has  been suggested  that rainwater  acidity in  remote  areas  is  controlled
principally by  the  presence of HC1  and  HNOa (Seinfeld  et  al . 1981).   More
data  are  needed to determine  the  relative  importance of these  reactions  in
the production of HC1 and their effect on acidic deposition.

4.2.1.5  Organic Ac ids-- Organic acids are expected to  occur  as  photooxidation
products of both natural and anthropogenic hydrocarbons.  In  general,  organic
acids  are  only weakly  dissociated in solution  (their  ionization constants
tend  to decrease with increasing chain length),  but the two  simplest  acids--
(HCOOH)  and acetic  (CHaCOOH)— have  appreciable ionization  constants  (pKa
~ 3.75 and 4.75, respectively).

The sources and sinks  for these  acids are not known at  this time.  HCOOH  is
expected as a  product of formaldehyde  (HCOH)  oxidation.  Su  et al .  (1979)
have  suggested  a  mechanism based  on  reaction  of HCOH with HO?  radicals  and
HCOOH  formation in  the  ozone-ethene reaction (Su et  al . 1980).    Similarly,
CH3COOH  is  formed  in  the cis-2-butene-ozone-H20 reaction  from   the  Criegee
intermediate (Calvert et al . 1978),

          CHaCHOO + H20 ^ CHaCOOH  + H20.                                [4-79]

The loss mechanisms for these acids are not known but  should  be a combination
of  reaction with  OH, wet and  dry  deposition,  and rainout.   Recent measure-
ments  (Dawson  et al .  1980)  indicate that  both acids  are present  in  the
troposphere at significant levels  (Table  4-10).

These acids can be assessed through further tropospheric measurements  (remote
and urban) and  rate data  for  their reactions with OH.   Thus far,  it  appears
they  should not be neglected as compounds affecting  acidity  of  rain in remote
areas.  These  and other organic acids will  contribute  to titratable H+.

4.2.2  Laboratory Simulations of  Sulfur Dioxide and  Nitrogen  Dioxide
       Oxidation

In addition to the aforementioned  work on the fundamental gas- phase reactions
germain to atmospheric acidity, a  number  of laboratory studies  have attempted
to  simulate atmospheric  conditions  in  controlled  experiments  and   thereby
obtain insight  into  the combined  effects of  simultaneous reactions.    These
experiments were  usually  conducted  in  "smog  chambers"  with  artificial  or
natural solar  radiation.

Numerous smog  chamber studies have described the evolution of sulfate  aerosol
from  S02  oxidation,  in  terms  of growth  and size distribution trends  (e.g.,
                                    4-17

-------
                TABLE 4-9.  ATMOSPHERIC HALOGEN COMPOUNDSa
Compound
(Natural)
CH3C1
CH3Br
CH3I
HC1
(Anthropogenic)
CHC13
C2C14
CHC12F
CH3CCl3b
Concentration
(ppb)

0.81
0.01
0.01
0.20

0.02
0.03
0.01
0.1
Lifetime*
(s x 107)

3.8
4.1
—
—

1.9
10.1
6.6
13.9
aFrom Seinfeld  et  al.  (1981),  assuming  an HO  concentration  of 3.7  x 10~5
 ppb.

bSource not clear.
                                    4-18

-------
TABLE 4-10.  TROPOSPHERIC HCOOH AND CH3COOH
             (DAWSON ET AL. 1980)
Acid pKa
HCOOH 3.75
CH3COOH 4.75
aAssuming removal by HO ~
x 10'12 cm3 molecule"1 s"1
Remote site
(ppb)
2
1
2 x 10-4 ppb
and k(HO + CH^C
Urban site
3.5
6.0
and assuming k(HO +
:OOH) - 10-12 Cm3 moiec
Lifetime3
(hr)
8
48
HCOOH) ~ 6
ule-1 s-1.
                   4-19

-------
 Kocmond and  Yang  1976, Friedlander  1978,  Whitby  1978,  McMurry  and  Wilson
 1982).   In  general,  sulfate  condenses to  form  particles with  a  relatively
 sharp  peak  in  mass distribution at  particle  diameters between 0.1  and 0.2
 ym.    Because other  S02  conversion  processes  (aqueous  and  heterogeneous)
 result in  particles of larger mean diameters, sulfate  particles < 0.2 ym in
 diameter are thought  to be characteristic of gas-phase S02 oxidation.

 Gas-phase  oxidation of  S02  to  sulfate  particles  has  been  detected  in  the
 absence of  sunlight when  olefins  and 03 reacted  (Groblicki  and Nebel 1971,
 Cox  and Penkett 1972,  McNeil's  1974).   As indicated earlier,  the significance
 of this oxidation  path has been assessed by  computer simulations  of the SOg
 reaction with the Criegee intermediate (Calvert et al. 1978).  This mechanism
 should  be  significant  only in highly polluted air.

 Smog chamber studies  also  have been  conducted  to  investigate  the  relative
 importance   of  S02   oxidation  via  the  free   radicals  OH,  H02,  and  CH302
 (Kuhlman  et al. 1978,  Graham  et al.  1979,  Miller 1980).   The experimental
 results,  aided  by  computer  simulations  of the  experiments,  indicated  that
 S02  is  oxidized predominantly by OH under urban-air conditions.

 Chemical  kinetics  and smog chamber results  indicate that the  OH  radical  is
 responsible  for the  majority  of  the  H2S04  and  HNOa  formed via  gas-phase
 reactions  in  the   atmosphere.   OH  concentrations  in  the  troposphere  are
 related to  a complex  and tightly coupled series of  reactions  involving  NOX,
 hydrocarbons (HC) ,   and 03.    Smog  chamber  experiments  have  been  used  to
 investigate,  on a  macroscopic  level, how  the HC-NOX-03  cycle affects  the
 OH population and the  formation of H2S04  and HN03.

 A  series of smog   experiments  focused  on S02 oxidation  indicated  that  the
 maximum   rate  of  S02  conversion   to   H2S04   depends   strongly  on   the
 HC/NOX  ratio,  increasing  with  higher   ratios   (Miller   1978).    Parallel
 reductions  in HC  and NO concentrations  in  these  experiments did  not  reduce
 the  average S02 conversion  rate.   Computer  modeling of these  experimental
 conditions  indicated  that  OH  was  primarily responsible  for S02  oxidation,
 and  the effects of HC and NOX concentrations on  the relative levels of  OH
 were  qualitatively  consistent  with   the  observed  trends in  S02  oxidation
 rates.   This  study  indicated  that  during  a diurnal  period the  gas-phase
 conversion  of  S02  to  sulfate   would  likely  be  10  to  20  percent  of  the
 initial S02  concentration for most urban  HC-NOX precursor  conditions.

 Outdoor chamber experiments using ambient air  in  St. Louis, MO,  supported the
 contention  that variations  in   OH  concentrations,  and thus  S02  oxidation
 rates,  are  more   strongly  affected   by  HC/NOX   ratios   than  by  absolute
 HC-NOX  concentrations  (Miller  1978).    Unfortunately,   neither   of   these
 studies  indicated   a  critical  concentration  region  for  HC-NOX below  which
 S02 oxidation might drop to rates typical  of the  background troposphere.

 Laboratory  simulations  aimed at  unraveling  the terminating reactions  of  NOX
 in the  atmosphere are  limited.   An early  breakthrough was  the identification
 of PAN  as  an important  product  of NOX  reactions  in irradiated  atmospheres
 (Stephens  et al.  1956).   The  development of  new  but imperfect methods  for
monitoring  HN03  (Miller and  Spicer  1975,  Joseph  and  Spicer 1978, Huebert


                                   4-20

-------
and  Lazrus 1979)  and  participate  nitrate  (Appel et  al .  1980) has  finally
enabled some assessments of the fate of NOx in the atmosphere.

Smog  chamber  experiments  with  HC  mixtures  representing  rural  and  urban
conditions  revealed that  the  conversion rate of N02  to products  depended
strongly  on  the HC/NOX  ratio,  increasing with  increasing  ratio  (Spicer  et
al.  1981b).   Here,  too,  the  HC/NOX ratio effect is most likely  the  result
of governing  the concentration  of hydroxyl   radicals.   The product  ratio  of
PAN  to  HN03  was   nearly  proportional  to   the  HC/NOX  ratio  and  the  more
reactive  "urban" HC's yielded  higher PAN/HNOa  ratios than  did  "rural"  HC
mixture.   Negligible amounts of  particulate nitrate were observed  in  these
experiments, and,  if certain assumptions  regarding wall  losses  are accepted,
reasonably good material  balances for NOX were obtained.

Regarding  absolute values for  conversion rates  for S02 and N02  to  acidic
products  it  should be noted that indoor smog chamber  experiments  generally
are  conducted  with a constant  radiation  flux, whereas true  solar  radiation
has  temporal  and spatial  variations in spectral   distribution and  intensity.
Winer  et  al .   (1979)  demonstrated  radiation effects  during  smog  chamber
simulations.   With this  caveat  in  mind, one  can discuss the  pseudo-first-
order  rates  for S02  and NOX conversion  to  acids,  as  presented in the  two
smog chamber studies with  similar HC components   (Miller 1978,  Spicer  et  al .
1981b).   For  HC/NOX ratios  near 5/1,  the  average  pseudo-first-prder  rate
for  S02  oxidation  was  ~  0.012  hr-1 ,  so  an  average  S02  lifetime  toward
        formation  would   be  83   hours.     For   similar  conditions,   the
pseudo-first-order  rate  for  N02  oxidation  to  HMOs   (given  PAN/HN03  ~
1/3)  was  ~  0.09  hr-1.   Thus,  a  lifetime  for N02  is estimated  to be  11
hours with respect to HN03 formation.

There  are important  transport  implications associated  with these  results.
S02,  having   an  average  lifetime  for  oxidation  of  3   to  4 days,  will  be
transported  over greater distances  than N02  and  would  be  expected to  be
removed  from  the atmosphere  by  dry deposition processes to  a greater extent
than  N02.   Likewise, the  sulfate produced  from S02  oxidation,  being  in
the  aerosol  phase, would be expected  to have a longer  atmospheric  lifetime
and  transport time  than  the  acidic  vapors  produced   from  NOg  oxidation.
Therefore,  both  the  precursors  and  acid  products  of  gas-phase  sulfur
transformations  will  have  substantially  greater  potential  for  long-range
transport than the precursors and products of nitrogen transformation.

4.2.3  Field Studies Of Gas-Phase Reactions

4.2.3.1   Urban  Plumes — Studies  of acid  formation  from gas-phase  reactions
under  actual  atmospheric  conditions  are  confounded by  many  difficulties.
Proper   assessments   of   expanding  mixing   volumes,   deposition   losses,
entrainment of  fresh  pollutants, and long  averaging  periods for  analytical
purposes  are  only some  of  the  problems.   In  addition, few  ambient studies
have attempted to  measure in detail the attendant pollutants and  conditions
(e.g.,   hydrocarbons,   aldehydes,  NOX,   03,   and   ultraviolet   radiation)
generally needed to interpret the data.
                                    4-21

-------
Many observations of  $03  oxidation  within  urban plumes and under  long-range
transport conditions are listed in Table 4-11.   The  cited  oxidation  rates  for
S02 range from 0 to 32 percent hr-1.

When  such  reports  are  examined,  it  is not  always clear  whether the data
pertained exclusively to  the gas-phase  reactions  or included  aqueous-phase
chemistry.   Another  reason  that  may  account,  in  part,  for  the  apparently
divergent rates  of  S02  oxidation  found  in these citations  is  the  tendency
to  compare  rates  derived  by  different  methods;   e.g.,  in  one  case  the
oxidation rates may represent 1-hour maxima,  while  in another  case,  the  rates
may represent averages taken over  periods of  a  day  or more.

As  might  be  expected,  the highest S02 oxidation  rates  have been  reported
for the more  highly polluted  atmospheres  associated  with  urban areas.    For
example (Table  4-11),  gas-phase $03 oxidation  rates as large as 32  percent
hr~l  have  been  inferred  for  St.  Louis,  MO,  13  percent  hr-1   for  Los
Angeles,  CA,   and  9  percent  hr-1   f0r Milwaukee,  VII.   In  contrast,  the
"average"   oxidation  rates  reported  for  distant  transport  situations  are
generally in the range of 0.5  to 2 percent hr-1.

The several studies conducted in  and around St. Louis,  MO,  offer  interesting
comparisons.   The largest SOg  oxidation  rates  reported  by  Breeding et  al.
(1976)  were  measured near  noon and on a  day  having the largest  nonmethane
hydrocarbon  concentration  for  their   study  period.    Two  Lagrangian-type
studies conducted  by  Alkezweeny   and   Powell  (1977) and  Alkezweeny (1978)
yielded fairly  consistent oxidation rates  in  the range of  10 to 12  percent
hr-1.   Measurements  taken  aboard  a manned  balloon (Forrest  et al.  1979)
resulted  in  upper-limit  estimates  of  4  percent  hr-1  for  S02   conversion
under stagnant  urban conditions.  The experiments of White  et al.  (1976)  led
to  similar  estimates of  S02  oxidation  rates for the  St.  Louis plume.
Numerical   simulations  of White's  data  by  Isaksen   et  al.  (1978)  indicated
S02  oxidation  rates  of about  5  percent  hr-1   and  a  diurnally  integrated
conversion of about 25 percent.

Perhaps the most puzzling aspect of  the data  regarding urban plumes is  the
widely  divergent  S02  oxidation rates observed  within single  studies;  e.g.,
a  range  of  1.2  to  13  percent  hr-1   for  LOS  Angeles,  CA  (Roberts  and
Friedlander  1975),  and  1  to  9   percent  for  Milwaukee,  WI   (Miller  and
Alkezweeny 1980).   In  the latter  study, such extreme rates were  observed on
two consecutive  days  of  nearly identical  relative  humidity and temperature.
The higher  rate occurred  when polluted  air moved through Milwaukee  from  the
southwest.   On  the  following  day, when  the  S02   oxidation  rate was  <  1
percent hr-1,  relatively clean "background"  air  passed  through  Milwaukee.
In  both  cases,  comparable  levels  of   fresh   pollutants  emitted  from  the
Milwaukee complex were entrained  in the  downwind   plume,  yet  the  previous
history  of  the  air  masses   seemed  to  govern  the  S02  oxidation rates.
Detailed kinetic modeling of the two cases was conducted,  taking into account
differences  in reactive  hydrocarbons,  NOX,  and  03.   The  associated  free-
radical chemistry  could  not  account  for the  observed  differences  in  S02
oxidation rates.  Thus,  the  agreement often claimed  between kinetic  modeling
results  and   data  observed   for   polluted  atmospheres  may  sometimes   be
fortuitous, and a  comprehensive  body  of  data  should be scrutinized before


                                    4-22

-------
     TABLE 4-11.  S02 OXIDATION RATES (% hr-1)  FROM STUDIES OF URBAN
                     PLUMES AND LONG RANGE TRANSPORT
Range
Average   Location/periods
                          References
  6-25
1.2-13
  0-4
  1-9
 16.6
 7.1
1.1
0.3-1.7
5.3-32
5
31
10-14
8-11.5
0.6-4
1.1
0.7
16
5
31
12
9.8
1.7
 2
 4
Rouen, France/W/D
Los Angeles, CA/S
 & F/D
British Isles/W/L
Western  Europe/S
 & W/L
St. Louis, MO/F/D
St. Louis, MO/S/D
Budapest,
 Hungary/S/D
St. Louis, MO/S/D
St. Louis, MO/S/D
Arnhem-Amsterdam,
 Nether!ands/S &
 W/D & N
St. Louis, MO/S/D
Milwaukee, WI/S/D
Benarie et al. (1972)b
Roberts  and Friedlander  (1975)b

Prahm et al. (1976)c
Eliassen and Saltbones (1975)

Breeding et al. (1976)d
White et al. (1976)e
Meszaros et al. (1977)c

Alkezweeny and Powell (1977)
Alkezweeny (1978)
El shout et al.  (1978)

Forrest et al.  (1979)
Miller and Alkezweeny (1980)
aSeason: W = winter;  S = summer;  F = spring or fall.   Time of day:
 D = daytime; N = nighttime;  L =  long term (>  24 hr)  averaging periods.
bHigher rates possibly related to aqueous-phase reactions.
ccalculated from their half-life  data.
^Calculated from their data by Alkezweeny and  Powell  (1977).
eBased on kinetic analysis of data by Isaksen  et al.  (1978).
                                    4-23

-------
existing  knowledge  of  gas-phase  chemistry  is  applied  to  predict  S02
oxidation in urban areas.

Information  on  the  gas-phase  transformations of NOX  to ac1d  products in
urban  plumes  is  scarce.   Spicer   (1980)   estimated  NOX   transformation/
removal  rates for the Phoenix, AZ,  urban plume  to  be less  than  5 percent
hr-1.    The  low  rates  were  attributed  at   least  in  part  to  the thermal
deposition  of  PAN-type  compounds  at the  high  ambient temperatures  of the
desert area.   Spicer (1977a)  reported  rates of  NOx conversion to products
of  about 10 percent  hr-1  for  Los Angeles,  CA,   if  certain  assumptions for
material  balances were granted.  In more recent measurements,  downwind  of Los
Angeles  (Spicer  et al.  1979),  typical  conversion rates of 5  to  10 percent
hr~l  were  observed.    Measurements  by  Spicer  et  al. (1981a)  resulted in
pseudo-first-order  rates  for  NOx removal  ranging  from   14  to  24 percent
hr-1  for  the  Boston,   MA,  plume.     The  average   lifetime  for  NOX  was
estimated to be  5.9  hr.   In the Boston  study, the  ratio  of PAN to HN03 was
1.8  and   the conversion  of NOX  to   particulate  N03-  was  < 1  percent of
the  total  product.   Given an  average  PAN/HN03  ratio of  1.8,  the pseudo-
first-order  rate  for N02   conversion  to acid would have  been  6.3 percent
hr-1,  and  the   NOX   lifetime  with  respect  to  HMOs  production   would be
about  16 hrs.   These  values  are similar  to estimates  given  earlier  with
respect  to global OH concentrations.

Somewhat different findings were recently reported by  Hanst et  al.  (1982) in
an  investigation  of Los   Angeles  smog  by   long-path  infra-red  absorption
spectroscopy.   Hanst et al. concluded  that  most of  the  N02  was .removed by
reaction   with   63   and    subsequent   reactions   of  ^05   and   N03   into
condensed products (particulate nitrates) not amenable  to  detection in  their
cell.

This  interpretation  conflicts with  the conclusion  reached  by the Battelle
researchers  (Spicer  et  al.  1981a)  which asserts  that 95  percent of the NOx
losses  in urban  plumes  can be accounted  for as gaseous  HNO^  and PAN, and
that  the amounts of  particulate  nitrate produced in  urban  plumes are  very
small.

As  indicated earlier, it  is apparent that more  research is needed concerning
the  fate  of PAN,  N^Os  and  N03 in  the  atmosphere  and  their  potential
contributions as  acidic species.

4.2.3.2   Power  Plant PIumes--The majority  of studies of S02 oxidation  in
the  atmosphere  have been conducted  in  association  with power  plant  plumes.
Compared to studies  of urban  air chemistry, power  plant plumes  offer the
advantages  of  higher pollutant concentrations, definitive  plume  boundaries,
the  presence of  inert tracers, and less severe deposition  losses.

In  general,  the gas-phase chemistry  pertaining to  reactions within  power
plant  plumes is  the  same  as for ambient air.  However, an important  concern
when  plume data  are  interpreted  and kinetics of the gas-phase  reactions  in
plumes are  modeled is adequate treatment of  the  turbulent exchange processes
(Donaldson  and Hilst  1972,  Lamb and Shu 1978, Shu et al. 1978).
                                    4-24

-------
Interpretations of  power-plant  plume data  show  that,  under most conditions
where plumes can be  discerned against  background,  the  rates of formation of
sulfate  and nitrate  are  slower  in  power plant  plumes compared  to  urban
plumes.  The main reasons  for this  are  imperfect mixing and an abundance of
NO  which  effectively  competes  with  S02   and  N02  for  hydroxyl   radicals.
Under  some conditions,  S02  and  N02  transformation  rates  in  power  plant
plumes  can exceed  those  in ambient  air (Miller  and  Alkezweeny  1980), and
under such conditions an excess  of 03 in the plume  can  be expected.

Selected  studies  of  power plant  plumes  are  listed  in Table  4-12.   The
selection  is  restricted  to  studies   where   gas-phase  S02   oxidation was
emphasized and/or NOX reactions  were investigated.

Studies concentrating on  heterogeneous  aspects  of plume reactions  have  been
reviewed  by  Newman  (1981)  and  are not  discussed  here.  As  is  the case in
studying urban plumes, one cannot  always distinguish  gas-phase  reactions from
other conversion mechanisms.

The  experiments  cited  in  Table  4-12   were  conducted  with  widely  varied
analytical  procedures,  transport  times,  ambient  pollutants,  meteorological
conditions, and emission  rates, all  of which greatly  influence the  results.
Considering all these factors in an interpretation of  the data is beyond the
scope  of  this  document.    In  general,  S02   transformation   rates   were
estimated   by   measuring   either  the   increase   in   submicron   particle
concentrations  (inferred   as   ^$04  mass)  or   the   actual   increase  in
filtered  sulfate  mass relative  to total  S concentration,  or to  an  inert
tracer,  such   as  sulfur  hexafluoride  (SFs).   In  the  few  cases  where NOX
transformations were  measured,  rates  of  NOX  loss  or  NOa"  formation were
based on total  S as  the conservative tracer of  plume  dilution.

Pueschel  and   Van  Valin  (1978)   measured   the  formation  of  new  particles
downwind  of the  Four  Corners,  NM,  plant  and estimated  a  flux  of  10lb
particles  s-1  of  H2$04   that  could   act  as  cloud  condensation   nuclei
(CCN) in the atmosphere.   Comparison of  the source strengths of CCN  from the
power  plant relative  to  those  for  natural  CCN  in  the  area  led  to the
assertion  that  the  photochemically  derived  CCN  from  power plants could  have
major effects on cloud structure and precipitation  processes  in the  West.

At about  the  same time,  experiments in  Canada (Lusis  et  al. 1978)  indicated
that,  under relatively  dry conditions,  S02 oxidation  was related  primarily
to  photochemical   reactions.    In  accord  with  photochemical   mechanisms,
oxidation  rates  were low  in  February   (<  0.5  percent  hr-1)  and relatively
high  in June  (1  to  3 percent  hr-1).    Increased  rates of oxidation  were
apparent at the leading edges of plumes.

Similar  "edge  effects"  were observed in early studies  of  the Labadie, MO,
plume  (Cantrell and Whitby  1978, Wilson  1978).   Another important feature of
the  Labadie  experiments  (Gillani  et  al. 1978,  Husar  et al.  1978)  was the
apparent diurnal variation  in the S02  oxidation rate  and the  inference that
solar  radiation  and  extensive  mixing  of  the  plume with ambient  air  were
required  for  substantial  S02 oxidation  rates.   During noon hours, the S02
conversion  rate was found  to be  1  to 4 percent  hr'1  compared to  nighttime


                                    4-25

-------
         TABLE 4-12.  SUMMARY OF POWER PLANT PLUME STUDIES WITH EMPHASIS  ON  GAS-PHASE TRANSFORMATION  RATES
Plant/location
Four Corners, NM
GCOS/Alberta
Labadie/MO
Season
October
Feb. & June
July
Range of SOg
conversion rates
(% hr-1)
2 - 8
0-3
0.41 - 4.9
Range of NOX
conversion rates
(% hr-1)
-
-
_
Reference
Pueschel and
(1978)
Lusls et al .
Cantrell and

Van Valin
(1978)
Whitby
PC
a*
Labadie/MO




Four Corners, NM


Centralia/WA


Leland-Olds/ND



Sherco/MN


Big Brown/TX
                             July
                             June
  0-4
0.9 - 5.4
                          Spring & fall    0.03 - 1.4
                             June
                             June
                             June
  0-0.7     0.2 as particulate
  0-3
0.2 as particulate
                              (1978)

                              Wilson  (1978),  Husar
                              et al.  (1978),  Gillani
                              (1978),  Gillani  et al.
                              (1978)
                              Hobbs et al.  (1978),
                              Hegg and Hobbs (1979a),
                              Hegg and Hobbs (1980)
0.4 - 14.9    0.2 as particulate

-------
TABLE 4-12.  CONTINUED
Plant/location
Colorado River
Basin/CO
TVA Cumberland/TN
Navajo/AZ /
f /
£3 Labadie/MO
Sherco/MN /
Cumberland/TN 1
Navajo/AZ _)
~~*\
Cobb/MI }
Andrus/MS >
Breed/IN \
Season
Summer
August
Summer & winter
July
-
August
Summer
May & Nov
May & Oct
Jun & Nov
Range of S02 Range of NOX
conversion rates conversion rates
(% hr-1) (% hr-1) Reference
1.5
0.1 -
0 -
0.08 -
2.3 -
1.1 -
0.3 -
0.1 -
0.1 -
0 -
Eatough et al . (1981)
4 3-12 Forrest et al . (1981)
0.8 3-10 times RSQ2 Richards et al . (1981)
5.4
14.2 - Whitby et al . (1980)
7.1
2.9
11 23 - 31 as NOX loss
5.9 5 - 21 as NOX loss Easter et al . (1980)
1.5

-------
rates <  0.5 percent  hr"1.    Mesoscale  modeling of  the Labadie experiments
(Gillani 1978,  Gillani  et al.  1978)  was an important attempt to budget the S
in a  dispersing  plume.   It was  concluded that, for the Labadie conditions,
some  20  to 40  percent  of   the  emitted  S02  may  be  converted   to  S04^"
while the remainder is lost by deposition  mechanisms.

Power plant experiments  conducted by  the  University of Washington  (Hobbs et
al.  1978,  Hegg  and  Hobbs 1979b)  employed a  variety  of  particle-measuring
techniques.    S02  oxidation   rates  derived  by  the  various  methods  showed
considerable scatter.   Higher S02  oxidation rates  generally  were found in
the  southwest  United  States,  and rates tended  to  increase with travel  time
and  ultraviolet  (UV)  intensity.     Measurements   of   particulate  NOs" at
three  of  the  plants  (Hegg   and Hobbs  1980)  showed   minimal  NOs-  in   the
condensed  phase  (generally   <  2  yg  m"3)  and  a  maximum  NOX   conversion
rate to particulate nitrate of 0.2  percent hr"1.

The employment of  different analytical  methods  by  Eatough et  al.  (1981)  has
led to interesting differences between the chemical composition of secondary
S042" particles,  depending  on regions  of the  United  States.   In  the East,
where  S02  conversion   rates  are  generally   high,   secondary    $04^"  is
predominantly  H2S04   and  ammonium   sulfate,   (NH4)2S04,  with   nominally
10 percent  as  an organic-S(IV)  compound.   In the  West,  25 to 75   percent of
secondary  S may  be organic-S(IV).   Furthermore,  in arid  western  states the
principal S042" salts formed  in plumes were metal  salts such as gypsum.

Reports  from the measurements of the Cumberland,  TN,  plume (Forrest et al.
1981)  are  similar  to   findings  from  the  Labadie  plume.    Nighttime   S02
conversion  rates  ranged  from  0.1 to  0.8  percent  hr"1,  while daytime rates
ranged from 1  to 4 percent hr"1.  Important  new information was obtained on
NOX  transformations.    Total   NOs"   formation   (gaseous   and   particulate
NOs")  rates were  0.1  to 3  percent  hr"1  at  night  and  3   to  12  percent
hr"1 during the  day.   The authors  point  out that the rate of plume  mixing
with  ambient  air might  have  been a  limiting  factor  for  N02 conversion to
S02  and  NOX  rates  of  conversion   reported   for   the  Navajo  Generating
Station in Arizona (Richards et al. 1981) were much lower than those reported
from  the  Cumberland  plant.    The  maximum  rate  for  S02  conversion in  the
summer  was  0.8 percent hr"1  and 0.2  percent hr"1  in the winter.   Rates of
gaseous nitrate  formation  (HNOs)  were  generally 3 to  10 times  larger  than
for S042" formation.

Experiments  conducted in Michigan, Indiana,  and Mississippi, where  SFs  was
used  to  trace   plume   dispersion,   resulted   in  generally  moderate   S02
conversion  rates,  0  to  3  percent hr"1,  with occasional  exceptions (Easter
et al.  1980).   S02  transformation  rates exhibited correlation  with ambient
HC reactivities  and  concentrations,  although for many cases  this could  also
be  interpreted  as   seasonal  variation  related  to   solar  intensity,  plume
dispersion,  or temperature.   For  example,  S02   oxidation rates  at Cobb,  MI,
were 2  to 11 percent hr"1 in  May  and 0.1  to 0.3 percent hr"1  in November.
Rates at  Breed,  IN,  were 0 to 1.5 percent  hr"1  in June  and 0 to 0.1 percent
                                    4-28

-------
hr-1  in  November.   At Andrus, MI,  the  rates were 0.5  to 4.9 percent  hr"1
in May and 0.1 to 3.7 percent in October.

Measurements   of   NOX  transformation  rates   in  the   above  study   were
inconclusive.   Chemical  analyses  indicated that transformations to  HN03  and
participate   N03~  were   minimal,   yet   large   NOX   losses   were   often
calculated  when  NOX was  compared  to SF^  or  total S.   The wide scatter  in
the data suggests analytical  problems.

4.2.4  Summary

Organic acids  generally  are  not regarded  as significant contributors  to  the
acidic deposition problem, mainly because  their ionization  constants  are  weak
relative  to  those  for  most  inorganic  acids.    However,  the  scarcity  of
information  on the  abundance  and fate of organic acids  in the atmosphere
makes it impossible to estimate their importance with  assurance.

Halogenated  compounds  (RX)   are  potentially   important   to   precipitation
chemistry,  but little information  is  available  on  the gas-phase reactions
that might  yield  HX.   Halocarbons of both natural and  anthropogenic  origin
exist  at  low  concentrations  and  react  slowly  or  not  at  all   in   the
troposphere.  Thus, their contribution to  the  production  of acid  compounds  is
potentially significant only  on a  global scale.

Most  of  the  concern  regarding  acidic  deposition  has  focused  on   S and  N
chemistry.   Measurements  of the  rates   of S02  and   N0£  oxidation  in  the
atmosphere  have  been  crude  and  imprecise.     This  relates  to analytical
difficulties,  extensive spatial and  temporal  averaging and, particularly  in
the case  of S02,  a  lack  of  distinction  between gas-phase  and  aqueous-phase
reaction paths.

Rates of  SOg  oxidation measured in  urban areas and  plumes  range  from  near
zero  to  30  percent hr-1.   yne  preponderance  of  data, however, indicates
upper-level  rates  of  12  percent  hr"l    for   midday,  summer  conditions.
Average daytime conversion rates  are in  a  range  of 3 to 5 percent  hr~l  for
summertime  conditions.   Systematic  measurements  of  seasonal  and   diurnal
variations  have not been made; peripheral  data  indicate that nighttime  and
wintertime conversion rates are <  1 percent hr'1.

Like the case  of  sulfuric  acid  formation,  the  rate of nitric acid formation
under various  atmospheric conditions is  not  well documented.   Most of  the
available data are  consistent  with  the  conclusion that  the reaction of N02
with hydroxyl  radicals is the  principal gas-phase route for HN03 formation,
although other reactions are  also  important.   In general,  N02 conversion
rates under  daylight,  summertime  conditions range from  <  5 percent  hr-1  to
24 percent  hr-1,  W1'th  at  least half  of the product yield  being nitric  acid
vapor.

There  is   conflicting  evidence  about   the   role  of  ^05   in   nitrate
formation;   its gas-phase  reaction  with   water  is  slow, but  it hydrolyzes
rapidly on moist  surfaces.  There is also  considerable uncertainty regarding
the fate of peroxyacetyl  nitrate (PAN) in  the atmosphere  and  its potential  to


                                   4-29

-------
contribute to acidic deposition.   Adequate  assessments  of  the  impact of  these
species  to  atmospheric  acidity   cannot  be  made,  and further  studies  are
warranted   involving   field  measurements   of  N03,  N205,   and  PAN   and
kinetic measurements of their hydrolysis  reactions.

Despite  some  conflicting   data  regarding   sulfur   and  nitrogen  oxides
transformations in  power plant  plumes,  a  few tentative conclusions  emerge.
Under  most  conditions,  rates of transformations  to  acidic  products  are
generally slower  in power plant  plumes  than  in  ambient air.   S02 oxidation
rates  under  daylight  conditions  fall  in the  range  of 1  to 6 percent  hr~l,
although some  exceptions exist.    SO?  conversion rates  in plumes from  some
plants in southwestern states are lower  than  in  other  parts of the country;
the basis for this trend is  not apparent.

A  paucity  of  data  exists  regarding  nitric  acid formation  in  power  plant
plumes.   A  few studies  in  which  this  measurement  was  attempted indicated
HN03  formation rate   in a  range  3  to  10   times  greater  than that  for
H2S04  formation.    This  result would  seem likely  if  the  hydroxyl   radical
was the principal  oxidant.

Overall,  field studies  of  $03   and  N02  transformations  in  air  have  not
provided conclusive evidence to  support  predominant reaction  pathways or  to
identify the  most  important  atmospheric variables  affecting  transformation
rates.   Most  of  the  information on  these  processes comes  from chemical
kinetic studies, model simulations and  smog chamber  experimentation.

A  survey of  fundamental  reactions confirms that the  rate  of gas-phase
oxidation  of  S02  is  governed  by  free-radical   concentrations   in   the
atmosphere,  primarily by the OH  radical  and  to a  much  lesser,  but uncertain,
extent  by   CH302  and  H02-    Of  the  reduced forms   of  sulfur  gases,  H2S
is by far the most reactive  in the atmosphere.  Its  reaction with  OH radicals
is  faster  than  is  the  rate  between  S02  and OH  and  the  product  of  the
reaction is  S02»    Other reduced sulfur compounds  such  as COS oxidize  much
more slowly  in the atmosphere, and their reaction  products have not been well
characterized.

A  survey  of the  fundamental reactions  of  nitrogen  oxides in  the atmosphere
indicates that gaseous  HNOs  formation  will be dominated  by the  reaction  of
N02  with  OH  radicals.   The  rate  for  this   reaction  is approximately  ten
times  faster  than the  rate  for  S02 oxidation by OH.    As mentioned above,
other  products of nitrogen oxides reactions  in air  are potentially important
to  acidic  deposition,  particularly  N205  and PAN  and  to a  lesser  extent
N03  and HN02,  and the fate  of these  species   in  the  atmosphere  must  be
better characterized before  assessments  can be made.

Smog chamber studies of  gas-phase transformations revealed  that the rates of
S02  and  N02  oxidation, under  simulated  urban  conditions,  were strongly
dependent on the  ratio of hydrocarbons  (HC)  to nitrogen  oxides  (NOX)«    The
findings were  qualitatively consistent with  kinetic models  that  predicted OH
concentrations to rise with increasing  HC/NOX ratios  but remain  relatively
constant with  proportional  variations  in  HC  and  NOX.   The product ratio of
PAN  to  HNOs  was  also  found   to  be   nearly  proportional  to  the  HC/NOx


                                    4-30

-------
 ratios.  Such relationships, however, have not been investigated under actual
 atmospheric conditions and other atmospheric variables will  undoubtedly muddy
 the water.

 The  number of  free radicals  and  competitive  reaction  paths  that  comprise
 atmospheric chemistry  is quite large  and many of  the reactions are  highly
 coupled.   Calculations  indicate  that  the  free-radical  concentrations  have
 pronounced  diurnal  and   seasonal   variations.     Unfortunately,   real-time
 measurements of free radicals have not been very successful, and knowledge of
 the  factors  influencing   the  concentrations  of  free radicals  is  largely
 theoretical.   In  polluted air, the  concentration of OH is considered  to be
 strongly  related  to  the  concentrations  of  hydrocarbons,   aldehydes,  carbon
 monoxide and  nitrogen  oxides,  whereas,  in relatively  clean  "background"  air,
 the OH  concentration  is dominated by  levels of  carbon  monoxide,  ozone  and
 water vapor.  In both cases, the characteristics of incident sunlight play an
 important role.  The  effect  of trace amounts of  anthropogenic  pollutants on
 "background"  OH  concentrations is  unknown  and  unlikely  to  be resolved  by
 computer modeling.

 If,  as  in  the  case  of  S02 and  N02>   oxidation  is  largely  limited by  the
 availability of free  radicals  such  as  OH, an assessment of the relationship
 between  precursor   concentrations  and   acid   formation  rates  requires  full
 knowledge  of  the  factors  governing  the  oxidizing  species.   While  there  is
 ample reason to expect  the relationships  to be nonlinear,  kinetic models  of
 the processes should  somehow be tested.  Such  applications,  when considered
 in  the  context of atmospheric  transport  and  other atmospheric   phenomena
 present many difficulties, as discussed in a later section of  this chapter.

4.3  SOLUTION REACTIONS (D. A.  Hegg  and P. V. Hobbs)

4.3.1  Introduction

The importance of  chemical  reactions within  cloud drops and  rain (hereafter
called hydrometeors) to the  formation  of  strong acids has been  suggested  on
 both  theoretical  (Scott  and  Hobbs  1967,  Barrie et al.   1974,  Larson  and
Harrison 1977)  and experimental (Junge  and  Ryan  1958,  Van  den Heuval  and
Mason 1963,  Penkett et al.  1979)  grounds.   Postulating  such  reactions  has
been necessary to explain  the  observed acidity of precipitation (Petrenchuk
and Selezneva  1970, Hobbs  1979,  Newman  1979,  McNaughton  and  Scott  1980).
Recent  studies  have  even  suggested  that  solution  reactions  may  play a
rate-limiting role  in S02  absorption  by  raindrops   (Baboolal  et  al.  1981,
Walcek et  al.  1981).   Most  of these studies have  dealt  exclusively with S
 species.   Even  in this  case,  considerable uncertainty  exists concerning
reactions  that  convert  the  precursor species,  aqueous  S02,   into  H2S04.
Moreover, a considerable body  of data  suggests  that  N and  Cl compounds  also
contribute significantly  to  precipitation acidity  (Gorham  1958, Petrenchuk
and Drozdova 1966,  Marsh 1978,  Hendry and Brezonik 1980, Galloway and Likens
1981).

Contributions to the acidity of rain by  various  aqueous  reactions   that  can
produce  HC1,  HN03,  and  H2S04  in  hydrometeors  are  evaluated   in   this
section.    During  this  evaluation,  the relative  importance  of direct  acid


                                   4-31

-------
vapor  absorption  reactions  and  acid-precursor oxidation  reactions is  con-
sidered.   In  addition,  the  importance  of  neutralization  in acidic  hydro-
meteors  is  assessed.    Whenever  possible,  detailed  discussion  of  kinetic
mechanisms is avoided and experimental rate expressions are employed.

The various  steps in the production of acidic precipitation, especially those
discussed in this chapter, are indicated schematically in Figure  4-2.

4.3.2  Absorption of Acid

The most direct means of  producing acidity  in  hydrometeors  is through  direct
absorption  of acid  vapors  and   the  collection  of  acidic  aerosol,  either
through nucleation capture  in clouds or  scavenging by hydrometeors.   While
both of these mechanisms  are  discussed  in detail  in Chapter  A-6, the  former
mechanism, involving gas scavenging,  lies on the borderline between  reactions
in solution and scavenging processes.   Because it sometimes involves solution
reactions and will be useful  in  assessing the  relative  importance of  various
reactions producing acids in  solution compared  with direct  absorption  of the
corresponding reaction  products,  acid vapor absorption will  also be  consid-
ered here.

With  regard  to particle  scavenging,  Chapter  A-6  shows  that scavenging  of
particulate  sulfuric acid by  cloud droplets  occurs  with  essentially the same
efficiency as scavenging of sulfuric  acid vapor.  Therefore,  despite the fact
that most of the  sulfuric acid in  the atmosphere  is in  particulate  form (due
to the very low vapor pressure of sulfuric acid),  we can  treat the scavenging
of sulfuric  acid  by considering  the scavenging  of sulfuric  acid  vapor  having
a pressure equivalent to a typical  mass  concentration of  atmospheric,  partic-
ulate sulfuric acid.  This procedure  allows  us  to treat  the incorporation  of
H2S04  into  hydrometeors  with  the  same   methodology  required   to   treat
HN03 and HC1  (both of which  are primarily  gases  in the atmosphere).

Two  steps  are necessary  to  evaluate  the importance of  absorption of  acid
vapors:     (1)  determining   the  solubilities  of  the  chemical   species  of
interest,   and  (2)   determining  their  concentrations  in  air.     Regarding
solubility,  the  Henry's  law constants  for the  three acids  identified  as
significant  contributors  to  the acidity of precipitation  (HC1,  HMOs, and
H2S04)  and   for  the   various   trace   gases   (Cl2,   N02,   NgCty,    HN02,
and 862) assumed  to  be the precursors of these acids in the atmosphere are
listed in  Table 4-13.

For these constants to  be suitable measures of solubility,  equilibrium  must
exist between the  gases  and  the liquid phase. While such equilibria no doubt
exist for  cloud droplets,  they may not for raindrops falling  through a  strong
concentration gradient  of  gases.   Furthermore,  the Henry's law  constants
shown in Table 4-13  are  based on measurements  at vapor  pressures  far  above
atmospheric values.   Thus,  gross extrapolations  must  be  used when they are
applied to atmospheric conditions.  Indeed,  the  very large  values  for  some  of
the  Henry's   law  constants   (_>  105   mol  fc"1   atnr1)  shown  in   Table  4-13
cannot  possibly  be  applied   to  conditions   in  the   atmosphere;  they  simply
                                    4-32

-------
END OF CONDENSATIONAL GROWTH,
START OF GROWTH VIA COLLECTION
PROCESSES. FOR WARM CLOUDS,
DILUTION EFFECTS CEASE.
i

PRODUCTION OF ACIDS IN DROPLET
FROM ABSORBED PRECURSORS.
CONTINUED ABSORPTION OF GASES.
*
'
L
CONDENSATIONAL GROWTH OF DROPLET
AND ABSORPTION OF VARIOUS ACIDS
AND ACID PRECURSORS.
*
t

SOLUBLE FRACTION OF CLOUD
CONDENSATION NUCLEI DISSOLVES
IN THE DROPLET. *


ACID PRODUCTION CONTINUES
(Ca and Mg BEGIN TO GO INTO
SOLUTION AND BUFFER THE
CLOUD DROPLET.). *
J
•
CLOUD DROPLET GROWS TO PRECIPITABLE
SIZE AND FALLS OUT OF CLOUD.
^
r
ABSORPTION OF VARIOUS ACIDS AND
ACID PRECURSORS IN DROP AS IT FALLS
FROM CLOUD TO GROUND. ALSO, DROP
SCAVENGES BOTH ACIDIC AND BASIC
PARTICLES FROM THE AIR.
-
i
PRODUCTION OF ACIDS IN RAINDROPS
FROM ABSORBED PRECURSORS.
             I
  NUCLEATION OF CLOUD  DROPLET
ON A CLOUD CONDENSATION NUCLEI.
            I
DEPOSITION OF DROP ON GROUND.
 Figure  4-2.   Schematic diagram of the steps in the production of acidic
              precipitation.  Steps discussed in this section are indicated
              by  asterisks in the lower right corner of the box.
                                   4-33

-------
        TABLE 4-13.  HENRY'S LAW CONSTANTS (H)  FOR GASES  OF  INTEREST
                     IN ACIDIC PRECIPITATION  FORMATION
Gasa
C12
(HC1)
N02
N204
HN02
( HN03)
S02
(H2S04)
H Temperature
(mol £-1 atm-1) (c)
6.2 x 10-2 25
2.5 x 103 25
2.48 x 10-2 15
2.15 15
4.76 x IQl 25
1.98 x 105 25
1.24 25
108 25
Source
Whitney and Vivian
(1941)
Calculated from vapor
pressure data in
International Critical
Tables (1928)
Komiyama and Inoue (1980)
Komiyama and Inoue (1980)
Martin et al . (1981)
Davis and de Bruin (1964)
Johnstone and Leppla (1934)
Calculated from vapor
pressure data in
International Critical
Tables (1928)
aThe strong acids are in parentheses  and  their  precursors  precede  them.
                                   4-34

-------
indicate large deviation from Raoults's law suggested by the exothermicity  of
acid solution reactions.  The  large  magnitudes of the  Henry's law  constants
also suggest  that the associated vapors are essentially completely  absorbed
by hydrometeors and that liquid-phase concentrations must be calculated  from
considerations of  mass  conservation;  we will  return to this subject  later.
Despite these  problems,  the values of  the Henry's  law constants listed  in
Table  4-13  are  useful  as  measures  of relative  solubility and  will  be  so
employed.

The values  shown  in  Table  4-13 illustrate the  very  high solubility of  HC1 ,
HN03,   and   H2S04   relative  to  their  gaseous   precursors.     This   high
solubility suggests that the direct absorption  of acid vapors might play  an
important role in acidic formation in  hydrometeors.   The range of  the species
listed  in  Table 4-13  is shown  in Table  4-14 to explore  this  possibility
further.

The information in Tables 4-13  and 4-14 permits estimates of the liquid-phase
concentrations of both directly absorbed acids and their absorbed precursors
in the atmosphere.  The ratio of these concentrations indicates the  potential
importance of aqueous-phase  acid  production  reactions.   For example,  if the
ratio of an acidic concentration  in the liquid  phase to the concentration  of
its absorbed precursor is high, very  high  reaction rates will be necessary  to
increase acidity  significantly during the  lifetime  of  a  hydrometeor.  For
HC1, this ratio is infinite under most atmospheric conditions.  Indeed,  only
Clg is  listed as  a precursor of HC1   in Table  4-14.   The implication  is not
that other precursors do not exist, for it  is well  known that in urban  areas,
large  quantities  of  chlorine and chlorinated  organics  are  emitted   into the
atmosphere  (MAS  1976).   However,  the  lifetime  of free  chlorine in  the
atmosphere is very brief, and  the reduced product is HC1.   Any chlorine  that
might  survive long enough to be  scavenged would  undergo absorption via the
very fast reaction (Whitney  and Vivian 1941):

     C12 + H20 (I)  +  H+  + Cl- + HOC1,                                   [4-80]

and could therefore be  considered the anhydride of  HC1.  Chlorinated  organ-
ics, on the other hand,  should  be stable  in  solution  and produce little acid.
For a more detailed discussion  of the  possible  inclusion of  free chlorine and
chlorinated organics  in  precipitation, see  Mills et al.  (1979).

Of more interest are the N  and S species,  which contribute substantially  to
the acidity  of precipitation.   The  concentration  of  H2S04  in  cloud  water
can be  taken  as the  mole  (mol)   concentration  of H2S04 Per cubic  meter  in
the gas  phase  divided by the  cloud water  concentration in liters per cubic
meter of air.   When  a concentration   of 1  ppb  for H2S04 (Table 4-14)  and a
cloud  water  concentration  of  ~  5  x  10-4  £  m-3  are  taken,  a   value of 8
x  10-5  mol  £-1   is  reached  for  the  maximum  concentration  of   directly
absorbed  H2S04  in  cloud water.   The concentration  in background air  is
almost certainly at least an order of magnitude less than  this value.  For
comparison,  the   concentration   of   S(IV)  (the  immediate  precursor   of
       in solution is  given  by:
                                    4-35

-------
                TABLE 4-14.   GAS-PHASE  CONCENTRATIONS OF  ACIDS
                    AND THEIR PRECURSORS  IN  THE  ATMOSPHERE
    Gasa
 Concentration
in "background"
   air (ppb)
Concentration
in urban air
   (ppb)
Source
  C12                -               -

  (HC1)              1               8          Kritz  and  Rancher  (1980), Okita
                                               et al.  (1974)

  N02             0.1-4            10-100       Robinson and Robbins  (1969),
                                               Noxon  (1975), Spicer  (1977b)

  N204           Negligible       Negligible     No  measurements available.

  HN02            0.003             2-4         Crutzen (1974), Winer (1979).

  (HN03)           0.02-5             10         Huebert and Lazrus (1978),
                                               Kelly  et al. (1979),  Spicer
                                               (1977b).

  S02               1-14            10-50        Georgii (1978), Hidy et al.
                                               (1978)

  (H~  S04)     < 1& (0.5)       < lb(0.5,4)     Commins (1963), Tanner et al.
    *                           ~              (1977), El shout et al. (1978),
                                               Yue and Hamill (1979)


aThe strong acids are in  parentheses and their  precursors precede them.

bThe   extremely   low  vapor   pressure   of   H2S04   results   in   extensive
 nucleation  of   H2S04-H20  droplets  under  atmospheric   conditions  when  the
 vapor  pressure  of H2S04   exceeds  ~  1   ppb  (Yue   and Hamill  1979).  The
 bracketed  concentration  of  4,  listed  under  urban  concentrations,  which
 appears to contradict this view, is derived from Commins (1963) and probably
 includes  substantial  particulate  H2S04.    The  bracketed concentrations  of
 0.5, for background and  urban  air,  are from  El shout  et al. (1978); these also
 include  particulate  H2S04.    Furthermore,  rapid  condensation  of  H2S04
 vapor  onto ambient  particles  may be  assumed  to  reduce  the  equilibrium
 concentration  of the vapor  far below 1 ppb.  The value of 1 ppb  is  used as an
 analog  for  approximately  4  ug  m~3   of   both  particulate  and  gaseous
 H2S04.
                                    4-36

-------
     CS(IV)] =  HS02.PS02
Kls    Kls K2s
        [H+]2
[4-81]
where  HS02  is the  Henry's  law constant  for  S02,  and KIS  and  K2s  are
the  first  and second  dissociation  constants  for SC^-HgO.    When  appro-
priate  values are  used for  those  constants  and  a cloud  water pH  of ~  5
(Petrenchuk  and Drozdova  1966, Hegg and  Hobbs 1981a)  is assumedt a  maximum
concentration  of  S(IV) in  urban air  is found  to  be  7.9  x  10~5  mol  jr*.
If we  assume a cloud droplet life  of  ~  1  hr, S(IV) oxidation rates  on  the
order  of 100  percent  hr"1  would be  required for  significant acid  produc-
tion.1   "Significant"  refers to acid  production  at concentrations at least
equal to those produced by direct absorption of acid vapor.

Furthermore,  assuming   a  background  concentration   of  H2S04  of  ~  0.1  ppb
and  a  background  concentration of  S02  of  - 10 ppb  in the northeast  United
States  (Hidy et al. 1978),  an  S(IV)  oxidation rate  of only  -  50  percent
hr"1 would be required  for significant acid  production  in background  air.

The  situation with  respect  to HN02  formation  in  solution is  quite  dif-
ferent.  Again, acid concentration  must be estimated from considerations  of
mass conservation.   Assuming  gas-phase  concentrations  of 5  and  0.5  ppb  in
urban and background atmospheres, respectively, the  same  procedure used above
for  S  yields  liquid-phase   HMOs   concentrations  of  4.1  x  10-4  mol  £-1
and  4.1 x  10~5  mol  £-1  for  urban  and  background   atmospheres,   respec-
tively.    The  corresponding  liquid-phase N(III)  (the  N  species generally
assumed  to  be the  precursor  of HN03  In solution)  concentrations  would  be
 ~ 8  x  IQ-s  mol   jT1   in   an  urban   atmosphere   and   3  x  10~8  mol  a -1
in the  background  atmosphere  (based on a pH of  5.0, concentrations  for N02
of 50  ppb  and 1  ppb in urban  and background atmospheres, and concentrations
of HN02 of 4  ppb and 0.003 ppb in  urban and background  atmospheres).   These
concentrations  suggest  that  oxidation  rates  of   ~ 5   x  10*  percent  hr-1
and      1   x  10^   percent   hr"1   in   urban  and   background  atmospheres,
respectively,  are  necessary  for significant  acid  production  to  occur via
precursors.  As shown later in  the  chapter,  these  rates  are far higher  than
are those of any known  reactions for N(III).

A possible  alternative  to the production of HNOa  in solution from absorbed
N(III)   is  its  production  from absorbed ^05 at  night  (Platt  et  al.  1981).
^y comparison,  raindrops  have lifetimes  from  1  to  5 min,  assuming cloud
 bases  from  1  to 3  km and  a  mean fall speed  of ~  10 m  s"1.   Solution
 reactions in raindrops will therefore make  a  relatively small contribution
 to  hydrometeor  activity   (although direct  absorption  of   acids  may  be
 substantial).   Attention is therefore focused on  solution reactions in cloud
 droplets.
                                   4-37

-------
However,  since  ^05   ^s  an  hydride  of  HN03,  this  mechanism  is  really
only an interesting variant on  the  direct  absorption  of  HN03;  therefore, we
will not treat it here as  a  solution reaction.

It may be tentatively concluded that liquid-phase oxidation reactions do not
play a  role  in HN03  formation  in  cloud droplets.  A  recent  modeling study
by  Durham et  al. (1981) suggests that  such  oxidation  also  plays  no role in
the acidity production  in raindrops.   The  principal reason for  the lack of
any contribution  to  the  formation  of  HNOs  from liquid-phase  oxidation in
hydrometeors  is  the low  rate  of N(III) formation  from  absorbed N02-   The
complex  nature  of   N02  absorption   by   water  has   led  to  considerable
misunderstanding and is discussed more  thoroughly  in Section 4.3.4.

4.3.3  Production of HC1 in  Solution

While little  evidence currently supports  the  formation  of HC1  in solution
from  gaseous   precursors,  HC1  has  long   been  thought  to be produced by
particles of sea salt dissolving in hydrometeors, either by absorption or by
production  in  solutions  of  HN03  and/or  I^SO^  (Robbins  et  al.  1959,
Eriksson  1960).     For both  HN03  and  I^SCty,   the reaction  is   simply  a
cation exchange  between chloride and the  less  volatile  nitrate  and  sulfate
anions.   The  HN03  reaction  has been shown  to convert  as  much  as  16  percent
of  initial  Nad  to HC1  within a  5-minute  reaction   time;  presumably,  the
H2S04  reaction  is  equally  fast.    However,  while  HC1   produced  in   this
fashion  will   contribute  to  the  acidity  of   hydrometeors  and  possibly
contributes a major fraction of  the background  gaseous Cl  in  the atmosphere
(Duce 1969),  it  obviously cannot increase  the  acidity  of hydrometeors above
what  would  be  produced  by  the  HN03  and/or  H2$04  from  which  it is
derived.

4.3.4  Production of HNO^  in  Solution

The  production of  HN03  in  solution  by   means  of nitrite  N0£-  (or HN02)
oxidation has been  proposed  as  a  significant  atmospheric  reaction.   The
oxidants  currently   considered  significant  are  03   (Penkett  1972)  and
H202  (Durham  et al.  1981).    While the oxidation  rates produced  by these
oxidants have  been  studied (Halfpenny  and  Robinson  1952,  Penkett 1972), the
results of the previous section  suggest that these  reactions  are not likely
to  be  important  in  the  atmosphere,  due  to the low levels  of  N(III) in
hydrometeors.   The  low levels of N(III) result from   the  low  solubility of
N02 in  hydrometeors and  the  relatively slow rate  of  N(III)  formation  from
the absorbed  N02.   This has  led to  some confusion.  For  example, Flack and
Matteson  (1979)  derive a  value of  100 mol A"1  atm~l for the  Henry's law
constant  of   N02,  compared  to  the  value  of  2.48 x 10~2  mol  £-1 given
in  Table  4-13.  The  higher value  is obviously  wrong  because  it exceeds the
constant  for  the   N02   dimer  (^04),   which    is   well   known   to  be
considerably  more  soluble than is  N02  (Andrew  and  Hanson 1961, Kameoka and
Pigford 1977,  Komiyama and Inoue 1980).
                                    4-38

-------
Much  of the  confusion over  this matter  is due  to the  complexity of  the
NOX  -  H20  system  at  the  high N02  concentrations  that  commonly  have  been
employed  in  laboratory experiments  (>_  5 ppm and  commonly >  200  ppm).   At
these concentrations the gas-phase reaction,

          3N02 +  H20   =   NO  +   2HN03,                             [4-82]

occurs  and  spontaneously  forms a  two-phase  system consisting of HNOa  vapor
and  droplets  of  dilute   HNOa  over  the  absorption  surface  (England  and
Corcoran   1974).      Also,   at   high   N02   concentrations   the   gas-phase
equilibrium,

          2N02 = N204,                                                [4-83]

results  in  appreciable N204,  which  can  then absorb into  solution via  the
fast disproportionate reaction:

          N204(g)  + H20U)  = HN02U)  +  HN03<*>

NO?   absorbs   in   a   straightforward  manner   but  then  forms  N204
which  undergoes  the  disproportionate  reaction   given  by  Equation   4-84
(Komiyama and  Inoue 1980).  This  reaction's rate is  slow enough  (k  ~ 4  x
105  s-1;  Kameoka and  Pigford  1977,  Komiyama and  Inoue  1980)  to  render  it
a rate-limiting step in formation  of N(III)  from absorbed N02 over  the  time
scale  of a  cloud  ( ~ 1 hr).    Recent  studies  by  Lee  and  Schwartz (1981)
support this viewpoint.

Finally,  because  of the low  surface-to-volume  ratios  of solutions used  in
laboratory  experiments compared  to  those existing  in  the atmosphere,  even
absorption  rates  measured  in laboratory  experiments at  relatively  low  N02
concentrations  can  be  limited by mass  transport.   For N02  concentrations
that  exist  in  the  atmosphere  (  ~ 1  to  100  ppb),  and   for  the surface-to-
volume  ratios  of  drops characteristic  of  clouds  ( ~ 3  x   105  nr1) ,   only
direct  N02  adsorption  is  of any  consequence.   Thus,  the  total  amount  of
N(III)  in solution  derived  from  N02  is governed by  the  Henry's  law  constant
for N02, given  in Table 4-13,  the equilibrium constant for the  liquid-phase
analog  Equation 4-83  (7.5  x  10^ a  mol~l;  Komiyama  and Inoue 1980),  and
the rate constant for Equation  4-84 (liquid  phase).  For  estimates of N(III)
used  in Section 4.3.2, we  assume  a  time scale  of one-half  the total  cloud
lifetime  in  determining  the  amount of  N(III)  formed  from  absorbed   N02-
Because of  the  disproportionate reaction  upon solution of N02,  each  mole
of N02  absorbed produces  1 mole  of  HN03 for each mole of N(III)   produced.
Therefore,  the  reaction  rates  for  N(III)  oxidation  necessary  to  produce
HN03  levels  rivaling  those due   to  direct  absorption,  either  of  N02  °r
HN03,  are  increased  to   roughly  103  hr-1  and  2  x  1Q5  hr-1 for  urban
and background atmospheres, respectively.

Of the two oxidation reactions  mentioned early in this  section,  the  oxidation
of N(III)  by 03  (Penkett  1972)  has been  studied  with direct  consideration
of atmospheric  applicability.   The  reaction was  studied in  a  stopped-flow
reactor,  the  rate  being determined  when  the 03  aqueous concentration  was
                                    4-39

-------
monitored  with  a  UV  spectrophotometer  at a  wavelength  of  255  nm.    Such
devices require reactant concentrations far exceeding  atmospheric  levels.

For  example,  the  03  concentrations  Penkett  employed  were  equivalent  to
gas-phase  concentrations   of  several  hundred   ppm,   103  to   ICr   times
atmospheric levels.   However, the  agreement  between the  oxidation  rate  for
S(IV)  by  03  measured  in  this  study  and  that  measured  by  wet-chemical
techniques  at much  lower  03  levels  (Larson  et  al .  1978)  suggests  that
extrapolation of  the  N(III)  rate  to atmospheric  levels may  be valid.   The
reaction  was  found to  be  first-order in  both  03  and  N(III).   The second-
order rate expression at 283 K and a pH of 5.9 was:


                                     k2 [03]  [Ndnj]                  [4-85]
                 dt           dt

with   k2   =   (1.60   +_  0.13)  x  10^   i  mol'1   s'1.     Assuming   that  the
ambient 03 concentration at  cloud level is  generally at or below 50  ppb  (at
STP),  the  characteristic time2  for  N(III)  oxidation  at 283 K  and a  pH  of
5.9  would  be  ~ 2  hr,  and  the  conversion  rate  (R)  50  percent  hr-l.J
Clearly, this reaction will  be of little importance in HN03 production.

The  oxidation of N(III)  in  solution by  H202  received  attention in  several
investigations (Halfpenny and Robinson 1952, Anbar and Taube 1954).   The rate
expression determined  by Halfpenny and Robinson  over the pH range of  -4.3
to 4.7 at  a temperature of 292 K was:


                        =  k [H202] [HN023 [H+]                        [4-86]
with   k  =  1.4  x   102   £2  mol"2   s"1.   These   investigators  considered
HNO^  to be  the  reducing  species in  solution,  although they point  out that
N02    might   still   be  the  reducing  agent   because   of  the  equilibrium
between   HN02   and   N02".      Anbar  and   Taube,   on   the   other   hand.,
determined   the  reaction  rate  by  monitoring  the  concentration  of  N02
spectrophotometrically  at a  wavelength  of  357 nm  and  imply  that N02~  is
the reducing agent in the  reaction.   Their  rate expression for  pH's from 4.6
to 5.1  at 298 K was:

                     2]     k3 k2 [H+]2  [N02~] [H202]                 j-
                                k_2 + k3 [H202]
 The e"1 decay time.

                        d ( £ In
 3Ri  (%  of  hr'1) = 100 x     crt      ,  where  Cn- is  the  concentration of the
 the  reactant under consideration.   Consequently, R-j  (%  hr-1) =  100  x k',
 where  k1  is  the pseudo-first-order rate coefficient.


                                    4-40

-------
where the  k's  are  rate constants as defined  by  Anbar and Taube, k3 =  5.8  x
10^  £3  mol~3   s~i,   and  k3/k_2  =  2.4.     For   atmospheric  levels   of
H202, this reduces to


             -  d[H2°2]  =  k'  CH+]2 [N02-] [H202]                     [4-88]
                  dt

with k1  = 1.4 x 107 £3 moT3 s'1.

The rate expression of Anbar and Taube must be converted to one with explicit
HN02  dependence  by  means  of  the  N02~  -  HNO?  equilibrium  to  compare
this value directly  with  that of Halfpenny  and  Rooinson.   This results in  a
rate  coefficient  of  6.3  x   102  £2  mol'2  s"1,  roughly  4.5  times  that
of  Halfpenny  and Robinson.   Given the different  experimental  temperatures,
methodologies, and concentrations  of  reactants,  this may be  considered good
agreement.    However,  both  experiments   were conducted  at  H202  concentra-
tions  (_>  0.05  mol   £-1)  and  N(III)  concentrations  (>  0.017  mol  £-1)
far  higher than those  encountered  in the atmosphere.   Tfiis  should be con-
sidered when  the rates are applied  to atmospheric  conditions,  particularly
because  no  activation energy  was  determined  for  the  reaction,  and  the
temperatures  at  which  these  rates were  made were  appreciably  higher than
those  typical  of  clouds   over  the United  States.   Nevertheless,  the rate
determined by Anbar and Taube  can  be  employed as a  rough  indication of this
reaction's importance.

For  typical  cloud  water pH's of 4.0  to  6.0, most of  the N(III)  in solution
will be  N02-, and  the values  of N(III)  calculated  in  Section 4.3.2 will
be so interpreted and inserted into the rate expression.  Once again,  a pH of
5.0  will  be  selected  for the mean  cloud water  oH.    For  the  H202  concen-
tration  in hydrometeors,   a value of  1.5 x  10"^ mol  £-1  will  be employed
(based on  measurements in  precipitation  [Kok 1980]  and  a  few,  as yet  un-
published, measurements in  clouds over the eastern United States [Kok, pers.
comm.]).    Inserting  these values  into  Anbar  and  Taube's  rate  expression
yields a  characteristic  time  for N(III)  oxidation  of 1.3  x  104  hr, sur-
prisingly  slow.   Clearly,  this reaction  can be  of  no  importance to HN03
production in hydrometeors.

The above  results support  the  tentative  conclusion reached  in Section  4.3.2,
i.e., that HN03  production in  solution  by  oxidation  of  N(III)  is  unim-
portant compared to direct  absorption  of  this species from the gas  phase. Of
course,  future  research  may  suggest  other  oxidation  reactions  appreciably
faster than the  two  that  have  been  suggested to  date, or future rate studies
may suggest higher rates  for these two reactions.   Our conclusion concerning
the  importance  of  N(III)  oxidation to HN03  formation  in  solution  is  highly
dependent  on  relatively few rate  studies, compared  to  the  case  for  H2S04
production.   This  dependence  should  be  considered when  the  influence  of
HN03 on acidic deposition  is assessed.

At this juncture, we  conclude  that HN03 concentration  in  solution  generally
is  determined  by HN03  production  in  the  gas  phase (or  possibly on  aerosol
particles) and its subsequent rate of absorption  into hydrometeors.


                                    4-41

-------
 4 .3 .5  Production of H?SQd In Solution

 4.3.5.1    Evidence  from  Field  Studies—From  analyses  presented  in  Section
 4.3.2,  it  appears  that  H2$04  is the  acid most  likely  to  be  produced  in
 cloud  droplets in  significant  quantities.   Furthermore, field studies show
 that sulfate   (S042~)  is  produced  in  clouds.    Such  evidence  has  been
 accumulating  for  some  time,  although early  data were  somewhat indirect.  For
 example,  Radke and  Hobbs  (1969),  Saxena et  al . (1970),  Dinger et al .  (1970),
 and  Radke (1970)  observed higher concentrations of cloud  condensation nuclei
 (assumed  to be mainly sul fates)  in  evaporating clouds than  in ambient air.
 Georgii  (1970)  found that while sulfate concentrations decrease with altitude
 in  dry  air, they  peak at cloud  levels in  air subject to  cloud formation.
 Similarly,  Jost  (1974)  found  anomalously  high  $042-   concentrations  in
 clear, subsiding  air  near the bases of cumulus clouds—the  sample air being
 considered  to  have passed through  the  clouds.  McNaughton  and Scott (1980)
 concluded,  on  the  basis of  mass  balance  calculations,   that  S042~  pro-
 duction   in  clouds  is  necessary  to  account for  the  acidity  and  $042-
 levels found  in precipitation.    Also,  recent  field  results (Lazrus  et al  .
 1983) suggest  appreciable sulfate formation in warm frontal  clouds.  Finally,
 Gillani and Wilson  (1983), in a  study of  power plant  plumes  interacting with
 clouds,  present particulate  and  gaseous S measurements  that  strongly  suggest
 that S042~  production   is  occurring  in  clouds.    The  in-cloud  SO?  to
 S042~  conversion  rates  observed  were  on  the  order  of 10  percent  hr~X  a
 significant  rate  even in light of  the analysis  in  Section 4.3.2,  because
 S02  concentrations  in power  plant  plumes  were  far higher  than  were values
 used in Section 4.3.2 and thus could produce considerable  acid even if only a
 relatively  small fraction of the SOg were converted to H2S04.

The  most direct  and  quantitative evidence for $042-  production  in  clouds
 has  come  from  recent  measurements of  S042~  concentrations  in  the  air
 entering  and  leaving  wave clouds (Hegg and Hobbs 1981a,b).  These measure-
ments  have yielded  S02-to-S042~  conversion  rates  typically  on  the  order
 of  102  percent  hr-1,  a  significant  value  according to   the analysis  of
 Section 4.3.2.   This in  situ data  set is  sufficiently  large  (18  cases)  to
 allow determination of an empirical  rate expression.  It is of the form:


               = ki [H+]a [S032-]  exp (EA/RT)                           [4-89]
 where kj = (3.3 x 105 +_ 6.2 x 105)  £1 .1  mol-Ll  s-1,
       a  = 1.1 _+ 0.1, and EA = (2-9 +_ 2.7)  kj  mol-1.

Section 4.3.3  shows  that the value of a  is  similar to that expected  if  the
$042-  is  produced   in   solution   via  03   oxidation.   However,   the   S042-
production  rates measured  in  these  field  studies  showed  no  significant
correlations with 03 concentrations.

These  field   measurements   dictate  examination  of  H2S04   production   in
hydrometeors in greater detail than for HC1  and HN03.
                                    4-42

-------
4.3.5.2  Homogeneous Aerobic Oxidation of S02-H20 to H2S04--
4.3.5.2.1  Uncatalyzed.  This reaction is the most extensively studied of any
of those to be dealt with.   It  has  been proposed for some time as  a  reaction
of considerable  importance  in the  atmosphere  (Scott and Hobbs 1967,  McKay,
1971, Miller and de Pena 1972).   However, some controversy  exists  concerning
its  atmospheric  importance.    For  example,  Beilke and  Gravenhorst  (1978)
dismissed  this  reaction as being  of  no  importance  in  the atmosphere.
However, Hegg  and  Hobbs (1978)  considered it currently impossible to  arrive
at  a  firm conclusion  as  to  its  importance,  due  to  the  wide  range  of
conversion rates and rate expressions  measured in the laboratory by different
workers (Figure 4-3).

While little has been  done  to resolve the  discrepancies  shown in  Figure 4.3
and  debate  continues  as to  its  atmospheric significance (see, for  example,
Penkett et  al.  1979;  Dasgupta  1980a,b),  Hegg and Hobbs  (1979a) employed an
updated version of the Easter-Hobbs  interactive cloud-chemistry model  (Easter
and  Hobbs  1974)  to demonstrate  that  most of the  rates shown in  Figure 4-3
would yield significant  sulfate  concentrations  in  the  atmosphere.    These
rates  will  therefore  be  included  in  the  evaluation   of  the  potential
importance  of  H2S04 production  reactions  in  clouds,  although,  as  pointed
out  by  Hegg and  Hobbs (1978),  these rate expressions could reflect a  low
level catalysis  of the aerobic  reaction  rather than a strictly uncatalyzed
reaction.

Larson  et  al.'s   (1978)   rate  expression  was  chosen   to   evaluate   the
significance  of this  reaction  in  the  atmosphere.    This study  has  been
selected because  it was conducted  with  great care.  For example, oxidation
rates  relative  to  sulfite  (SOs2")    were   measured   by  monitoring   S03
(and  sometimes  sulfate)  concentrations, and SOg degassing  from solution was
evaluated quantitatively.   Such procedures obviate criticisms made of  other
laboratory  studies   of  S032~  oxidation   rates  with  respect  to   mass-
transport  limitation  of  the oxidation  (Kaplan  et al.  1981, Schwartz  and
Freiberg 1981).  Similar  procedures were  employed  by Fuller and Crist (1941)
and  by  Brimblecombe and Spedding (1974).   Hence,  the  disparities shown  in
Figure 4-3 are not entirely due to mass-transport problems.

Because it is unlikely that the reaction is much faster  than that measured by
Larson et al.  (1978) (and it may be  appreciably lower due to inhibitors;  Hegg
and  Hobbs  1978),  the  Larson et al.  rate may  be  considered  an  upper  limit to
the atmospheric oxidation rate.  The  rate  expression  for  this  reaction  at pH
_< 7.0 is:


           S 4    = (|q + k2 EH+]1/2)  [S032-]                          C<
           dt

with kx = (4.8 +_ 0.6)  x 10-3 s-l and

     k2 = (8.9 + 1.0)  £l/2 mol-1/2  s-l.
                                    4-43

-------
      10
        -1
   «/»
   —  10
      10
        -3
      10
        -4
             LARSON ET AL
                 (1978)
          .MILLER AND
           de PENA (1972)'
                (pH = ?)
                                FULLER AND CRIST  (1941) -
                                 AS MODIFIED BY McKAY (1971)
RIMBLECOMBE AND
 SPEDDING (1974)
                SCHROETER  (1963)
            WINKELMANN  (1955)
                                    — SCOTT AND HOBBS  (1967)
                                       (pH = ?)
                      BEILKE ET AL. (1975)
                                    6           8
                              pH OF THE SOLUTION
                              10
12
Figure 4-3.   Pseudo first-order rate coefficients  ("K0")  for the non-
             catalyzed aerobic oxidation  of S0^~  in  solution (Hegg and
             Hobbs 1978).
                                   4-44

-------
 Activation  energies  for these two coefficients are 40 +_  10  kJ  mol~l  and 7 +_
 6  kJ  mol"1,  respectively.    Assuming  a  hydrometeor  pH  of  5.0  and  "a
 temperature  of  278  K  (henceforth  all  rates  will   be  evaluated  at  this
 temperature,  because  it  is  representative  of  those encountered  in  warm
 clouds),  this  expression yields a characteristic  time  for  sulfate oxidation
 of - 44  s,  implying a conversion rate of ~ 8 x 103 percent hr~l.

 Before  Equation  4-90  and the  criterion  rate4  calculated  in  Section  4.3.2
 can  be  compared,  Equation 4-90  must  be  changed  from  a Spa2-  to a  S(IV)
 dependence.   This change  can  be  done by multiplying  the righthand  side  of
 Equation  4-90  by the ratio of  SOa2' to S(IY)  in solution  at  the giveji pH.
 For  a pH of  5.0  at  278 K,  this  is essentially  the  ratio of SOs2'  to
 bisulfite (HSOa')  and  equals  1  x  10-2.    This ratio implies  an  S(IV)
 oxidation rate  and  thus  an   ^$04  production  rate,   of 80  percent  hr"1.
 Comparing this  to  the  rates  calculated in  Section  4.3.2  for  significant
 H2S04  production (50  to 100  percent  hr'1),  shows  that Equation 4-90  can
 produce  significant ^$04 under background atmospheric  conditions.
4.3.5.2.2   Catalyzed.   The catalyzed  aerobic oxidation  of S(IV) to
has received nearly as much laboratory study as has the uncatalyzed reaction
Reviews by  Beilke  and Gravenhorst (1978) and Hegg  and  Hobbs (1978)  indicate
the  range of  rates  measured  for such  a  reaction.   However, most  of  the
studies  conducted  have  involved  catalyst  and  reactant  concentrations  far
exceeding those  encountered in the  atmosphere.   Furthermore, Kaplan et  al .
(1981) and  Freiberg  and  Schwartz (1981)  have suggested that in most, if  not
all, laboratory studies the oxidation rates have been limited by  mass trans-
port  and are  therefore  not  applicable  to the  atmosphere.   Freiberg  and
Schwartz  specifically  cite the  study  of Barrie  and Georgii  (1976)  as  one
where mass transport may have compromised measured  rates because of the  large
size of the droplets employed  as  the reaction medium.   However, Freiberg  and
Schwartz observe that the droplets used by Barrie and Georgii were ventilated
at an unspecified rate and  that  if this  rate were high enough, the  reaction
rate  would  not  have  been  limited by mass  transport.    Because  Barrie  and
Georgii 's study was conducted  with both  reactant and  catalyst concentrations
approaching  atmospheric  levels,  it is  worthwhile to  attempt to  establish
whether rates  these workers measured accurately reflect the  chemical kinet-
ics.   This  can  be  done by  comparing  the rates of Barrie and Georgii  with
chemical  rate  data derived  from  experiments where mass transport  definitely
did not limit reaction rates.

If one extrapolates the  results  of Kaplan et al .  (1981) for  Mn catalysis  to
the low catalyst levels  Barrie and  Georgii  employed, assuming the reaction
rate is first-order in catalyst concentration  (Hegg and  Hobbs 1978), the rate
derived is much slower than what Barrie  and  Georgii  observed.   Because Kaplan
et  al .  performed  their  study  under  conditions  free  from  mass-transport
limitations   (according  to  the   theory  of  Freiberg  and   Schwartz),   the
relatively fast  rate  of  Barrie and  Georgii must also be  considered free  of
      rate  necessary  to  produce  a  sulfate  concentration  similar  to  that
 obtainable by direct adsorption of
                                    4-45

-------
this constraint.  Comparison of the Barrie and  Georgii  rate  for  Fe  catalysis
with  that of  Brimblecombe  and Spedding  (1974),  from  which  mass- transport
effects  were  eliminated by  direct measurement  of both  S(IV)  and  S(IV)  in
solution, again reveals that the Barrie and Georgii rate is  the faster  of the
two.

It  may  be concluded that the  rates measured  by Barrie and Georgii  were not
significantly limited by mass transport and should therefore  be applicable to
cloud droplets.   Reactions  in  large raindrops, on the  other hand,  will  most
likely be limited by mass transport.

Barrie  and  Georgii  (1976)  studied  three catalysts:   Fe, Mn  (the  two  most
widely  accepted  catalysts  of   atmospheric  significance), and an  equimolar
combination of these two elements.   From Table 1 and Figure  2 of  their  paper,
the following rate expressions  for  these three catalysts have been derived:
     For Mn:                = kMn [Mn+2]  [H+]0-46[S032-]                [4-91]


     For Fe:       d[S^2"]  = kFE [Fe+2]  [S032-]                        [4-92]



     For Mn^       d[S°42"]  = kmix[Mn+2 + Fe+2][H+]°'64  [S032-]         [4-93]



with kMn = 1.6 x 108 £1<46 mol1-46 s'1, kFe= 5.8  x 106 £  mor1

s-1, and km1x = 1.8 x 109 d1-64 mol1 -64  s-l, all  at 298  K.


The activation energies were not determined explicitly in this  study,  but the
data shown are in accord with previous determinations of the  activation ener-
gies of the  Mn-  and Fe-catalyzed  reactions   ( ~ 113  and  ~  126  kJ  mol'1,
respectively; Hegg and Hobbs 1978).  The Mn  plus  Fe catalyst not only showed
a  synergistic  effect relative  to  individual  catalysts,  but also displayed
negligible temperature dependence.  The  catalyst therefore  could be  of  con-
siderable importance, at least in an urban atmosphere.   The  relatively large
temperature dependence of the two single metal catalysts, on the other hand,
somewhat decreases their potential  atmospheric importance. ^
5The cited  activation  energy for  the Mn  reactions  for example, lowers  the
 given  rate  coefficient  for  this  reaction  to 6  x  106   &1'46 mol"L*^b
 s"1 at  a temperature of  278 K,  a reasonable temperature for  clouds.   In
 general, the  relationship  between activation  energy  and rate  coefficient,
 which determines the temperature sensitivity of  a rate expression, is  given
 by  the  Arrhenius  equation:  ki  = Aj  exp { -Ei/RT},   where  kj  is  the rate
 the  rate  coefficient  with  activation energy  E-j,   and  A^  is  a  constant
 determinate from measurements  at  several  temperatures.


                                    4-46

-------
The  major  problem in evaluating the  significance  of catalyzed reactions  in
the  atmosphere  is in estimating concentrations of possible catalysts  in  the
atmospheric hydrometeors.  Assume the maximum  concentrations of Mn  and Fe in
urban  air  to  be  - 0.2  and  ~ 6  yg  m"-3,  respectively  (Miller  et  al.
1972,  Lee  and von Lehmden 1973, McDonald  and Duncan 1979, Lewis and  Macias
1980).   The  soluble  fractions for  the  Mn  and  Fe  species  found  in  the
atmosphere  are  ~  0.25 and 0.15  percent,  respectively (Gordon et al.  1975).
For  a  liquid water  content of  ~  0.5  g  m"3,  these figures  yield  cloud
water  concentrations  of  -  2  x   10~8  mol  a~l  of  Mn  and  -  3  x  10"'
mol  &"1  of Fe, with  perhaps  an order of  magnitude  of uncertainty in  these
values.  These  values compare reasonably well with  the maximum levels .of Mn
and  Fe found  in  Florida  rainwater,  which  are reported  to be 6  x 10~°  mol
jT1  of  Mn and  4  x  10~7  mol  rl  of  Fe  (Tanaka  et   al.  1980).    How-
ever,  these values are somewhat lower than  rainwater concentrations  reported
by  Liliestrand  and Morgan  (198L)  for southern  California (Mn:  -  2  x  10"7
mol  i'1;  Fe:   - 10"°  mol  £"1  and  by   Drozdova  and   Makhon'ko   (1970)
for   the   Soviet   Union   (Mn:   -  5  x  10"7   mol   A'1;   Fe:   ~  10"6   mol
JT1).   On   this basis,  and assuming  some  variability in  liquid  water  con-
tent,  upper  limits   for  Mn  and  Fe  of   - 10"6  mol &"1  will  be assumed.
The  dependence  of these  rates  on  cloud liquid water content are examined
later.   Employing these  concentrations at  a  temperature  of 278  K and pH  of
5.0, yields characteristic oxidation times  for S(IV)  of:    0.93 hr (Mn),  0.19
hr  (Fe), and  0.01 hr  (Mn  +  Fe).  The corresponding conversion  raies are ~
100  percent  hr"1  (Mn),   500   percent  hr"1  (Fe),  and  ~ 5  x  10-5  percent
hr"1  (Mn +  Fe).   These values certainly suggest that  the  catalyzed  reaction
will  be considerably  important,  at  least in urban air.   However,  a word  of
caution is required.

It  is not  clear  that the  Mn rate  or  the  mixed  catalyst rate  Barrie  and
Georgii (1976) measured can be extrapolated to the atmospheric case.   Barrie
and  Georgii   observed  negligible  oxidation  with 10~6 mol a~l  of Mn as  a
catalyst.  No clear evidence  shows that  the mixed  catalyst effect occurs  at
concentrations  below  10~5  mol  A"1.    Furthermore,  these estimates  have
yielded  rates  that   produce  substantial   HpS04  in  solution relative   to
initial concentrations of H2$04.  One  would  therefore  expect the solution
pH  to  drop  substantially.   Given the  inverse square dependence  on H+  con-
centration  of the  S032~  concentration  in  solution,  the rate  expressions
for the catalyzed (and the uncatalyzed as well)  reactions  suggest they  may  be
self  limiting  in  hydrometeors.   Hence, the  rates  calculated  above from  the
characteristic times,  based  on initial  pH's,  will  be  upper  limits  to  the
time-average rates.  Finally,  the mixed catalyst  rate is  so fast that  it  will
be almost certainly  limited  by mass  transport,  even in  raindrops of  modest
size, as suggested by Freiberg and Schwartz (1981).

4.3.5.3  Homogeneous  Non-aerobic  Oxidation  of SO?'H?0 to H?SOa--SO? absorbed
into  atmospheric  hydrometeors can be oxidized  by  oxidants other  than 03-
Indeed,  recent  work  on   H2S04  production  in clouds  and  rain   has   tended
to   emphasize   the   oxidation  rates  by  03   and   H202  (Penkett  et  al.
1979,  Durham  et  al.  1981).    Recently,  interest  has also revived  in the
classic  reaction  involving   S032"   oxidation   by  N(III)    in   solution
                                    4-47

-------
(Martin  et  al.  1981, Chang et  al.  1981).   Of  these three oxidants, 03  has
been the most widely studied and will  therefore  be examined first.

The  relevance   of  03  to  S042"   formation   in  hydrometeors  was   first
examined  by  Penkett  (1972),   who  studied  $032-   oxidation   by   03   in   a
stopped-flow reactor  at  a solution pH  of 4.65  and  a  temperature  of 283  K,
values  representative  of the atmosphere.   However,  the reactant concentra-
tions  employed  were  far higher than  those encountered  in the atmosphere.
More recently,  several other  studies  have been  conducted on the 03 reaction
with reactant concentrations closer to those in  the atmosphere.  These studies
are summarized  in  Table  4-15.   The  study  by Penkett  et al.  (1979) contains a
number of errors in the  derived rate expression.   It is  therefore preferable
to  show the rate  expression  derived by  Dasgupta (1980a)   from  the data  of
Penkett et al.   However, the rate for atmospheric conditions (last  column  in
Table 4-15)  is  that directly measured  by Penkett et al.

Examination  of rates shown in Table 4-15  suggests nearly as much uncertainty
about  the 63  oxidation  rate  as for  uncatalyzed aerobic  oxidation.    Rates
tend to  increase  as  the ratio  of  03  to S(IV)   in  solution increases, sug-
gesting that oxidation rates measured in the laboratory were limited by mass
transport of  03.    However,  03  concentrations  in  solution  were   measured
directly in experiments  of Penkett et  al.,  thus precluding any limitations
due to  mass  transport.   In  any case, the mole  ratios  of  03  to  S(IV) used
in the studies  with the higher derived rates are far  above  atmospheric values
(  ~  10"4).    Because  the  rates  derived  for  atmospheric conditions  from
measurements of  Penkett  (1972)  and  Larson  et  al.  (1978)   differ only  by  a
factor  of  3,  despite extrapolations  over   several  orders of  magnitude   in
reactant concentrations,  the higher of the two rates {Penkett 1972)  has been
selected to  estimate the  importance  of  this  reaction  in  H2$04 production
in hydrometeors.   While  the  relatively  conservative  nature (compared to the
upper end of the range in rates  given  in Table 4-15)  of  this estimate should
be considered,  Hegg and Hobbs's  (1981b)  observations  discussed  in Section
4.3.5.1 cast doubt on the applicability  to the atmosphere of the higher  rates
shown in Table  4-15.

Table 4-15  shows that the characteristic time for  S(IV)  oxidation  is ~ 1  hr
for the  Penkett rate, and  the  conversion rate   is ~ 100  percent hr'1,  which
should be significant in  the  atmosphere.6

It has  been  proposed (Penkett et al.  1979)  that the  03 reaction  mechanism
is a free-radical  chain, similar to that of the 02  oxidation  reaction.   If
so,  like the aerobic oxidation,  it should be both catalyzed and inhibited  by
certain trace metals and organics  in  solution  (Hegg and Hobbs 1978). Inter-
estingly, Barrie   and  Georgii  (1976)  reported  a  substantial  enhancement
6The characteristic or e"1  folding time is  given  by  /T7
                                                                  "
                                                    /T7~T —  .
 in the atmospheric pH range of  ~  3  to 6,  HS04-  s    I Si IV)   at

                              1      d S(IV)-1         _!
  SUV) and this becomes:   [HSOa-]   dt      = { Klto3]  K
                                    4-48

-------
           TABLE  4-15.   LABORATORY STUDIES OF S(IV)  OXIDATION BY 03 IN  AQUEOUS  SOLUTION
                        Rate expression
                                           Experimental
                                               pH
                  Molar  ratio of
                    reactants
                   C03]/[S(IV)]
                 Reaction rate3
               (In mol I'1 S'1)
             at 278 K, 1 ppb S02
              40 ppb 63, and a
                  pH of 5.0
Penkett (1972)
Barrie (1975)

Erickson et al.
 (1977)
                       i = 3.3 x 105  i mol"1 s'1
                       at 283 K
k2C03]CHS03-] + k3
  [03][S032-]
k2 =  3.1 x 105 i moT1 s'1
k3 =  2.2 x 109 i moT1 s'1
  at  298 K
   4.65



   4.0

-1.3 - 4.02
                                                             0.03 - 0.5
                                                            10-6-5 x lO'5
5-50
aShows derived rates for atmospheric  conditions.
bThe  measured rate at pH = 4 and 283  K was converted to that at pH = 5 and 278 K  by assuming that the
 rate is proportional to [HS03-]j and changes negligibly with temperatures over 5 K.
1.5 x  10-9



  5 x  10-llb

  2 x  10-7
Larson et al .
(1978)

Penkett et al .
(1979) as
modified by
Oasgupta
(1980a)
k4[03][HS03-] [H+]-0-l
l<4 = 4.4 x lO* *0.9 mol-0.9 s-l
at 298 K
k2[03][HS03-] + K3
[03][S032-]
k2 = 3.73 x 105 £ mol-1 s"1
k3 = 3.12 x 108 s. moT1 s'1
at 298 K
4.0 - 6.2 6 x 10-4 5 x 10"10
-2 x 10-3

1-5 0.1 - 0.4 6.6 x 10-9


-------
 in sulfite oxidation  rate  by 03 when Mn  ions were present  at  roughly 10-5
 mol  n~l.   However,  no  data or  discussion of  this result  was  given,  and
 only  recently  has a  study  of  the catalyzed 03  reaction  appeared  in  the
 literature.   This  study,  by Harrison et al. (1982), found that  Mn and Fe on
 the   order  of  ICT5  mol   Jr1 enhance  the  oxidation  rate,   though  over  a
 relatively  narrow  pH  range  centered at -4.4.   The maximum  enhancement is
 roughly a factor of 2  for Fe  and about 5 for Mn.  Given the large uncertainty
 in the  uncatalyzed 03 rate,  and that at a pH  of 5.0 the Mn  and Fe enhance-
 ments were  negligible for Fe and about  a factor of 3 for Mn at the high con-
 centration  of  10~5  mol  £-1, this  rate  will  be  considered  indistinguish-
 able  from the uncatalyzed rate already discussed.

 Oxidation  by  H202 has  only  recently  been  considered  important for  acid
 production in hydrometeors.   While early laboratory work on this reaction was
 done  by Mader (1958),  the first study relevant to the atmosphere was reported
 by Hoffmann  and Edwards (1975).   Penkett  et  al.'s  (1979)  study essentially
 repeated the  study of Hoffmann  and Edwards,  with  explicit extrapolation to
 atmospheric conditions.   Martin  and Damschen  (1981) have attempted to inte-
 grate all  extant  measurements  on  the  reaction  within  the framework  of  the
 nucleophilic displacement mechanism, first advocated by Hoffmann and Edwards.
 While this  approach  has the  advantage of  producing  both  a  simple and widely
 applicable rate expression,  it is not yet clear whether all   the objections
 Dasgupta (1980a,b) raised to the  Hoffmann  and  Edwards mechanism  have  been
 met.    However, from   the  point  of  view  of this  document,  details  of  the
 mechanism are unimportant as  long  as a  rate expression  is available that can
 plausibly  be  applied  to  the  atmosphere.    In this  regard,   the  relatively
 simple  rate expression derived by  Martin and Damschen  (1981)  is  adequate and
 appealing.  It  is


             d[S043  = k [H202] [S02.H20]                               [4-94]
                dt

 with  k   =  8.3  x   105  a moT1  s'1  at  298  K  and  an   activation  energy  of
 ~  28  kJ mol'1.

 This  expression is independent  of pH for  a constant S02 partial  pressure.
 However, as the pH of the solution  increases, less  and  less  S(IV)  in solution
 will   be  in  the   form  of  S02«H20.    Thus, the effective  S(IV)  oxidation
 rate decreases  rapidly with increasing pH.

 Before  the  above  rate expression  is employed,  the H202  concentration  to
 be used must  be determined.   Many recent  calculations  of the importance  of
 the   H202  oxidation   reaction   have employed  gas-phase  H202   concentra-
 tions of 1  ppb  or  greater (based on actual measurements) and  a value  of  the
 H202    Henry's   law    constant,   based   on   H202   vapor    pressure   data
 (Scatchard et al.   1952) taken under  conditions far  removed from  atmospheric.
 While the rather careful  extrapolations  on such data appear plausible,  they
 cannot  be applied  directly  to atmospheric  conditions.   For example,  Martin
 and Damschen calculate a  value for  the  Henry's  law constant  of  6.07 x  105
mol  £-1  at  273   K.    At  273  K,  1   x   10'9  atm  H202  is   equivalent   to
                                    4-50

-------
4.46  x  10-8  mo-|  m-3  Of  H202.    For  a  cloud  water  content  of 0.5  q
m-3,  and  assuming   all  of  the  H202  goes  into  solution,   the   resultant
concentration  would  be  only 8.9  x  10-5  mol  £-1,  close to an  order of
magnitude less than  the concentration predicted by the Henry's law  constant.
Hence,  as  was the case  for several  of the strong  acids,  the H202 concen-
tration in solution cannot be based on Henry's  law equilibrium.  Furthermore,
H202  is  reactive  in  solution  with  a  variety  of  organic   and   inorganic
species  (Ardon  1965) that could rapidly  deplete  it without producing  acid.
Kok  (1980)  found  concentrations   of  H202  in  precipitation considerably
lower  than  those predicted  for  Henry's  law  equilibrium.   Because of  this
uncertainty   in  the  value  of  the   H202  concentration  in  hydrometeors
derived from gas-phase measurements, values derived from direct measurements
of  this species in  rain and  cloud water  (Kok 1980, pers.  comm.) will be
employed.    The value  selected  is  0.5   ppm   or  ~  1.5  x  10~b   mol   x.'1.
Employing this  value in  the Martin  and Damschen  rate expression for atmos-
pheric  conditions  results  in  a characteristic time  with  respect  to  S(IV)
oxidation of 0.14  hr at  a pH of 5.0, which yields a  highly significant con-
version  rate  of 700 percent  hr"1.    Indeed,  this rate  is  high  enough  that
limitations  due to  mass transport  are  likely to be  important  for larger
hydrometeors.

The  last oxidant considered  in  this section  is  N(III)  (i.e.,  either  N02~
or  HN02 in  solution).    The reaction(s)  between N(III)  and  S(IV) species
in solution has been known for many years because it  was integral to the old
lead-chamber   process    for   producing   H2S04   (Duecker   and  West   1959,
Schroeter  1966)  and remains considerably  important in  flue-gas  scrubbing
technology  (Takeuchi et  al.  1977).    Because N0x's  and  S02  commonly co-
exist in polluted air, several recent studies  have attempted to evaluate the
possibility of a significant aqueous reaction between these two species  (Nash
1979, Chang et al. 1981).  Oblath et al. (1981) and Martin  et  al.  (1981) have
presented  explicit   rate  expressions  they use  to  evaluate  the    reaction's
significance in  the  atmosphere.  The Oblath et al. study  contains  consider-
ably more  information on the  conversion mechanism.   Furthermore,  it was con-
ducted  in  the  pH range of 4.5  to 7.0, whereas Martin et al.'s was  conducted
at  pH's less than 3.0.  On  the  other hand, the sulfite and nitrite concen-
trations employed  by Martin et  al.  were closer  to  atmospheric levels  than
were  those  used by Oblath et al.  Also,  Martin et al.'s rate  expression  is
relatively simple and easily applied to atmospheric conditions.  In  any  case,
the  two rates agree within a factor  of 3 at pH's near atmospheric.  There-
fore,  Martin  et al.'s expression  will  be  employed  as  a  significance  test.
This expression  is:


    d[SQ4  ]   =  ki[H+]l/2 {[HN02] + [N02]}{[S02.H20] + [HS03]}       [4-95]
       dt

with  ki  =  142 £3/2  mol~3/2   s"1  at 298  K.   No  activation  energy was
determined  by Martin et  al.  (nor  by  Oblath et  al.  for atmospheric  condi-
tions);  it will  be assumed  to  be negligible.   Employing this  rate expression
with  the appropriate values of  N(III)  from Section  4.3.2  yields  a charac-
teristic  time  with  respect  to  oxidation  of  S(IV)  of   70  hr  for  urban
                                    4-51

-------
conditions.   This reaction's  importance  to the  H2S04  production in  hydro-
meteors is therefore negligible.

Finally, we note that, based on their interpretation of the  data  of  Takeuchi
et  al .  (1977),  Schwartz  and  White  (1982)  have  suggested  that aqueous  N02
may oxidize S(IV)  at  a significant rate under somewhat polluted  conditions.
However,  more  work  must  be  carried   out  on   this   reaction  before   its
atmospheric significance can be assessed.

In closing this section, it should be noted  that  aerobic oxidation of sulfite
is subject to inhibition by numerous atmospheric  constituents (Hegg and Hobbs
1978).  Presumably, the  same will  be true  of the 03 reaction,  if it is  in
fact produced  by a free-radical  chain  mechanism.   Furthermore, both  03  and
H202 are  highly reactive  in  water and  can suffer  either  catalytically  or
photochemical ly induced decay.  The rates discussed do not  account  for  such
inhibition or decay.   Therefore,  in some cases these rates  may  overestimate
those in the atmosphere.

4.3.5.4   Heterogeneous Production  of HpSOg.  in  Solution—Few  heterogeneous
reactions  in  solution have  been  proposed  for H2S04 production.   The  only
such reaction that has been  studied extensively is the oxidation  of  S(IV)  on
graphitic carbon suspended in  solution (Brodzinsky et al . 1980, Chang  et al .
1981).  Before  this reaction is discussed in  detail, heterogeneous reactions
involving metal oxides are discussed briefly, prompted by the fact that  many
trace metal  catalysts commonly  invoked for  homogeneous  oxidation of $032-
occur in relatively insoluble  form in the atmosphere.   Heterogeneous  oxida-
tion processes  involving trace metals could therefore be of  some  importance.
Certainly,  gas-solid   heterogeneous  reactions  involving   trace  metals   are
treated  extensively  in  the  literature on   atmospheric  $042-   production
(Urone et al .  1968).  However,  in  solution,  only  one  such  reaction  appears  to
have been examined:   the study by  Bassett and  Parker (1951)  of  the oxidation
of  S032~  to  H2S04 by various  oxides   of  Mn.   While  not  a   quantitative
rate study,  this  work  suggests  that substantial  H2$04 can  be produced  by
this reaction  relative to aerobic  oxidation, at least for  high concentrations
of metal oxides.

Recent  modeling studies of  the  heterogeneous carbon-sulfite  reaction  have
concluded that this reaction may  play an important role  in  sulfate  production
in water films  on  atmospheric  particles (Middleton et al . 1980,  Chang  et al .
1981).  Both studies  emphasize that  the  reaction  would  require  quite  low  pH
solutions  and   a  long reaction  time to  be competitive  with  other  sulfate
production mechanisms.  The  rate  expression of Brodzinsky et  al . (1980)  is
employed  to  evaluate the  significance  of  this  reaction  for  H2$04  pro-
duction in atmospheric hydrometeors:

    dS(IV)  *  k [Cx] r.02]°'69          a[S(IV)2] _             [4-96]
      dt                        (1  +B[S(IV)]  +ct[S(IV)]2)
where  k  =   1.69   x  10'5  mol -03  £0.69   g-l  s-l§   a  =   1.50  x   1012
jr  mol -2,    0  =  3.06  x   106   a  ml-1,  [Cx]  =   grams  of  carbon   per
                                    4-52

-------
 liter,  and [02]  and [S(IV)]  are  in molar  concentrations.   The  activation
 energy of  the reaction is given as 48 kJ mol'1.

 It  should  be noted that the graphitic carbon used to derive Equation 4-96  was
 Nuchar C-190, a commercial product with a well -character!' zed elemental  compo-
 sition  and BET  surface  area  (550  m2  g-1).    However,   soot  generated  in
 various combustion processes (i.e., combustion of acetylene,  natural  gas,  and
 oil)  was  also employed.   Chang et  al . (1981)  report  an average  Arrhenius
 factor  five times larger  for  these soots than  for Nuchar-90.   This  higher
 value  will be employed  in these  calculations.   Another   novelty  concerning
 Equation 4-96  is that it  is  nonlinear in [S(IV)] and  therefore has  charac-
 teristic times that  are  functions of  the  concentration of S(IV).    Finally,
 use of Equation 4-96  requires  an estimate  of the graphitic carbon  concentra-
 tion  in  hydrometeors.  A  recent  direct measurement  of elemental  carbon  in
 rainwater  collected   in  Seattle  that   was  2.4  x 10-*  g £-1  (Ogren  1980)
 has been  employed.    All  of the  elemental  carbon is  assumed to  act as  an
 efficient  catalyst.

 Assuming a  temperature of 278  K, a cloud water pH of 5.0, and an urban S(IV)
 concentration  in  solution of  7.9 x  10~5 mol £-1,  the  rate expression  of
 Brodzinsky  et al .  yields  a  characteristic  time  for  S(IV)  oxidation   of
 ~  10^ hr.   Therefore,  this  reaction  should  be  of  little  importance   in
 H?S04  production  in  precipitation,  although  it  might  be  important   in
 fogs of low liquid water content in urban areas.

 4.3.5.5     The  Relative   Importance   of   the   Various   H^SjU   Production
 Mechanisms—In sharp  contrast to  HC1   and  HN03  production in  hydrometeors,
 numerous  reactions  are  capable  of  producing  significant levels  of  H2$04
 in  solution.   It  is  therefore  important to  assess the relative  magnitudes  of
 these  reactions  under differing  atmospheric conditions.    To  do  this,  two
 relatively extreme cases that can produce precipitation  are considered.

Much has been made of production of  acid in  mists and fogs,  which  is  of some
 importance  from   the standpoint   of  S042"  production  in  the atmosphere.
 However, it is of  little consequence  to acidic  deposition  because  even a
modestly precipitating cloud will  deposit far more  acid  on  the ground than
 will a fog.  As an example of a "polluted"  case,  a low-lying  stratus cloud  in
 urban  air  with a  liquid water  content of  -  0.1  g nr3   (about  the  lowest
 liquid water  content that can  produce precipitation  in a  warm  cloud)   is
considered.    H2S04  production  by  02  (catalyzed  and  uncatalyzed) ,   by
03,  and  by  h^Og oxidation  of  S(IV)   in  solution  is  considered.    Values
of the various parameters to  be employed are given in Table 4-16.  The  value
 for the  partial  pressure of 03  is based on  numerous measurements in  urban
air, the  concentration  of HgOg  is derived  from Kok's  (1980) measurements,
 and the cloud  water  pH  range  is based  on measurements reviewed by Falconer
and Falconer  (1979).    The mechanisms  considered have different  pH  depen-
 dencies, so the production rates  over the pH  range of polluted clouds must  be
considered.
Figure 4-4  plots  the production rates for  the various oxidants.   The
reaction  dominates  ^$04 production  in  polluted clouds,  with  the possible
                                    4-53

-------
            TABLE 4-16.  VALUES OF PARAMETERS USED TO ESTIMATE
                  H2S04 PRODUCTION IN A POLLUTED CLOUD
            Parameter                                   Value
Partial pressure of H2$04                          1
Partial pressure of S02                            50 ppb
Temperature                                        288 K
Cloud liquid water content                         0.1 g m~3
pH of cloud water                                  3.5 - 4.5
Partial pressure of 03                             100 ppb
Concentration of H202                              4.7 x 10'5 mol
Concentration of Mn                                10~6 mol  £-*
Concentration of Fe                                10"6 mol  ir1
                                    4-54

-------
                                                  PRODUCTION  RATE OF  H2S04   (Mole i
                                                                                           -1   -1
       IQ
       c
CJl
CJ1
     3 CO
    <£1 O
     ro -^

     O T3
     -h 1
       o
    -a Q.
     o c
     — ' n
c+ O
(D 3
Q.
  -S
O Q>
— • rt
O  -h
•  O
  -s
       O)
       O
       C
       to

       o
       X
       _J.
       Q.
       a>
       t/>

       o
       ro
       -s
       ro
       -a
             re
             o
             oo
             o

-------
exception  of  the upper  end  of the  pH range  (where  the rather  speculative
mixed-catalyst rate becomes comparable to  that of H202).

We next  consider a more typical mid-level  cloud (at the  ~  800-mb .pressure
level) with  a more  substantial  liquid water content  of  ~ 1  g m~^,  situ-
ated in  a  moderately  industrial  region.   The  parameter  values  used in  this
case are listed  in  Table 4-17.  The pH range  is  again derived  from Falconer
and  Falconer   (1979)  and  the  H202 concentrations  from  rainwater measure-
ments by Kok  (1980).   The metal  concentrations were estimated  by  employing
typical  (rather than peak) metal  concentrations in clear air, divided by  the
cloud  liquid   water  content given  in Table  4-17,  using  the  same  percent
solubilities  as  previously employed.   The  resultant low  metal  concentra-
tions preclude consideration  of  catalytic  oxidation by Mn or  Mn plus  Fe.
Because some experimental support exists for  Fe-catalyzed oxidation at  these
levels (Brimblecombe and Spedding 1974),  it  is considered here.

Figure 4-5 plots  the  rates for  the oxidants considered.   While the  H202
reaction again appears to be the single most  important reaction  over much of
the pH range,  the most striking result revealed  by  Figure  4-5  is that all of
the  oxidants  can  contribute  significantly  to H2S04  production  above  a  pH
of -  5.2.   Of course,  this result is quite  sensitive  to the  concentration
of  H202   employed;  further  data  on  this  important   parameter  would  be
highly desirable.   Nevertheless,  it is important to  note that,  on  the  basis
of available  field data  and  rate  studies,  no one  oxidant  dominates  H2S04
production in all atmospheric situations.

Another important point that can be addressed with the aid  of Figures  4-4 and
4-5 is the time  scale  for  acid produced  in  solution  to  reach  the concentra-
tions produced by direct absorption of gases into cloud  drops.   This question
was approached in  the  derivation  of the  S(IV) conversion rates  necessary to
produce  significant acid in  solution.  However,  Figures 4-4 and 4-5  allow a
more precise estimate.

The  maximum  concentration of  directly absorbed  H2S04  in  an  urban polluted
cloud   should  be  -  4.2   x   10'4  mol   
-------
           TABLE 4-17.  VALUES OF PARAMETERS USED TO ESTIMATE
                  H2S04 PRODUCTION IN A MID-LEVEL CLOUD
        Parameter                                          Value

Partial pressure of H2S04                             0.1  ppb
Partial pressure of S02                               5  ppb
Temperature                                           278  K
Cloud liquid water content                            1  g  m~3
pH of cloud water                                     4.5  - 6.0
Partial pressure of 03                                40 ppb
Concentration of ^02                                 5.9  x 10~6 mol
Concentration of Mn                                   2  x  10~9 mol 2.
Concentration of Fe                                   3.3  x 10~8 mol
                                  4-57

-------
                                                                                                 -1  -1
                                                         PRODUCTION RATE  OF  H2$04 (Mole  Jf   s  )
en
CD
                                en
                             euro
                             ^ oo
                             to o
                             o -a
                             -h -S
                                o
                             3 Q-
                             -•• c:
                             Q. O
                              I  C+

                             fD O
                             < 3
                                o>
                             O rt-
Q- -h
(/)  O
•   -S
                                O
                                C
                                to
                                o
                                X
                                3
                                rf
                                O

                                fD
                                -S
                                n>
                                                  o

                                                    vo
                                                  o
                                                   (»
                                               CT>
                  -p.
                  •

                  00
                  en
             -o

             o
                  en
                                               en
                                               •
                                               01
                                               en
                                               •
                                               00
                                                                                              n>

-------
   10
     -3
d)
"o
    10
      A
     '4
                                                   ABSORBEDH^Sp.
o
o
o
o
    10
      -5
                                                  — -TOTAL H2S04
                                                  PRODUCTION  IN SOLUTION
                                         H2S04 PRODUCED IN SOLUTION
                                             BY H202 ALONE
       0123456
                                    TIME (Min)
   Figure 4-6.   Time dependence  of H2S04 production  in  an  urban  polluted
                cloud (cloud water pH  =  4.0).
                                      4-59

-------
                                                                                      -1
                                                  CONCENTRATION OF  H2$04 (Mole £   )
CTi
O
        C

        fD

        -^

        ~~J
     O ->•

     O fD

     Q. O.
        fD
     £ -a
     OJ (D

     fD Q.
     -S fD
        3
     XJ O
     3: n>

      II O

     en
       •LO
        o
        o
        a.
        O
        Q.
        I

        ro .

        ro
        o

        o
        c
        Q.

-------
range  listed  in  Table  4-17,  oxidation  by Q£  and (h  produces  sufficient
additional  H?S04  to reduce  this  time to  ~ 1  min.   These results  suggest
that  not  only the  rate, but  also  the pH  dependence  of the  H2S04  pro-
duction  in  solution,  will   depend  on the  H202  concentration  and  the  pH,
because  these two  parameters  determine  how  much of  the  H2S04 produced  in
solution  is  due  to the  non-pH-dependent  H202  reaction and  how much to  the
other highly pH-dependent reactions.

One  final  point is  suggested  by Figures 4-6  and 4-7.   The rates  shown  in
these figures  produce  substantial  quantities of  acid  in a  relatively  short
time.  Furthermore, a major  component  of  this  production  is a  pH-independent
reaction  (H202  oxidation)   that  will  not   be  self- limiting   in  the  usual
sense  of  the  term.   If absorbed  concentrations  of  H2S04,  HMOs,  and  HC1
are considered as well, within a few minutes of cloud  formation, cloud  water
pH1 s in  urban  air  might be expected to reach  a  value  of 2.0 or even  lower.
Because such low pH's are not observed and because the  anion levels predicted
by direct  absorption and the rates shown in Figures 4-5 and 4-6 are similar
to those observed in urban precipitation  (Larson et al . 1975, Liljestrand and
Morgan 1981), acid neutralization must play  a role.

4.3.6  Neutralization Reactions

4.3.6.1   Neutralization  by  NH3--Probably the most  important single  neutral-
ization  process  Tn  the  atmosphere  is the  absorpti on-hydrati on  of NH3  by
acid aerosols  and  hydrometeors  and, in the case  of hydrometeors,  the  subse-
quent dissociation reaction

     CNH3]gas + CH20]liq = [NH4OH]

                [NH4OH]  = [NH4+]  +  [OH-].                               [4-97]
The  preeminence  of  this  neutralization process  arises because  NHs  is  the
only basic gas of widespread, substantial occurrence in the atmosphere.   The
hydration and  dissociation  reactions are generally assumed  to be fast  com-
pared to acid production reactions in solution (Scott and  Hobbs  1967,  Beilke
and  Gravenhorst  1978).    Therefore,  the   concentration  of  NH3  (and  OH-)
consequently is given  by  the equilibrium expressions  for NH3  absorption  and
dissociation in solution.

This appears  to  be the case even  for  the  fastest of  the reactions shown  in
Figures  4-3  and  4-4.    For  example,  the  H202 reaction   in  urban   air
produces  -   2.3  x   10~6   mol   5,"1    s'1   of   ^04,  or   9.6  x  10-18
mol  s"1  in  a  10 pm  radius droplet.    If  a background  concentration  of
NHa  of  1 ppb  (Levine  et  al . 1980)  is  assumed,  the  rate  of NH3  scavenging
due to collisions with a 10  pm droplet  will  be 8.25 x  lO"15 mol  s*1.

Recent work  by Huntzicker  et al .  (1980)  suggests that the  reaction coef-
ficient  for the collisions  will  be close to unity for  acidic  droplets 10  ym
in radius.   In this case,  the  collision frequency becomes the  rate  of  NH3
delivery  to  the   droplet.    The  NH3   is  hydrated  virtually  instantly  in
solution, and  the product   ammonium  hydroxide  (Nh^OH)  dissociates  with  a


                                   4-61

-------
rate  constant of  kd  =  6  x  105  s-l  (Eigen  1967).    Thus,  after  ~  10-6
s,  the  rate  of  OH"  production  equals  the  collision  frequency  and  NH3
neutralization will  not  be transport limited.   It is therefore possible  to
estimate  the  NH4+   concentration  (and  the  associated  OH"  concentration)
in solution from equilibrium considerations,  even for  these  fast  reactions.

When  the equilibria are  employed  for  an  NH3  solution,  NH40H  dissociation
and  water dissociation,  the  concentration  of  NH4+   in  solution  is given
by


     [NH4+] =   Ha pa Ka  ^+]                                         [4.93]


where  Pa  is  the  partial   pressure  of  NH3,  Ha  the  Henry's  Law constant
for  NH3,  and Ka   and  Kw  the  equilibrium  constants  for NH40H  and  H£0
dissociation,  respectively.

Recent measurements  of  ambient NH3 concentrations range  from  0.5  to 25 ppb
(McClenny  and Bennett  1980,  Levine et al.  1980).   While the  values for  Ka
and Kw are well  known,  recent work by Hales and Drewes (1979) has suggested
that  the  commonly  accepted  value  for  Ha  of  55  mol  a~^-  atnr1  at 298  K
is too  high by  about a  factor of  - 5  for  atmospheric hydrometeors  (due  to
interaction between  dissolved NH3  and  CO?  at atmospheric concentrations).
When  this  is  taken  into account,  the  NH4"^ concentration at 278 K is given
by
     [NH4+] * 3.3 x 1011 Pa [H+].                                       [4-99]

This yields  a  range  of NH4+  concentrations from  1.65  x  10"4 to  0.8 mol
£  .   Thus,  1.65 x  10~4  to  0.8 equivalent  of acid could  be neutralized
by NH3  alone.   However, a word  of caution  is in  order.   While concentra-
tions of  NH4+  found in cloud  water  lie toward  the  lower  end of this  range
(Petrenchuk  and  Drozdova 1966, Sadasivan 1980,  Hegg  and Hobbs 1981a), most
rainwater  samples  have substantially  lower  NH4+  concentrations  than are
predicted  by the above calculations  (Lau  and  Charlson  1977).   While this
discrepancy is well known,  it remains unresolved.

4.3.6.2   Neutralization by Particle-Acid Reactions—Reactions  between  strong
acids produced  in  hydrometeors and  particles incorporated  into these  hydro-
meteors by scavenging (either nucleation or  below cloud  scavenging)  are well
known.    But these  generally have  been considered  from the  standpoint  of
initially alkaline droplets produced  from,  say,  sea  salt  nucleation  acidified
by absorption or  production  of strong acids  (Robbins  et al.  1959,  Eriksson
1960,  Hitchcock et  al. 1980).    The  initial   "alkaline"  salt for  such   a
reaction is predominantly NaCl.

However,  the widespread occurrence  of CA2+  in rainwater and  the  fact that
calcite   (CaCOs)   and   dolomite  (CaC03-MgC03)   are   often  substantial
components of  the atmospheric  aerosol  have  led to  the assertion  (Winkler
                                    4-62

-------
1976)  that  these  minerals  will   act  to  neutralize H2$04  in  hydrometeors
via the substitution reaction:

          CaC03 + H2S04 = CaS04 + H2C03.                                [4-100]

The relative weakness of carbonic  acid ensures  that  this  reaction  produces  a
net decrease  in  acidity.   Certainly, CaS04 has been measured  in significan|
quantities  in  urban atmospheres (Sumi et  al.  1959, Kasina  1980), and  Ca^
and Mg2+  are known  to be important  components  of  the ionic  precipitation  in
the United States (Chapter A-8).  Therefore, observational  support exists for
this  idea.    Indeed,  Sequeira (1981) recently  found that excess Ca  in  pre-
cipitation  (in excess  of  that attributable  to sea  salt and  thus of  soil
origin) correlates  much better with excess  sulfate  than does NH3, and  that
Ca and Mg  concentrations  in  precipitation  are  often more than sufficient  to
offset  observed  S042~  loadings.    Sequeira  also  suggests   a   role   for
calcium oxide  (CaO)  derived  from  fly  ash  as  well as  for  CaC03  and  MgC03.
The interesting  point  about  these three  minerals  is their low solubility  in
water  (e.g.,  compared  to sea  salt) and  their  increasing  solubility  with
increased  acidity.   They may,  therefore,  act  as  hydrometeor buffers  in the
atmosphere, much like  NH3.   The absolute amount of  Ca  and Mg available for
such  buffering is  highly  variable,  with Ca  ranging from 10~7  to 10~4 mol
jT1 and  Mg  fairly  uniformly a factor of  5 to  10 lower  in both  rainwater
and cloud water (Petrenchuk and  Drozdova  1966,   Hendry  and Brezonik  1980,
Sadasivan  1980,  Liljestrand  and Morgan  1981).   Clearly, Ca,  at  least, can
substantially contribute to acid neutralization in  hydrometeors.

4.3.7  Summary

The  three  acids that  dominate  the acidity   of  precipitation  are   H2S04,
HN03, and HC1, in decreasing order of importance.   The methodology employed
to assess  the  importance  of  their formation within clouds and rain has  been
to compare  the  solution  concentrations  of  these  acids  produced by  direct
absorption of  their respective acidic vapors  from the  gas phase  with  those
generated by plausible solution reactions over the lifetime of the  cloud and
raindrops.   If  an  aqueous-phase  reaction  produced  solution  concentrations
comparable to  those  resulting from  absorption, the  reaction was  considered
significant.    In   cases  where  several   reactions  were  found   capable of
producing  significant  concentrations of  a  particular  acid,  their relative
importance has been evaluated.   Finally,  because the  potential acidity of
precipitation  far  exceeds that  commonly observed,  plausible aqueous-phase
neutralization reactions have been examined.

4.4  TRANSFORMATION  MODELS (N. V.  Gillani)

4.4.1   Introduction

Secondary  products  of  chemical  transformations  of  SOX and  NOX  emissions
are generally more  acidic than their precursors.   In the context of acidifi-
cation of  lakes, vegetation, and  soil,  however,  the chemical  form in which
the deposition arrives  at the surface is  of relatively little  significance
(because precursor  depositions  are rapidly  converted to the secondary  forms
following deposition) compared  to  the  fact that the rate of the  deposition


                                    4-63

-------
process  itself  depends strongly  on  its chemical  form.   Thus, for  example,
sulfate  particles  are  believed  to  have  a  considerably   longer   average
atmospheric residence  than  S02,  and hence  a larger range of  impact.  Nitric
acid,  on  the  other hand, is  likely  to be  removed  from the atmosphere  more
efficiently and  rapidly  than  its precursors.  Consequently, it is necessary
for transport-deposition models to distinguish between primary and secondary
pollutants, and  to facilitate  atmospheric  chemical transformations  through
appropriate modules.

The  chemical   transformation  module  is  an  integral  part  of  the   overall
transport-transformation-removal   model.    The   framework  within  which  the
larger  model  is  formulated  and  solved may  be  Lagrangian  (trajectory),  or
Eulerian (grid), or some hybrid  scheme  (details  in  Chapter A-9).   Lagrangian
or trajectory models simulate the changing concentration  field  within a given
polluted  air  parcel  (e.g.,  a  puff  or plume release)  as  a  result of  the
combined effects  of dilution, chemistry,  and depositions.   Typically,  the
concentration field  as well as  meteorological  variables are  assumed to  be
homogeneous within  the air  parcel.   Recent  attempts have  also been  made  to
obtain  simulations with  finer  spatial  resolutions within  the  air  parcel.
Lagrangian models  are  tailored  for  simulations  of pollutant kinetics  at the
plume  scale.   Regional Lagrangrian  simulations  are commonly based on  simple
linear  superpositions  of  individually-calculated concentrations  of  multiple
plumes.   Individual  plumes may  be  referred to  as  point  sources  or  area
sources.   For the  modeling  of nonlinear  processes in multiple  interacting
plumes over regional scales, Eulerian grid models are more  appropriate.   They
are based  on  the  solution  of  coupled  transport-transformation-removal  mass
balance  equations  of individual  species   over  specified  two-   or  three-
dimensional spatial grids.  Typical  grid  sizes  in regional  models vary  from
50 to  100 km  to  a  side.   Within  each grid cell,  pollutant  concentrations,  as
well  as  meteorological variables, are  assumed to be  uniformly distributed.
In a  pure grid  model, emissions  within a  grid cell  are  considered in  an
aggregate  sense,  and  are instantaneously  homogenized  over  the  entire  cell
volume.   The  error of  this approximation  is  particularly  severe  in  two-
dimensional grid models  which  lack  vertical resolution.   The  effects  of
sub-grid scale processes are sometimes  included  in  terms of  bulk  parameteri-
zations.  Alternately, a  hybrid scheme  may be used in which  individual  plumes
may  be modeled  in a  Lagrangian sense and  detail  until   they  acquire  the
spatial dimensions  of  the Eulerian  grid  size,  and subsequent simulation  is
within the Eulerian framework.   The  output  from  a  grid model  is  an  evolving
series of snapshots of the  deposition  field over the entire modeled  region.
This is clearly  very  desirable  in regional  modeling.   Grid models,  however,
require far more extensive input information, computations,  and computational
resources  than   trajectory  models,  and  are generally  quite  expensive  to
implement.  The  chemical  transformation module  does not depend,  per  se,  on
the framework of the  larger model formulation.   However,  its validity  does
depend  on  the  spatial-temporal  resolution  of  the  simulation,   and  on  the
facility for  accommodating  nonlinear processes  and plume interactions  with
its  chemically  different environment.    The remainder  of this  section  is
focused on the transformation module.

An objective  of this  section  is to review  and  assess  briefly  our  present
ability to predict the rates of chemical transformations  of  primary emissions


                                    4-64

-------
of  SOX  and  NOX  to  secondary  acidic  products  (sulfates  and  nitrates)
during atmospheric  transport.   Such predictions are based  on  transformation
models,  which  are  mathematical  formulations  relating  secondary  pollutant
formation  rates to  concentrations  of  the precursor  gases  (e.g.,  S02,  NO),
and to any other chemical and meteorological  factors considered to contribute
to  the  transformation processes.   The  principal  approaches  in  formulating
such  models  are discussed for  S  and N compounds, for power plant  and urban
plumes,  and  for  each of  the  major conversion  mechanisms believed  to  be
important.   Specific  formulations of practical  interest  are reviewed briefly
along  with  their  applications,  and  major   outstanding  problem  areas  are
identified.   An overall  assessment  is  presented of our present  standing in
terms of the desired goals of transformation  modeling.  Emphasis is  placed on
formulations  believed to be  suitable for inclusion  as transformation modules
in  current long-range transport-transformation models  aimed  at  simulating
regional-scale acidic depositions.

The atmospheric transformation processes are  very complex,  involving multiple
parallel  pathways  (mechanisms)  of   physical   diffusion and  homogeneous  and
heterogeneous chemical  reactions of a  wide  variety  of  reactants  and  cata-
lysts.  The reactants may be of  primary  or background  origin or intermediate
or  secondary  products of concurrent reactions.   A variety  of  meteorological
factors—UV radiation, temperature,  relative humidity, clouds, fogs,  atmos-
pheric turbulence,  and  others—also have important influence  on  atmospheric
transformation processes.  Many of these factors are interdependent; e.g., UV
radiation, temperature,  clouds,  and turbulent mixing are closely related to
insolation.   Furthermore,  a given  factor may  simultaneously  have  opposite
effects on different chemical reactions;  e.g., the effect of plume dispersion
should  be  to "quench"  reactions between  co-emitted  species  (Schwartz  and
Newman 1978), but also  to  promote reactions  of  primary  emissions with back-
ground species  (Wilson  1978, Gillani and  Wilson 1980).   Given the  complex
array of reactants and their reactions  influenced in  a  complicated  manner by
interdependent environmental  factors, one must  recognize that  no single  and
simple mathematical  expression  can describe  adequately the  transformation
processes of  a  given pollutant.  Realistic  transformation  models  should be
capable of distinguishing among  the  different conversion mechanisms and,  for
each mechanism,  should  reasonably reflect the  dependence  of  the conversion
rate on current plume, background, and  environmental conditions.

Historically, the science of transformation  modeling  is young.  As recently
as 1977, the  state  of the  art was  such  that in a widely acclaimed  regional
monitoring  and   modeling program,   the  conversion  rate  of  S02  to  S042~
was represented  by  a single constant number over a regional  scale, regardless
of time of day, season, or prevailing meteorological conditions (OECD  1977).
Even today, such practice  is not uncommon in  regional  models, perhaps with
some justification.   Since 1977, however, significant progress  has  been made
in  developing  transformation  modules  appropriate  for   regional   models,
particularly for the gas-phase  mechanism of S conversions.   Applicable models
for  the  liquid-phase  mechanism  are  still   rare  and  primitive.    Current
transformation  models for   N  compounds  are  generally complex,  requiring
extensive computational  resources even  for  mesoscale  applications.    Their
validations are  limited.
                                    4-65

-------
4.4.2  Approaches to Transformation Modeling

Basically  two approaches  to  transformation  modeling  exist—a  fundamental
approach and an empirical  approach.

4.4.2.1  The  Fundamental  Approach—The  fundamental  approach consists of  the
so-called "explicit mechanisms method"  and its simplified counterparts.   In
theory, the explicit mechanisms method involves  consideration of all  signifi-
cant reactants and  their  elementary reactions involved in each mechanism  of
sulfate  or  nitrate formation.   Concentration  changes by  all  chemical  re-
actions are calculated simultaneously for all  species  at short-term  intervals
(typically a few seconds).  Reactants include not only  the  precursors (e.g.,
S02,  and NO),  their  principal  oxidizing  agents  (e.g.,  OH,   HOo,  and  R02
in  the  gas-phase   mechanism,  and  02,  03  and  H202  in   the   liquid-phase
mechanism),    and   the   secondary   products  of  concern  (e.g.,  H2S04   and
HNOj)  but  also catalysts and  significant intermediate  species involved  in
the mechanisms.  Of particular significance are the multitude of  reactive  HC
species  and  their derivatives involved in  gas-phase  chain  reactions   that
contribute to photochemical  smog  formation, as well  as to sulfate  and nitrate
formation.  In a  spatially  homogeneous  system (well-mixed plume) consisting
of  n  species, a total  of 2n  first-order,  nonlinear, ordinary  differential
equations must be  solved  simultaneously at each  time  step  to  evaluate  the
changing species concentrations in  the plume and in  the background with  which
the plume interacts.   Plume-background interactions  must  be  facilitated  in
the model.  If spatial  inhomogeneities are important and need to be  resolved,
the system of equations becomes substantially  larger.  Also,  because a  wide
range  of  reaction-time scales are  generally  involved, computations for  the
equations'  solutions at each  time  step  are  quite  involved, time-consuming,
and expensive.

Implementation of  the  explicit mechanisms method  has many associated  prob-
lems.  The  list  of possible reactants is  long, and sometimes  there is  even
disagreement  about what  the  products  are  in given individual reactions.
Values  of  many  elementary  reaction  rate  constants   have  either  not   been
measured or are  not quite reliable.   Model  input requirements also include
specification of initial concentrations  of all  species  in  the  plume and  in
the  background.    While  primary  emissions  from  major  point  sources  are
reasonably  well  characterized,  area source  emissions  are not.    This  is
particularly  true  for  the  hydrocarbons.  Also, the spatial-temporal resolu-
tion of the current area source emissions inventories  is generally inadequate
to  verify model  performance based  on the  available mesoscale   field data  of
power  plant and  urban plume  transport  and  transformations.     Atmospheric
measurements  are  either rare  or  nonexistent  for  many  short-lived  species,
some  of crucial   importance  (e.g.,  OH,  H02,  R02,  and  H202).     Detailed
HC  and  aldehyde  measurements  in   the  atmosphere  are  not common.    Input
specifications  and  model   validations  are  thus  only  partial   and   very
approximate.

Perhaps the best example of an attempt to simulate  smog chemistry  by explicit
mechanisms is  the  work of Dernerjian et  al.  (1974),  which incorporated  more
than 200 species, the great majority of them arising from the explicit use  of
specific  reactive  HC  and  corresponding  organic  intermediates  and sinks.


                                    4-66

-------
Despite  this  model's comprehensiveness,  the authors warn that  its  represen-
tation of the real atmosphere, which undoubtedly contains hundreds of organic
compounds, may  be an oversimplification.   Such complex  chemical  modeling  is
currently impractical  for  application in  regional  models.    Simplifications
and further approximations are necessary.   The key is to achieve a reasonable
condensation  of  the  vast number of HC and  aldehydes,  and their reactions,
while adequate  representation is maintained.'  Such  condensation is  attempted
either  by  "lumping"   groups  of species  by  some  common  criterion  and  then
treating each group  as a single species  in the model,  or by substituting  a
single surrogate species either for all  HC (e.g.,  propylene  by Graedel  et al.
1976, "nonmethane HC" by Miller et al. 1978) or for a particular lumped group
of  HC  (e.g.,  xylene for aromatics,  by Hov  et  al.  1977).    Two  principal
methods  of  "lumping"  have  been developed:    the  HSD  method (Hecht et  al.
1974), and the carbon bond mechanism (CBM) method (Whitten et al. 1980).   In
the HSD  method,  organic species of like  reactivities  are grouped  into  four
main classes: paraffins, aromatics, olefins, and  aldehydes.  Many models  use
a modification  of this  in  which  the following  six lumped  classes  are  used
after Falls and  Seinfeld  (1978)  and Falls et  al.  (1979):  ethylene,  higher
molecular  weight  olefins,  paraffins,  aromatics,  formaldehyde,  and  higher
molecular weight aldehydes.   In the CBM method, similarly bonded  C  atoms  are
lumped  into  four or  more classes.   In principle,  the  CBM  is  closer  to  the
explicit mechanism and is  also easier  to use  in conjunction  with measured
data than  is the  HSD mechanism.    Such formulations have been   further  con-
densed  in  specific  simulations  by reducing  the  number  of  species modeled
through  the  use of  surrogate  reactions and  rate coefficients  which  effec-
tively include the role of the omitted species (Levine  and Schwartz  1982).

Validation of simulations performed by detailed chemical  models  has,  to date,
been generally  based  on  matching   calculated  concentrations  of certain  key
aspects  of  photochemical  smog  formation  (e.g.,  HC  loss,  and OH  or  03
formation)  with  those measured  in controlled  smog chamber  studies  in  the
laboratory.  The roles of such meteorological variables as sunlight,  tempera-
ture, and  relative  humidity  are  simulated  directly in  the experiments  and
included in  the  calculations through the  dependence of  elementary  reaction
rates on them.  The role of other meteorological  variables such  as turbulence
and inhomogeneous  mixing generally is  not  simulated  in laboratory experi-
ments.  This  is probably a serious  limitation.

In  the   real  polluted atmosphere,  the deficiency  of certain  key   reactive
ingredients in a primary emission may  well  be overcome  through entrainment  of
such ingredients  from the  background air.   The formation of ozone and  sul-
fates in HC-poor  power plant emissions in  the eastern United  States  during
summer  afternoons is  thus  almost  as rapid  as  in  HC-rich  urban  emissions
(Figure  4-8  on  p. 4-73; also  see Gillani  and Wilson  1980).    Appropriate
background characterization  and treatment  of plume-background  interaction can
be of critical importance in realistic modeling of transformation processes.

An important  positive  feature of detailed chemical  models is that  nonlinear
chemical couplings between  species, including the  coupling between sulfur and
nitrogen chemistry,  is explicitly  retained.    In  this  sense,  the same model
can, in  principle, perform  simulations of  SOX and NOX  transformations,  as
                                    4-67

-------
well as  of  urban or power  plant  plume chemical evolution. With  appropriate
spatial-temporal  resolution, the  effect  of plume-plume and plume-background
interactions can also be performed.

One of  the  major undesirable features of  the detailed chemical approach  is
the necessity of performing extensive computations.   However, considerable
differences exist in amounts of computation necessary,  depending on choice  of
numerical method and degree of chemical approximations involved.  The number
of  species  in the  chemical  schemes  commonly used varies between 10 and  100.
The amount of computations increases nonlinearly and rapidly with increasing
number  of  species.   For  any given  chemical  scheme  of smog simulation, the
main  numerical   problem  arises  from  the   fact that  the  various  chemical
reactions occur  at  speeds which vary  by  several  orders  of magnitude.   This
wide range  of time  scales involved  in this problem  makes the corresponding
set  of  differential   equations   quite  "stiff."    Standard  techniques  for
integrating  sets of differential equations (e.g.,  the  Runge-Kutta Method)
cannot  provide  stable solutions  of such  stiff systems  at  realistic cost.
Special techniques such as those  developed by Gear (1971) provide much  more
efficient  numerical  integrations  by  performing  automatic  time  and error
control, and are capable of providing  accurate numerical  solutions, albeit  at
considerable cost and  requiring the use  of large high-speed computers.  The
Gear technique  has  been  used  widely  in simulations  of  photochemical smog.
Other  attempts  to reduce  computations have  resorted  to  the  use of quasi-
steady-state assumptions for certain very  reactive  species.  Such  assumptions
are not always  justified and have  been  shown to lead to  large inaccuracies
not only under  polluted conditions  but also  in relatively clean background
conditions  (Farrow  and Edelson 1974, Dimitriades  et  al.  1976, Jeffries and
Saeger  1976, Hesstvedt  et  al.   1978).    Judiciously  invoked steady-state
approximations  (QSSA),  based  on  continuous  monitoring  of  pollutant  time
scales  during on-going  simulations, can  permit locally analytical solutions
(Hesstvedt  et  al.  1978)  and  even locally  linearized   analytical solutions
(Hov 1983a).   Such numerical techiques can provide  solutions comparable  in
accuracy  to the Gear  solutions  at  a  fraction of  the  cost,  and  can  be
implemented on smaller computers.

Examples of specific  detailed  chemical  model  calculations  for  atmospheric
applications are considered in Section 4.4.4.1.  A recent review paper by Hov
(1983c)  is  also recommended  for those  interested in further details  per-
taining to the fundamental approach  of transformation modeling.

4.4.2.2  The Empirical  Approach—Given the  substantial  uncertainties and  gaps
in  the  input information  needed  for detailed chemical  models, and given the
discrepancies in reported transformation  rates of SOX  and  NO^,  the use  of
detailed kinetic models continues  to  be  questioned,  and simpler empirical
rate expressions are often favored.   A great deal of  experimental  research  on
chemical transformations  has been  directed  at obtaining estimates  of the
conversion  rates of SO?  to  sulfates, and  of NO  to N02  to nitrates in the
laboratory and in the field.  In recent years, some success has been achieved
in  relating field  estimates of the  conversion rates to  specific  conversion
mechanisms and to specific, measured influencing factors.  A  large number  of
source-related and  environmental  factors have been implicated  as  influencing
transformations.   They include the time  and  height of  source release, the


                                    4-68

-------
nature  and  amounts of  the acid  precursors,  other  co-emitted species,  the
reactivity of  the  air mass in  which  emissions are  transported,  as well  as
such  meteorological   factors  as  sunlight,  temperature,  absolute  humidity,
clouds and fogs, and atmospheric stability.

In  the  empirical   approach,  an  attempt  is  made  to  identify   the   rate-
controlling  factors   for  each mechanism  and  to  formulate  and  validate  an
overall  rate  expression  for  measured  sulfate or  nitrate  formation by  each
mechanism directly  in terms of these  factors,  which are also measured.   In
other words, the effect of the multiple elementary reactions  is parameterized
in terms of pertinent, measurable chemical and meteorological  factors.   Such
parameterizations of  the conversion rate  are  simple rate expressions,  which
can be  inserted  directly  into regional models as  the  transformation module.
They entail  very few  computations and  require inputs that are, for the  most
part,  relatively  readily  available even  on  a regional  scale.   In spite  of
their simplicity, they often yield quite reliable  estimates  of actual  atmos-
pheric  formations of  such  final  products  as sulfates.   This is particularly
true  when  their  formulation  is  based  directly  on  field  data  and   their
application is based  on measured  input data.   Their  principal  disadvantage
is that they lack generality,  being   applicable  mainly  under environmental
conditions reasonably close to  those  for which  they have been successfully
validated.   In specific  applications  for which relevant parameterizations are
available,  their simplicity and reliability make them immensely valuable.

The  existing  empirical  parameterizations of  sulfur  chemistry  are largely
based  on  mesoscale  plume   data.     At  least   three  important   practical
implications of this  limitation may  be identified.  First of  all,  given the
dominance of  source-specific  characteristics  in  mesoscale  plume  transport,
empirical parameterizations which are  mesoscale in origin must be  considered
to  be  specific  to  the   source   type (e.g.,   power plant  plumes   versus
urban-industrial  plumes)  for  which they  were developed.  Secondly, because
the characteristic  time scales  of the  atmospheric  residence of  secondary
pollutants  such  as  sulfates  and  ozone  are   significantly longer  than
mesoscale,  it must be presumed that the parameterizations for plumes would be
sensitive to boundary conditions.   In  fact,  empirical  observations  have shown
that sulfate and ozone formation rates  in  power  plant as  well  as urban  plumes
are strongly sensitive to  the chemical condition of the background air, and
to the rate of plume  dilution by entrainment  of this background air (Gillani
and Wilson 1980, Miller and Alkezweeny 1980).   Plume-background interactions
can sometimes even obscure  the initial chemical distinctions between a power
plant  plume  and  a petroleum  refinery  plume  (see Figure  4-8).  Finally, one
must question the validity of empirical parameterizations of  mesoscale  origin
in synoptic scale applications.   On  the positive  side,  however, it has  been
demonstrated  empirically   that  pollutant  plumes  evolve  to  the chemical
maturity characteristic of regional  air  masses within  only a few hours  of
transport during  sunny  convective  conditions  typical  of summer  days   in the
eastern  United  States  (Gillani  and   Wilson  1980).   At  least  under  such
conditions,  chemical  parameterizations  derived  from data of  chemically-aged
plumes may have validity even during  long  range.
                                    4-69

-------
The reactions governing S02 oxidation have the general  form

      S02 + Ox + (M) ->  products -* S042-,                               [4-101]

where Ox  represents  the  principal  oxidizing agents;  i.e.,  OH and  possibly
H02  ana  R02  for  gas-phase  oxidation  (Calvert  et  al .  1978),   and  H202,
03,  and 02  for liquid-phase  oxidation   (Penkett  et al .  1979);  (M)  repre-
sents the catalysts,  if and  when  any   are  involved.   With  the  possible
exception  of  catalyzed  reactions   (Freiberg  1974),  the  rate  of  sulfate
formation, rs, may be expressed as:


     rs = — (S042-)  =  ks • (S02),                                     [4-102]
          3t

where  the fractional  conversion  rate,   ks,  depends  on  Ox,  the  oxidizing
species.   Such  a relationship  is  valid  as long as  S02 is not in  stoichio-
metric  excess.   The  validity  and  linearity of  this  equation  are  further
discussed  in  a separate  section  (Section 4.4.3).   Parameterization of  ks,
which is  the  goal  of  empirical  transformation models,  is  thus a  represen-
tation  of  the weighted contributions of  factors which  effectively  determine
the Ox concentrations.   It may be broken  down by mechanisms into:

     ks = kS  + k$  + ks,                                          [4-103]
where  components  on  the  right  hand  side  represent,  respectively,  the
fractional  conversion  rates  by gas-phase,  liquid-phase, and  heterogeneous
aerosol  surface  reaction  mechanisms.    No   parameter!' zations   have   been
attempted  for  the  heterogeneous  mechanism,   partly  because  reliable  and
particular  atmospheric  data  are  lacking  and  partly  because  the  mechanism
generally  is  not  considered  important on  the  regional scale.    Specific
parameterizations  of  S   conversions   are  most  developed  for   kSg,   and
efforts  to parameterize  ks.  have  just  begun.   These  are discussed in  the
next section.

Similarly,  the  formation  of  the two  principal  secondary  nitrates  (HN03  and
PAN) are largely governed by the reactions

     N02 + OH -»- HN03                                                 [4-104a]

and  N02 + RC002 ->• PAN.                                              [4-104b]

Hence, their formation rates may be expressed as:

     rHN03 = kHN03 • (N02)                                           [4-105a]

       rPAN = kPAN ' (MOz).                                           [4-105b]
where  the fractional  conversion  rates, k^  (N =  HN03,  PAN), depend  on the
concentrations  of  OH  and  RC002,  respectively.    The parameterizations  of
ku   would  represent  the  weignted  contributions   of   the  factors  which
                                    4-70

-------
effectively  determine  these free  radical  concentrations.   Empirical  param-
eterizations of kN  based  on field data have  not  been formulated or  tested.
Sensitivity  of k^  to  the  HC-NOX  mix  has  been  studied  in  smog  chamber
experiments.  Some of the most recent specific results and their  implications
will be discussed in a later section.

4.4.3  The Question of Linearity

A much debated matter, and  one  of  considerable  practical importance  in terms
of regional modeling and  control  strategy,  is the question  of linearity  (or
lack of  it) in the source/receptor relationships  between  emissions of  SOX
and NOX and  their depositions.   An important subset of this larger  question
pertains  to the  linearity  of  relationships between  rs   and  S02,  and  r^
and NOX.    In  this  section, the  discussion  is limited  to the  question  of
linearity  of the  chemical  transformation  processes.   If the transformation
chemistry  is  nonlinear,  certain  common   modeling  practices  based  on  the
assumption  of  linearity must  be viewed with caution.  For  example,  regional
models typically have a  spatial  resolution over grids of 50 to  100  km to  a
side.  The assumption of uniform  species  concentrations  within a grid  cell
that includes  concentrated  emissions  sources  may  give erroneous transforma-
tion estimates unless some  appropriate  parameterization of  sub-grid  scale
processes is included.  Distinctions in the chemical mix of different  source
emissions are  also  presumably  important in the case of nonlinear chemistry.
Linear superpositions of  species  concentrations,  calculated  for individual
plumes assumed to  be isolated,  will  also  give erroneous  estimates of  non-
linear secondary formations  in regions with multiple plume interactions.   The
validity  of the  linearity   assumption  is  also crucial  to the  success  of
attempts to control  secondary pollutants by a strategy of linear rollback  of
precursor emissions.

The lack of consensus on  the question of linearity,  particularly  with respect
to  sulfur  chemistry,  is  probably due  to  different  interpretations of  the
definition of  the term linear relationship.   By definition, the  relationship
between rs  and S02  is linear  if it can be  stated  in the  form  of  Equation
4-102,   and  if  ks  1S   independent  of   S02.    Clearly,  k_   is   variable
through its  dependence  on  species,  such  as  the  OH free radical,  that  are
responsible  ultimately for  the  oxidation  of S02«   Therefore,  the  critical
question is  whether these oxidizing  agents are themselves dependent on  SOo.
There is  no doubt  that  in a fresh  plume with high  concentration of S02»  OH
level  is  significantly controlled  by S02  itself,  and the  oxidation of  S02
is a nonlinear process.   Such conditions,  however,  are short-lived.   Subse-
quently,  if  there  are no further  fresh injections  of SOo  into  this  plume,
the formation  of  OH will  be  governed by  the HC-NO*  chemistry  in  the  plume
and by entrainment  from  the background of OH itself and of  other  reactive
species contributing to OH  formation.  The  direct  dependence of plume  HC-NOX
chemistry on local  S02  concentration  is  very weak  in  this stage  of  plume
transport.   Consequently,  one  commonly finds  in  the  published literature
explicit or implicit statements  about linear  sulfur chemistry  under  such
conditions.  If the mathematical  definition of linearity is  to be interpreted
strictly,  such  statements  are correct within the context of the transport  of
a particular plume  release.   In the  broader context of modeling of  longer-
term averages  or  continuous  emissions,   possibly  varying  with time,   and


                                    4-71

-------
with  inevitable  plume-plume and plume-background  interactions,  an indirect
form  of nonlinearity  does exist  because  of the  correlation  between  SO?
emissions  and  the  co-emissions  of  NOX  and  HC.    A broader  definition of
linearity which requires  ks to be independent not only of SC^  but also of
co-emitted species  is  implicit in the works  of  Cahir et  al.  1982 and  Hidy
1982.

The significance of the role of the co-emitted species is  illustrated  in the
following  practical  example.    Suppose   we  wish  to  answer  the   following
question:  "Will  a 50 percent reduction  of S02  emission  from  source A (or
region A)  result  in  a  corresponding  50 percent decrease in downwind  sulfate
formation?"  There is no  unique answer to  this question.   First,  the  manner
in which the emission  reduction  is achieved is important.   If source  A  is  a
coal-fired power  plant, and the  reduction  in S02  emission  is  achieved by  a
50  percent reduction in  the  amount  of  fuel  burned,  there may  also be an
accompanying  reduction  in NOX emissions  which,  in  turn,  will  cause ks to
be different.  The answer  to  the question,  therefore, is "no".   The cause of
this apparent or  effective  nonlinearity  is  the  indirect dependence of ks on
S02  through   the  correlation between  co-emitted  S02  and  NOX.     The 50
percent reduction in S0£  emission  could  also have been  achieved by  the use
of  fuel  of 50  percent lower  sulfur content  or  by scrubbing  SO? from the
combustion products prior to stack  emission.  To  the  extent that  these latter
procedures may  not  have   changed  NOX emissions,  kc will  remain  unchanged
except during initial transport, and  the downwind  sulfate  formation would be
expected  to  decrease by  about 50  percent, all  other  conditions being the
same.  The answer to the question is  therefore "yes".
A second factor that will profoundly  influence  downwind  sulfate  formation  is
the  composition  of the  air  that the  plume encounters during mesoscale  and
long-range transport.   Field evidence  suggests  that  the role of  co-emitted
species may  be substantially enhanced,  or overwhelmed,  by  the  role of  the
background  air which  the plume  entrains  by  mixing  processes.    Like  the
co-emitted species, a polluted background can also serve as  the source of the
oxidizing agents.  Figure 4-8 shows an  example  of the  side-by-side transport
of  two St.  Louis  plumes of very different  emission  composition, yet  com-
parable  secondary  formations.     The  Labadie  power   plant   emission   is
characterized  by  a very  low HC/NOX  ratio.   The  urban  plume of  St.  Louis,
including the  emissions  from a  large  petroleum  refinery  complex,  by contrast
is much richer in reactive HC emissions.  The secondary formation of ozone in
large  plumes on summer  days  is  closely related to the  formation  of sulfates
(White et  al.  1976, Gillani  and Wilson 1980).   The  formation of  ozone  and
sulfates in  power plant plumes  at rates comparable to those  in  urban plumes
is  due to  the entrainment  of  polluted background  air.  During  long-range
transport, the role of  the background air may well predominate as a source of
reactive  species which oxidize  S02-    In laboratory  measurements with  no
role of a  variable  background,  a first-order dependence of  sulfate formation
on  SO? concentrations   has  been observed for  gas-phase reactions  (Miller
1978)  as well  as for liquid-phase reactions (Penkett et al.  1979).  Mesoscale
field  measurements  are  also  generally  consistent  with pseudo-first-order
dependence between rs and S02, except during early transport.
                                    4-72

-------
                CO
                C


                fD
                CO
                                                               CONCENTRATION  (ppb)
 CT C
 Cu  ~*3
 3  fD
    CO
T3  •
    fD
 CL 3
 CU  O
•a  cu
             
             to
O  fD
3  -5
f+ <-+-O O
3" -S —' -5
fD CU CU O
   <
<-+• fD
-s -s
CU CO T3
< fD —'3
fD    CO.
-5 <-*• 3
Co 3" fD "d
fD -S    -S
   O CU O
S C 3 -(,
cu CQ CL -••
       3  rt- 3 w
       Cu  3-
       Q. (T> el- O
       fD     =r -t,
         -a ro
       cu  — i    -a
       -s  c c -s
O  3
c:  fD
3  co
Q.
    cu
ro
O
o cu
          -s
          cr
          a>
                -s
                <<
                oo
       00-30
       •   o
ro  c
c*t* ^3

Cu
   "O

1— > fD
t£) to
-~j n>
CO 3
-- "C+
•   CO

    CU



    ro
    ~~4
    ai
    O
    cr -fi
    co  tn
    fD  O

    <  3
    ro
    fD
CO CL

   CO
   m
r~ n
o o
C 3
-J. Q.
CO CU
n       <<
o  —)

"O  fD  3" O
Cu     fD CO
-j -a    --^
Cu  —i CL
cr c  cu -a
     • fD
    ft>
      ' fD
       <
          rt- O
       _  CU —i
       co     —i

          5S
          -5 CU
    -*>

    fD
    CO
 .sn.
fD  -S  fD
—'  fD     r+
CO     n  CO
    -s  o
O  fD  —i -"•
-h  —' —i 3
    fD  fD
fD  Cu  O  rt-
X  CO  <-h 3-
O  fD  fD  fD
fD  O. CL
CO        I—
CO  CU  Q. CU
    -s  c  cr
O  O  -S  Cu
N  C  -"• CL
O  3  3  -••
3  cud  ro
fD
    3  cu -a
c+  3  a.    s
3-     Q. CU fD
ro  cr cu  -"• -s
    O <<  -5
Cu  rt >••  O
CQ  3-    -S
fD        CU
Q-        -h
                        ro
                        o
                 O
                 73
                 O
                 CO
                 •z.
                 a
                    o
                    HH
                    CO
                    O
                    m
o
                            00
                            O
                            O
                            O
                                    »—O
                                    UD
                                    cn
                                                             S
                                    ro
                                    o
                                    o
                                    o
           ro
           o
           o
           en
        m

        o
                                    ro
                                    o
                                        ro
                                        o
                                        i—*
                                        en
                                                                           o
                                                                           o
                                                                           en
                                                                           O
                                                                                                            ro
                                                                                                            O
                                                                                                            o

-------
A common practice in detailed models of sulfur chemistry  in  the  atmosphere  is
to  represent  the  S02  +  OH  reaction  as  a   terminal  reaction,  effectively
leading  to  the  formation  of  H2S04  and   depletion  of  the   OH   radical
concentration.    This  dependence  of  OH   on  S02  therefore  contributes  to
nonlinearity of the  sulfur  transformation  process.   It has been  pointed out
recently (Stockwell and Calvert  1983)  that the S02-OH reaction may  initiate
a chain  of reactions  which may  lead not only to  formation  of H2$04, but
also to  regeneration of  OH in the  presence  of NOX  and  hydrocarbons.   Such
recycling of OH would have the effect of weakening  the nonlinearity of sulfur
chemistry.    An  important  conclusion  of  the  recent  MAS  report   on  acid
deposition  (NAS  1983)  was  indeed  that nonlinearity  of  sulfur chemistry  is
probably  quite  weak,  and  that  even  the  broader  coupling  between  sulfur
emissions and depositions may be substantially  linear on a  long-term average
basis over the  spatial  scale of eastern North  America.

Based  on  theoretical  considerations,  the  relationship  between  r^  and NOx
is  expected  to be  nonlinear,  because kjj depends on  OH,  for example,  which
depends  directly  on the  NOx  chemistry.    Results  of  recent  smog   chamber
experiments  suggest,  however,  that  the  nonlinearity  of  rjj  may   also  be
short-lived relative to the time  scale of long-range  transport (Spicer 1983).
Pseudo-first-order  parameterizations  of r^ may be justifiable,  but kw may
also need  to  reflect the make-up  of the air  an  emission  encounters  during
transport.

4.4.4  Some Specific Models and Their Applications

4.4.4.1   Detailed  Chemical Simulations--Detai1ed  chemical  modules   based  on
the explicit mechanisms approach  have been used within  Eulerian as  well  as
Lagrangian formulations,  and in model applications  at the plume  scale as well
as  the regional  scale.   Such transformation modules  differ  principally  in
terms  of their  representations of  the hydrocarbons,  and in the  methods used
for  the  numerical  solution of the  set of nonlinear differential equations
describing  the  species concentration changes  by  chemical  reactions.   The
following discussion outlines some specific representative models and  is not
intended as an  extensive review of chemical models.

The LIRAQ model (McCracken et al. 1978, Duewer  et  al.  1978) is  an example  of
a  two-dimensional   grid  model  (single  well-mixed  vertical   layer).   The
transformation  module attempts to simulate  photochemical  smog  formation based
on the HSD scheme (Hecht et al. 1974), and  the numerical  solution is  based  on
the  Gear technique.   The  SAI  Airshed  Model  (Reynolds et al.  1979)   is  a
three-dimensional   grid  model   which  permits  initial  isolation  of   elevated
point  sources  from surface sources.   It  uses  the carbon-bond mechanism  of
photochemical  smog simulation  (Whitten and  Hogo 1977),  and numerical  solution
is  by  a finite  difference technique  (SHASTA)  developed by  Boris   and  Book
(1973).  An ambitious  three-dimensional  regional  grid model  currently  under
development  at  EPA (Lamb  1981)  presently  uses  the  chemical scheme  of
Demerjian and Schere (1979) which uses  four hydrocarbon classes  of different
reactivities.    In  some  regional  models  (e.g., McRae et  al.  1979),  point
source plumes are simulated in a Lagrangian sense within the  framework  of  an
Eulerian grid  network  until  they attain  the  dimensions of  the grid  cell.
Thereafter, the simulation is  continued in  the Eulerian  frame.


                                    4-74

-------
 On  a global basis,  the  troposphere  is presumed to be clean  and  the organic
 species most relevant  to  smog  formation  are  carbon monoxide (CO)  and methane
 (CH4).   Recently,  a two-dimensional  global  model was  employed  by  Fishman
 and  Crutzen  (1978)  to  predict  the  global  distribution  of OH,  H02,  and
 CH302  radical   concentrations.   Predicted  OH  concentrations  were  reason-
 ably comparable with recent, measured atmospheric concentrations (Sheppard et
 al.  1978).   Altshuller (1979)  used  this  model  for OH to investigate the var-
 iability of the sulfate formation rate by the homogeneous gas-phase mechanism
 with respect to latitude  and altitude.  His results showed that in the clean
 enviroment, OH  is the principal  oxidizing  agent,  and that,  at higher lati-
 tudes, e.g., in the  northeastern  United  States,  Canada,  and northern Europe,
 large seasonal differences  in  sulfate formation by this mechanism  are to be
 expected.   Very  little  sulfate formation is  likely in winter by  gas-phase
 mechanisms.

 The regional model of Carmichael  and Peters  (1979)  is  based on  the  chemistry
 of  a clean  background in  which  the  only  organic species  are  CO  and  C02-
 They invoke the pseudo-steady-state  assumption  for the oxidizing  species OH,
 H02,  H202,  and   03,  and  use  their iterative  solution  for these  species
 in first-order expressions  for the  oxidation of S02 to  estimate  the sulfate
 formation rate.

 Most  plume simulations  are based on  trajectory-type  models.  Calculations
 made for polluted industrial  regions and urban  areas  have simulated certain
 observed  phenomena  related particularly  to  03  behavior  (Graedel  et  al.
 1978) but at the  same time have yielded conflicting results concerning impor-
 tant control strategies.  Results by Graedel et  al.  (1978) suggest  OH levels
 to be  directly  proportional to N02  levels,  implying  that reduction  of  NOX
 emissions would help  control  nitrate and sulfate  production.  Miller (1978)
 showed rather  that NOX emissions tend to delay S0£  oxidation  and that  the
 ratio  (NMHC/NOX)   of  initial  concentrations of nonmethane  HC's  and  N0x's
 dominates  the  S02  oxidation   rate.    Miller's  conclusions  were   verified
 experimentally.   Actually,  as suggested by Miller  (1978),  precursor  effects
 may  significantly  differ  in   the   first  several  hours  of  daytime  plume
 transport from their effects during  subsequent regional  transport.

 Detailed chemical  calculations also  have been applied to  simulate  sulfate and
 nitrate formation  in urban plumes  (Isaksen et al. 1978, Miller and Alkezweeny
 1980, Bazzell and  Peters 1981)  and in power plant plumes  (Miller et  al. 1978,
 Bottenheim and Strausz 1979, Levine  1981,  Hov  and Isaksen  1981, Stewart  and
 Liu 1981).   In  these calculations,  proper simulations of  the changing back-
 ground air  and of plume-background  interactions were necessary for at least
 qualitative  agreement with field  observations.   Levine  (1981)   neglected
 plume-background  interactions  and,  as a  result,  his  conclusion  that  power
 plant  plume  dilution  inhibits  sulfate  formation  is   contrary   to  field
observations in moderately  polluted  regions  (Gillani  and Wilson 1980).   Hov
and  Isaksen  (1981),  on  the other  hand,  treated  crosswind spatial  inhomo-
geneities in  sulfate formation resulting  from   plume-background interaction
and  succeeded  in  simulating,  at  least qualitatively, many  features of  the
crosswind plume data  of Gillani and  Wilson.   Stewart and  Liu (1981)  similarly
provided cross-wind  resolution  and  plume-background interactions  with  their
reactive  plume model  which  was based  on  the carbon-bond  mechanism for  the


                                    4-75

-------
 simulation  of chemical  kinetics.   Recently,  Hov  (1983b) performed  a plume
 simulation  in which vertical  stratification  of the concentration  field  was
 considered.    In  general,   plume  simulations  have  indicated  that  03  and
 aerosol  formation  are  greater when the background is polluted, that OH is the
 dominant oxidizing  species  for  S02  and  N02,  and  that  OH  and  peroxy
 radical  (H02,  R02)  concentrations,  which  play  an  important role  in  03
 formation, peak at midafternoon in polluted regions.

 In  all of the above simulations, only the homogeneous gas-phase chemistry was
 included.    Rodhe  et  al.   (1979)  added  reactions  of  S02  and  N02  with
 H202  in the  presence  of  "clouds" to  a  highly  lumped  gas-phase  chemistry
 model.   H202  generation was  calculated  based on  the gas-phase  reactions.
 The  authors  recognized  qualitatively that the effective rate  constants  for
 cloud  reactions  must  include  not   only  the  effect  of  the  liquid-phase
 transformations occurring in cloud droplets  and in  precipitating  clouds,  but
 also  exchange rates of  the  reacting species  between  the  droplets and  the
 surrounding  air,   and  the  frequency  and  occurrence of clouds and  precipi-
 tation.  They then proceeded to choose rate constant values  such that overall
 gas-  and liquid-phase  oxidation  rates of  S02  became  comparable  and  the
 liquid-phase  oxidation of N02  became  relatively  insignificant compared  to
 its  gas-phase counterpart.    This procedure  for  the liquid-phase  mechanism
 represents  a  highly parameterized  approach, with  parameter values  assumed
 rather  subjectively.    Their  calculations  were  applied regionally  to  the
 European industrial environment under  summertime conditions.   The  relative
 contributions of gas-phase and liquid-phase mechanisms to  sulfate  and nitrate
 formation,  of course,  reflected their assumptions.

 4.4.4.2  Parameterized Models—For many years,  no consensus  could  be  reached
 concerning the  relative importance of  the many chemical  and  meteorological
 factors  implicated  as  influencing   gas-to-particle  S  conversion.    Most
 transport-transformation models  used constant  pseudo-first-order  rates  for
 the  oxidation  of  SOg.   Documentation  of  sunlight  as a dominant  environ-
 mental factor governing  sulfate  formation  in power plant plumes (Gillani  et
 al.  1978) has since been verified  and widely  accepted and  used.   In par-
 ticular, in a recent review  of field  data on  sulfate formation in  power plant
 plumes during all  seasons in  the United  States,  Canada,  and Australia,  Wilson
 (1981) observed that the outstanding  common  pattern in this  broad  data base
 was  the diurnality  of  the  sulfate  formation directly  related  to  solar
 radiation.    Such  a role of sunlight  is  also  consistent with the  observed
 distinct summer  peak  in regional  S042~ distribution in  the eastern  United
 States (Husar  and Patterson 1980), even though corresponding S02  emissions
 are distributed fairly  uniformly over  all seasons  (U.S.  DOE 1979).

 A  sunlight-dependent   model   of  the   form  ks   *   Rj»  the   total   incoming
 solar  radiation   flux  at ground  level, was  used  by  Gillani  (1978)  in  a
 diagnostic  mesoscale plume model  and by Husar  et  al.  (1978)  in  a multiday
 plume S  budget  study.    A similar  parameterization  has  been used by  Shannon
 (1981) and  by others.   Gillani  found  that such  a model based  only on  sunlight
 could not  simulate the  observed day-to-day  variation  in sulfate  formation.
 Evidently,  factors other than sunlight must be  included.  Also,  the manner  in
which  sunlight  influences  the conversion  process  must  be  more carefully
considered.   As Wilson (1981) noted, observed correlations of the conversion


                                   4-76

-------
rate  with  sunlight, or  with  air temperature (Eatough  et  al.  1981), do  not
imply the direct  role of these  factors in the underlying mechanisms.   These
two  factors  are  highly  correlated,  as  are   both  to  turbulent  mixing,
convective cloud  formation, and a number of  other  factors, which  alone  can
exert   rate-controlling   influences   on   specific   conversion   mechanisms.
Accordingly,  formulation of meaningful  parameterizations  must  be based  on
mechanistic considerations.

Gillani  et  al.  (1981)  recently advanced a  parameterization  of  the  gas-to-
particle  S  conversion   by the  gas-phase  mechanism  based  on  plume   data
collected during  the summer in the Midwest  (Missouri  and Tennessee).    The
motivation  for  their  gas-phase  parameterization  was  derived  from  their
earlier  identification of a recurrent pattern  of  03 and aerosol  generation
in  power plant plumes,  which  evidently  involved  participation of  reactive
species entrained from the background  (Gillani and Wilson 1980).   Gillani  et
al. argued that accelerated photochemical  generation of the radical  species
OH,  H02,  and  R02   that  oxidize gas-phase  S02  would be  facilitated  by
reactions involving  NOX  emissions,  entrained reactive  HC,  and free  radical
species.  Consequently,  the quality  of the background air  and  the  extent  of
plume dilution  by its  entrainment were  judged  to  be important  contributing
factors, in  addition to  sunlight which  powers  the  photochemical  reactions.
Given  the  lack  of  detailed  data  of  the  oxidizing  species,  the  authors
resorted  to   using  03  as a surrogate for,  or an  indicator  of,  air   mass
reactivity.    Vertical  plume  spread, Azp,  was  chosen  as  a measure  of  the
extent of plume dilution.  The  resulting  gas-phase  parameterization  is:

     kSG-(.03 +_ .01)RT  • (AZ)P  •  (03)0,                               [4-106]

where   kSG   is   in   percent    hr-l,   Rj  is   in   kW  m~2,   (Az)p  is   in
meters,  and  background   ozone,  (03)0. is  in ppm.    The coefficient  0.03 +_
0.01 was chosen on the basis of  the  best  fit between  the calculated  (Equation
4-106)  and  measured  values of  kSQ.    The  measured  values  were  for  dry
(relative  humidity  <   75  percent),  cloudless  conditions when  gas-phase
reactions may  safely be  assumed  to predominate.   The parameterization was
validated successfully  by data collected  in  the  plumes  of three  large  central
power  generating  stations in Missouri and  Tennessee during  two  different
summers.  The empirical   coefficient  (0.03) thus pertains to such  large  power
plant plumes in which the initial  NOX/S02 ratio  is about 1:3.

The  above  parameterization is  believed  to  provide  good   estimates  of  the
gas-phase sulfate  formation rate under  the  moderately  polluted conditions
characteristic of the eastern United States  in summer and appears  to be  valid
even under more polluted  conditions  during stagnation episodes.   Its validity
in winter, even  in  this  region,  remains  to be  tested.   Its  performance  in
clean regions such as the Southwest, and  in extremely polluted  areas  such  as
Los  Angeles,  CA,  on a   smoggy  day  is   also unproven.    Furthermore, the
parameterization has no  validity  for urban plumes  and  possibly also plumes
from small  power  plants  owing to  substantially  different composition of the
emissions.     In  spite  of these  restrictions,  the  parameterization is  of
practical significance.    Its   input requirements   are  minimal  and  can  be
satisfied presently over a regional   scale in  the eastern United States.  Its
explicit inclusion of plume-background interactions  and air mass conditions


                                    4-77

-------
probably  gives  It some  validity  even during  long-range  transport when  the
role  of  the  background  is  expected  to  be dominant.   Application  of  the
parameterization  based  on  1976 St.  Louis,  MO,  data  of the input  variables
yields the  diurnal   and  seasonal   pattern  of  kSG as  shown in  Figure 4-9.
The  magnitudes  and  temporal  variations  shown  are  plausible and  consistent
with  available  field data, as well  as  with expectations  based on  detailed
chemical   calculations  (Calvert  et al. 1978, Altshuller 1979).   The  results
predict that  in the  Midwest, gas-phase mechanisms may  be expected  to  convert
about  10  to  20   percent  of  the  SOg  in  a   power   plant  plume  to  S042-
during an average summer day, while corresponding conversion in  winter may be
about an  order  of magnitude  smaller.   By comparison,  measured  values  of  S02
to S042"  conversion  by all mechanisms range between   15 and 35 percent  for
summer conditions  in the same region (Gillani  and  Wilson  1983).    It may  be
inferred, therefore, that  liquid-phase mechanisms may convert about 5  to  15
percent of the SOg to S042' per day during summer in the Midwest.

Gillani et al.  (1983) have recently also made  a first attempt to  formulate a
parameterization  of  liquid-phase  S042~   formation   resulting   from  plume-
cloud interactions.   The formulation explicitly recognizes  that  the  overall
conversion  rate,   ks, ,   depends   not  only  on  the  chemical reaction rate
within  cloud  droplets,  K$L,  but  also  on  the  physical  extent  of  plume-
cloud interactions.  Because" clouds are discrete entities  in space  and  time,
and  plume-cloud interactions  are  somewhat random events,  the authors  choose
to describe  plume-cloud interactions in  probabilistic terms.   The  overall
formulation has the general form

     kSL = P  • K$L                                                    [4-107]

where P  represents  a measure  of  the  probability and  extent of  plume-cloud
interactions.  All three quantities in the  equation are time dependent.   The
dependence  of  P   on local  plume  and cloud   dimensions  has  been  derived
explicitly  (details  given  in  original   reference),  and  its  values  are
determined  during an actual  power plant plume  model  run based on current,
calculated  plume  dimensions  and  local  cloud  data   from  surface  weather
observations  of the  National Weather  Service network  of stations,  as  well  as
on local  lidar and  aircraft measurements.  P represents  a measure of  the
fraction of a given plume volume which is  in contact with the liquid phase.

The  authors  did   not  attempt  to  parameterize K$,.    It  depends on such
variables as  liquid  water concentration, droplet pfl,  and  concentrations  of
dissolved  S,  oxidizing  agents  (HgOg,   03,  and  Og),  and catalysts  (Fe
and  Mn).   No data  were  available for such  cloud chemical  composition.  The
authors  did,  however,  obtain  an  average  daytime estimate for  K$,   under
typical summertime fair-weather convective  cloud  conditions  in  the Kentucky-
Tennessee  area.     The  inferred  value   of  K$,   (summer  daytime  average
conversion  rate  within  clouds)  was  12 ± 6  percent hr"1.    This  value
compares  with  values  of  0  to 104  percent hr"1  estimated by  Hegg et  al.
(1980),   based  on   ambient   S02   and  S042~   measurements  in   wave  cloud
situations  and  with predicted values ranging  from  10  to  20 percent  hr'1  in
large storm  cloud systems in  the summer  based  on  an  indirect mass  balance
technique (Scott  1982).  Also, the value of P averaged over 24 hr is expected
to be significantly  less than 0.1  during summer  as well  as winter.   In other


                                    4-78

-------
        IQ
        C
        -s
                                            AVERAGE SULFUR CONVERSION  RATE  (%  hr-1)
a>
cu  n
— ' CU
oo
   o
  ^ c
  > —
   O)
     o a
     o -••
     3 c
     < -s
     fD 3
     -S CU
     CO —i
i—> —" O
i-« -•• O
        3
        a.
  -o
   3
   CO
   fD
   O_

   O
   3

   rt-
   CU
     co cu

     —' O
     -h 3
     £= CU
     -S —•

     CTT3
     << CU
        rt-

     CU fD
     (/> ~s

     ~a co
     3-
     Cu O

     fD
   rt fD fD
   fD O
   -S 3-lQ
   —J* QL) &)
   N 3 CO
   CU ->•  I
   rt- co rt-
   -'•30
   O CO
   3 >.  T3
   CO    CU
     CU -S
   D" O rt-
  «< O -"•
     O O

   ^2 ^
   -i. O- fD

   —i 3
   O) CD
                                                                  a
                                                                  f*

-------
words, the  average  bulk  plume  conversion rate by liquid-phase mechanisms  is
likely to be less than the  local  droplet-phase conversion rate by more  than
an order  of magnitude.   All of these estimates involve  several  assumptions
and approximations  and must be used  with  caution.   Values  of  K$i  at  night
and in winter  are  believed  to  be substantially smaller as a  result of  lower
concentrations of  the photochemically-generated  oxidizing  species,  03 and
H202.

Based on the above parameterizations and St. Louis,  MO,  data,  it  is estimated
that the 24-hr average,  overall  sulfate formation  rates  in July  are likely  to
be 0.8  ^ 0.3  percent hr-1  by  gas-phase  reactions  and at  least 0.4 +_ 0.2
percent  hr'1  by   liquid-phase   reactions.     Winter  rates  by gas-phase
reactions are  estimated  to  be  an order of magnitude smaller than in  summer
and by liquid-phase reactions are estimated to be comparable during  the two
seasons.

A variety of  empirical data suggest that liquid-phase conversions in  wetted
aerosols may be significant  at  relative  humidity  between  75 and  100  percent
(Dittenhoefer and de Pena 1980, McMurry  et  al .  1981).   Winchester (1983) has
formulated  the following  empirical  parameterization of  ks which  highlights
the role of absolute humidity and temperature:
where  PH2Q  denotes  the  partial  pressure  of  water  vapor,   and  Pn?0,sat
denotes  the  saturation  vapor  pressure  of  water  vapor  (a  measure   of
temperature).

No  comparable  parameterizations  of  NOX  transformations  have  been  formu-
lated.    Summertime  plume   measurements  suggest  that N03"   formation   is
primarily in  the  form of  nitrate vapor (Forrest et al . 1979, 1981;  Hegg  and
Hobbs 1979b;  Richards et  al .  1981)  and that  oxidation of NOe  to HNOa  may
proceed  about  three  times  faster  than  does  oxidation  of  S02  to  H2S04
(Forrest et  al . 1981,  Richards et  al . 1981).   Gas-phase  mechanisms of HN03
formation are believed to  predominate  in the  summer.

Whitby  recently  used  a  simple model  assuming the  total  accumulation mode
aerosol  formation  rate to  be directly  proportional  to  UV radiation  intensity,
to  simulate  observations  of aerosol  formation in  the  St. Louis, MO,  urban
plume of 18 July 1975.  He  estimated  that  about 1000 tons of secondary fine
aerosol  may be produced in the St. Louis plume in one summer irradiation  day
(Whitby 1980).  For the  same plume transport, Isaksen  et  al .  (1978) used  a
detailed  chemical   model  to  simulate  the measured  data  of  03  and  504?-
formation presented  by White  et al .  (1976)   and  estimated  peak  H2S04  and
HN03  formation  rates of  5  and 20  percent hr-1,  respectively,  to occur  in
the early  afternoon.  Alkezweeny  and Powell  (1977)  also measured  the  St.
Louis plume  and estimated  afternoon  $042- formation  rates  to be  10 to  14
percent  hr-1.    Miller   and  Alhezweeny  (1980)   measured S042'   formation
rates in  the Milwaukee urban  plume,  particularly related  to the  quality  of
the background air mass, to  range from 1  to 11  percent hr~l.
                                    4-80

-------
Spicer (1977a)  estimated  the  N02-to-products  transformation  rate in the Los
Angeles urban plume  as 10 ±  5 percent  hr-1.   In  more recent measurements
downwind of  Los Angeles  (Spicer  et al.  1979), the  observed  lower limit of
NOX conversion  rates  ranged  from 1  to  16  percent hr-1, with  typical  rates
in  the  5   to   10   percent   hr-1  range.     Spicer  (1980)  estimated  NOv
transformation/removal  rate for  the Phoenix urban  plume  to  be  less  than 5
percent hr-1, while  data  for  Boston showed rates  in the 14  to 24 percent
hr-1  range.   Transformation  products  of   NOX  transformations  include  not
only  inorganic  nitrate  (e.g.,  HNOs), but also  organic  species (e.g., PAN).
Spicer attributes  the low conversion rate in  Phoenix  at least  partly to
thermal decomposition of PAN  and its analogs at  the  high ambient  temperatures
of the desert area.

Recently,  Middleton  et al. (1980)  performed  a model  study of relative amounts
of  sulfate  production  in  wetted  aerosols  in  a  polluted  environment by two
different mechanisms: condensation  of S02  gas-phase oxidation products, and
catalytic  and  noncatalytic   S02  oxidation  in  the  liquid  phase.    The
microphysical vapor  transfer  to  the aerosols  and  the  chemical  conversion
within the  aerosols  were  treated as  coupled  kinetic  processes.   Concentra-
tions  of  the oxidizing species  (e.g.,  OH, and  ^02) and of  the catalysts
(e.g., Fe,  Mn,  and  soot)   were  assumed  known,  and representative values for
day and night and summer  and winter  were used.   The study  concluded that in
the  daytime, photochemical  reactions  and liquid-phase  oxidation  by  ^0?
are likely  to predominate, with  particle acidity playing a minor  role.   At
night, sulfate  production  rates  are low, being  principally by catalytic and
noncatalytic  liquid-phase mechanisms  involving  03   and  02-    The  daytime
H202 reaction rate was enhanced bv the lower winter  temperatures.

4.4.5  Summary

Transformation models can, at  best,  be  only as good  as our understanding of
the transformation processes.   Significant  gaps  in this  understanding remain,
particularly  with  respect  to  the   physical  and chemical  kinetics  of  the
liquid-phase processes.  The  validity and extrapolation  of laboratory results
to  real atmospheric  conditions  are often questionable.   Field measurements,
in general, are insufficient, particularly for wet conditions.  For example,
simultaneous  physical  and chemical  measurements pertaining  to plume-cloud
interactions are almost nonexistent.

Detailed chemical models  are  not yet practical  for application in regional
models to predict acidic  product formation  and deposition.  Many individual
pieces of information—mi crophysi cal  pathways and  chemical  reactions—must be
put together  correctly  and we are  still  struggling to  assemble an adequate
information  base about   the  individual  pieces.    To   complicate  matters,
important couplings exist between the different major mechanisms of sulfate
and  nitrate  formation  (e.g.,  H202  formed by  gas-phase  photochemistry is
of   paramount   importance  in   liquid-phase   chemistry),   and  significant
interdependences exist  among  the  major influencing  environmental  factors.
Detailed  chemical  models already   can  simulate  qualitatively many  field
observations, but the  validity of  quantitative  predictions based  on these
                                    4-81

-------
models is  questionable.   Furthermore,  their  application  requires  substantial
computational resources.

It appears that, for the  foreseeable future,  empirical  parameterizations will
serve  as  transformation  modules  in   regional  models.   Preliminary  param-
eterizations have  been developed only for sulfate  formation  in power plant
plumes,  and  will   undoubtedly  continue  to  be  improved.     No   practical
parameterizations  exist  yet  for  nitrate formation  or  for  urban  plumes.
Adherence  to mechanistic considerations  is  recommended  in  formulating  the
parameterizations.   More, and  more reliable, measurements of such  important
variables   as   the  atmospheric   concentrations  of   OH,  ^Og,   NH3,   HC,
sulfate,  and nitrate and  of  cloud dimensions and cloud chemical composition
are needed direly.

H2$04  and  HMOs  formation apparently  peaks  during  daytime  and  in  summer.
Gas-phase  mechanisms  are  believed to contribute  a  larger  share,  on   the
average,  to  these  secondary  formations under warm,  sunny conditions.  Typi-
cally, on  a  summer day  (24  hr) in  the eastern  United  States,  about 25 +_ 10
percent of  the  airborne  S02  in power plant  plumes is likely to be  converted
to sulfates.  Nighttime conversion is  a small part (about  5 percent  or less).
S  transformations  may be somewhat  higher  than  these in  the southeastern
United States.   HN03 formation  rate  in  power  plant plumes is  about three
times as fast as the sulfate  formation rate by gas-phase mechanisms.   Aerosol
N03~  formation   rate  is   apparently very  small, at  least  in the  summer.
Both sulfate and nitrate  formation are faster in urban  plumes.

4.5  CONCLUSIONS

The discussion  of  homogeneous  gas-phase reactions has led  to  the  following
conclusions:

 0   Organic acids  produced  during gas-phase oxidation  of  hydrocarbons  are
     expected   to   make   only  minor   or   insignificant  contributions  to
     precipitation   acidity because of  their relatively  small  dissociation
     constants.   More information is needed for  assessment (Section  4.2.1).

 0   Acids  (HX)  produced from  gas-phase reactions of halocarbons  are also
     expected to make  insignificant  contributions  to  regional  disposition
     problems;   their  effects  on  global  precipitation  chemistry  are more
     plausible  but uncertain.  Direct  anthropogenic  emissions  of  HX   are
     potentially important (Section 4.2.1;  Chapter A-2).

 0   Oxidation of  reduced forms  of sulfur in the atmosphere generally leads
     to sulfur dioxide  (S0£)  formation (Section  4.2.1).

 0   SO?   oxidation in  air  is  dominated by   reaction  with  hydroxyl  (OH)
     radicals,  and although  the  reactions  of   the  HOS02 adduct  and other
     possible intermediates are  unknown,  the final  product  is  sulfuric acid
     aerosol (Section 4.2.1).

 0   The  average   lifetime  of   SO?   with  respect  to   this  reaction  is
     approximately  3 to 4 days  (Section 4.2.2).


                                    4-82

-------
     Of  the remaining  free-radical  processes  for  S02  oxidation,  only  the
     reaction  by  peroxyalkyl  radicals  appears  to have  possible  atmospheric
     significance; additional information  is  needed for  assessment  (Section
     4.2.1).

     Gas-phase  oxidation  of  nitrogen  dioxide  (N02)  leads to  a  variety  of
     products;  nitric  acid,  dinitrogen  pentoxide  (NzOs)  and peroxyacetyl
     nitrate (PAN) are  in greatest  abundance.   Nitrogen  tri oxide  and  nitrous
     acid play  active roles  in  photochemical  cycles but  make  smaller  direct
     contributions to acid deposition.   Further research on the  fate  of PAN
     and NgOs is direly needed (Section 4.2.1).
 «   The  average lifetime  of N02  W1'th  respect  to  reaction  with  hydroxyl
     radicals  is approximately  one-half day and  the product is nitric  acid
     vapor (Section 4.2.2) .

 °   Field data  tend to confirm overall  transformation rates for  nitrogen and
     sulfur  oxides,  as  established  in  laboratory  experiments,  but fail  to
     give conclusive  evidence about dominant  reaction  pathways  and meteor-
     ological  effects.   Gas-phase transformation  rates  in  power  plant  plumes
     are usually smaller than in urban plumes because of imperfect  mixing and
     an  abundance of  nitric  oxide  which  suppresses  the concentration  of
     hydroxyl radicals (Section 4.2.3).

 °   The concentrations  of  hydroxyl  radicals  in the atmosphere  are  governed
     by  a tightly  coupled   reaction  cycle involving  HC-CO-NOX-03,  but  not
     S02, and  the OH concentrations are  not  satisfactorily defined except,
     perhaps, on a global scale.   In  polluted air, the ratio of hydrocarbons
     (hC) to  nitrogen  oxides (NOX)   is  expected to  be  a  dominant variable
     for  the  OH radical  concentration.    The  cause-effect relationships
     governing the free  radical  composition  of the atmosphere need  further
     clarification (Section  4.2.1).

 0   Overall,  the  kinetics  and mechanistic  details of gas-phase  chemistry
     affecting acidic  species are  understood, albeit  some important gaps
     remain.    Adequate models of gas-phase chemistry can  be formulated  but
     their application  to  real  atmospheric  situations remains  a  problem
     (Sections 4.2.1, 4.2.2, and 4.2.3).

The review of  the current understanding of the production  of acidity  within
hydrometeors  has  led to the  following conclusions:

 0   The production of both  HNOs  and HC1  within hydrometeors is negligible
     compared with direct absorption of these  species  from the  gas phase.
     Here, the concentration of  these  species  in precipitation  will   be
     influenced strongly by  homogeneous gas-phase chemistry  (Sections  4.3.3
     and 4.3.4).

 0   Production  of  H2S04  in  solution   within   hydrometeors,   by   any   of
     several   different  mechanisms,  can  rival  or  even  surpass  direct
     absorption of H2S04  by  hydrometeors (Section  4.3.5).


                                   4-83

-------
     Of   the   various  production   mechanisms  for   H^SCty   in   solution,
     catalyzed  and  uncatalyzed  aerobic oxidation  and  oxidation  by  H202
     appear to be most important (Section 4.3.5).

     While  oxidation   by  ^02  appears  to  be the  single  most  important
     reaction  producing   H2S04,  the  extent  of  its  contribution  to  the
     acidity  of hydrometeors  will  depend  directly  on  the  H202  available
     in solution, a parameter  not  well  characterized at this time  (Section
     4.3.5).

     The amount of acid absorbed  and produced in hydrometeors is  such that
     the pH's of precipitation particles should be much  lower than  observed
     (Section 4.3.5).
 0   Neutralization  of  hydrometeor  acidity  by  NHs  absorption  and  by
     reaction with  scavenged  parti cul ate CaCOs,  MgCOs,  and CaO may  be of
     considerable importance (Section 4.3.6).

Considerable  progress  has  been  made  in transformation modeling  in  recent
years.    Significant  gaps  remain,  however,   in our  ability  to  predict
transformation  rates of  SOX  and  NOX  under  atmospheric  conditions.    The
following observations summarize the current status of the principal aspects
of transformation modeling:

 0   It  is  now  possible   to  simulate  the  principal  features of  the   smog
     chamber  chemistry  of  the   SOX-NOX-HC  system   rather  accurately  by
     detailed   modeling    of   the   chemical   kinetics   based   on  lumped
     representations  of  the  hydrocarbons,  even though  details  of the
     chemical mechanisms are not  fully  understood (Section 4.4.4).

 0   Detailed chemical  models of  plume  transformations  under  atmospheric
     conditions   have  successfully simulated  many qualitative  features of
     field  observations,   including  some  details of  crosswind  profiles
     influenced   by  plume-background  interactions.   These  simulations are
     mainly restricted to  gas-phase chemistry (Section 4.4.4).

 0   The  principal  current limitations  in  detailed  chemical  modeling are
     probably related to inadequate  characterization  of  the emission  field
     and  of  the ambient polluted  regional  background.    Improved  and more
     detailed inventories   of  the  emissions  of  SOX,   NOX,   and  HC   from
     major   sources,   including   the  urban  area  sources,  and  reliable
     measurements  of  reactive  species  (e.g.,  OH,   R02,  H202)   in  the
     ambient  atmosphere  are needed  before  reliable  conclusions concerning
     regional -scale  transformation processes  can be  made.    The  relative
     importance   of  co-emissions  vs  background  entrainment as  sources of
     oxidizing  agents   (OH,  R02,  H202>  etc.)   is  not  understood  at the
     present time (Section  4.4.4).

 0   Current detailed chemical models generally  do not include liquid-phase
     chemistry.   Quantitative  descriptions  of the liquid-phase  environment
     (e.g.,   cloud   dynamics,   plume-cloud   interaction,   etc.)   are  not
     adequately   incorporated  into transformation models.    Cloud  and fog


                                    4-84

-------
chemistry measurements are sparse and much needed.  Coupled modeling of
gas- and liquid-phase chemistry is necessary, particularly under summer
conditions.   First steps in  this  direction have been  taken  (Sections
4.4.2 and 4.4.4).

For  the  near future, it  appears  that transformation modules based on
empirical parameter!'zations will continue to predominate in operational
regional models.  All models, to varying degrees, use parameter!zations
based on laboratory  and  field data.   Currently,  regional  models  mostly
employ  pseudo-first-order  or  constant  first-order  bulk  conversion
rates.  The  basis  for refining these estimates  to reflect at least the
gross diurnal and  seasonal variations, and  even  the  role  of a changing
background,  exists.   Increasingly,  new  models   are  incorporating  such
empirical  expressions,   which  are   constantly  being  improved.     The
state-of-the-art of  such  parameterizations  will  be  further advanced as
more data  are obtained  and  analyzed,  particularly  for NOx  precursors
and  products, for  urban  plumes, and for other than  summer conditions.
Detailed  chemical   models also serve  to  improve  our  understanding
and  basis for  the  formulation of  empirical  parameterizations which
reflect the  underlying  physical-chemical  processes  rather  than  merely
expressing statistical correlations.   At this time, the  major  sources
of   uncertainty   in  determining  atmospheric   transport  ranges   of
pollutants  are  probably  associated  with  transport   and  deposition
processes  rather  than with  transformation  processes  (Sections 4.4.2
through 4.4.4).
                               4-85

-------
4.6  REFERENCES

Alkezweeny, A. J.   1978.   Measurements of aerosol particles and trace gases
in METROMEX.  J.  Appl .  Meteorol .  17:609-614.

Alkezweeny, A. J. and D.  C. Powell.   1977.   Estimation of transformation rate
of  S02  to  S(k   from  atmospheric   concentration  data.    Atmos.  Environ.
11:179-182.

Altshuller, A.  P.   1979.    Model  predictions  of  the  roles  of homogeneous
oxidation of  sulfur dioxide  to  sulfate in  the troposphere.  Atmos. Environ.
13:1653-1661.

Anbar, M.  and H. Taube.    1954.   Interaction of  nitrous  acid  with hydrogen
peroxide and with water.  J.  Am.  Chem.  Soc.  76:6243-6247.

Andrew,  S. P.  S.  and D.  Hanson.     1961.    The  dynamics  of  nitrous  gas
adsorption.  Chem. Eng. Sci.  47:105-113.

Appel , B. R.,  S.  M. Wall, Y.  Tokiwa,  and  M.  Haik.  1980.  Simultaneous nitric
acid,  particulate  nitrate  and  acidity measurements in  ambient  air.   Atmos.
Environ. 14:549-554.

Ardon, M.   1965.   Oxygen,  Elementary Forms  and  Hydrogen Peroxide.   W.  A.
Benjamin Pub.  Co., New York.

Baboolal,  L.  B., H. R. Pruppacher,  and  J.  Topalian.    1981.   A sensitivity
study  of a  theoretical  model  of S02  scavenging  by  water drops  in  air.   J.
Atmos. Sci. 38:856-870.

Bandow,  H.,  M.  Okuda, and  H.  Akimoto.   1980.   Mechanisms  of  gas-phase
reactions of CsHg and N03 radicals.   J. Phys.  Chem. 84:3604-3608.
Barrie, L.  A.    1975.   An  experimental  investigation of  the  adsorption of
sulphur dioxide  by  cloud  and  raindrops  containing  heavy  metals.   Ph.D.
Thesis, University of Frankfurt.

Barrie, L. and H.  W.  Georgii.   1976.  An  experimental  investigation of the
adsorption of  sulphur dioxide by  water  drops containing  heavy  metal  ions.
Atmos. Environ. 10:743-749.

Barrie, L., S. Beilke,  and  H.  W.  Georgii.   1974.   SOe  removal  by cloud and
fog  drops  as  affected  by  ammonia  and   heavy  metals.    In  Precipitation
Scavenging, Proceedings of  a Symposium held  at Champaign, ITT.,  Oct. 14-18.
R.  G.  Semonin  and  R.  W.  Beadle,  eds.   Technical  Information  Center,  ERDA
(NTIS No.  CONF  -  741003).

Bassett, H. and W. G. Parker.  1951.  The  oxidation of  sulphurous acid.   J.
Chem. Soc. 47:1540-1560.
                                   4-86

-------
Baulch,  D.  L., R.  A.  Cox, R.  F.  Hampson, J.  A.  Kerr,  J. Troe,  and  R. T.
Watson.   1980.   Evaluated kinetic  and photochemical  data  for atmospheric
chemistry.  J. Phys. and Chem.  Ref.  Data 9:295-471.

Bazzell,  C.  C. and  L.  K. Peters.   1981.   The transport of photo-chemical
pollutants to the background troposphere.   Atmos. Environ. 15:957-968.

Beilke,  S.  and G. Gravenhorst.   1978.   Heterogeneous S02-°xidation  in the
droplet  phase.  Atmos. Environ. 12:231-239.

Beilke,  S., D.  Lamb,  and J. Muller.  1975.  On the uncatalyzed oxidation of
atmospheric   S02   by   oxygen   in   aqueous   systems.     Atmos.   Environ.
9:1083-1090.

Benarie,  M.,  A.  Nonat,  and T. Menard.   1972.   The transformation  of sulfur
dioxide  into  sulfuric  acid in  relation  to  the  climatology  of  an  urban/
industrial  area  (Rouen,  France).    Intern.  Clean Air  Conf.   May  15-18,
Melbourne, Australia,  pp. 176-186.

Boris, J. P. and D. L. Book.  1973.  Flux corrected  transport.   I. SHASTA, an
algorithm that works.  J. Comp. Phys.  11:38-69.

Bottenheim, J. W.  and 0. P. Strausz.   1979.  The effect of a polluting source
on  the  air  quality downwind  of  pristine  northern  areas. Atmos.  Environ.
13:1085-1089.

Breeding, R. J., H. B. Klonis,  J.  P.  Lodge, J.  B. Pate, D. C. Sheesley, T. R.
Englert, and D. R. Sears.   1976.  Measurements of  atmospheric pollutants in
the St. Louis area.  Atmos. Environ.  10:181-194.

Brimblecome,  P.  and  D.  J. Spedding.    1974.   The  catalytic  oxidation  of
micromolar aqueous sulphur dioxide-I.  (Oxidation of dilute solutions by iron
(III)).  Atmos. Environ. 8:937-945.

Brodzinsky,  R.,   S.  G.  Chang, S.  S.  Markowitz,  and  T.  Novakov.    1980.
Kinetics  and  mechanisms  for the  catalytic oxidation  of sulfur dioxide  on
carbon in aqueous  suspensions.  J. Phys. Chem. 84:3354-3358.

Cahir, J.  J., J.  N.  Pitts, J.  Ross,  S. A.  Twomey,  and J.  R.  Wiesenfeld.
1982.  The source-receptor relationship  in  acid precipitation:   Implications
for generation of  electric  power  from coal.   Physical  Dynamics, Inc.  Report
No. PD-LJ-82-268R  on Workshop of January 18-22.

Calvert, J.  C., S. Fu, J.  W. Bottenheim, and 0.  P. Strausz.  1978.  Mechanism
of the  homogeneous  oxidation of sulfur  dioxide in the troposphere.   Atmos.
Environ. 12:197-226.

Cantrell, B. K. and K. T.  Whitby.    1978.   Aerosol  size  distributions  and
aerosol volume formation for a  coal fired power  plant plume.  Atmos. Environ.
12:323-333.
                                   4-87

-------
Carmichael, G.  R.  and  L.  K.  Peters.   1979.   Numerical  simulation  of the
regional   transport  of  S02   and   sulfate   in  the  eastern  United  States.
Preprint, 4th. AMS Symp.   Turbulence, Diffusion, and  Air Pollution.   Reno,
NV.  January 15-18.

Castleman, A.  W.,  R.  E. Davis,  H.  R.  Munkelwitz,  I.  N.  Tang, and W. P. Wood.
1975.  Kinetics of association reactions pertaining to sulfuric acid aerosol
formation.  Int. J. Chem. Kinet. Symp.  1:629.

Chang, S.  G.,  R.  Toorsi,  and T.   Novakov.   1981.   The  importance  of soot
particles  and  nitrous  acid   in  oxidizing   SOg  in   atmospheric  aqueous
droplets.  Atmos.  Environ.  15:1287-1292.

Clark, W. W., D.  A.  Landis,  and A.  B.  Harker.   1976.   Measurements  of the
photochemical  production of  aerosols  in  ambient air  near a  freeway  for  a
range of S02 concentrations.   Atmos.  Environ. 10:637-644.

Commins,   B.  T.   1963.    Determination  of  particulate   acid  in  town air.
Analyst 88:364-367.

Cox, R.  A.  and  S.  A. Penkett.  1972.  Aerosol  formation  from  sulfur dioxide
in the presence of ozone and olefinic hydrocarbons.  J.  Chem. Soc. Faraday
Trans. I. 68:1735-1753.

Cox, R.  A., and M. J.  Roffey.  1977.  Thermal  decomposition of peroxyacetyl
nitrate in the presence of  nitric  oxide.  Environ. Sci. Technol. 11:900-906.

Cox, R.  A.  and  D.  Sheppard.    1980.   Reactions of  OH  radicals with gaseous
sulfur compounds.   Nature 284:330-331.

Criegee,  R.  1957.  The course of  ozonation  of unsaturated compounds.   Record
Chem. Progr. 18:111-120.

Crutzen,  P. J.  1974.   Photochemical  reactions  initiated by and influencing
ozone in unpolluted tropospheric air.   Tellus 26:47-56.

Dasgupta, P. K.   1980a.   Discussion  of  the importance of atmospheric ozone
and  hydrogen peroxide  in oxidizing sulphur dioxide  in  cloud and  rainwater.
Atmos. Environ. 14:272-274.

Dasgupta, P. K. 1980b.   Further discussion  of  the  importance  of atmospheric
ozone  and  hydrogen  peroxide  in   oxidizing sulphur  dioxide  in  cloud  and
rainwater.  Atmos. Environ. 14:620-621.

Davis, D.  D.  and  G.  Klauber.   1975.    Atmospheric  gas  phase  oxidation
mechanisms for the molecule S02«  Int.  J. Chem. Kinet. Symp.  1:543-556.

Davis, D. D., A. R.  Ravishankara,  and S. Fischer.   1979.   S02 oxidation via
the  hydroxyl  radical:    Atmospheric  fate of  HSOX  radicals.   Geophys. Res.
Lett. 6:113-116.
                                    4-88

-------
Davis, W. Jr. and H. J. de Bruin.  1964.  New activity coefficients of 0-100
percent aqueous nitric  acid.  J.  Inorg. Nucl. Chem. 26:1069-1088.

Dawson, G. A.,  J.  C. Farmer, and  J.  L.  Moyers.   1980.   Formic  and  acetic
acids  in  the  atmosphere of  the southwest  USA.    Geophys.  Res.   Lett.
9:725-728.

Demerjian, K. L. and K.  L.  Schere.  1979.  Applications  of  a photochemical
box model for ozone air quality in Houston, Texas.  Proc. APCA Conference on
Ozone/Oxidants:   Interactions  with the Total  Environment II,  Houston,  TX.
October 14-17.

Demerjian, K.  L.,  J.  A. Kerr, and J.  G.  Calvert.  1974.  The mechanism of
photochemical  smog  formation.    J^n Advances  in  Environmental  Science  and
Technology, Vol. 4, John Wiley, N.Y.  262p.

Demore, W. B.,  L.  J. Stief,  F.  Kaufman,  D. M. Golden, R. F.  Hampson,  M. H.
Kurylo, J. J.  Margitan,  M.  J. Molina, and  R. T.  Watson.   1981.   Chemical
kinetic  and  photochemical  data  for  use  in  stratospheric modeling.    JPL
Publication 81-3, California  Inst.  of Tech., Pasadena, CA.

Dimitriades,  B., M.  C.  Dodge, J.  J.  Bufalini, K.  L.  Demerjian, and  A. P.
Altshuller.    1976.    Letter  to  the  Editor.    Environ.   Sci.  Technol.
10:934-936.

Dinger, J. E., H. B. Howell, and T. A. Wojciechowski.  1970.   On  the  source
and  composition  of cloud nuclei  in  a subsident air mass  over the  North
Atlantic.  J. Atmos. Sci.  17:791-797.

Dittenhoefer, A. C. and R. G.  de  Pena.  1980.  Sulfate aerosol  production and
growth  in coal-fired power  plant  plumes.   J.  Geophys.  Res.  85,  No.  C-8,
4499-4506.

Donaldson, C. D. and G.  R. Hilst.  1972.   Effects of inhomogeneous mixing on
atmospheric photochemical  reactions.  Environ. Sci. Technol. 6:812-816.

Drozdova, V. M.  and E. P. Makhon'ko.   1970.   Content of  trace elements in
precipitation.  J.  Geophys. Res.  18:3610-3612.

Duce,  R.  A.    1969.    On the  source of  gaseous  chlorine   in the  marine
atmosphere.  J. Geophys.  Res. 74:4597-4599.

Duecker,  W.  and J. West  (eds.).   1959.   The Manufacture of  Sulfuric  Acid.
American Chemical Society  Monograph Series No. 144.

Duewer, W. H.,  M.  C.  McCracken,  and  J.  J. Walton.   1978.    The  Livermore
Regional Air Quality Model:  II. Verification  and sample application  in the
San Francisco Bay area.  J. Appl. Meteorol.  17:273-311.

Durham, J. L.,  J.  H. Overton, Jr., and V.  P. Aneja.   1981.    Influence of
gaseous  nitric  acid on  sulfate  production and  acidity in  rain.    Atmos.
Environ. 15:1059-1068.
                                    4-89

-------
Easter,  R.  C. and  P.  V.  Hobbs.   1974.   The formation of  sulfates  and the
enhancement  of  cloud  condensation   nuclei   in  clouds.    J.  Atmos.  Sci.
31:1586-1594.

Easter, R. C., K. M. Busness, J. M. Hales, R. N. Lee, D. A. Arbuthnot, D. F.
Miller,  G.  M.  Sverdrup,  C.  W. Spicer,  and J.  E.  Howes.    1980.    Plume
conversion  rates  in the  SURE  region.   EPRI  EA-1498, 1-2.   Electric Power
Research Institute,  Palo Alto,  CA.

Eatough,  D.  J.,  B.  E. Richter,  N.  L.  Eatough,  and L.  D. Hansen.   1981.
Sulfur  chemistry  in  smelter  and  power  plant  plumes  in  the  western  U.S.
Atmos. Environ. 15(10)-.2241-2253.

Eigen, M.  1967.  Proton  transfer  and  general acid  base  catalysis.   JUi Fast
Reactions  and  Primary   Processes  in  Chemical   Kinetics.    Interscience
Publishers,  New York,  NY.

Eliassen, A.  and  J. Saltbones.   1975.   Decay  and  transformation  rates  of
SOg  as   estimated  from   emission  data,  trajectories   and  measured  air
concentrations.  Atmos. Environ. 9:425-429.

Elshout,  A.  J., J. W. Viljeer,  and  H.  Van Duren.   1978.    Sulfates  and
sulfuric  acid  in  the  years  1971-1976  in the Netherlands.   Atmos.  Environ.
12:785-790.

England,  C.  and W. H. Corcoran.    1974.   Kinetics  and  mechanisms  of  the
gas-phase reaction  of water  vapor and  nitrogen dioxide.    Ind.  Eng. Chem.
Fundam. 13:373-384.

Erickson, R.  E.,  L.  M.  Yates, R. L. Clark,  and  D.  McEwen.    1977.   The
reaction of sulfur  dioxide with ozone  in water  and its possible atmospheric
significance.   Atmos.  Environ.  11:813-817.

Eriksson, E.  1960.  The yearly circulation of chloride and  sulfur in  nature;
meteorological, geochemical and pedological  implications.   Part II.   Tellus
12:63-109.

Falconer, R.  E.  and P. D.  Falconer.   1979.   Determination of cloud water
acidity at  a  mountain observatory in  the Adirondack Mountains of  New York
state.  Publication  741, Atmospheric  Sciences Research Center,  SUNY,  Albany,
NY.

Falls, A. H.  and  J. H. Seinfeld.   1978.   Continued development of a  kinetic
mechanism for photochemical  smog.   Environ. Sci. Techno!. 12:1398-1406.

Falls,  A.  H., G. J.  McRae,  and  J.   H.  Seinfeld.    1979.    Sensitivity  and
uncertainty of reaction mechanisms for photochemical  air pollution.    Int.  J.
Chem. Kinetics 11:1137-1162.

Farrow, L. A. and D. Edelson.   1974.   The steady-state  approximation:  Fact
or fiction?   Int.  J. Chem.  Kinetics 6:787-800.
                                   4-90

-------
Fishman, J.  and P.  J.  Crutzen.   1978.   The distribution of  the hydroxyl
radical  in  the troposphere.   Atmospheric Science  Paper No. 284, Colorado
State University,  Boulder.

Flack, W. W.  and  M.  J.  Matteson.   1979.   Mass transfer  of  gases to growing
water droplets.  Jji  Polluted  Rain,  T.  Y. Toribara, M. W. Miller,  and  P.  E.
Morrow, eds.  Plenum Press,  New York, NY.

Forrest, J., R. Garber,  and  L.  Newman.   1979.  Formation of sulfate, ammonium
and nitrate in an  oil-fired  power  plant  plume.  Atmos. Environ. 13:1287-1297.

Forrest, J.,  R.  Garber, and  L.  Newman.   1981.   Conversion  rates in  power
plant plumes  based on filter  pack data  - Part I:  The coal-fired  Cumberland
plume.  Atmos. Environ.  15:2273-2282.

Forrest J. F.,  S.  E.  Schwarts, and L.  Newman.   1979.   Conversion  of sulfur
dioxide to sulfate during  the  Da Vinci flights.   Atmos. Environ. 13:157-167.

Freiberg,  J.   1974.    Effects  of relative  humidity  and   temperature  on
iron-catalyzed  oxidation  of  S02  in atmospheric aerosols.   Environ.  Sci.
Techno!. 8:731-734.

Freiberg, J.  E. and  S. E.  Schwartz.   1981.  Oxidation of  SOg  in aqueous
droplets:   Mass-transport limitation  in laboratory studies and  the ambient
atmosphere.  Atmos. Environ. 15:1145-1154.

Friedlander,  S. K.   1978.   A review of the  dynamics  of sulfate containing
aerosols.  Atmos.  Environ. 12:187-195.

Fuller, E. C. and  R.  H.  Crist.   1941.   The rate of oxidation of sulfite ions
by oxygen.  Am. Chem. Soc. 63:1644-1650.

Galloway, J. N. and G. E.  Likens.   1981.  Acid precipitation:  The  importance
of nitric acid.  Atmos.  Environ. 15:1081-1086.

Gear,  C.  W.   1971.    Chapter  11  in  Numerical   Initial  Value   Problems  in
Ordinary Differential  Equations.  Prentice-Hall,  Englewood Cliffs,  NJ.

Georgii, H.  W. 1970.   Contributions to the atmospheric sulfur  budget.   J.
Geophys. Res. 75:2365-2371.

Georgii, H.  W.   1978.    Large scale spatial  and temporal  distribution  of
sulfur compounds.   Atmos.  Environ. 12:681-690.

Gill am', N.  V.   1978.    Project  MISTT:   Mesoscale  plume  modeling of  the
dispersion,  transformation  and  ground   removal  of  S02«   Atmos.  Environ.
12:569-588.

Gillani, N. V. and W. E. Wilson.  1980.   Formation and transport of ozone and
aerosols in power  plant  plumes.  Annals  N.Y.  Acad. Sci. 338:276-296.
                                   4-91

-------
Gillani, N. V. and W.  E.  Wilson.   1983.  Gas-to-particle conversion of sulfur
in power plant plumes:   II.  Observations of  liquid phase conversions.  Atmos.
Environ. 17(9):  1739-1752.

Gillani,  N.  V.,  J.  A.  Colby,  and W.  E.  Wilson.   1983.   Gas-to-particle
conversion  of  sulfur  in  power  plant  plumes:    III.  Parameterization  of
plume-cloud interactions.  Atmos.  Environ. 17(9): 1753-1764.

Gilliani,  N.  V.,  R.  B.  Husar,  D. E.  Patterson,  and  W. E.  Wilson.  1978.
Project MISTT:   Kinetics of particulate  sulfur  formation in  a  power  plant
plume out to 300 km.   Atmos. Environ. 12:589-598.

Gillani,  N.   V.,  S.  Kohli, and  W. E.  Wilson.   1981.    Gas  to  particle
conversion of sulfur  in  power  plant plumes:    I.  Parameterization of  the
conversion rate for  dry,  moderately  polluted  ambient conditions.   Atmos.
Environ. 15:2293-2313.

Gordon, G. E., D. D. Davis,  G.  W.  Israel,  H. E. Landsberg, T. C. O'Haver, S.
W. Staley, and  W. H. Zoller.  1975.  Atmospheric impact of major sources and
consumers of energy.   Progress Report-75,  EPA Grant No. ESR75-02667.

Gorham, E.  1958.  Atmospheric  pollution  by  hydrochloric acid.   Quart.  J. R.
Met. Soc. 84:274-276.

Graedel, T. E.,  L. A.  Farrow, and  T. A. Weber.  1976.  Kinetic  studies of the
photochemistry of the  urban  troposphere.   Atmos.  Environ.  10:1095-1116.

Graedel,  T.   E.,  L.  A.  Farrow,  and T.  A.  Weber.    1978.    Urban  kinetic
calculations with altered source conditions.  Atmos. Environ. 12:1403-1412.

Graham, R. A. and H.  S.  Johnston.   1978.   The photochemistry of N03 and the
kinetics of the  ^05 -  system.  J.  Phys. Chem. 82:254-268.
Graham, R.  A.,  A. M.  Winer,  R.  Atkinson,  and J.  N.  Pitts.   1979.   Rate
constants  for the  reaction of  H02, SO?,  CO, N?0,  trans-2-butene  and  2,
3-dimethyl-2-butene at 300 K.  J.  Phys.  Chem.  18:1563-1567.
Graham,  R.  A.,  A.  M.  Winer,  and  J.  N.  Pitts,  Jr.    1977.   Temperature
dependence  of the  unimolecular  decomposition   of pernitric  acid  and  its
atmospheric implications.   Chem.  Phys. Letters 51:215-220.

Groblicki, P. J.  and G. J.  Nebel.   1971.   The  photochemical  formation of
aerosols in urban atmospheres, pp.  241-267.   In Chemical Reactions in Urban
Atmospheres.  C.  S.  Tuesday,  ed.   American Elsevier, New  York.

Hales, J. M. and  D.  R. Drewes.   1979.   Solubility of ammonia in water at low
concentrations.   Atmos. Environ.  13:1133-1147.

Halfpenny, E.  and P.  L.  Robinson.   1952.   Pernitrous  acid.   The reaction
between  hydrogen peroxide  and  nitrous  acid,   and  the properties  of  an
intermediate product.  J.  Chem. Soc. 48:928-938.
                                   4-92

-------
Hampson, R.  F.,  Jr.  and  D. Garvin.  1977.  Reaction rate  and  photochemical
data  for  atmospheric  chemistry.    NBS  Special  Publication,  U.S.  Dept.  of
Commerce.

Hanst, P. L., N. W. Wong, and J. Bragin.   1982.   A long  path  infra-red  study
of Los Angeles smog.  Atmos. Environ.  16:969-981.

Harrison,  H., T.  V.  Larson,   and  C.  S.  Monkton.   1982.    Aqueous  phase
oxidation of  sulfites by ozone in the  presence  of  iron and manganese.  Atmos.
Environ.  16(5):1039-1042.

Hecht, T. A., J. H. Seinfeld, and M. C.  Dodge.   1974.  Further development of
generalized kinetic mechanism for photoche  ical  smog.  Environ.  Sci. Technol.
8:327-339.

Hegg, D. A. and  P. V.  Hobbs.   1978.  Oxidation  of sulfur dioxide in  aqueous
systems  with  particular  reference  to  the  atmosphere.   Atmos.  Environ.
12:241-253.

Hegg, D. A.  and P. V.  Hobbs.   1979a.  The  homogeneous  oxidation of sulfir
dioxide in cloud droplets.  Atmos. Environ. 13:981-987.

Hegg,  D.  A.  and P.  V.  Hobbs.   1979b.   Some observations  of  particulate
nitrate  concentration  in coal-fired  power  plant  plumes.   Atmos.  Environ.
13:1715-1716.

Hegg,  D.  A.  and  P.  V.  Hobbs.    1980.    Measurements of  gas-to-particle
conversion in the plumes from five  coal-fired electric power plants.  Atmos.
Environ. 14:99-116.

Hegg,  D.  A.  and  P.  V.  Hobbs.    1981a.    Cloud water  chemistry and the
production of sulfates in clouds.   Atmos. Environ.  15:1597-1604.

Hegg, D. A. and  P. V.  Hobbs.   1981b.  Field  Studies of  the  oxidation  of SO2
in  clouds.    Quarterly  Progress  Report for April  1-June  30.    EPA Grant
R805263010, Dept. of Atmospheric Sciences, University of Washington, Seattle,
MA.

Hegg, D. A. and  P. V.  Hobbs, and  L. F.  Radke.  1980.  A preliminary study of
cloud chemistry, pp. 7-10.  In  Preprint Volume, E  hth  Intern.  Conf.  on the
Physics of Clouds,  Clermont-Ferrand, France.

Hendry, C,  D. and  P.  L.  Brezonik.  1980.   Chemistry  of  precipitation  at
Gainesville, Florida.  Environ.  Sci. Technol. 14:843-849.

Hesstvedt,   E.,  0.  Hov,   and   I.  Isaksen.     1978.     Quasi-steady-state
approximations  in  air  pollution  modeling:   Comparisons  of  two numerical
schemes for oxidant prediction.   Int.  J. Chem. Kinetics 10:971-994.

Hidy, G. M.   1982.   Potential  fallacy  in assuming  linear proportionality
between  S02   emissions  and  acid  deposition.    Paper  presented  at  Second
National  Symposium  on Acid Rain, Pittsburgh, PA, October 6-7.
                                    4-93

-------
Hidy, G.  M.,  P. K.  Mueller,  and E.  Y.  long.   1978.   Spatial  and temporal
distributions of airborne  sulfate in  parts  of the  United States.   Atmos.
Environ. 12:735-752.

Hitchcock, D. R.,  L. L. Spriller, and  W.  E. Wilson.   1980.   Sulfuric acid
aerosols and HC1 release in coastal atmospheres:   Evidence of rapid formation
of sulfuric acid particulates.   Atmos. Environ.  14:165-182.

Hobbs,  P.  V.    1979.   A reassessment of the mechanisms  responsible  for the
sulfur content of acid rain.  In Proceedings:  Advisory Workshop to Identify
Research  Needs  on  the  Formation  of  Acid  Precipitation.    Electric  Power
Research Institute  Rpt. EA-1074.

Hobbs, P. V., D. A. Hegg, M. W.  Eltgroth, and L. F. Radke.  1978.  Evolution
of  particles  in the  plumes of coal-fired  electric  power plants.   Atmos.
Environ. 12:935-951.

Hoffman,  M.  R.  and  J.  0.  Edwards.    1975.   Kinetics  of the  oxidation of
sulfite   by   hydrogen   peroxide  in   acidic  solution.    J.   Phys.   Chem.
79:2096-2098.

Hov, 0.   1983a.   Numerical solution  of a simplified form of  the diffusion
equation  for chemically  reactive  atmospheric  species.   Atmos.  Environ.
17:551-562.

Hov, 0.   1983b.   One-dimensional vertical model  for ozone and other gases in
the atmospheric  boundray layer.   Atmos.  Environ. 17:535-550.

Hov,  0.   1983c.   Aspects of  the  parameterization  of  transformation  and
removal  processes in  air quality modeling.   Paper presented at the 14th NATO
International   Technical   Meeting   on   Air   Pollution  Modeling   and  Its
Applications, Copehagen, September 27-30.

Hov, 0.,  I.  S.  A.  Isaksen.  1981.  Generation  of secondary  pollutants in  a
power plant plume:   A model study.  Atmos.  Environ. 15:2367-2376.

Hov, 0.,  I.  S.  A.  Isaksen, and E. Hesstvedt.   1977.   Diurnal  variations of
ozone and  other pollutants in an  urban area.  Report No.  24,  Institute for
Geophysics, Univ. of Oslo.

Huebert B. J. and  A.  L. Lazrus.  1978.  Global tropospheric measurements of
nitric acid vapor and particulate  nitrate.  Geophys. Res. Lett. 5:577-580.

Huebert, B. J. and  A. L. Lazrus.  1979.  Tropospheric measurements of nitric
acid  vapor and  particulate nitrate.    In  Nitrogeneous  Air Pollutants.   D.
Grosjean, ed. Ann  Arbor Science,  Ann  ArTJor,  MI.

Huntzicker, J.  J.,  R.  A.  Gary,  and  C.  Ling.    1980.    Neutralization  of
sulfuric acid aerosol by ammonia.  Environ. Sci. Techno!. 14:819-824.
Husar,  R.  B.  and  D.  E. Patterson.    1980.   Regional  scale  air pollution:
Sources and effects.   Annals N.Y.  Acad.  Sci.  338:399-417.
                                   4-94

-------
 Husar,  R. B. and  Patterson.   1980.   Regional scale air pollution:   Sources
 and effects.  Annals N.Y. Acad. Sci. 338:399-417.

 Husar,  R. B. and  D. E.  Patterson,  J.  D.  Husar,  N. V. Gillani,  and W.  E.
 Wilson.    1978.    Sulfur budget  of  a  power  plant  plume.   Atmos.  Environ.
 12:549-568.

 International Critical Tables  3.   1928.  Washburn, E. W., ed.   McGraw-Hill,
 New York, NY.

 Isaksen,  I.  S.  A.,  E.  Hesstvedt,  and 0. Hov.   1978.   A chemical model  for
 urban  plumes:   Test for  ozone and particulate  sulfur  formation  in  the  St.
 Louis urban plume.  Atmos. Environ. 12:599-604.

 Jeffries,  H.  E.  and M.  Saeger.  1976.  Letter to  the Editor.  Environ.  Sci.
 and Technol. 10:936-937.

 Johnstone, H. J. and P.  W.  Leppla.   1934.   The  solubility of sulfur  dioxide
 at  low partial pressures.  J. Amer. Chem.  Soc. 56:2233-2238.

 Joseph,  D.  W.   and C.  W.  Spicer.    1978.   Chemiluminescence  method   for
 atmospheric  monitoring  of  nitric acid  and  nitrogen  oxides.    Anal. Chem.
 50:1400-1403.

 Jost,  D.   1974.    Aerological  studies on  the atmospheric  sulfur   budget.
 Tell us 26:206-212.

 Junge, C.  and T. G.  Ryan.   1958.   Study of  S02 In  oxidation in solution  and
 its role  in atmospheric chemistry.  Quart. J.  R.  Met. Soc. 84:46-55.

 Kameoka,  Y.  and  R. L. Pigford.   1977.   Adsorption of nitrogen dioxide  into
 water, sulfuric  acid, sodium  hydroxide,  and alkaline sodium sulfite   aqueous
 solutions.  Ind.  Eng. Chem. Fundam.  16:163-169

 Kan,  C.  S., J.  G.  Calvert,  and J.  H. Shaw.   1981.    Oxidation  of sulfur
 dioxide by methylperoxy radicals.   J.  Phys.  Chem. 85:1126-1132.

 Kan, C. S.,  R. D.  McQuigg, M. R. Whitbeck, and  J.  Calvert.   1979.   Kinetic
 flash  spectroscopic  study  of   the  CHsO?  and  CHa02  -   SO?  reactions.
 Int. J. Chem. Kinetics  11:921-933.                              L

 Kaplan, D.,  D.   Himmelblau,  and  C.  Kanoaka.     1981.   Oxidation  of sulfur
 dioxide  in  aqueous ammonium  sulfate  aerosols  containing  manganese as a
 catalyst.   Atmos. Environ.  15:763-773.

 Kasina, S.  1980.   On precipitation  acidity in southeastern  Poland.   Atmos.
 Environ. 14:1217-1221.

Kelly, T.  J.,  D.  H. Stedman,  and G. L. Kok.   1979.   Measurements  of
 and HNOs in rural air.   Geophys. Res.  Lett.  6:375-378.
                                    4-95

-------
Kocmond, W.  C.  and J. Y. Yang.   1976.   Sulfur dioxide photoxidation rates,
and aerosol  formation mechanisms, a  smog  chamber  study.   EPA 600/3-76-0900,
U.S. Environmental  Protection  Agency,  Research  Triangle Park, NC.

Kok, G.  L.   1980.   Measurements  of  hydrogen  peroxide  in  rainwater.   Atmos.
Environ. 14:653-656.

Komiyama, H. and H. Inoue.   1980.  Adsorption of nitrogen oxides into water.
Chem. Eng. Sci.  35:154-161.

Kritz, M.  A. and J. Rancher.   1980.   Circulation of Na, Cl,  and  Br in the
tropical marine  atmosphere.  J.  Geophys. Res. 85:1633-1639.

Kuhlman, M.  R.,  D.  L. Fox,  and  H. E.  Jeffries.  1978.  The  effect of CO on
sulfate aerosol  formation.   Atmos. Environ. 12:2415-2423.

Lamb, R.  G.   1981.   A  regional  scale (1000 km)  model  of photochemical air
pollution:   I.  Theoretical  formulation.   U.S.  EPA.   Technical Report.   In
press.

Lamb, R. G.  and W.  R. Shu.  1978.  A  model of  second-order chemical reaction
in turbulent fluid  -  part I.  Atmos.  Environ. 12:1685-1694.

Larson, T. and H. Harrison.   1977.  Acidic sulfate aerosols:   Formation from
heterogeneous oxidation  by 03  clouds.  Atmos. Environ. 11:1133-1141.

Larson, T.,  R.  Charlson, E. Knudson, G.  Christian,  and  H.  Harrison.   1975.
The influence of a sulfur  dioxide point  source on the rain chemistry  of a
single storm in  the Puget Sound  region.  Water,  Air,  Soil Pollut. 4:319-328.

Larson, T.,  N.  Horikp,  and  H.  Harrison.   1978.  Oxidation of sulfur dioxide
by oxygen and ozone in  aqueous  solution:   a  kinetic  study with significance
to atmospheric rate processes.   Atmos. Environ.  12:1597-1611.

Lau, N. C. and  R. J. Charlson.   1977.  On the  discrepancy between  background
atmospheric ammonia gas  measurements  and  the  existence of acid sulfates as a
dominant atmospheric aerosol.  Atmos.  Environ.  11:475-478.

Lazrus, A. L.,  P.  L.  Haggenson, G.  L. Kok, B. J. Huebert, C. W. Kreitzberg,
G. E. Likens, V. A. Mohnen, W.  E. Wilson, J. W.  Winchester.   1983.  Acidity
in  air  and water in a  case of  warm  frontal  precipitation.   Atmos. Environ.
17:581-592.

Lee, R.  E.  and  D.  J.  von  Lehmden.   1973.    Trace  metal  pollution  in the
environment.  J. Air.  Pollu. Control  Assoc. 23:853-857.

Lee, Y. N.  and  S.  E.  Schwartz.    1981.  Evaluation of the rate of uptake of
nitrogen dioxide by atmospheric  and sulfate liquid water.   J.  Geophys. Res.
86:11971-11983.
                                    4-96

-------
Leu,  M.  T. and  R.  H. Smith.   1981.   Kinetics of the  gas phase  reactions
between hydroxyl and carbonyl sulphide over the temperature range 300-517 K.
J. Phys. Chem. 85:2570.

Levine, J.  S.,  T.  R. Augustsson,  and J.  M.  Hoell .   1980.    The vertical
distribution of tropospheric ammonia.  Geophys.  Res.  Lett.  17:317-320.

Levine, S.  Z.  1981.   A model  for  stack  plumes  reactions with  atmospheric
dilution (SPREAD).  Atmos. Environ. 15:2573-2581.

Levine, S.  Z.  and  S. E.  Schwartz.    1982.   Construction  and  testing of  a
_S_urrogate   CHEmical   MEchanism   (SCHEME)   for   tropospheric    photochemical
reactions.  Chap. 11.  In Trace Atmospheric Constituents,  S. E.  Schwartz, ed.
Vol.  12 in  Advance's  in "Environmental  Science and  Technology.   John Wiley  &
Sons, Inc., New York.

Lewis,  C.  W.  and  E.  S.  Macias.    1980.    Composition  of  size- fractionated
aerosol in Charleston, West Virginia.  Atmos.  Environ. 14:185-194.

Liljestrand,  H.  M.  and J.  J. Morgan.   1981.   Spatial  variations of acid
precipitation in southern California.  Environ.  Sci .  Technol . 15:333-338.

Lusis,  M.  A., K. G.  Anlauf,  L.  A.  Barrie, and H.  A. Wiebe.   1978.    Plume
chemistry  studies  at a  northern  Alberta power  plant.   Atmos.  Environ.
12:2429-2437.

Mader,  R.  M.   1958.   Kinetics  of the hydrogen peroxide- sulfite reaction in
alkaline solution.  J. Amer. Chem. Soc. 80:2634-2639.

Marsh, A.  R. W.  1978.  Sulphur and  nitrogen contributions  to the acidity of
rain.  Atmos. Environ. 12:401-406.

Martin, L.  R.,  and  D.  E. Damschen.   1981.   Aqueous  oxidation  of  sulfur
dioxide by hydrogen peroxide at low pH.  Atmos.  Environ.  9:1615-1621.

Martin, L.  R., D. E.  Damschen, and H.  L. Judeiker.   1981.   The reactions of
nitrogen oxides with SOe in aqueous aerosols.  Atmos.  Environ. 15:191-195.
McClenny,  W.  A. and  C.  A. Bennett, Jr.   1980.   Integrative technique  for
detection of atmospheric ammonia.  Atmos.  Environ.  14:641-645.

McCracken,  M.  C.,  D. J.  Wuebbles, J. J.  Walton,  W.  H.  Duewer,  and  K. E.
Grant.   1978.   The  Livermore  Regional Air  Quality  (LIRAQ) model:   I. Concept
and development.  J. Appl .  Meteorol .  17:254-272.

McDonald, C. and H. J. Duncan.   1979.   Particle size distribution of metals
in the atmosphere of Glasgow.   Atmos.  Environ.  13:977-980.

McKay, H. A. C.  1971.  The atmospheric oxidation  of sulfur  dioxide  in water
droplets in the presence of ammonia.   Atmos.  Environ.  5:7-14.
                                   4-97

-------
McMurry,  P.  H., and  J. C.  Wilson.    1982.    Growth  laws for  formation of
secondary ambient aerosols:   Implications  for chemical conversion mechanisms.
Atmos. Environ. 16:121-134.

McMurry, P. H., D. J. Rader, and J. Stith.  1981.  Growth laws for secondary
aerosols  in   power  plant  plumes:    Implications  for  chemical  conversion
mechanisms.  Atmos.  Environ.  15:2315-2327.

McNaughton, D.  J.  and  B.  C. Scott.   1980.    Modeling evidence  of  in cloud
transformation of sulfur dioxide to sulfate.   J. Air. Pollut. Control Assoc.
30:272-273.

McNeils, D. N.   1974.   Aerosol  formation  from gas-phase  reactions  of ozone
and  olefin in  the  presence of  sulfur dioxide.    EPA  650/4-74-034,  U.S.
Environmental  Protection Agency,  Research Triangle Park, NC.

McRae, G.  J.,  W. R.  Goodin, and J. H. Seinfeld.   1979.   Development  of a
second-generation airshed model  for photochemical air pollution.   Proc.  4th
AMS Symp.  on Turbulence, Diffusion  and Air Pollution,  Reno,  Nevada, January
15-19.

Meszaros, E., D. J. Moore,  and  J.  P.  Lodge.   1977.   Sulfur dioxide-sulfate
relationships in Budapest.   Atmos.  Environ. 11:345.

Middleton, P., C. S.  Kiang,  and  V. A.  Mohnen.   1980.  Theoretical estimates
of  the relative  importance of  various  urban  sulfate  aerosol  production
mechanisms.  Atmos.  Environ.  14:463-472.

Miller, D. F.   1978.    Precursor effects on S02  oxidation.   Atmos.  Environ.
12:273-280.

Miller,  D. F.   1980.    A  model  of  S02  oxidation  in  smog.    EPA,  U.S.
Environmental  Protection Agency,  Research Triangle Park, NC.

Miller, D. F.  and A. J.  Alkezweeny.  1980.  Aerosol formation in urban plumes
over Lake Michigan.  Annals.  N.Y. Acad.  Sci. 338:219-232.

Miller, D. F.  and C. W.  Spicer.   1975.   Measurements  of nitric acid in smog.
J. Air. Poll.  Control  Assoc.  25:940-942.

Miller, D. F., A. J.  Alkezweeny,  J.  M. Hales,  and R. N.  Lee.   1978.  Ozone
formation related to power  plant  emissions.  Science 202:1186-1188.

Miller, J.  M.  and  R.  de  Pena.    1972.    Contribution  of scavenged  sulfur
dioxide to the sulfate of rainwater.  J. Geophys. Res. 30:5905-5916.

Miller, M. S., S. K.  Friedlander, and G. M. Hidy.  1972.  A chemical element
balance for the Pasadena aerosol.  In Aerosols  and Atmospheric Chemistry, G.
M. Hidy, ed.   Academic Press, New York,  NY.
                                   4-98

-------
Mills, E.  L.,  C. B. Murphy  Jr.,  and J. A.  Bloomfield.   1979.  Oxidants  in
precipitation.   Jji  Polluted  Rain.   T.  Y.  Toribara,  M. W.  Miller  and P.  E.
Morrow, eds.  Plenum Press, New York,  NY.

Moller,  D.   1980.    Kinetic model  of atmospheric  $62  oxidation  based  on
published data.  Atmos. Environ.  14:1067-1076.

Nash, T.  1979.  The effect of nitrogen dioxide and of  some  transition metals
on   the   oxidation   of  dilute  bisulphite   solutions.     Atmos.   Environ.
13:1149-1154.

National Academy  of Sciences.  1976.   Chlorine and hydrogen  chloride.    Ir±
Medical  and  Biologic  Effects of Environmental  Pollutants.  National Academy
Press, Washington, DC.

National Academy of Sciences.  1983.   Acid  deposition -  Atmospheric  Processes
in  Eastern North  American,   National  Academy  of Sciences  Report,  National
Academy Press,  Washington,  D.C.

Newman,  L.    1979.    General considerations  on  how  rainwater  must obtain
sulfate  nitrate  and acid.   In^ Proceedings  of  the International  Symposium  on
Sulphur Emission to the Environment,  London,  UK.

Newman,  L.   1981.  Atmospheric  oxidation  of  sulfur dioxide  as viewed  from
power plant and smelter plume studies.  Atmos.  Environ.  15(11): 2231-2239.
Noxon, J.  F.   1975.   N0£  in the  stratosphere  and  troposphere  by  ground
based absorption spectroscopy.   Science 189:547-549.

Noxon,  J.  F.,  R.  B.  Norton,  and  E.   Marovich.     1980.     NOa   in   the
troposphere.  Geophys. Res. Lett.  7:125-128.

Oblath, S. B., S.  S. Markowitz, T. Novakov, and  S. G.  Chang.   1981.   Kinetics
of the formation  of hydroxyl ami ne disulfonate by  reactions of nitrites with
sulfites.  J. Phys. Chem.  85:1017-1021.

Organization  for  Economic  Cooperation and  Development.    1977.   The OECD
programme  on long  range  transport  of  air  pollutants:    Measurements   and
findings.    Final  Report.    Organization   for  Economic  Cooperation   and
Development, Paris.

Ogren, J.  A.   1980.   Deposition  of particulate  elemental carbon  from  the
atmosphere.     Presented   at  General   Motors  International   Symposium  on
Particulate Carbon:  Atmospheric Life Cycle,  Warren, MI, October  13-14.

Okita, T.,  K.  Kaneda, T.  Yanaka, and  R.  Sugai.    1974.    Determination of
gaseous  and  particulate  chloride  and  fluoride in  the atmosphere.    Atmos.
Environ.  8:927-936.

Penkett,  S.  A.    1972.   Oxidation of  SOz and other atmospheric  gases by
ozone in aqueous solution.   Nature Physical  Science 240:105-106.
                                    4-99

-------
Penkett, S. A.,  B.  M.  R. Jones, K. A.  Brice,  and  A.  E. J. Eggleton.  1979.
The importance of atmospheric ozone and hydrogen  peroxide  in oxidizing sulfur
dioxide in cloud and rainwater.   Atmos. Environ.  13:123-137.

Perner, D. and  U.  Platt.  1979.  Detection of nitric acid in the atmosphere
by differential  optical  adsorption.  Geophys.  Res.  Lett. 6:917-920.

Petrenchuk, 0. P. and V. M. Drozdova.  1966.  On the chemical composition of
cloud water.  Tell us 18:280-286.

Petrenchuk, 0.  P.  and   E.  S.  Selezneva.   1970.   Chemical  composition of
precipitation    in   regions  of  the   Soviet   Union.      J.   Geophys.   Res.
75:3629-3634.

Platt, U., D.  Perner, J. Schroder, C.  Kessler, and A. Toennissen.  1981. The
diurnal variation of N03.  0. Geophys.  Res.  86:11965-11970.

Platt, U., D.  Perner,  A.  M.  Winer,  G. W.  Harris, and  J. Pitts Jr.   1980.
Detection  of   N03   in   the  polluted  troposphere  by   differential  optical
adsorption. Geophys. Res.  Lett. 7:89-92.

Prahm, L. P.,  U. Torp,  and  R. M. Stern.  1976.  Deposition  and transformation
rates  of  sulfur  oxides  during  atmospheric  transport  over the  Atlantic.
Tellus 28:355-372.

Pueschel, R. V. and  C.  C.  Van  Valin.   1978.   Cloud nucleus  formation  in a
power plant plume.   Atmos.  Environ. 12:307-312.

Radke, L.  F.    1970.   Field and  laboratory  measurements with  an  improved
automatic cloud condensation nucleus counter.   Preprints of Papers Presented
at  the American Meteorological   Society,  Conference  on Cloud  Physics,  Fort
Collins,  CO, August 24-27,  1970:7-8.

Radke, L.  R.  and P. V.  Hobbs.    1969.  Measurements of  cloud  condensation
nuclei, light  scattering coefficient,  sodium-containing  particles, and Aitken
nuclei in the  Olympic Mountains  of  Washington.  J.  Atmos.  Sci. 26:281-288.

Reynolds, S. D., T. W.  Tesche, and  L.  E.  Reid.  1979.  An  introduction to the
SAI airshed model  and  its usage.   SAI Technical Report  EF  79-31.   Systems
Applications,  Inc., San  Rafael,  CA.

Richards,  L.  W.,  J.  A. Anderson,  D.  L. Blumenthal,  A.   A.  Brandt, J.  A.
McDonald, N. Watus, E. S.  Macias, and  P. S. Bhardwaja.  1981.  The chemistry
aerosol physics, and optical  properties of a western coal-fired power plant.
Atmos. Environ.  15:2111-2134.

Robbins,  R. C., R.  D.  Cadle, and D. L. Eckhardt.  1959.   The conversion of
sodium chloride to  hydrogen chloride in  the   atmosphere.   J.  Meteorol.,
16:53-56.
                                   4-100

-------
Roberts, P. T.  and S. K. Friedlander.   1975.   Conversion  of  SOg to sulfur
particulate 1n  the  Los  Angeles  atmosphere.    Environ.  Health  Prospective
10:103-108.

Robinson,  E.  and  R.  E.  Robbins.    1969.   Sources,  abundance and  fate  of
gaseous  atmospheric  pollutants.    Report 6755, Stanford  Res.   Inst.,  Menlo
Park, CA.

Rodhe, H.  1978.  Budgets and turnover  times of atmospheric  sulfur compounds.
Atmos. Environ.  12:671-680.

Rodhe, H., P. Crutzen,  and  A.  Vanderpol.  1979.   Formation of sulfuric and
nitric  acid  in the  atmosphere during  long  range  transport.   Proc.  WMO
Symposium  on  Long  Range Transport  of  Pollutants, Sofia,  Bulgaria,  October
1-5.

Sadasivan,  S.    1980.   Trace  constituents  in cloud  water,  rainwater and
aerosol  samples collected near the  west  coast  of  India during  the southwest
monsoon.  Atmos. Environ. 14:33-38.

Sander, S. P. and R. T.  Watson.  1981.   A kinetics study  of the reactions of
S02 with CH302-   Chem.  Phys.  Lett.  77:473-475.

Sanhueza,  E.,  R.  Simonintis,  and  J.  Heicklen.   1979.    The reaction  of
CH302 with S02.   Int. J.  Chem.  Kinetics 11:907-914.

Saxena,  V.  K.,  J. N.  Burford, and J. L.  Kassner.   1970.   Operation  of a
thermal  diffusion chamber for  measurements on  cloud  condensation  nuclei.  J.
Atmos. Sci. 27:73-80.

Scatchard,  G.,  G. M.  Havanagh,  and L.   B.  Ticknor.   1952.   Vapor-liquid
equilibrium.  VIII.  Hydrogen  peroxide-water mixtures.  J.  Amer.  Chem.  Soc.
74:3715-3720.

Schroeter, L. C.   1963.   Kinetics of air oxidation of sulfurous acid salts.
J. Pharm. Sci. 52:559-563.

Schroeter, L. C.   1966.   Sulfur Dioxide:  Application  in Foods,  Beverages,
and Pharmaceuticals.   Pergamon  Press, Oxford, UK.

Schwartz, S. E.   1982.   Gas-aqueous reactions  of sulfur and nitrogen oxides
in liquid water clouds.  Presented at Acid Rain Symposium,  American Chemical
Society, Las Vegas, NY, March-April, 1982.

Schwartz, S. E.  and L.  Newman.   1978.   Processes limiting  oxidation of sulfur
dioxide in stack plumes.   Environ.  Sci. Technol. 12:67-73.

Schwartz, S. E.  and J. E. Freiberg.  1981.  Mass-transport  limitation to the
rate  of reaction in  liquid  droplets:    Application  to oxidation of  S02  in
aqueous solutions.   Atmos. Environ.  15:1129-1145.
                                   4-101

-------
Schwartz, S. E. and W. H. White.  1982.   Kinetics of  reactive dissolution of
nitrogen oxides into aqueous solution.  Jji Advanc. Enviro. Sci. Technol., 12,
S. E. Schwartz, ed. Wiley and Sons,  Inc.,  New  York.

Scott,  B.  C.    1982.   Predictions  of  in-cloud conversion rates  of -S02 to
$04  based  upon  a  simple  chemical  and  dynamical  model.    Atmos.  Environ.
16:1735-1752.

Scott,  W.  D.  and  P.  V.  Hobbs.   1967.   The  formation of  sulfate  in water
droplets.  J.  Atmos. Sci. 24:54-57.

Seinfeld, J. H., F. Allario, W.  R.  Bandeen, W.  L. Chaimeides, D. D.  Davis, E.
D. Hinkley, and  R.  W.  Stewart.   1981.  Report of  the NASA working group on
tropospheric program planning.   NASA Reference Publication  1062.   National
Aeronautics and Space Administration,  Washington, D.C.

Sequeira, R.   1981.  Acid  rain:   Some preliminary  results from global data
analysis.  Geophys. Res.  Lett. 8:147-150.

Shannon, J.  D.    1981.     A model  of  regional   long-term  average   sulfur
atmospheric pollution,  surface  removal,  and  net  horizontal flux.    Atmos.
Environ. 15:689-701.

Sheppard, J. C., M. J. Campbell, and B. Au.   1978.   Boundary layer hydroxyl
measurements  by  a  14C  tracer  technique.    Presented  before   the Div.
Environmental   Chemistry,  American   Chemical Society,  Miami, FL.    September
11-14.

Shu,  W.  R., R.  G.  Lamb,  and J.  H.  Seinfeld.   1978.  A model of second-order
chemical  reactions  in  turbulent   fluid  -   part  II.      Atmos.  Environ.
12:1695-1704.

Spicer,  C. W.   1977a.   The fate of nitrogen oxides in the atmosphere.  Adv.
Environ. Sci.  Technol.  1:163-261.

Spicer,  C.  W.    1977b.    Photochemical  atmospheric pollutants  derived from
nitrogen oxides.  Atmos.  Environ. 11:1089-1095.

Spicer,  C.  W.    1980.    The  rate of NOX  reaction  in  transported  urban air.
In Atmospheric Pollution 1980, Proc. 14th Int. Colloq.,  Paris,  May 5-8.  M.
"H7 Benarie, ed.  Elsevier Scientific Publishing Company,  Amsterdam.

Spicer,  C.  W.    1983.   Smog chamber studies  of NOX  transformation rate and
nitrate-precursor relationship.   Environ.  Sci.  Technol. 17:112-120.

Spicer,  C. W.,  D. W. Joseph, P.  R.   Sticksel, G. M. Sverdrup, and G. F.  Ward.
1979. Reactions and transport of nitrogen  oxides and ozone in the  atmosphere.
Battelle-Columbus Report to EPA.
                                    4-102

-------
Spicer,  C.  W.,  J.  R. Koetz, G. W.  Keigley,  G.  M.  Sverdrup, and G. F.  Ward.
1981a.   A study of  nitrogen oxides  reactions  within  urban  plumes  transported
over the ocean.  EPA, U.S. Environmental  Protection Agency, Research Triangle
Park, NC.

Spicer,  C.  W., G.  M.  Sverdrup, and  M.  R. Kuhlman.   1981b.   Smog  chamber
studies  of  NOX   chemistry  in  power   plant   plumes.     Atmos.  Environ.
15:2353-2366.

Stephens, E. R., W.  E.  Scott,  P.  L. Hanst, and  R.  C.  Doerr.   1956.   Recent
developments  in  the study  of  the  organic chemistry of  the atmosphere.  J.
Air. Pollut. Contr. Assoc. 6:159-165.

Stewart, D. A. and M.  K.  Liu.   1981.    Development  and  application  of a
reactive plume model.  Atmos.  Environ.  15:2377-2393.

Stockwell,  W.  R.   and  J.  G. Calvert.   1983.   The mechanism  of  the HO-S02
reaction.  Atmos.  Environ. 17:2231-2236.

Su,  F.,  J.  G. Calvert,  J. H.  Shaw,  H. Niki,  C. M.  Savage,  and  L.  D.
Breitenbach.   1979.   Spectroscopic and kinetic  studies  of a new metastable
species  in  the photooxidation  of  gaseous formaldehyde.    Chem.  Phys.  Lett.
65:221-225.

Su, F., J.  G.  Calvert, and J. H. Shaw.   1980.  A FTIR  spectroscopic study of
the ozone-ethane  reaction mechanism  in  Op-rich mixtures.   J.  Phys.  Chem.
84:239-246.

Sumi, L., A. Corkery, and J. L. Monkman.   1959.  Calcium and sulfate  content
of urban air.  Amer. Geophys.  Union, Geophys.  Mon.  3:69-80.
Sze,  N.  D. and M.  K. W.  Ko.   1980.   Photochemistry of  COS,  C$2,
and H?S:   Implications for the atmospheric  sulfur  cycle.    Atmos. Environ.
14:1223-1239.

Takeuchi,  H.,  K.  Takahashi ,  and N.  Kizawa.   1977.    Absorption  of nitrogen
dioxide  in sodium sulfite solution from air as  a diluent.  Ind. Eng. Chem.
Process Des. Dev.  16:486-490.

Tanaka,  S.  M.  Dargi ,  and  J.  W. Winchester.   1980.   Short term  effect of
rainfall on elemental composition and size distribution of aerosols in north
Florida.  Atmos.  Environ.  14:1421-1426.

Tanner, R. L., R.  Cederwall ,  R.  Garber,  D. Leahy,  W.  Marlow,  R.  Meyers, M.
Phillips,  and  L.  Newman.   1977.   Separation and analysis of aerosol sulfate
species at ambient concentrations.   Atmos.  Environ. 11:955-966.

U.S. Department of Energy.   1979.   Monthly  Energy Review, DOE/E1A0035/10(79) ,
October.

Urone, P., H.   Instep, C. Noyes, and J. T.  Parcher.   1968.  Static studies of
sulfur dioxide reactions in air.  Environ.  Sci . Technol . 2:611-618.
                                    4-103

-------
Van den  Heuval,  A. P.  and B. J. Mason.   1963.  The  formation  of ammonium
sulphate in water  droplets exposed  to gaseous  sulphur  dioxide  and ammonia.
Quart. J. R. Met.  Soc.  89:271-275.

Walcek, C., P. K.  Wang, J. H. Topalian, S. K.  Mitra,  and H. R.  Pruppacher.
1981.   An experimental  test of a theoretical model  to  determine  the rate at
which  freely  falling  water  drops  scavenge  S02  in  air.   J.  Atmos.  Sci.
38:871-876.

Whitby, K.  T.   1978.  The physical characteristics of sulfur  aerosols. Atmos.
Environ. 12:135-159.

Whitby, K.  T.  1980.   Aerosol formation in urban plumes.  Annals N.Y. Acad.
Sci. 338:258-275.

Whitby, K.  T., R. Vijayakumar, and G. R. Anderson.  1980.   New particle and
volume formation rates  in  five  coal-fired  power plant plumes.  Presented at
the Symposium on  Plumes:   Measurements and Model  Components, Grand Canyon,
AZ.

White, W. H., J.  A.  Anderson, D. L. Blumenthal, R.  B. Husar, N.  V. Gillani,
J. D. Husar, and W. E.  Wilson.   1976.   Formation  and  transport of  secondary
air pollutants:  Ozone  and  aerosols  in  the St. Louis  urban  plume.   Science
194:187-189.

Whitney, R. P. and J.  E.  Vivian.   1941.   Solubility  of chlorine  in water.
Ind. Eng. Chem.  33:741-744.

Whitten, G. Z.  and  H.  Hogo.   1977.   Mathematical  modeling  of  simulated
photochemical  smog.  U.S. EPA Technical  Report,  EPA-6003-77-011.

Whitten, G. Z.,  H. Hogo, and J.  P. Kill us.   1980.  The  carbon bond mechanism:
A condensed kinetic mechanism for photochemical  smog.   Environ. Sci. Technol .
14:690-700.

Wilson,  W.  E.   1978.   Sulfates  in  the atmosphere:   A  progress  report on
Project MISTT.  Atmos.  Environ.  12:537-547.

Wilson, W.  E.  1981.   Sulfate formation in  point-source  plumes:  A  review of
recent field studies.   Atmos. Environ. 15:2573-2582.

Winchester, J. W.  1983.    Sulfur,  acidic aerosols,  and acid  rain  in the
eastern United States.  Ch.  6 in Adv.  Environ.  Sci. Technol. Vol.  12.  John
Wiley and Sons,  New York.

Wine, P. H., R.  C. Shah, and A.  R. Ravishankara.  1980.   Rate of reaction of
OH with CS2.  J.  Phys.  Chem. 84:2499-2503.

Winer, A.  M.   1979.    Detection of  nitrous acid  and  nitrate radical in the
polluted  atmosphere  by differential  optical   absorption  spectroscopy.  In
Proceedings  of  the  Workshop  on  the  Formation  and  Fate  of  Atmospheric
Nitrates.  EPA,  Research Triangle Park,  NC.
                                    4-104

-------
Winer, A. M., G. M. Brewer, W. P. L. Carter, K. R. Darnell, and J. N. Pitts.
1979.  Effects of ultraviolet spectral  distribution on the photochemistry of
simulated polluted atmosphere.  Atmos. Environ. 13:989-998.

Winkelmann,  D.     1955.    Die  electrochemishe  messung   von  oxidations
geschwindigkeit von  NOaSOs  durch gelosten  sauerstoff.    Z.  Electrochemie
59:891-895.

Winkler, E.  M.    1976.   Natural  dust and  acid  rain.  Water,  Air,  and Soil
Pollut. 6:295-302.

Yue,  G.  K.  and P.  Hamill.    1979.    The homogeneous  nucleation rates  of
          aerosol  particles in  air.  J. Atmos. Sci. 10:609-614.
                                    4-105

-------
               THE ACIDIC DEPOSITION PHENOMENON  AND  ITS  EFFECTS

              A-5.  ATMOSPHERIC CONCENTRATIONS AND DISTRIBUTIONS
                         OF CHEMICAL SUBSTANCES

                              (A. P. Altshuller)

5.1  INTRODUCTION

Air quality measurements of those substances  that may contribute directly  or
indirectly  to  acidic  deposition processes  are discussed  in  this chapter.
Substances  such  as  sulfur  dioxide  and  nitrogen dioxide  may  contribute  to
acidic deposition in two ways:  (1) They can undergo  dry  and wet deposition  to
soil and subsequently undergo reactions to acidic species  in soils; (2)  They
can  undergo atmospheric  chemical  transformations  to   particle  sulfate and
gaseous and particle forms of nitrate  which,  in turn, can  undergo  deposition
to  soils,  lakes,  and   streams.   These  substances  may  be acidic  in  their
original   forms  as  are NH4HS04,  H2S04,  and  HN03,  or  they  may   undergo
reactions in  soil  that result in release  of hydrogen  ions.   Ammonia is  an
important nitrogen  species  that can neutralize airborne acidic substances,
but in soils in the form of ammonium ion it can  react to form hydrogen ions.

A number of other elements are of interest as airborne  substances.  Alkaline
earth metals such as calcium can  react as  calcium  ions  to  neutralize acidic
substances.   Iron  and  manganese  ions  are of  significance to the extent  that
they can  be demonstrated to  participate in  catalytic  reactions  in  aqueous
droplets to  enhance the conversion of sulfur  dioxide to  sulfate (Chapter  A-4,
Section 4.3.5).  Other  airborne metallic elements may,  upon deposition,  have
possible adverse biological effects in soils, lakes,  and streams.   Aluminum
and manganese ions have been  identified  as possible causes of toxic  effects
in soils (Chapter  E-2,  Section  2.3.3.3.2).   Aluminum ions  are of particular
concern in causing adverse  effects in lakes and  streams  (Chapter E-4,  Section
4.6.2).   Zinc, manganese,  cadmium,  lead,  and  nickel   at  sufficiently  high
concentrations also can have toxic effects  in lakes  and  streams (Chapter  E-5,
Section 5.6.4.2),  and indirect health effects have been  associated with lead,
aluminum, and mercury (Chapter E-6).

Ozone  and hydrogen peroxide participate  in  oxidation  of  sulfur  dioxide  to
sulfate in aqueous droplets  (Chapter A-4,  Section 4.3.5.3). The ambient air
concentrations of  both  of  these oxidants  will  be  considered,  although  sub-
stantial  difficulties  have  been encountered  in the measurement  of hydrogen
peroxide.

The effect  of  light scattering  by  submicron aerosols  such as  sulfates and
nitrates is  significant in the  areas  of eastern North America  impacted  by
acidic deposition.  Particle sulfate appears  to be  particularly important  in


                                    5-1

-------
Its adverse  effects on visibility  when  suspended in  air  and a  significant
contributor to acidic deposition to soils, lakes, and streams.  Therefore, a
discussion of visibility  degradation  effects  of these aerosol  species  is
included in this chapter.

Measurements of airborne substances that may contribute to acidic deposition
are of particular  interest  in  rural  areas.    However,  in  the  past,  most
measurements of airborne substances were  made in urban  areas.   Cities  were
the major  sources  of pollutants of concern until after  World War II.   They
still   contribute  substantially  to the  total  burden of  airborne  sulfur and
nitrogen compounds.  Urban plumes  also  are significant because, through dry
and wet  deposition  processes,  they contribute directly  to  the loading  into
soils,  lake,  and  streams  substantially  downwind  of  cities  (Chapter  A-3,
Section 3.4.2).

5.2  SULFUR COMPOUNDS

5.2.1   Historical  Distribution  Patterns

Substantial changes in the geographical  and seasonal distributions of  sulfur
oxides and  in the  stack heights  of emission sources  of sulfur oxides  have
occurred  over time.   Many  of these  changes  occurred  before  air   quality
monitoring networks were established.

Wood was  the  predominant fuel  used in the United States until the late  19th
century  (Schurr et  al. 1960)  when coal  use  began  to increase.   The coals
burned, unlike wood, contained substantial  amounts of sulfur, emitted  to the
atmosphere as sulfur oxides.   Before and during World War II, the major  uses
of coal  included residential/commercial  heating,  production of coke,  and the
operation  of  railroad  locomotives (Schurr  et   al.  1960).    Most  of these
sources  of sulfur  oxide  emissions,  except  for locomotives,  were  in  the
cities.   In  addition,  small  coal-fired  power plants were  often  located  in
cities.   Thus, most sulfur oxides were  emitted  from sources  near the  ground
surface.  These  near-surface  emissions   plumes  impacted   on  the   adjacent
countryside resulting in high  sulfur  oxide concentrations  in and near urban
centers.

Coal usage declined in the  United States immediately after World War  II.   By
the late  1940's  and 1950's,  coal  use  in residential/commercial  heating and
railroad locomotives dropped  off rapidly  as coal  was  replaced  by oil  and  gas.
In cities, coal  for residential/commercial  heating was  replaced by gas, which
reduced sulfur oxide emissions  substantially,  and by  fuel oil  containing  high
sulfur contents,  which did  not reduce  sulfur  oxide emissions appreciably.
Sulfur oxide  emissions  increased  in the 1960's  from industrial  sources and
the  rapid growth  of  electric  utility  sources.    However,  emissions  from
industrial sources decreased  in the 1970's  (Chapter  A-2,  Figure  2-6).   In the
late  1960's  and  early 1970's,  regulations  were  enacted  to limit the  sulfur
content of fuels, thus reducing emissions  from fuel  oils.   These  regulations
were applicable in particular to cities  in  the northeastern  United States.

The spread of cities into  suburban areas after  World War  II  resulted in  more
diffuse  sources  of urban  plumes, although  emission  sources  in   suburban


                                     5-2

-------
areas  usually  used  low-sulfur  fuels.   Coal-fired  electrical  utility capacity
in  the midwestern and southeastern United  States increased rapidly.   These
power  plants  were constructed outside of  cities  and with  increasingly  tall
stacks.   By  the 1970's,  numerous large power  plants with stacks of  varying
heights  were  distributed  throughout   nonurban  areas of  the United  States.
These  complex  and varied  emissions  sources contributed  to  the loadings  of
sulfur oxides  in rural areas on a seasonal  and annual basis.

Where  local  contributions  are  negligible,  the impact of  urban  plumes on re-
mote areas is  unclear, although long-range  transport is  more likely  in winter
(Chapter  A-3,  Section  3.4.2)  because  of atmospheric conditions.  The plumes
from sulfur oxide emission sources with tall stacks  can be  isolated from the
surface  for  varying  diurnal  periods   depending  on  the hour of release  and
season of the year (Chapter A-3, Figures 3-19,  3-20, 3-21, and  3-22).   During
these  diurnal  periods, these  sources  contribute to  the total  sulfur  loading
of  the lower troposphere,  but  not  to the  sulfur  oxides  measured   at  ground
level.   Therefore,  ground-level monitoring  alone is inadequate to evaluate
the total sulfur  loading of  the atmosphere available to  participate  in  sub-
sequent wet  and dry deposition.  Chapter A-8  presents  further  discussion  of
deposition monitoring.

5.2.2  Sulfur Dioxide

5.2.2.1   Urban  Measurements—Most of  the  sulfur content of  fuels is  emitted
to  the atmosphere in the  form  of  sulfur dioxide  ($03).  Sulfur dioxide  was
monitored in various large cities  in  earlier years, but  no  nationwide  moni-
toring network existed until  the 1960's.

Jacobs  (1959a)  reported  ambient air  concentrations  of  S02  in Manhattan  and
several other  sites in the New  York,  NY, area  for 1954-56, with higher  con-
centrations in winter than  in summer.   The  diurnal profiles  showed midmorning
and  late  afternoon  peaks  or  early  morning  peaks in  S02 concentrations.
Jacobs  reported  hourly  S02  concentrations  as  high   as 2500  to  3000  yg
m~3  during  some  winter  and  fall  air  stagnation episodes.   On  an  annual
average basis,  S02  concentrations  at  the Manhattan  monitoring site averaged
420, 520, and  500 yg  m"3  in  1954, 1955,  and  1956, respectively.    Methods
of  sampling and chemical  analysis were reported also (Jacobs  1959b).

A National Air  Sampling Network  (NASN) was  initiated in the  United  States  in
the 1950's, but sulfur dioxide  was not measured until  the early 1960's.   In
comparison with the  S02  concentrations reported by Jacobs (1959a), the  NASN
measurements in Manhattan  in  1964  and  1965  averaged  450  and  370 yg  m~3,
respectively (Dept.  of Health, Education and  Welfare 1966).  These  results
appear to indicate relatively little  change  in  concentration from the 1950's
to  the  mid-1960's.   This  is  not unexpected because fuel  sulfur content was
not restricted during this  time.

In  the 1963-72  period  the  decreasing  order  of annual   average   SO?  con~
centrations  was  (1)  East  Coast,  (2)  Midwest (east  of  Mississippi),  (3)
Southeast, (4)  West Coast,  and  (5) Midwest (west  of the  Mississippi  River),
and (6) western states.  Many  urban   sites  west of the Mississippi   River had
                                     5-3

-------
S02 concentrations averaging only  10  to  20 percent of the concentrations at
sites on the East Coast (Altshuller 1973).

Trends in the annual average, seasonal, and episodic concentration levels of
S02 with  time have  been evaluated by  geographical  region  and  in specific
urban  areas  (Altshuller 1980).    Between  1963-65  and 1971-73,  S02 concen-
trations (3-year  quarterly  averages)  at  urban  sites decreased  by about 80
percent in the northeastern United States  (Figures 5-1 to  5-4)  and by 30 to
50 percent in the midwestern United States (Altshuller 1980).  The declining
SOg concentration levels in cities appear  to  relate  better to reductions in
local   sources of  sulfur  oxide  emissions than  to  regional-scale  utility
emissions.

S02  concentrations   in   the  northeastern  United  States,   in the earliest
period  (1963-65)  for which  measurements  are available,  by  quarter  of the
year, were  in the order:   fourth  quarter >  second  quarter  > third quarter
(Figures 5-1  to  5-3).    In 1971-73,  the  same  order  prevailed (Altshuller
1980).

Trends in S02 concentrations  in urban areas  in  the  1970's are available on
an annual  average basis for the  United States  and geographical  regions within
the United  States (U.S. EPA  1977a,  1978b).   Based  on  1,233 U.S. sampling
sites,  the  composite average  of  urban  S02  concentrations  decreased  by 15
percent between  1972 and 1977  from the  1972 level  of  23 yg nr3  (U.S. EPA
1978b).  The  90th percentile concentrations  of  S02  decreased by  23 percent
between 1972  and  1977   from  a   1972  level of  52 yg  nr3.   There  were no
significant changes  in  either the 90th  percentile concentrations  or in the
composite average concentrations during the last  few  years  of the 1970's.

By  the latter part  of  the  1970's,  ambient  air concentrations of  S02 nad
been  reduced  to relatively low  levels.   In  1976  the  composite annual average
(and  90th  percentile)   concentrations  were:    United States—20 yg  nr3 (40
yg  m-3),   New   England--25  yg  m-3  (40   yg   m-3);   Great  Lakes--28  yg
nr3  (50  yg  nr3)   (U.S.  EPA   1977a,  1978b).    These   concentrations   were
well  below  the S02  concentrations experienced  in  the  1960's  or  the early
1970's. During the last few years, S02  concentration levels appear to  have
stabilized.

5.2.2.2    Nonurban  Measurements—Measurements   for  S02   concentrations  at
nonurban sites  in the  United  States  are  more  limited  than  those at urban
sites.  In addition, the concentrations measured  often are  near the limits of
detectability.  Measurements of  six nonurban  sites in the  United States  over
a  period of years for  which results are available in the  NASN data bank are
listed in Table 5-1.

The  annual  average  concentrations  range  near  10  yg  nr3.      First-  and
fourth-quarter concentrations  often  exceeded second-quarter  concentrations,
and concentrations during  the  third quarter  of  the  year were almost  always
the  lowest values  at  each site.   No  clear trends  in  nonurban S02  con-
centrations with  time  are evident on  an  annual  average  or  quarterly  basis
(Figures  5-1  to  5-3).   Although average  S02  concentrations  at nonurban
sites were much lower  than  at urban  sites during the 1960's,  the  difference


                                     5-4

-------
       25
       20
  o>
oo
z
o
I—1
tn  i-u
i   o
01  Z
   o
   o
  2
  _)
       15
       10
        0
Figure 5-1.
                                                                  LEGEND


                                                        NORTHEAST - URBAN, SECOND QUARTER

                                                          O  SULFATE


                                                          D  SULFUR DIOXIDE

                                                        NORTHEAST - NONURBAN, SECOND QUARTER

                                                          A  SULFATE


                                                          O  SULFUR DIOXIDE
                                                                                           I
                                                                                                     125
                                                                                                    100  -
                                                                                                           0>
                                                                                                       75
                                                                                                     50
                                                                                                     25
            1963-65
                           1965-67
                                    1967-69
1969-71
1971-73
1973-75
1975-77
                                           THREE YEAR AVERAGE
               Three year running  average  sulfur dioxide and  sulfate  concentrations  during second quarter

               of year for urban and  nonurban sites  in the  northeastern  United  States.   Adapted from

               Altshuller (1980).
                                                                                                          o
                                                                                                          o
                                                                                                          X
                                                                                                          o
                                                                                                          (SI

-------
                                                9-S
     to
     c

     ID

     cn
      i
     ro
1-4 fD 3-

   -s ft)
     QJ
   C -S
   -s
   cr -s
   01 c
   at ->•
   3 3
   Q.IQ

   3 Q)
   O <
   3 fD
                                   SULFATE  CONCENTRATIONS  (yg  m3)
Q) fD
3
   CO
CO C
—J. «J
n- -b
fD c
co -5

-i. Q.
3 -"•
   O
rl- X
3- ->•
fD Q.
   fD
3
O Q>
-S 3
<-(• Q.

fD CO
Oi C
CO —i
a- -h
fD O)

3 fD

cr o
3 O
-•• 3
<-+ o
fD fD
Q. 3
   c-h
CO -S
rt- QJ
O) r+

fD O
CO 3
   to

   Q.
> C
CX -S
0, _i.
-O 3
ct'Q
fD
Q. rt-
   -s  -s
   o  a.
   J> c
   — ' Q)
   r+ -S
   CO <-h
   3- fD
   C -S

   — ' O
   fD -h
73
m
m
          m

          i
          m
          co
                                 cn
                                                                                    no
                                                                                    o
                VO
                CT>
                CO
              VO
              cn
               i
                VO
                cn
                •^i
                 i
                cn
                vo
                vo
                cn
                 I
                •Vl
                co
                co

                »vi
                cn
                 
-------
     IQ
     -5
     n>
     tn
      i
i-* n> 3~
vo n> -5
oo -s n>
o   ID
   c -s
   -s
   cr -s
   01 c:
   3 3
     3
   QJ _..
   3 3
   CLtQ

   3 DJ
   O <
   3 
   C -5
   -S 0>
   CTlQ
   cu n>
   3
     in
   10 c
   n>
   VI
   -i. a.
   a> o.
     n>
   3
   O CU
n- -h
n> o>
-S c*
3 (D

c: n
3 o
-"• 3
<-*• o

   S
   «-»•
00 -5
c-t- O;
fa c+
<-*• -••
n> O
w 3
   tn
   S.
   Q. -S
   (D _..
  TD 3
   rt-to
   n>
   Q. -h
     O
   -h C
   -S -5
   O rt
                                                      L'S


                                    SULFATE  CONCENTRATIONS  (yg m3)
                                                                              ro
                                                                              o
                                                                                              fNJ
                                                                                              tn
             10
             Oi
           tn
           vo
           en
           tn
   -I  -<
   o  ^
   x  2
           i
           o»
           vo
             «£>
             cn
             vo
              i
         oo
              i
              ^j
              oo
              "•J
              oo
              i
              >j
              tn
           vo
           >*i
           tn

           -vj
   in -5
   3- rt-
   C O>
CD  O
-S  -h
                 V
              -I
                                ro
                                en
                                             en
                                             o
                                                                tn
                                                                              O
                                                                              o


                                  SULFUR DIOXIDE  CONCENTRATIONS  (yg  m3)
                                                                                                 ro
                                                                                                 tn

-------
   CO
   a.

   ce.
   o
   Q.
       20
Ul

I
   O


   O
   o
   o
   o
   ec.
   co
   co
X
o
(—1
o

OL
   to
       10
                                                                                   LEGEND

                                                                            NORTHEAST S02 AIR QUALITY


                                                                            MIDWEST S02 AIR QUALITY

                                                                            NORTHEAST (EXCLUDING PENNSYLVANIA)

                                                                             S02  POWER PLANT EMISSION

                                                                            MIDWEST (EXCLUDING PENNSYLVANIA)

                                                                             S02  POWER PLANT EMISSION

                                                                            U.S. S02 EMISSION FROM POWER

                                                                             PLANTS
                                                                                                 •4-
   <   QL-L.
                                           200
                                                                                                                       O)
                                                                                                               100
                                                                                                                    o

                                                                                                                    o
                                                                                                                    CJ
                                                                                                                       X
                                                                                                                       o
                                                                                                                       a:

                                                                                                                       13
                                                                                                                       U-
                                                                                                                       co

                                                                                                                       LU


                                                                                                                       i
          1960
                                     1965
1970
1975
                                                           YEAR
       Figure 5-4.  Annual average urban ambient air concentrations and emissions  (million  tons)  of  sulfur  dioxide

                    in northeastern and midwestern United States.  Adapted from Altshuller  (1980).

-------
         TABLE 5-1.  SULFUR DIOXIDE CONCENTRATIONS AT  NQNURBAN  SITES
                 IN THE EASTERN UNITED STATES (in yg nT3)
                        (ADAPTED FROM NASN DATA BANK)
Site
First
quarter
Second
quarter
Third
quarter
Fourth
quarter
Annual
average
Acadia National  Park, ME
1968
1969
1970
1971
1972
1973
Coos County, NH
1970
1971
1972
1973
Calvert County,
1970
1971
1972
1973
8
12
15
19
6
9

ND
12
7
13
MD
ND
20
5
12
Shenandoah National
1968
1969
1970
1971
1972
1973
20
16
16
15
10
18
7
9
7
11
6
NDa

ND
10
6
ND

ND
15
6
9
Park, VA
5
7
6
8
5
8
                                         5           9           10
                                         889
                                         8          15           11
                                         7           9           13
                                         677
                                        ND          ND
                                        12           8           -
                                         799
                                         499
                                        ND          ND
                                        10          18
                                         8           9          13
                                         697
                                        ND           8           -
                                         6          11           10
                                         9          11           11
                                        11           8           11
                                         7          10           11
                                         5          19            9
                                         679
                                     5-9

-------
                         TABLE 5-1.  (CONTINUED)
First
Si te quarter
Second
quarter
Third
quarter
Fourth
quarter
Annual
average
Jefferson County, NY
1970
1971
1972
1973
Monroe County,
1967
1968
1969
1970
1971
1972
1973
ND
8
3
8
IN
19
13
19
13
11
15
30
ND
5
5
19

5
7
10
8
8
10
11
16
6
5
ND

6
7
8
16
7
7
10
ND
7
9
25

33
12
18
10
14
15
10
_
7
6
—

11
10
14
12
11
11
15
aND = not detectable.
                                     5-10

-------
between  urban  and nonurban S02 concentrations narrowed substantially  in  the
1970's.

Mueller  et  al.   (1980)   reported  measurements  from  the  Sulfate  Regional
Experiment  (SURE)  obtained  from  a 54-station  nonurban  network operated  in
August and  October  1977  and mid-January, February, April, July, and  October
1978.    The  S02   concentrations  measured in  New England  and the  Southeast
were  almost always  below 26  yg  m~3, except  during January-February  1978.
Monthly  average   isopleths  for S0£  of  between  26  and  52  yg m-3  included
varying  portions of  several midwestern and mid-Atlantic States  from month  to
month  during the  study.   Monthly average  S02 concentrations  of about  80
vig  m~3  were shown  for  small  areas  in August  1977  and  January-February
1978.   The highest  SOg  concentrations tended  to  be in portions of the  Ohio
River Valley and  western Pennsylvania.   These concentrations of S02  at  SURE
sites were  substantial compared to those reported  at  urban  sites in  the  late
1970's.   However,  other measurements in  western  Pennsylvania  in July  and
August   1977  resulted   in   average   SOg  concentrations   of  18  yg  m~3
(Pierson et al. 1980a),  which  are substantially lower than  those reported  by
Mueller  et al. (1980).

S02  measurements  at  rural  sites in  Union  Co.,  KY,  Franklin  Co.,  IN,  and
Ashland  Co.,  OH,  were reported between  May  1980 and August 1981.   Monthly
average  S02 concentrations  ranged  from as  low  as 8  to 10  yg nr3  during
summer months  to  as  high  as 30  to  40 yg  nr3  during  the  winter  months
(Shaw and Paur 1983).

A number of  Canadian monitoring  networks were established during  the  1970's
(Whelpdale and Barrie  1982).   While precipitation measurements  have  received
the  greater  emphasis in  these networks,  air quality measurements  for  sulfur
dioxide  are available from the Air and Precipitation Monitoring Network (APN)
(Barrie  et al. 1980,  1983;  Whelpdale  and Barrie  1982).  Six monitoring sites
east of  Manitoba  are in operation at rural  locations.   Sulfur dioxide  is  col-
lected on  a  24-hour integrated basis on a chemically impregnated  filter.   A
low-volume  sampler   operates  at  a  flow rate  of   about  20  a  min'1  at  an
elevation  of 10  meters.   The geometric means of  24-hour  average S02  con-
centrations  on  a yg m~3  basis  for  the period  November  1978  to December
1979 are:   Long  Point, Ontario,  11;    Chalk River,  Ontario,  5.5; ELA-Kenora,
Ontario, 0.86; Kejimkujik,  Nova  Scotia,  0.86  (Barrie et al.  1983).   Large
concentration fluctuations are observed  at these sites and  are  attributed  to
the  alternating presence of clear background air  and  air polluted by large
S02 sources in the Lower Great Lakes  area (Barrie et al. 1980).

In  Europe,  annual mean  SOo concentrations  range  from about 20  yg  nr3  in
rural areas of the United Kingdom, the Netherlands, and the  Federal Republic
of  Germany  to concentrations  of  2  yg nr3  or lower  in the remote areas  of
northern and western  Europe  (Ottar 1978).   This range of S02  concentrations
over rural  areas in  Europe  is  close to the range  of concentrations discussed
above for rural areas of North America.

Georgii  (1978) has  reviewed aircraft measurements  of  S02 over the European
Continent.    The  average  concentration  of  S02  decreased  from about  5  yg
nr3  at  2  to 3  km   altitude  to  1  yg m~3  at  5  km  altitude.   From other


                                     5-11

-------
aircraft flights, Georgii  and Meixner  (1980)  obtained  a mean concentration of
1.3 yg nr3 above 6 km over Europe.

5.2.2.3   Concentration  Measurements  at  Remote  Locations--Meszaros   (1978)
reviewed remote  measurements  of  S02 concentrations.  Several   investigations
had been reported of  SOg  concentrations  as a  function of latitude over the
Atlantic  Ocean.    Concentrations of  S02  ranging  from  0.1  to  0.2  yg itr3
were observed at latitudes  above  60°N and below  10°N  in  the  northern  hemi-
sphere as well as in  the  southern hemisphere.   Between latitudes of 10°N and
60°N  over  the Atlantic  Ocean SOe  concentrations increase to  1  yg  nr3 at
25°N  and  at  55°N  latitude and  peak  at  about 3 yg  nr3  at  40°N  latitude.
These large  increases in  S0£ concentrations at midlatitude were  attributed
to continental emission  sources.   Other  investigations resulted in measure-
ments  of  S02  averaging  0.3  yg  nr3  over  the  Pacific  Ocean  and  0.2 yg
m-3 over tne Indian  Ocean  (Meszaros  1978).

Measurements  of  S02  concentrations were  obtained  in  aircraft flights  over
remote areas  as  part  of the 1978 Global  Atmospheric  Measurements  Experiment
of Tropospheric Aerosols and Gases  (GAMETAG) by Maroulis  et al.  (1980).  The
areas sampled were between 57°S and 70°N and included the  central  and  south-
ern Pacific  Ocean and the western  section  of  the United States and  Canada.
The average  $03  concentrations  reported  in pptv  were as  follows:   northern
hemisphere,  boundary  layer,  89;  free troposphere,  122; southern hemisphere,
boundary layer,  57;  free troposphere, 90.   The  S02  concentrations in  pptv
over  marine  and  continental  environments were  as follows:  marine boundary
layer, 54; free troposphere, 85;  continental  boundary  layer, 112; free tropo-
sphere. 160.   The boundary layer S02 concentrations  were in  the 0.1  to 0.3
yg  m~3  range   in  reasonable  agreement  with  other  remote   measurements
(Meszaros 1978).  Bonsang et al. (1980)  reported S02 concentrations  ranging
from  0.03  yg itr3  over  the  tropical   Indian  Ocean  to  0.3  yg   nr3  over
the Peruvian upwelling.   A  relationship was  identified  between  the  atmos-
pheric S02  concentrations  and the biological  activity in  sea  surface  waters
(Bonsang et al. 1980).

The S02  concentrations measured  at many  remote  sites are  factors of  10  to
100 less than those measured at rural  sites in  eastern North America (Section
5.2.2.2).    However,  the  S02  plume  from  eastern North  America  appears  to
cause  large increases  in the S02  concentrations  measured at  midlatitudes
well  into  the Atlantic  Ocean (Meszaros  1978).   A similar impact of  large
plumes from  strong  source  areas  has been observed at several  rural Canadian
sites (Barrie et al. 1983).

5.2.2.4  Comparison of Sulfur Dioxide Emissions and Ambient Air Concentration
— In  Chaper  A-2,  Section  2.3.2,  the historical  trends  in  sulfur  dioxide
emissions  are discussed.   Total  sulfur  dioxide  emissions  increased  rapidly
during the 1950's, more slowly during the 1960's, peaked  at or somewhat after
1970,  and decreased  somewhat by 1975.   The  sulfur  dioxide  emissions from
electric  utility fossil-fuel  power  plants continued  to increase until  1975.
The  sulfur  dioxide  emissions from industrial  sources  started to  decrease
rapidly  after about  1965 and continued  to decrease into the  1970's.   By
state,  the historical  trends  varied  substantially.   After 1970  the  sulfur
dioxide  emissions in  New England decreased substantially.  In Kentucky and


                                     5-12

-------
West Virginia  the sulfur dioxide  emissions  continued to increase  from  1950
into the 1970's.   In  the area consisting of the states  of  Pennsylvania,  New
York,  Ohio,  Indiana,  and Illinois, the  sulfur dioxide  emissions  increased
rapidly between 1950 and 1960, remained about constant from 1960 to 1970,  and
decreased after 1970.

As discussed  in Section 5.2.2.1, sulfur dioxide concentrations  within  urban
areas  started  decreasing  during  the  1960's  and  continued decreasing into  the
1970's.  These  decreases  in  sulfur  dioxide  concentrations within urban  areas
are consistent with the decreases in sulfur dioxide emissions from industrial
sources.   Within  some  urban areas  on  the  east coast  of the  United  States
emission regulations  also  resulted  in  either the  use of low-sulfur coal  or
residual fuel  oil in  power  plants  (Altshuller 1980).   As a  result,  sulfur
dioxide emissions  from power plants as well as industrial  sources  decreased
in these urban areas from the late 1960's onward.

No clear trends are evident  in  the  sulfur dioxide  concentrations in nonurban
areas  on a  regional  average basis.   This  lack of  trend in  sulfur  dioxide
concentrations  in nonurban  areas  does not  appear consistent  with the  in-
creases in sulfur  dioxide emissions  which did occur  after 1965  from electric
utility plants constructed  in  nonurban  areas  (Chapter  A-2,  Section  2.3.2).
However, the varying  patterns of trends  of  emissions from state  to  state  do
complicate the relationships between emissions  and  ambient air concentrations
in nonurban areas.  In general,  there was a  shift  from  emissions from  indus-
trial  sources  discharged  near ground level  to emissions  from  tall  stacks  of
power  plants.   A  substantial  fraction of sulfur dioxide  emissions from  elec-
tric utility  power plants  with  tall  stacks in nonurban areas  are  emitted
aloft  and remain aloft over long distances.   A  portion of these emissions  are
eventually removed by  wet scavenging while another portion  can  pass on  aloft
into Canada or  over the  Atlantic.   These sulfur dioxide  emissions  would  not
contribute to  sulfur  dioxide concentrations  measurable  at ground-level  moni-
toring sites in the United States.  As  a result, the  increase in ground  level
sulfur dioxide  concentrations should not be proportional to  the incremental
increase in the power plant emissions after  1965.   Therefore the increment in
terms  of  sulfur dioxide  concentrations  would  be  relatively  small  and  dif-
ficult to measure at nonurban monitoring sites  by  the sulfur dioxide sampling
and analysis procedure used.

5.2.3  Sul fate

5.2.3.1  Urban  Concentration  Measurements--In 1963 the  National  Air Sampling
Network collected particulate  matter  on  high-volume  (HIVOL)  samplers  and
began  analyzing  for sulfur  as  water-soluble sulfate at urban sites in  the
United States.

The potential  for  a  positive sulfate artifact resulting  from  collection  and
conversion  of  SOg on  glass-fiber filters  was  discussed by  Lee and Wagman
(1966).  Subsequent laboratory studies have  shown that the magnitude of  such
an artifact  depends  on  temperature, $03  concentration,  the  air volume  per
unit area  of filter surface,  and other parameters  (Coutant  1977,  Meserole  et
al. 1976).   The  conversion  of  S02  to sulfate on clean glass-fiber filter
                                     5-13

-------
surfaces was sensitive to  temperature  but showed little dependency on  humi-
dity.  A substantially smaller artifact was obtained on surfaces coated with
ambient air  particulates  than  on uncoated  filter  surfaces.   Coutant  (1977)
estimated sulfate loading  errors  from  the use of untreated glass-fiber fil-
ters under usual flow conditions in HIVOL samplers to be in the range  of 0.3
to 3.0 yg m~3.

The results reported from field observations have varied widely from small or
negligible to large  artifact effects  (Appel  et al.  1977, Pierson et al.  1976,
Stevens  et  al.  1978).     However,  differences  in  sampling  techniques and
analytical procedures used complicated comparisons.  It will be assumed that
sulfate artifacts are not large enough to influence substantially the  trends
in sulfate concentrations  observed.   If  the sulfate artifacts were substan-
tial, part of the decreases in ambient air  sulfate concentrations would have
to be attributed to  the concurrent reductions  in  sulfur dioxide.  Conversely,
increases also  occurred  in ambient  air  sulfate concentrations.   These in-
creases were even larger than indicated, if they occurred at the same  time a
positive sulfate artifact was  decreasing.

At most urban  sites  in  the  western  United  States  in the  1960's,   sulfate
concentrations were  below  10 pg  m-3;  at three-quarters of  the  urban  sites
in   the   eastern  United  States  concentrations  were  above   10  yg  m"3
(Altshuller 1973).   The general  order  of  decreasing  sulfate concentrations by
geographic region in the  1960's and 1970's was:  (1) East Coast, (2)  Midwest
(east of  Mississippi),  (3) Southeast, (4)  West  Coast,  (5)  Midwest (west of
Mississippi), and (6) western  states.   Average  sulfates for  urban sites in
the  western  United  States  ranged  from 30 to  50  percent of the concentration
of sulfate at urban  sites on the East Coast.

The  excess  in  urban sulfate concentrations  over the  regional  background of
sulfate  is  a  measure of  the  contributions  by  local  primary  sources and
atmospheric  transformations within  the urban  area  (Altshuller  1976,  1980).
Although  regional  background  levels  of  SOg   were  small  compared  to  urban
concentration levels, regional  background levels of sulfate  have  been sub-
stantial in  the eastern United States compared to  urban concentration  levels
(Altshuller  1976, 1980).    These  regional  background levels  of  sulfate are
formed  from  atmospheric  transformations  of  sulfur dioxide  to  sulfate (see
Chapter A-4).

Control of local sulfur oxide emissions by  reductions in fuel sulfur  content
resulted  in  a substantial  reduction  in  ambient air sulfate concentrations,
particularly in the first  and fourth quarters  of the year (Altshuller 1980).
The  largest  decreases occurred  in urban  areas  in the  northeastern  United
States, but  smaller decreases also occurred in urban areas in  the Midwest and
Southeast.   In contrast, during  the  third  quarter of the  year, ambient air
sulfate concentrations increased in the 1960's and  1970's, and then decreased
somewhat  at  some  sites.   Increasing sulfate  concentrations during the third
quarter occurred well into the 1970's at  some  sites in the Ohio  River Valley
region and at sites in the South.
                                     5-14

-------
The  urban  excess,  the difference  between  the average urban and  the  average
regional  (nonurban)  sulfate  concentration  in a  region,  decreased  substan-
tially  between 1965-67  and  1976-78  in  the  North,  Midwest,  and  Southeast
during  the  first and  fourth quarters of the year  (Altshuller  1980).  Smaller
decreases in the urban excess occurred in the second and third  quarter in the
Northeast  and Midwest,  but  increases  occurred  in  the   southeastern  urban
areas.

The  increase  in  third-quarter sulfate  concentrations  at urban sites  in the
late  1960's  into  the  1970's occurred on  the  average   in  the  northeast,
southeast,  and midwestern regions, indicating geographic-scale  processes  at
work.   The  increases occurred consistently at sites  in  the Ohio  River Valley
area and adjacent areas in the Southeast.  Regional-scale sulfate episodes  or
potential episodes increased  in  frequency during the  same period.   Most  of
these  episodes occurred  in  the  June-through-August  period  of  each  year
(Altshuller  1980).    Therefore,  the  higher  sulfate concentrations  in  the
summer months at urban sites are  likely to be associated with large regional-
scale processes (Altshuller 1980, Hidy et al. 1978,  Mueller et  al.  1980).

In the  late 1970's,  the average urban  sulfate  concentrations  by  quarter  of
the year in the northeastern, southeastern, and midwestern United  States had
the order:  third quarter >  second quarter  > first quarter >  fourth  quarter
(Altshuller  1980).    The  first-   and  fourth-quarter average  urban  sulfate
concentrations in  the Northeast  and Southeast  were below  10 yg  nr3; the
third-quarter  average  urban  sulfate  concentrations  in   the  Southeast and
Midwest were  at  15  ug m~3.   The  urban excess,  the  difference  between the
average urban  and average nonurban sulfate concentrations, had decreased  by
the late 1970's compared to  earlier years,  except in  the  Southeast.  Regional
trends at urban sites in the  United  States also have  been  discussed by  Frank
and  Possiel  (1976).   Plots  of  the  regional  distribution  of  sulfates  were
developed.

5.2.3.2   Urban Composition  Measurements--The composition  of  the sulfate  in
urban areas has been the subject of  a  number  of  investigations.    In  several
investigations of  aerosol  composition within  urban areas, including Secaucus,
NJ, Philadelphia,  PA, Chicago, IL, and Charleston,  WV, the sulfate appeared
to be   in  the form  of  ammonium  sulfate [(NH^SO^  (Wagman  et  al.  1967,
Lee and Patterson 1969,  Patterson  and Wagman 1977,  Lewis  and  Macias  1980).
However, no special  precautions were taken  to  preserve  sample acidity.

Tanner et al.  (1979)  using a  coulometric modification of the Gran  titration,
reported aerosol  samples in New York City to  be  slightly on the acidic  side
of  (NH4)2S04   in  winter  (February  1977),  but   to  have   the more  acidic
average   composition   of   letoricite,   (^4)3*1(504)2,    in    the    summer
(August 1976).   These  investigators also  found  sulfate  to be  highly  cor-
related with  ammonium in  both   summer  and  winter  aerosols.   Lioy  et al.
(1980)   during  a  high sulfate episode  in  the east  on  3  to 9 August  1977,
observed high  acidities  at  nonurban sites,  as  did Pierson  et al. (1980a).
However, in New  York City  the aerosol  appeared  to  be  nearly  neutral,  sug-
gesting higher ammonia fluxes in  and near New York City.
                                     5-15

-------
Coburn et al .  (1978) measured  the  acidity of sulfate aerosols In St. Louis,
MO, by an in situ thermal analysis technique  during  a  16-day period in late
April to early May 1977.  Although the acidity reached a one-to-one ratio of
[NH4+]  to  [H+]  on   one  morning,  for the  most  part  the  sulfate  aerosol
tended to be in the  form of
In earlier measurements  in  the Los Angeles  area  during  1972  and 1973, suf-
ficient ammonium  ion  appeared to  be  present  to  neutralize the  sulfate  to
(NH^oSCk  except  near  strong  local  sources  of  sulfur  oxides  (Appel  et
al . 1978).  However,  the  authors  did point  out  that the techniques used could
not  distinguish  between  neutralization  of  acidic  constituents  before  and
after collection.   In  subsequent measurements  in  July 1979 at  Lennox near
strong  sulfur   sources,   significant  levels  of   ^$04  and  particulate
acidity were obtained (Appel et al . 1982).   Sulfuric  acid constituted 10  to
20 percent of the total  sulfate.

It would  appear that the  sulfate aerosol  in  urban  areas  tends  toward the
composition  of  (NH4)2$04,   but  that  its  composition  is  variable  with
more of a tendency toward acidic  species  in the summer.

5.2.3.3   Nonurban  Concentration  Measurements—Altshuller (1973)  pointed out
large differences in the  range and average  concentrations  for  sites in the
eastern  compared  to  the western  United  States  based  on measurements  of
sulfate  concentrations  at  nonurban   sites  in  1965-68.    Relatively  little
overlap  occurred  in  frequency ranges,  with  the  sulfate concentrations  at
eastern  sites  averaging  8.1 yg  nr^,  and  those at western sites averaging
2.6  ug  m-3.   At 10  percent  of  the  western  sites,  annual average  concen-
trations  were as  low  as  0.5 to  1.0 yg  m-3.   The  eastern and western sites
appeared  to  represent separate  and  distinct  populations as  far as  sulfate
concentrations were concerned (Altshuller 1973).  A continental background of
less  than  1 yg  m-3 was  indicated  by  the  minimum  sulfate  concentration
levels at eastern and western nonurban sites.  A more  detailed  stratification
of results  on  sulfate concentrations  at nonurban  sites in  the United States
indicates  the  order  of  decreasing  sulfate  concentrations in  the  1965-72
period to be:   (1) East  Coast and Midwest (east of Mississippi River), (2)
Southeast,  (3)  Southwest,  (4)  Midwest (west  of Mississippi River)  and West
Coast, and (5) Mountain  States.

Between 1963-65 and 1976-78, sulfate concentrations at nonurban sites  (Acadia
National   Park, ME; Coos  County, NH; Orange County,  VT;  Washington  County, RI;
Calvert County, MD; and Shenandoah National Park,  VA)  varied only  slightly in
the  first,  second,  and  fourth  quarters  of the  year (Figures  5-1   to 5-3)
(Altshuller  1980).   The first-  and  fourth-quarter  trends  showed both  small
increases  and decreases in sulfate concentration at the nonurban  sites in the
Northeast,  Southeast,  and  Midwest (Altshuller 1980).   The  second-quarter
trends either were positive or showed no change in these  three  regions.

At the nonurban  sites in the  northeastern and midwestern  United  States, the
third-quarter  sulfate concentrations  increased during the  1960's,  peaked  in
the  early 1970 's,  and subsequently decreased,  just as at the  urban  sites  in
these  regions  (Altshuller  1980).   This upward trend occurred most  consis-
tently for nonurban sites in the Ohio Valley area.


                                     5-16

-------
Although  urban  sites showed  decreases in  sulfate  concentration during  the
winter  quarters,  presumably due  to  local-scale reductions  of sulfur  oxide
emissions  (Altshuller  1980),  no  substantial  changes  were  experienced  at
nonurban  sites   distant  from  such  local  influences.     Conversely,   since
third-quarter trends  were presumably influenced strongly by  larger  regional
processes, both urban and nonurban sites  in the same region  and even  across
regions should show similar behavior.  The second quarter showed intermediate
behavior.   Despite  the  large  upward trends  in  sulfur  emissions from  power
plants  during the 1960's  and  1970's  (Figure  5-4), very  small  increases were
measured  at  nonurban sites in the  Midwest  or  East.   The  only  substantial
upward  trends were  in the third quarter of the year at nonurban sites.   The
trend downward after the early 1970's at the midwestern  nonurban sites  during
the third quarter of the  year appears consistent with the  downward  trend of
sulfur  emissions  in most midwestern states  between 1970  and  1978  (Chapter
A-2, Table 2-14).

A  plot of  the  regional  distributions  of  nonurban  sulfate  concentrations
averaged from months  in 1977 and  1978  are  shown in  Figure 5-5  (Hi 1st  et al.
1981).   Sulfate  concentrations  were  the highest  in  the  Ohio Valley  area
followed  by  other  parts of the  Midwest,  mid-Atlantic  states  and  Southeast.
During  summer  months in  1977  and  1978,  Mueller et al .  (1980)  observed  a
broader regional  distribution  of  sulfates  than observed  during  the  entire
study period, with high  sulfate concentrations extending  all  the way  from the
Ohio River Valley to the Atlantic Seaboard.

In the late 1970's the average nonurban sulfate  concentrations in  the eastern
and midwestern United States had  the same ordering by quarter of  the year as
at urban  sites:   third quarter  > second  quarter > first  quarter >  fourth
quarter (Altshuller 1980).  Based on sulfate  measurements made  from  May 1980
to August 1981  at  three  rural sites  in  the Midwest,  Shaw and Paur  (1983)
reported  monthly average  concentrations  ranging from  as  low as 3 vg  m~3
in some winter  months to 12  to  15 pg nr3  in the summer months.   The  sea-
sonal  variations in  sulfate concentrations were  just  the  opposite of  those of
sulfur  dioxide.   As  a  result, the  percentage  of particle  sulfur  of  total
sulfur measured ranged from 5  to 10 percent in the winter months to more than
40 percent in the summer months.

Diurnal  sulfate  concentrations  were measured  at two  rural   sites,  one  in
Kentucky and the other  in Virginia,  during the summer of 1976  (Wolff  et al.
1979).    Two types  of  diurnal   patterns   for  sulfate   concentrations  were
observed.   On  one  group  of days,  the  sulfate concentrations peaked in  mid-
afternoon at about  the  same time  the ozone concentrations peaked.   Downward
mixing of sulfate from the layer aloft, as the noctural  inversion  layer broke
up, was suggested  as being  responsible for  a  substantial  fraction of  the
sulfate  in   these  afternoon  peaks.    The  second  diurnal  pattern  involved
sulfate concentration peaking  between  2000 and  0400 hours at night.    This
type of diurnal  behavior  appeared to be most  pronounced  on  clear  nights when
ground  fog developed.   A  few  days fell into  neither of these  two patterns.
These latter days were characterized by very  low  sulfate concentrations,  < 5
ug nr3, and occurred after passage of a cold front.
                                     5-17

-------
                                                         1-HOUR
                                                      S02 (ppb)
                                                       24-HOUR
                                                     r\         O
                                                        (yg nT
Figure 5-5.   Sulfur dioxide (arithmetic mean)  and sulfate (geometric
             mean) concentrations.   Data obtained during 5 months
             between August 1977 and July 1978.   Adapted from
             Hilst et al.  (1981).
                                   5-18

-------
The sulfate concentrations measured at rural monitoring sites outside of St.
Louis, MO,  were 80  and  90 percent  of  the sulfate  concentrations  at  urban
sites within St. Louis during the years 1975 through  1977  (Altshuller 1982).
These results also are consistent with a  strong regional  influence on sulfate
concentration distributions.

Vertical  profile measurements  were obtained from aircraft  flights over south-
eastern Ohio  in early August 1977  and  January 1978  (Mueller  et al. 1980).
Measurements were made  in the layer between 0.3  and  1.5  km and at a higher
layer between 1.5 and 3 km above mean sea  level.  On the  average, the sulfate
concentrations  in the lower  layer were similar to those obtained at ground
sites.  The  sulfate  concentrations  in the upper  layer were smaller than  in
the lower  layer.   In August  1977,  the  aircraft measurements indicated that
the sulfate concentrations in  the lower  layer were about  twice as high in the
afternoon  hours  as  in the morning  hours.   In  a  winter  period, the sulfate
concentrations varied little  between  the  morning and afternoon hours in the
lower  layer  aloft.    The sulfate concentrations  in the  lower  layer  in the
winter were about one-third of those in  the afternoon  in  the summer.

Twenty-four-hour average sulfate concentrations were measured in the Canadian
APN  concurrently  with  S02  concentrations   (Barrie  et   al.  1980,   1983;
Whelpdale and Barrie  1982).   Atmospheric particulate matter  was collected  on
a  Whatman  40 particulate  filter, which  preceded the chemically impregnated
filter used  to  collect  sulfur dioxide.   Sulfate  was  determined by  means  of
ion chromatography.   The geometric means  of the 24-hr average sulfate con-
centrations  on  a yg  m~3  basis for  the   period  November  1978  to  December
1979 are:  Long Point,  Ontario, 1.0; Chalk River,  Ontario,  1.9; ELA-Kenora,
Ontario,  1.0; Kejimkujik,  Nova  Scotia,  1.8  (Barrie et al.  1983).   Sulfate
concentrations  do  not  decrease as  rapidly as  do S0£  concentrations  with
distance  from  major  source  regions.   Sulfate concentrations, just  as S02
concentrations,  show  large fluctuations attributed  to the  alternate presence
of clean  air and polluted air  from large  source regions (Barrie  et al. 1980).

Concentrations of sulfate  as  a  function  of percentage  cumulative  frequency
are plotted in Figure 5-6  (Barrie et  al.  1983).   Results  from Canadian  sites
for the  period  November 1978 to  December  1979 are compared with  those ob-
tained in  the eastern United  States during 1974-75.  Except for the highest
sulfate concentrations experienced  at Canadian sites in  lower Ontario, the
sulfate concentrations at  Canadian  sites  fall   well  below those at sites  in
the United States.    This is  particularly so  for the  Canadian  sites more
remote from large source regions.

5.2.3.4  Nonurban  Composition Measurements—Char!son et al. (1974) reported
evidence  obtained with a semi quantitative  humidographic  technique  of acidic
sulfate  species frequently present at  a rural  site outside  of  St.   Louis
during September 1973.   The acidic composition  was  variable  (Charlson et al.
1974, 1978a).  The sulfate aerosols  were acidic more  frequently at  the  rural
site than  at the urban site.  There  was  no dependence on wind  direction nor
on  synoptic  conditions,  consistent  with  regional  sources of the sulfate
aerosol (Charlson et al.  1974).
                                     5-19

-------
    100
     50
CO
 I

 01
 a.
 UJ
 UJ
 =3
 O
 •— «

 fe

 g
     10
    0.5
   0.1
LEGEND

  LONG POINT
  TORONTO
  CHALK RIVER
  KEJIMKUJIK
  ELA-KENORA
           0.1
            10           50         90

            PERCENT CUMULATIVE FREQUENCY
99    99.9
  Figure 5-6.  A comparison of the cumulative frequency distribution of
               daily sulfate concentration at several rural locations in
               eastern Canada for the period Nov. 1978 to Dec. 1979 with
               that for the 'SURE' region in the north eastern United
               States for 1974-75.  Adapted from Barrie et al. (1983).
                                        5-20

-------
Samples were  obtained  at 125 m above ground  level  on  a  meteorological  tower
at  Brookhaven  National  Laboratory  from  May through November 1975  (Tanner  et
al.  1977).    The  ratio  of  [H+]  to  [NHA+]   in  ng  m-3  varied  from  0  to
1.6:1.   In 9  of the  11  samples  taken  [NH4+] was  substantially  in  excess
of  [H+],  particularly  for  the three  samples collected  in  October and No-
vember,  which  were  predominantly  in  the  form  of  (NH4)2$04«    Use  of   a
diffusion  battery  sampling   technique  indicated  that  particles  below the
optical range  were  more acidic than  the particles that  effectively  scatter
light.   It also was  observed that air mass  passage  over water from  source
areas resulted in more acidic particles  in the suboptical range than air mass
passage over land.

Aerosol measurements were made at a rural  site at Glasgow, IL,  during  a 9-day
period late in July 1975  (Tanner and Marlow 1977).   During the earlier por-
tion of the sampling period  with little  or  no strong  acidity measurable, the
air mass backward trajectories indicated  reasonably  direct transit  from urban
and/or  power  plant sources.   Stagnation conditions occurred  on  29-30 July
with movement of the air mass from St. Louis past the  vicinity  of large power
plant  sources.   Significant  strong  acidity was  measurable  in the  aerosols
reaching the Glasgow, IL, site during this period.

Measurements of  sulfate aerosol  composition were made in Research  Triangle
Park, NC,  during 4  days in July 1977 (Stevens et al.  1978).   Care  was taken
to  preserve the acidity of  the  samples  with use of  a diffusion  denuder  to
remove ammonia  during  collection  and with  preservation  of the samples  over
nitrogen before analysis.  The  amount of strong acidity measured was  highly
variable among the 16 samples.  In  about  half  the  samples, the  strong  acidity
was  zero  or  near  zero.   In three  of  the  samples,   the  ratio of  [H+]  to
[NH4+] in neq nr3 was near 1:1.   The  highest ratio of [H+]  to [NH4+]
occurred concurrently with the highest sulfate concentration.

Measurements of aerosol composition were  carried  out  at a site in  Tennessee
at  646  m  altitude in the Great Smoky Mountains  National Park  in the  latter
part of September 1978 (Stevens et al, 1980).   Each  of  the 12 aerosol  samples
collected and  analyzed  for  strong  acidity were acidic.   The average  acidity
was  close  to  that  of  N^HSCty.    The  higher  ratios  of  [H+]   to   [NH4+]
occurred with the higher sulfate concentrations.   Because no denuder was used
to remove ammonia, some neutralization could have  occurred.  Therefore,  it  is
possible  that  the  samples  were  even  more  acidic than  indicated  by the
measurements.

Weiss  et  al.  (1982)  at   the  Shenandoah   Valley  site  obtained   (NH4+)/
($042-)  molar   ratios  ranging   from   0.5   to  2.0   with   strong  diurnal
variations. The particles were  most  acidic  in midafternoon and least  acidic
between 0600 and 0900 hours.

Sulfate composition measurements were made  on samples collected at 853 m  on
top  of  a  tower  on  the  summit  of  Allegheny  Mountain in southeastern
Pennsylvania between 24 July and  11  August 1977  (Pierson et al. 1980a).   On
the  average,   the  [H+]  was  slightly  in  excess  of  [NH4+],   corresponding
to  a  composition  near  that  of  NHAHS04.    The  concentrations   of the
othercations  were  so  low  that  [H+J   and   [NH4+]   were  the  predominant


                                     5-21

-------
cations  associated  with  [S042-],  and  the  sum  of  [H+]   and  [NH4+]  was
essentially  stoichiometric  with  [S042~].     For  sulfate  concentrations
above  15  ug m~3  the  [H+]  to  [S042~]  mole  ratio  was  between  1:1  and
2:1 and  approached 2:1  for several  samples.   Therefore, appreciable amounts
of 1^504  must have  been  present at the  high  sul fate concentration levels.

Lioy et al.  (1980) reviewed in detail the high sulfate  episode during August
3-9, 1977.  The  occurrence of a strong acid distribution on a regional  scale
was identified  by  these workers,  based  on measurements at High Point,  NJ,
Brookhaven, NY,  and Allegheny  Mountain (Pierson et  al. 1980a).

Measurements of sulfate  composition  have  been made  by  an  infrared spectro-
photometric  technique  (Kumar et al. 1982)  at five nonurban sites in  the
eastern and midwestern United States near Rockport,  IL; Racquette,  NY;  State
College, PA;  Charlottesville, VA;  and Upton, NY.   The average acidities by
season of the year at these locations ranged  from  slightly  more  acidic than
(Nfy^SCty to that  equivalent to  (Nfy^HfSCh^,  although acidities equivalent
to that  of NH4HS04 were observed  also.   The acidities were  reported  to be
higher in summer and winter than  in spring and fall.  Varying diural patterns
were observed but  acidities tended  to  be higher in  the daylight hours than at
night.

Acidity measurements were made on samples collected  from aircraft and on the
ground at a  number of locations  in the midwestern  and eastern United States
during the months  of April, July,  August,  and November (Ferek et al. 1983).
The higher acidities were obtained  aloft and  at Whiteface Mountain, NY during
the spring and summer months.  Acidities at these locations  often correspond-
ed  to   compositions  between   (Nfy) 3^504)   and   NH4HS04  and   in   some
samples  were more  acid   than  NfyHSC^.    Acidities  were  lower at  ground
level  locations  than aloft, in the  midwest than  in the east, and in November
compared to the  summer months.

In several investigations,  the tendency was for the higher acidities to occur
concurrently with  the higher sulfate concentrations (Stevens et  al.  1978,
1980;  Pierson et al.  1980a).

In summary, there  appears to be  substantially more  evidence  for strong acidic
species at rural than  urban  sites.   The highest acidities  in aerosols have
been measured at  mountain  locations  and in  samples collected from aircraft
aloft.

5.2.3.5   Concentration and  Composition  Measurements  at  Remote  Locations—
Meszaros (1978)  reviewed available sulfate measurements at  remote locations.
He  estimated an   average  sulfate  concentration   of 1.3 yg m~3  over  the
Atlantic Ocean.   The  sulfate concentration as a function of latitude has two
maxima.    One  of   these  occurs  near 40°N  latitude  where  S02  also  has  a
maximum concentration and the other occurs  south of the  equator.  Around 40°N
the  sulfate  concentration is   2  yg  m~3,  but  decreases   to   below  1  yg
m~3 above 50°N.   Meszaros  estimated sulfate  concentrations of  about  0.3
yg m~* over clean  areas in  the Northern Hemisphere.
                                     5-22

-------
Gravenhorst  (1978)   obtained  an  average  sulfate  concentration  of  excess
sulfate  (excluding   the  contribution  of sea  salt)  of 0.9  pg  nr3  ± 0.5
yg m-3.  The excess sulfate tended to be acidic.

Measurements of  sulfate were  made  at  a  remote  sampling site  in the  Faroe
Islands during February 1975 (Prahm et  al. 1976).   During  a period when air
masses were crossing  the  site  after traveling only over the North Atlantic,
excess  sulfate  averaged  0.14  ug  nr3.   During  another  period when air
masses had passed over the British  Isles  upwind, the  excess sulfate  averaged
1.07 yg nr3.

An  excess  of  submicron  sulfur  particles also  was measured  at a  site  in
Bermuda (Meinert and  Winchester  1977).   The  excess sulfur was  attributed  to
long-range transport from the North  American  Continent.

Aerosol  samples  were  collected  from  aircraft  flying   in  the  central and
southern Pacific Ocean  and remote areas  of  North  America during GAMETAG  by
Huebert  and  Lazrus  (1980a).    The  ranges  of  sulfate concentrations   in
different  environments   in  yg  nr3   were:   continental  boundary layer,  <
0.25 to 0.5; marine boundary layer,  0.36 to 3.6; free troposphere, < 0.06  to
0.35.

As  indicated   by  the results  of  Meinert and  Winchester  (1977),   Meszaros
(1978), and by Prahm et al.  (1976), remote sites can  presumably  be fumigated
by continental  sources well upwind.

5.2.3.6  Comparison of Sulfur Oxide Emissions and Ambient Air Concentrations
of  Sul fate—Ambient air  concentrations of  sulfate are  tfie  result  of fTT
primary sulfate emissions and (2) the secondary sulfate  formed  by conversion
of a portion of  the sulfur dioxide  emissions to  sulfate in the  atmosphere.
The relative contributions of  primary and of secondary  emissions to ambient
air sulfates will vary with geographical location and  time of year.

Altshuller  (1980)  concluded  that reductions  of  fuel  sulfur  content  within
urban areas in the  northeastern United States caused substantial reductions
in primary sulfate emissions.  This  reduction  in  primary sulfate emissions  in
turn appeared  to account  for the decrease in ambient air sulfate concentra-
tions during the  first  and  fourth  quarters  of  the year.   In  contrast, the
ambient air sulfate concentrations increased during the third quarter  of the
year in urban  areas.  These  increases in ambient air sulfate concentrations
appear  to  relate  to the  increases  in  regional  scale   emissions  of  sulfur
dioxide and sulfate.

In nonurban areas in  the  eastern United States during  the  late 1960's into
the  early  1970's   small  increases   in   ambient  air  sulfate  concentrations
occurred  in  the  first  and  second   quarters  of  the  year.    A  substantial
increase in ambient air  sulfate concentrations  occurred from the mid-19601s
into the early 1970's during the third quarter of the year at nonurban  sites
in the northeastern and  midwestern United States  (Altshuller 1980).

Less  than  a  proportional   increase   in  ambient  air  sulfate  concentrations
should  occur   because part  of  the  sulfate  formed  from  the  incremental


                                     5-23

-------
emissions of sulfur dioxide remains aloft (Chapter A-2, Section 2.3.2).  The
sulfate aerosol  formed from the sulfur dioxide emissions in plumes from tall
stacks remains aloft long distances,  particularly  during the cooler months of
the year  (Chapter  A-3,  Section  3.4.1).   A  portion  of the  sulfate emitted
aloft will be removed  by wet scavenging  and a portion will pass into Canada
or over  the  Atlantic.    This  portion  of  the sulfate  will  not contribute to
sulfate concentrations measured at ground level monitoring sites in  nonurban
areas in  the  United States.   During  the third quarter of the year, deeper
mixing within the boundary layer occurs more frequently and brings the plume
down  to  the  ground  shorter  distances downwind of  the stack  (Chapter A-3,
Section 3.4.1).   Therefore, a larger  portion of the incremental sulfur oxide
emissions as  sulfate  should be measurable  during the third  quarter of the
year at ground-level monitoring sites in  nonurban  areas in  the  United States.

Increases in  visibility  and turbidity with  increases  in  sulfur oxide emis-
sions from the 1950's or  1960's into  the 1970's have  been  observed,  particu-
larly during  the  third quarter of the year (Husar  and Patterson 1980). The
trends with time and the  seasonal  patterns of airport  visibility measurements
over horizontal  ranges and turbidity  measurements  through  the  entire  air mass
closely resemble those of the sulfate concentrations.   This is  to be  expected
because  sulfate accounts for a large part  of  the light extinction  at rural
sites in the eastern United States (Section  5.8).

5.2.4  Particle Size Characteristics  of Particulate Sulfur  Compounds

5.2.4.1   Urban  Measurements--Particle size  distributions  have been  reported
in a number of urban locations for sulfur as sulfate in collected particulate
matter.   Similar  results do  not appear  to  be  available  for  sulfur in any
other valence state.   Stevens et  al.  (1978)  attempted to  analyze for sulfite
in samples from South  Charleston,  WV, Research  Triangle  Park, NC,  New York,
NY, and Philadelphia, PA.  The  sulfite content  of the samples  did  not exceed
the  minimum  detection limit of  8  ng m-3.   By comparison  with  the   fine
particle  sulfur concentration,  this  results  in  less  than 0.1  percent of the
extractable sulfur as sulfite or 2 percent of the total fine  particle (< 3.5
urn) sulfur as sulfite.

A  five-stage  impactor  with  stage mass median diameters (MMD's) of 1.9,  3.6,
and 7.2 ym with a backup filter was  used at two sites in  Pittsburgh, PA, in
1963-64  to separate particul ate matter into size  fractions (Corn and Demaio
1965).  Sulfate was measured by a turbidimetric  method.  A substantial amount
of the sulfate  was  reported  to  be  in larger particles with MMD's between 1.9
and 3.6 ym.

Size distribution of sulfate  in  particulate  matter was determined  by Roesler
et  al.  (1965)  at  sites   in  Chicago, IL,  and Cincinnati, OH.   A  six-stage
Andersen  cascade impactor  was used for particle  size distributions. Sulfate
was measured  by a  turbidimetric method.   The MMD's obtained at the  sites in
Cincinnati and  Chicago were  0.4 ym   and  0.3 vm,  with nearly  90  percent of
the sulfate less than 3.5 urn.

Wagman et al. (1967) obtained sulfate size  distributions  at sites in  Chicago,
IL,  Cincinnati,  OH, and  Philadelphia, PA,  during 1965.   Lee and  Patterson


                                     5-24

-------
(1969) reported ammonium size  distributions  during  the  same time periods at
these  sites.    A  six-stage  Andersen  cascade  impactor  was  used  for  size
separations.  Sulfate was analyzed by the turbidimetric method, and ammonium
was determined by the Nessler method  with alkaline potassium mercuric iodide.
The  average  MMD's for  sulfate and  ammonium were  similar, with  an overall
range  from  0.35  to 0.66 urn.  The  higher MMD  in  Philadelphia  was attributed
in  part  to dust  generated from  road  construction  near  the   site.   Eighty
percent of the sulfate was  less than  2 ym at all  of  the sites.

Sulfate particle  size  increased with humidity  at  all  sites  (Wagman  et al.
1967).  Substantial  scatter occurred with MMD ranging  from less than 0.2 ym
at  lower  humidities  to 0.6  to 0.8 ym  at  higher  humidities   at  three  mid-
western sites.   At the site  in Philadelphia, PA,  the MMD  exceeded 1  ym at
higher humidities.  Correlation of MMD's with absolute humidities was poor.

Ludwig and  Robinson  (1968) obtained particle size  distribution  of samples
collected in  the  Los Angeles  and San  Francisco Bay areas  of California in
1964-65.   A Goetz aerosol  spectrometer was  used.   The  analytical  procedure
involved high-temperature reduction of  the  sulfur  in the sample to hydrogen
sulfide in  a  microcombustion  furnace and iodimetric microcoulometric titra-
tion for the hydrogen sulfide.  Average  MMD's were computed  from measurements
at  several  sites  in  Los Angeles and the  San  Francisco Bay  area.   Except at
the Lennox, CA,  site,  the  MMD's ranged from 0.2 to 0.4 ym.  The Lennox site
is  directly  downwind  of  a  number of  emission  sources,   including an oil
refinery and  a  sewage  treatment plant,  and is 2 miles from the ocean,  which
may account for the higher  MMD at this  site.

Ludwig and Robinson  (1968) reported  that at these West  Coast  sites, samples
collected during  periods  of  higher  relative  humidity  (RH)  had  the higher
MMD's  for sulfur-containing particles.   The  weighted average MMD varied from
0.1 ym in  the 12.5  to  27.5  percent  RH class to 1.1 ym  in  the 72.5 to 87.5
percent RH class.

Ludwig and  Robinson  (1968) also  observed diurnal   decreases  in  the sulfate
size distribution  by time  of day  as follows:   forenoon  >  afternoon >  early
morning > evening.   Wagman et  al.  (1967)  did not observe consistent diurnal
changes in  sulfate size distribution from site  to  site.   In  fact, only the
Chicago,  IL,  site showed  significant  changes in sulfate  size distribution
with sulfate  size  decreasing  by time of day as follows:   morning > midday  >
evening.   Therefore,  in Chicago and at the West Coast  sites, sulfate par-
ticles tended to  be  smaller  during  the  evening  hours.    Both  groups  of
investigators reported no  relation between diurnal variations  in sulfate size
and humidity  changes,  but  no explanation in terms  of atmospheric  processes
was suggested.

Particle  size  distributions  for sulfate  and  other  species  were obtained in
Riverside,  CA,  during  the  first half  of  November  1968  (Lundgren 1970).
Samples were  collected  on  a  four-stage Lundgren impactor.   The average MMD
for  sulfate  was  about  0.3 ym  with  the  range  of  MMD's  for  the  10 samples
collected varying  from  0.1 to 0.6 ym.    On  the  average,  about 90 percent of
the  sulfate  in the  collected  particles  was below  1.7  ym.   Particle size
                                     5-25

-------
distributions of  sulfate  also  were reported by Appel  et  al.  (1978)  for the
Los Angeles, CA, Basin area as  0.3  to  0.4 ym  for most samples.

Patterson and Wagman (1977) obtained particle size distribution of collected
samples for a number  of  species  including  sulfate and ammonium in Secaucus,
NJ, near New York, NY, between  29 September and 10 October 1970.  Seven-stage
Andersen  cascade  impactors  were   used  at 28  £  min-1,  with  either  Gel man
type A glass-fiber or Millipore® backup  filters.   Sulfate was analyzed by the
methods used  previously  (Wagman  et al.  1967, Lee and  Patterson  1969).   The
air masses traveling across the site  were  classified  into four visual  range
classes.  For sulfate and ammonium, the  MMD's, by visual range class, were:

     Visual  range  (mi)       Sulfate (ym)         Ammonium (ym)

           > 26                 0.60                0.26
         13 to 26               0.39                0.34
          8 to 13               0.46                0.38
           < 8                 0.40                0.36

The MMD's  for sulfate and ammonium were  reasonably  similar except  for the
background case of > 26 miles.  For this condition, much more of the mass of
the sulfate was  in  the range  0.54  to 0.95  ym than was  the case  for  ammo-
nium.   Almost all  of the sulfate and ammonium in  the collected particles was
below 1.5 ym.

Tanner et al.  (1979)  measured  sulfate  in August 1976 and February 1977 in New
York,  NY, using  a  diffusion battery along with HIVOL sampling.  The diffusion
battery was  used  to  classify  particles by  size  less than  0.25 ym  before
filter sampling and analysis.  During the  summer  month,  about 50 percent of
the sulfur-containing  aerosols were  less  than  0.25 ym;  during  the  winter
month only 25  percent were  less than 0.25 ym.

Stevens et al. (1978)  concluded from measurements for sulfur along with other
metals in  New York, NY,  Philadelphia,  PA,   Charleston,  WV,  St.  Louis,  MO,
Portland, OR, and Glendora,  CA,  that sulfate in  the  fraction  less  than 3.5
ym  had  to be associated  predominantly  with ammonium  and hydrogen  ions in
urban areas.   If  all  of the metals  were  assumed  to be in the  form  of sul-
fates, only 10 to 32 percent of the sulfate  would be  accounted  for  as  metal
sulfates at these urban sites.  Because  it is likely that most of the metals
would be in the form of oxides, halides, or carbonates rather than sulfates,
these estimates  would form upper limits.

Separation of particles into two fractions with a  fine  fraction consisting of
particles less  than  3.5  ym involves  use of  a virtual impactor  or dichoto-
mous  sampler  (Stevens et  al.  1978).   The  percentages  of  sulfur found in the
size  range  less  than  3.5 ym  at  various sites  were:   New  York,  NY—93%;
Philadelphia, PA—85%;  Charleston, WV—92%;  St.  Louis,  MO—79%;  Portland,
OR~83%; Glendora, CA—87%.  Sampling was done in the  winter months  of 1975
and 1977.  In additional  measurements  reported from a site in Charleston, WV,
91 percent of the sulfur measured  during a period in  the  summer  of 1976 was
in  the fine particle  size  range  (Lewis  and Macias 1980).  Altshuller (1982)
analyzed  data on  particulate  sulfur  measured  with  dichotomous  samplers at


                                    5-26

-------
urban sites in St.  Louis,  MO.   From 80 to 90 percent of sulfur measured was
fine particle sulfur with  no  substantial  seasonal  pattern  between the third
quarter of 1975  and the fourth quarter  of  1976.

5.2.4.2  Nonurban Size Measurements—Junge (1954, 1963) reported on the par-
ticle size of sulfate aerosols at Round Hill, MA,  50  miles south of Boston,
and at a site south of Miami,  FL.   He  found  most of the particles containing
sulfate to be in the 0.08 to  0.8 ym range  rather  than  in the  0.8  to 8 ym
range.  Junge (1963)  found the average composition of the particles between
0.08 and 0.8  ym  to correspond  to  a mixture of (1^4)2504 and (NH4)HS04-

Charlson et al.  (1974)  found  strong  acidity in particles at Tyson Hollow, MO,
35  km  WSW  of the Arch in  St.  Louis, using an integrating nephelometer with
humidity control  (humidograph).   Because  the nephelometer would respond to
particles  predominantly  in the  optical   range,  0.1  to 1 urn,  the technique
associates acidity with submicron-size  acid sulfate  particles.   In subsequent
work  in  the St.  Louis area,  well   over  90  percent of  sulfur in particles
measured at rural sites near  St.  Louis  were found to be in the  fine particle
size range with  little, if any, seasonal variation  (Altshuller  1982).

Measurements of  particle  size distribution  of  sulfates  were made  with a
diffusion  battery  technique  at Glasgow,   IL,  104 km NNW of the Arch  in St.
Louis, from July  22-30,  1975  (Tanner and  Marlow 1977).  About  50 percent of
the  sulfate  containing  particles  were less than  0.25 ym  in size.   The
higher acidities were associated with the  submicron  particles.

In  the previously mentioned sulfate  measurements  in  the Great Smoky Mountains
National Park,  strong acidity was   associated with  the fine  particle size
fraction (Stevens  et al.  1980).    It  was estimated that ammonium bisulfate
constituted 61 percent of the fine particle mass.

Pierson  et al.   (1980a)  used  an Andersen  eight-stage cascade impactor to
obtain particle  size  distributions  for sulfate  and hydrogen  ions at a tower
on  Allegheny Mountain  in  southwestern  Pennsylvania.   The particle size dis-
tribution  curves for sulfate and hydrogen  ion were almost  identical, with an
average HMD  of  0.8 urn.   About 90  percent of the  sulfate and hydrogen ion
content  was less  than  3 ym.   The  [H+]-to-[S042~]  ratios  were  somewhat
higher for particles  between  0.7 and  1.1 ym than  for those  less  than 0.7
urn,  or  between  1 and 2 ym.    Acidity  was measured  in even larger particles
but  the  [H*] to  [S04^~]   ratio  was lower  than  for  particles  less  than 2
ym  (Pierson et al. 1980a).

Aircraft outfitted  with particle sizing equipment were flown across portions
of  Arizona, Utah, Colorado, and New  Mexico on 5 and  9  October 1977 (Macias et
al.  1980).   The  MMD for sulfur in  the collected particles was  not reported,
but  can be approximated  as less than 0.5 ym.   Sulfur particles less than 1
ym  constituted 92 percent of the sulfur content.

5.2.4.3  Measurements  at Remote Locations--Gravenhorst (1978)   found the ex-
cess  sulfate  in  marine aerosols to  be present   in submicron-size particles.
The  sulfate  associated  with  sea salt  was present in supermicron particles.
                                     5-27

-------
Meinert and Winchester (1977) also found the excess sulfate to be present in
submicron-size particles  in  samples collected  in  Bermuda.   Similarly, the
excess  sulfate   in  samples  collected  off  the  West  African  coast  was  in
submicron-size particles  and the larger  particles  appeared to  contain the
sulfate associated with sea salt (Bonsang  et al.  1980).

5.3  NITROGEN COMPOUNDS

5.3.1  Introduction

The nitrogen oxides  and their atmospheric  reaction products constitute a more
complex group  of chemical  species  than  do sulfur dioxide  and particulate
sulfates.    Unlike sulfates,  nitrate composition frequently  is  dominated by
volatile species, nitrous  acid,  nitric acid,  and organic nitrates,  particu-
larly peroxyacetyl nitrates.   Nitrous oxide, although present in significant
trace  concentrations  in  the   atmosphere,  does  not   react   within  the
troposphere.

Nitric oxide, the predominant nitrogen oxide in emissions  can  be converted
rapidly to nitrogen dioxide by  reactions with oxy  radicals  and  ozone in the
atmosphere.   Subsequent  atmospheric  reactions  result  in the  formation  of
nitric  acid.   Nitric  acid  and ammonia  are in  equilibrium with  ammonium
nitrate.   Ammonium  nitrate  formation  is  favored by  lower  temperatures and
sufficiently high levels of  ammonia.   Mixed nitrate-sulfate aerosol   systems
also play  a significant role  in  determining  the  nitric acid concentration, as
does  relative  humidity.    Nitrous acid  can  form  at night  but  is   rapidly
photolyzed in daylight.  A wide variety  of volatile organic nitrates can be
synthesized in the  laboratory;  however,  many are short-lived in  the atmos-
phere  or,   if  present,  occur  at parts-per-trillion  concentrations.   The
exceptions are the  peroxyacetyl nitrates  (PAN),  which  are present  at sig-
nificant concentration  levels relative  to  the  other nitrogen oxides and their
acids.  Because the peroxyacetyl nitrates and their precursors are in rever-
sible equilibrium, nitrogen dioxide can be regenerated and nitric acid may be
formed as  these species undergo  atmospheric  transport.

As  a consequence  of  the  atmospheric  reactions  discussed above,   several
species containing nitrogen  can contribute  directly or  indirectly to acidic
deposition.

5.3.2  Nitrogen Oxides

5.3.2.1    Historical  Distribution Patterns  and  Current  Concentrations  of
Nitrogen Oxides—Nitric oxide is the most  commonly emitted oxide of nitrogen.
Less  than  10 percent  of  nitrogen  oxides  are  emitted  as  nitrogen   dioxide
(N02).  Exceptions are  found in emissions from some types of diesel   and jet
turbine engines and  tail  gas from nitric acid plants,  which can contain from
30  to 50  percent  nitrogen  dioxide.   Because  nitric  oxide  (NO)  converts
rapidly to  N02 in the atmosphere, N02  is the  predominant  form  of nitrogen
found outside cities.

Historical  trends for  NO and N02 are  not  available  from nonurban sites but
are  available  from a  limited  number  of  urban  sites.    Because of these
                                     5-28

-------
limitations,  it is  not  useful  to  separate historical  trends  from  current
measurement results.

5.3.2.2  Measurements Techniques- Nitrogen Oxides—Most of the nitrogen  oxide
measurements  made  during  the  1970's involved  use  of chemi luminescent  ana-
lyzers.  While  the  chemiluminescent  technique  can be used to analyze  nitric
oxide  directly  and  specifically,  analysis  of  nitrogen  dioxide or  nitrogen
oxides (NO +  NOg) requires  a  converter  to reduce nitrogen dioxide to  nitric
oxide.   However, it has been  found  that such  converters  also will  reduce
other nitrogen compounds to nitric oxide.  Winer  et  al .  (1974)  reported  that
commercial chemiluminescent analyzers equipped  with  either molybdenum or  with
carbon  converters   quantitatively  reduced  peroxyacetyl   nitrate   to  nitric
oxide.  Nitric acid also was observed to cause  a response in chemiluminescent
analyzers, but the response to nitric acid was  not determined  quantitatively.

Spicer and  coworkers discussed the  use of various  converters  or scrubbers
(Spicer 1977, Spicer et  al . 1976b,  Spicer and Miller 1976).  Nearly quanti-
tative, but somewhat variable chemiluminescent responses to nitric acid  have
been obtained (Spicer and  Miller  1976,  Spicer  et al . 1976b).   The reduction
of nitric acid  to  nitric oxide  by a  stainless  steel  converter  was  shown  to
increase rapidly from below 10  percent  to over 90 percent between 400 C and
550 C.  However, the use of the  lower temperature also reduces the  efficiency
of conversion of nitrogen  dioxide  to nitric  oxide  by stainless  steel  con-
verters,  so  lowering the  temperature would not  be  a satisfactory  approach
(Spicer  et  al .  1976b).    Although  carbon  converters will  reduce  nitrogen
dioxide  to  nitric  oxide  efficiently at  lower  temperatures  than stainless
steel, the nitric acid reduction also continues  to occur efficiently down  to
140 C.   Nylon filters or scrubbers remove  nitric acid but not peroxyacetyl
nitrate and provide  a basis for analyzing nitric  acid  differentially (Spicer
et al .  1976b).   Use  of  ferrous sulfate as  a  scrubber was  found to  remove
nitric acid with high efficiency, but it also  removed  a  variable  fraction  of
peroxyacetyl   nitrate (Spicer et  al .  1976b).   Use  of  such  scrubbers  with
chemiluminescent instruments permits  the analysis not only of nitrogen  oxides
but also of other nitrogen compounds (Kelly and Stedman 1979b,  Spicer et al .
1976b, Spicer 1979).

5.3.2.3   Urban  Concentration  Measurements — The Air  Quality  Criteria for
Oxides of Nitrogen  (U.S. EPA 1982) contains detailed compilations of ambient
air concentrations  of nitrogen  dioxide  in U.S. urban areas.  Pertinent  data
from  the  criteria  document  are  summarized  in  the  following discussion.
Average  NO  and  N02  concentrations   at Continuous  Air  Monitoring  Program
(CAMP)  sites  were  comparable,  while peak  concentrations  of  NO  tended  to
exceed peak  concentrations  of
Trends  in  NO?  concentrations  at  the  six CAMP  sites in  Philadelphia,  PA,
Chicago, IL, Cincinnati, OH, Denver, CO,  St.  Louis,  MO,  and Washington, DC,
and at  other  sites  in  Los Angeles, CA, Azusa, CA, Newark, NJ , and Portland,
OR, have been tabulated and statistically  analyzed.

The annual mean  concentrations of N02  at the  sites  ranged from  50  to 150
|jg nr3  with  the  higher concentrations occurring  at the  sites  in downtown
                                     5-29

-------
Los Angeles  and in Chicago.   The maximum 1-hr NC>2  concentrations  at these
sites ranged  from  200  to 1500 yg  m-3.   Peak  1-hr  concentrations  above 750
yg  m-3  were  frequently measured  in downtown  Los  Angeles  and Azusa,  CA,
but infrequently, if at all,  at other sites.   Both upward and downward trends
with time were measured at  these  sites.

At  31 urban  sites  during 1976,  the  maximum  1-hr  concentrations ranged from
216 to  815  yg m~3.  The annual  mean concentrations at  two-thirds  of these
sites ranged from 50 to 100 yg  nr3.

Seasonal  behavior   in  N02   concentrations  varied  at  urban  sites,  with  a
summer  peak  occurring  at a  site in Chicago,  IL,  winter peaks at  sites  in
Denver,  CO,  and Lennox, CA, but no significant seasonal trends at other sites
in California.

The diurnal   patterns of  N02  concentrations are available by  quarter of the
year at eight sites (Trijonis 1978).   Except  for the  two  sites in the western
part of the  Los Angeles  Basin, the diurnal  patterns show  two peaks—one  in
the morning  hours, the  other  late in  the afternoon or during  the evening
hours.   At   the  two  sites  in  Los  Angeles, only a  single  peak late in the
morning hours was observed.   These peaks varied in size from site to  site and
with the quarter of the year.

5.3.2.4   Nonurban  Concentration  Measurements—Measurements  made  of nitric
oxide and nitrogen  dioxide at  suburban  and at rural locations in the United
States  are   tabulated  in Table  5-2.    Mean   and  maximum  concentrations  of
nitrogen oxides  are listed.  At  eastern nonurban  locations the mean concen-
trations  of  nitric  oxide  ranges  from 1  to  10   yg  m-3  while  the  mean
concentrations of nitric oxide  at western rural locations were at or below 1
yg  m~3.   Maximum  concentrations  of  nitric  oxide  at  a  number  of  sites
exceeded mean  concentrations by   factors of  10 to 30.   At eastern  nonurban
locations the mean concentrations of  nitrogen dioxide  ranges  were  from 2  to
27  yg  m-3,  but most  of  the  mean  values   ranged  from 4  to  14  yg  m"3.
At  two  western  rural sites the mean concentrations  of  nitrogen dioxide were
at  or  below 3 yg  m-3.   Maximum  concentrations of  nitrogen dioxide  at most
sites listed in  Table  5-2 exceed mean concentrations by  factors  of 5 to 10.
Although  mean  concentrations  of  nitrogen  dioxide  at  a  site exceed  mean
concentrations of  nitric oxide,  maximum concentrations of  nitric  oxide at a
number  of sites  equal  or exceed  maximum concentrations of nitrogen  dioxide.
This  latter effect  suggests  that  occasional fumigations  by  strong  local
sources of nitric oxide can occur at many rural locations.

The  range of mean nitrogen  dioxide  concentrations  of  4  to 14  yg  m-3 given
above compares with the 50  to  100  yg  nr3   range  obtained  for many  urban
sites (Section  5.3.2.3).  Additional measurements related to the gradient of
nitrogen dioxide concentrations  between  urban and  rural  sites are available
from the RAPS/RAMS monitoring  results in the  St. Louis area (U.S.  EPA 1982).
During  an air pollution episode  in  St. Louis during 1  and  2 October  1976,
nitrogen dioxide as well as  other  compounds  including ozone were elevated in
concentration.   The  diurnal  patterns and concentrations  of nitrogen dioxide
at  rural  compared  to  urban sites were  substantially different.  The diurnal
patterns at  urban  sites included two peaks  in nitrogen  dioxide,  one in the


                                     5-30

-------
TABLE 5-2.  MEASUREMENTS OF CONCENTRATIONS OF NITROGEN OXIDES AT SUBURBAN AND RURAL SITES
Site (Type)
Montague, MA (R)
Ipswhich, MA (R)
Scranton, PA (S)
DuBois, PA (R)
Ul
co
[ - »
Bradford, PA (R)
McHenry, MD (R)
Indian River
DE (S)
Lewisburg, WV (R)
Shenandoah, VA (R)
Research Triangle
Park, NC (S)
Period of
measurement
(method)
Aug.-Dec. 1977
(chemilumin.)
Dec. 54-Jan. 55
(colorimetric)
Aug.-Dec. 1977
(chemilumin.)
June-Aug. 1974
(chemilumin.)
July-Sept. 1975
(chemilumin.)
June-Aug. 1974
(chemilumin.)
Aug.-Dec. 1977
(chemilumin.)
Aug.-Dec. 1977
(chemilumin.)
July-Aug. 1980
(chemilumin.)
Nov. 65-Jan. 66
Sept. 66-Jan. 67
Nitric
Mean
3
ND
3
ND
2.4
ND
3
1
1
2.3
NA
oxide,
Max.
78
ND
70
ND
34
ND
114
33
NA
NA
NA
Nitrogen
Mean
7
2.6
11
19
5.1
11
5
4
4
10.6
14.3
dioxide,
Max.
73
3.8
64
70
68
60
48
28
NA
NA
NA
Reference
Martinez and Singh
1979
Junge 1956
Martinez and Singh
1979
Research Triangle
Institute 1975
Decker et al . 1976
Research Triangle
Institute 1975
Martinez and Singh
1979
Martinez and Singh
1979
Ferman et al . 1981
Ripperton et al .
1970
                 (colorimetric)

-------
                                                  TABLE 5-2.  CONTINUED
en
i
oo
ro
Site (Type)
Research Triangle
Park, NC (S)
Green Knob, NC (R)
Appalachian Mt.
Florida, southeast
coast
DiRidder, LA (R)
Wilmington, OH (S)
McConnelsville, OH
(R)
Wooster, OH (S)
New Carlisle, OH (R)
Ashland, Co., OH (R)
Period of
measurement
(method)
Aug. -Dec. 1977
(chemilumin.)
Sept. 1965
(colorimetric)
July-Aug. 1954
(colorimeteric)
June-Oct. 1975
(chemilumin J
June-Aug. 1974
(chemilumin.)
June-Aug. 1974
(chemilumin.)
June-Aug. 1974
(chemilumin.)
July-Aug. 1974
(chemilumin.)
May- Dec. 1980
Nitric
ug nr
Mean
10
2.7
ND
1.9
ND
ND
ND
6.0
4.3
oxide,
o
Max.
249
NA
ND
17
ND
ND
ND
64
NA
Nitrogen
vg
Mean
13
6.4
1.8
4.9
13
12
13
27
15.6
dioxide,
nr3
Max.
145
NA
3.7
43
90
70
90
NA
NA
Reference
Martinez and Singh
1979
Ripperton et al .
1970
Junge 1956
Decker et al . 1976
Research Triangle
Institute 1975
Research Triangle
Institute 1975
Research Triangle
Institute 1975
Spicer et al . 1976,
Shaw et al . 1981
                            (chemilumin.)
                                                                                         Shaw et al.  1981

-------
                                            TABLE  5-2.   CONTINUED
Site (Type)
Franklin Co., IN
(R)
Union Co., KY (R)
Giles Co., TN (R)
Creston, IA (R)
en
i
CO
Wolf Point, MT (R)
Pierre, SD (R),
site 40 km WNW of
P i erre
Jetmore, KA (R)
Period of Nitric Oxide, Nitrogen Dioxide,
measurement yg m-3 pg m-3
(method) Mean Max Mean Max Reference
May-Dec. 1980 3.0 NA 14.3
(chemilumin.)
May- Dec. 1980 2.5 NA 12.3
(chemilumin.)
Aug.-Dec. 1977 5 96 11
(chemilumin.)
June-Sept. 1975 4.7 28 4.3
(chemilumin.)
June- Sept. 1975 < 1 .0 NA 1.5
(chemilumin.)
July- Sept. 1978 < 0.25 NA 2.3
(chemilumin.)

April -May 1978 1.2 NA 7.5
(chemilumin.)
NA Shaw et al .
NA Martinez and
1979
55 Martinez and
1979
25 Decker et al
NA Decker et al
NA Kelly et al .

1981
Singh
Singh
. 1976
. 1976
1982

NA Martinez and Singh
1979
R = Rural.
S = Surburban.
ND = Not determined.
NA = Not available.

-------
late morning hours and the other during  the  evening  hours.   At suburban  sites
only an evening  peak  in  nitrogen dioxide occurred, while  at rural  sites no
peak in nitrogen dioxide concentration was observed.   The evening peaks in
nitrogen dioxide  concentration  within  the  city  ranged  from 250 to  500 ]jg
nr3, while  the  concurrent concentrations of nitrogen  dioxide  at the outer-
most rural  sites, 40  km  from the center  of the  city, ranged from  20 to 40
pg  Fir3.    Similarly the  24-hr   average  concentrations of  nitrogen  dioxide
ranged  from 200  to 265  vg m~3  at  urban  sites  but averaged  only  20 yg
m~3  at  rural   sites.    These  results   demonstrate  the  rapid   decrease in
nitrogen dioxide concentrations  that  can occur from urban  sites to adjacent
rural sites.

The  cumulative  frequency distributions  of  hourly  nitrogen dioxide  concen-
trations  reported  in  two studies  (Decker  et  al.  1976,  Research  Triangle
Institute 1975)  are reproduced  in part  in  Table 5-3.   Except  at the  sites
evaluated as suburban (Table 5-2), nitrogen dioxide concentrations exceeding
40  yg  nr3  occur very infrequently at nonurban sites.   Even at those  sites
considered to be in suburban locations,  nitrogen dioxide concentrations  were
infrequently  above  60  yg  nr3.    The  highest nitrogen  dioxide concentra-
tions  at  nonurban  locations infrequently  fall   within   the range  of  mean
nitrogen dioxide concentrations  at urban sites.

The  distinction  between  suburban and  rural  sites was made on  the  basis of
three  factors:  (1)  geographical  location,  (2)  frequency  of  elevated  concen-
trations  of nitric oxide,  and   (3)  the ratio of nitric  oxide  to nitrogen
oxides (NO + NOg).  The  third of these  factors was  discussed in some detail
by  Martinez  and Singh (1979).   They  found  this  ratio  tended to be lower at
rural than at urban or suburban  sites.   At  the four SURE sites  they  consid-
ered rural,  the ratios of NO to NOX ranged from 0.11 to 0.33  and averaged
0.23.  At the five  SURE  sites they considered  suburban,  the ratios  of NO to
NOX ranged from 0.21 to 0.43 and averaged 0.33.

Some  of the  relationships  discussed  above may  be somewhat biased  by the
tendency in  a  number of  the  studies involving nonurban  sites  to  limit the
measurements  to the warmer  months of  the  year.   Nitrogen dioxide  concen-
trations during  the winter months  have  been reported to exceed  those  during
the summer months by 50 to 100 percent (Shaw et al.  1981).   Nevertheless, the
measurements available do indicate a rapid  decrease  in  nitrogen oxide  con-
centrations  from urban to suburban  to rural locations in  the eastern  United
States.

5.3.2.5   Measurements  of Concentrations at Remote Locations—The results of
measurementsfor nitrogen  oxidesfrom  anumber of  studies carried  out at
remote locations are tabulated in Table  5-4.   The distinction between remote
and  rural locations is  somewhat arbitary.  In this discussion,  locations at
which  concentrations  of  nitrogen   dioxide  of  less  than  1  ug   m~3  were
frequently  measured are  considered   to  be remote.    However,  substantially
higher concentrations of  nitrogen  oxides were  observed at  a number of  these
locations  on those occasions  that   polluted  air  masses  crossed  over  the
measuring sites.
                                     5-34

-------
                  TABLE 5-3.  CUMULATIVE  FREQUENCY  DISTRIBUTION  OF  HOURLY  CONCENTRATIONS  OF
                              NITROGEN DIOXIDE  AT RURAL  AND  SUBURBAN  LOCATIONS
oo
Site/reference
DuBois.PA
Research Triangle
Institute 1975
Bradford, PA
Decker et al . 1976
McHenry, MD
Measurement
period
June-Aug. 1974
July-Sept. 1975
June-Aug. 1974
Percent of hourly average concentrations
greater than stated concentrations
20 yg m-3
13.2
2.1
6.9
40 yg m-3
1.0
0.1
0.2
60 yg m-3 80 yg m-3
0.2 0.0
0.0 0.0
0.1 0.0
Research Triangle
Institute 1975
Wooster, OH
Research Triangle
Institute 1975

McConnelsville, OH
Research Triangle
Institute 1975

Wilmington, OH
Research Triangle
Institute 1975
Creston, IA
Decker et al. 1976

Wolf Point, MT
Decker et al. 1976

De Ritter, LA
Decker et al.  1976
                             June-Aug.  1974



                             June-Aug.  1974



                             June-Aug.  1974


                             July-Sept.  1975


                             July-Sept.  1975


                             July-Sept.  1975
23.8
 5.6
14.9
6.9
0.5
2.6
1.9
0.1
1.1
0.3
0.0
0.5
0.2
0.4
4.8
0.0
0.0
0.3
0.0
0.0
0.0
0.0
0.0
0.0

-------
TABLE 5-4.  CONCENTRATIONS OF  NITROGEN  OXIDES  MEASURED  AT  REMOTE  LOCATIONS
Sites
Colorado, USA
Niwot Ridge
Colorado, USA
Niwot Ridge
Colorado, USA
en Fritz Peak
CO
(Ti
Island of Hawaii
Mauna Kea
Laramie, WY
Ireland, Adrigole
Co. Cork
Ireland, Loop
Head
Ireland, Loop Head
Measurement
period
(method)
Jan. and April
1979 (chemilumin.)
Dec. 1980 to Jan.
1981 (chemilumin.)
Fall 1974; Summer
Spring 1975-76
(absorption
spectroscopy) Dec.
1977 (chemilumin.)
Nov. 1954
(colorimetric)
Summer 1975
(chemilumin.)
Aug. -Sept. 1974
(chemilumin.)
April 1979 (Diff.
opt. abs. uv)
June 1979
(chemilumin.)

NO
0.02-
0.06
NA

NA
NA
ND
0.01-0.
<_ 0.2
ND
< 0.01
Concentrations
in yg m~^
N02 NO
NA 0
NA <

< 0.2
NA 0
2
06 NA 0
0.8
0.3
0.16

xa
.4-0.5
0.1

NA
.2-0.5
ND
.2-0.8
NA
ND
NA
Remarks Reference
Kelly et al . 1980
Bellinger et al .
1982

Noxon 1978
Kley et al . 1981
Junge 1956
Drummond 1976
Maritime Cox 1977
air
Maritime Platt and Perner
air 1980
Maritime Helas and Warneck
air 1981

-------
                                                 TABLE 5-4.   CONTINUED
CJl
I
co
Concentrations
Measurement in yg nr^
Sites (method) NO N02 N0xa
Tropical Areas 1965-1966 0.1-0.6 0.4-0.8
(colorimetric)
0.3-0.5 0.6-0.9

0.3-0.8 0.6-0.1
0.3-0.8 0.6-0.9



Remarks Reference
Under Lodge and Pate
canopy of 1966, Lodge et al
forest 1976
Above
canopy of
forest
Riverbank
Seashore
and
maritime

-------
At Niwot Ridge in the Rocky Mountains 20 miles west of Boulder, CO, Kelly et
al.  (1980)  reported  average concentrations of  0.4 to 0.5 yg  m-3  in  clean
air,  while  Bellinger et  al.  (1982) reported  nitrogen  oxide concentrations
below 0.1  yg  m-3 in  a  number of  clear air masses  passing  this site.   In
contrast, Kelly et al. (1980)  observed nitrogen oxide concentrations up to 40
yg m-3  when polluted  air  arrived from  the  east.   At Adrigole  on  the  coast
of  Ireland, Cox  (1977)  measured  nitrogen dioxide  concentrations below  1 yg
m-3  in  maritime  air  but  also reported  measuring  maximum  hourly concentra-
tions of  nitrogen  dioxide of  10  yg nr3 and  a maximum  daily  average  value
of about  3  yg m-3.   Similarly at  Loop Head,  the  concentrations  of  nitro-
ogen  dioxide measured in  maritime air by Platt and Perner (1980) were below
0.3  yg  m-3,  in  other  air masses  they measured  nitrogen  dioxide  concen-
trations  from 4  to  5  yg nr3.    Therefore,  although the  sites  listed in
Table 5-4 are  listed as  remote, it  was not  uncommon for air masses containing
nitrogen  oxide  concentrations  overlapping  those at  rural  locations  to  pass
across these sites.

In aircraft flights  up  to 5  to  6 km over West Germany,  Drummond  and  Vol z
(1982)  measured  nitrogen  dioxide  concentrations  in the  0.1  to  1 yg  m-3
range.  Kley et al.  (1981) measured  nitrogen  oxide concentrations  as  low as
0.1  yg  m~3 at 7  km  over  the  vicinity  of  Wheatland,  WY.    During  the  1977
and  1978 GAMETAG flights,  nitric  oxide  concentrations  equal  to or  below 0.1
ug m-3 were measured in  maritime  and in  continental  air at 6  km.

The  measurements at  the surface and aloft at remote locations  result in  very
low  concentrations of nitrogen oxides  in clean air masses.   The background
concentrations at the surface and aloft at  remote  locations can be 10 to 100
times lower than at  rural  locations in eastern North America (Tables 5-2 and
5-3).   The  higher concentrations measured at remote locations are attributed
by  the  various investigators to  polluted  air masses  from  populated areas.
Therefore,  natural  sources  of nitrogen  oxides do  not  appear likely to  con-
tribute significantly to  the  nitrogen oxide concentration  levels  in eastern
North America.

5.3.3  Nitric  Acid

5.3.3.1   Urban Concentration  Measurements- -Nitric acid  (HNOs) measurements
have been limited to short studies within urban  areas.   Continuous coulometry
(Spicer  et  al. 1976b,  Spicer  1977)  with a  detection  limit of  about  2 ppb
(5.16  yg m-3)  and  Fourier  transform  infrared  spectroscopy (FTIR)  with  a
detection  limit  of 6 ppb  (15.48  yg  rrr3) (Tuazon et al .  1978,  1980,  1981a ,
b;  Hanst  et  al.  1982)  were used to obtain the  ambient  air measurements for
HN03 listed  in Table 5-5.  An intercomparison study was conducted on the  10
different techniques  for  measuring nitric acid  in  Claremont,  CA,  during  an
8-day period in August and September 1979 (Forrest et al.  1982, Spicer  et al.
1982a).  The methods compared included chemi luminescence,  infrared, diffusion
denuder,  and filtration techniques.   The  nitric acid concentrations  ranged
from 1.8 to 37.0 yg m-3  or 0.7 to 14.4 ppb based  on the  median  values  of
the 10 methods (Spicer et al. 1982a).

The average  HN03  concentrations  in the Los Angeles Basin area ranged  from 7
to  40  yg m-3  (Table 5-5).   The  Riverside  site where the highest ammonia


                                     5-38

-------
                       TABLE 5-5.   CONCENTRATIONS OF  NITRIC  ACID,  PEROXYACETYL NITRATE,
                                AND AMMONIA  AT URBAN SITES  IN THE UNITED STATES
GO
Concentrations, yg m-3
Site
West Los Angeles, CA
(Cal. State Univ.)
West Covina, CA
Claremont, CA
(Harvey Mudd College)
Clareraont, CA
(Harvey Mudd College)
Riverside, CA
(UC Riverside)
Riverside, CA
(UC Riverside)
St. Louis, MO
Dayton, OH
ND = Not determined.
Period of
year
June 1980
Aug-Sept. 1973
Oct. 1978
Aug-Sept. 1979
Oct. 1976
July- Oct.
July-Aug 1973
July-Aug 1974

Avg
18.1
7.7
41.3
20.6
5.2-12
12.9-18
7.7
15.5

HN03
Max
30.0
103.2
126.4
56.8
.9* 20.6
.13 51.6
206.41
139.31

PAN
Avg
35
10
25
20
45
30
a 10
3 ND

Max
80
95
185
55 0
90
90
95
ND

aMany individual values were below detectability limits (DL); lower concentrations
assuming values below DL equaled zero; upper concentration values listed based on
DL equaled following concentrations: HN03, 12.9 yg m~3; PAN, 10 yg nr3; NH3, 2.1
NH3
Avg Max
2.1 5.6
2.1 9.1
5.6 21.0
.7-2.83 8.4
14.0 42.0
14.7 92.4
2.8 11.2
ND ND

listed based on
assuming values
yg m~3.
References
Hanst et al . 1982
Spicer
Tuazon
1981b
Tuazon
1981a
Tuazon
1978
Tuazon
1980,
Spicer
Spicer
1976a

below
1977
et al.
et al.
et al.
et al.
1981a
1977
et al.


          ''These values appear unusally high when compared with NOX, PAN and 03 concentrations reported  as
          present during same time periods.

-------
concentrations  were  measured  had   the  lower  HN03  concentrations.    This
follows  from  the  equilibrium  between nitric acid and ammonia, with  ammonium
nitrate  aerosol  being shifted  toward  aerosol  formation  in  the  presence  of
high ammonia concentrations.

                      NH4N03  t  NH3 +  HN03-

The  maximum HN03  concentrations  reported  at  several  midwestern  sites  are
higher  than those at Los Angeles area  sites.   These maximum  concentrations
also are unusually  high  in  comparison  with the NOX. ozone,  and  peroxyacetyl
nitrate  concentrations  measured concurrently.    Therefore,  these values  are
suspect.

The  averages  of   24-hr  HN03  concentrations  are  small  compared  with  the
corresponding  NOX  concentrations.    The  NOX  concentrations  averaged  over
the  study  period  were:   St.  Louis, MO,  111 yg m-3;  West  Covina,  CA,  343
yg nr3 and Dayton, OH, 134 yg  nr3  (Spicer et al.  1976a, Spicer  1977).

The  diurnal  patterns at  the  Los  Angeles  area  sites for HN03 concentration
are similar to that of the ozone with peaking in the afternoon hours  (Spicer
1977;  Tuazon  et  al.  1981a,b;  Hanst et  al. 1982).   Nitric acid decreases
appreciably in  concentration  during the  night.   In  Dayton, OH,  and in  St.
Louis, MO, the diurnal profiles  of nitric acid  showed both morning and  after-
noon peaks, unlike  ozone and  PAN, which  peaked  only in  the afternoon  hours
(Spicer et  al. 1976b, Spicer  1977).   However,  the nitric  acid  concentrations
frequently were near the limits  of detectability.

5.3.3.2  Nonurban Concentration Measurements--Measurements of nitric acid  at
suburban and  ruralsites  are  listed  in Table  5-6.   Some  of the earliest
measurements of nitric acid  in ambient air were made at two  sites outside  of
Dayton,  OH—Huber Heights,  a  surburban  location,  and  New  Carlisle,  OH, a
small  town  (Spicer  et al. 1976a).  Analyses were made  by continuous  coulo-
metry.   The average concentrations  of  nitric acid  were  in the 2.6 to 5.2 yg
nr3  range.    The  maximum  concentration   of 116.1   yg  m~3  reported  at  New
Carlisle appears to be too high.

Nitric acid measurements were  obtained  at Pittsburg,  a small  town in northern
California  (Appel  et al. 1980).   Tandem  filter  technique  was used  with a
Teflon prefilter  for collection of particulate nitrate and either a nylon  or
Nad-impregnated  filter  was  used to collect  HN03.    Positive interference
problems are  known  to  occur  because  of nitrate loss from  the  particulate
collected on  the  prefilter, due  to  volatilization  onto   the  filter  used  to
collect  HN03.   The  range of  nitric acid  concentrations  was 0.7  to 3.9 yg
nr3 (Table 5-6).

Nitric  acid was measured by  Spicer et  al.  (1982c)  at Beverly  Airport,  MA
(Table 5-6).  The  nitric  acid concentrations usually were below the limit  of
detection of  2 ppb  (5.16 yg  nr3)  of  the  chemiluminescent  technique  used.
An integrated  filter technique  also was used  for  nitric  acid involving  the
use of a Teflon prefilter and  a  nylon backup filter.
                                     5-40

-------
TABLE 5-6.  MEASUREMENTS OF  CONCENTRATIONS OF NITRIC ACID,  PEROXYACETYL
          NITRATE  AND  AMMONIA AT SUBURBAN  AND RURAL  LOCATIONS
Concentrations, yg m~3

Site
Beverly Airport,
MA (S)
Van Hiseville, NJ
(R)
en
£ Luray, VA (R)
Research Triangle
Park, NC (S)
Huber Hts., OH (S)
New Carlisle, OH (R)
Croton, OH (R)
Warren, MI (S)
Period
of
measurement
July- Aug.
July- Aug.
July-Aug.
June-July
July-Aug.
July-Aug.
1978
1979
1979
1980
1974
1974
August, 1980
Sept.-Oct
Jan. -Feb.
May-J une
. 1979
1980
1980
HN03
Avg
2
< 2
1
2
2
5
1
0
1
2
.6
.1
.0
.1
.1
.2
.8
.8
.3
.4
Max
£ 5.2
11.6
2.1
2.4
38.7
116.1
9.8
< 2.6
5.2
15.5
PAN
Avg
9.0
2.5
ND
ND
< 5
ND
ND
ND
ND
ND
Max
110
32.5
ND
ND
50
ND
ND
ND
ND
ND
NH3
Avg
ND
ND
1.3
0.4
< 0.7
ND
0.4
0.8
0.6
0.9
Max
ND
ND
2.9
0.6
11.9
ND
0.6
2.8
< 1.4
5.6
References
Spicer et al
1982c
•
Spicer and
Sverdrup 1981
Cadle et al .
McClenny et
1982
Spicer et al
1976b
Spicer et al
1976b
McClenny et
1982
Cadle et al .
1982
al.
•
•
al.
1982

-------
                                               TABLE 5-6.  CONTINUED
Site
Concentrations, ug ~3
Period of HNOa PAN NH3
measurement Avg Max Avg Max Avg Max
References

ro
    Abbeville, LA (R)
Commerce City, CO
(S)

Thurber Ranch, AZ
(35  mi. SE Tucson)

Pittsburg, CA (S)
June-Aug. 1979    1.8       NA


Nov.-Dec. 1978    2.1       NA


July-Aug. 1981    1.6      5.2


February 1979     2.1      4.1
ND      ND    2.1     NA     Cadle et al.  1982
                                                                ND      ND    1.3     2.9    Cadle et al. 1982
ND      ND    0.8     1.5    Farmer and Dawson
                             1982

ND      ND    0.4     0.8    Appel et al. 1980
    ND = Not determined.

    NA = Not available.

-------
 In  this same  study  (Spicer et al.  1982c),  aircraft flights were made  fol-
 lowing  the urban plume of Boston, MA, over the Atlantic Ocean.  On one flight
 it  was possible  to  measure  the  nitric  acid  formed not  only  in the  urban
 plume,  10.3   yg  nr3,  but  also  in  the  Salem  power  plant  plume,  15.5  yg
 m"3.  The plumes were over the Atlantic Ocean north of Cape Cod.

 Measurements  of  nitric  acid concentrations were made during July  and August
 1979  at Van  Hi Seville,  NJ,  in  the  New Jersey  pine  barrens  (Spicer  and
 Sverdrup 1981).  Nitric acid was measured by the chemiluminescence technique,
 and  inorganic  nitrate  (HNOs  and  N03")  was  determined  by   use   of  the
 Teflon  prefliter  and nylon backup  filter collection method.  These  authors
 suggested that the potential  for  loss of nitrate  off the  Teflon prefilter
 onto the nylon filter,  resulting in  a  positive  interference  problem,  made it
 desirable to  consider the filter method  as acceptable only for measuring  the
 concentrations of  total  inorganic  nitrate.   On the  average,  the  total  inor-
 ganic  nitrate during the  study was 5  yg  m-3 and  the estimate  of  nitric
 acid concentration was  less  than 0.8  ppb or  2  yg  m-3  (Table  5-6).    The
 average diurnal profile for nitric acid  peaked at 1500  hours.  The ozone  and
 PAN concentrations peaked at about the same time in the afternoon.

 McClenny et   al.  (1982) reported  measurements of  nitric  acid  in  Research
 Triangle Park, NC, and  a rural  area near Croton,  OH (Table 5-6).   Analyses
 were made  by the tungstic  acid  integrative  sampling method,  which has  a
 sensitivity of 0.07   ppb (0.18  yg  m-3).   Nitric  acid is  effectively  ad-
 sorbed  on a   tungstic acid  surface,  subsequently  desorbed  into carrier  gas,
 and passed  on to  a  NOX chemiluminescent analyzer.   Maximum concentrations
 of  nitric acid and of  ozone  occurred  near midday at both sites, with  lower
 nighttime concentrations  for  both  but not as  large a  decrease for  nitric
 acid.

 Measurements  of  nitric  acid  by filter  techniques  at  several  suburban  and
 rural   sites   (Table  5-6) were  reported  by  Cadle  et al.   (1982).    At  the
 Abbeville, LA, and the  Commerce  City,  CO, sites, nitric acid concentrations
 were obtained by  difference  between  the inorganic  nitrate  collected on  a
 microquartz  filter and  particulate nitrate  collected on  a Teflon  filter.
 However,  subsequent   tests  indicate  that  the nitric   acid  may  have  been
 overestimated.   The  second method involved  removal  of nitrate  on a  Teflon
 filter  followed by removal  of nitric acid on a nylon filter.  The  positive
 interference  problem  possible with  this second  technique has  already  been
 discussed.

 The average diurnal  profile for  nitric acid from measurements at  Abbeville,
 LA, shows a  single late morning peak  for nitric  acid and  an afternoon  peak
 for ozone.  Nitric acid concentrations were found  to increase  from  fall  to
 winter to spring  in 1979-80  at the  Warren, MI,  site  (Cadle  et al.  1982).

 Both Appel  et al.  (1980)  and  Cadle et al. (1982) concluded  that  the  concen-
 trations of nitric acid and ammonia at their measuring  sites were too  low to
 result in the formation  of ammonium nitrate in  particul ate  matter.

Kelly and Stedman  (1979b)  measured  nitric acid by a chemiluminescent  tech-
 nique at a rural  site about 15 miles  east of  Boulder,  CO.  The nitric  acid


                                     5-43

-------
concentrations  during  February  1978  usually  were  in  the  1.3  to  12.9  yg
in" 3 range with  many  of the concentrations of nitric acid  in  the  2.6 to 5.2
yg m-3 range.

A  collection method  involving  condensation of  water  vapor  onto  a cooled
surface was  used  by  Farmer and  Dawson (1982) to  collect  nitric  acid (Table
5-6).   During  part  of  the sampling  period in  early August  1981, sulfur
dioxide and  nitric  acid concentrations  were well correlated.   The authors
associated this behavior with  transport and chemical transformations occur-
ring within smelter plumes  fumigating  the site.

The  average  nitric  acid concentrations  at most  of  the  suburban  and rural
sites  were  at or  below  2.6   yg m-3  with  the  concentrations  frequently
occurring in the  0.7  to 2.1 yg  nr3  range  (Table 5-6).    These concentra-
tions  of  nitric acid  are  about a factor of 10  lower  than  the  nitric acid
concentrations measured at  urban sites (Table 5-5).  The  nitric acid concen-
trations at suburban and rural  sites  also are about a factor of 5 to  10 lower
than the nitrogen dioxide  concentrations  at  surburban and  rural sites (Table
5-2).

5.3.3.3   Concentration  Measurements  at  Remote  Locations—Measurements  of
nitric  acid  also  are available at a  number of  remote  or relatively remote
locations  (Huebert  and  Lazrus  1978,  1980a,b;  Huebert 1980;  Kelly  et al.
1980).  Kelly  and coworkers measured nitric acid concentrations  at a rela-
tively  remote  site,  Niwot Ridge,  in the  Rocky  Mountains 20 miles  west of
Boulder,  CO, between  December  1978  and  August   1979.    A high  sensitivity
chemiluminescent  instrument  was used  with nitric acid measured  by thermal
decomposition  to  nitrogen  dioxide   followed  by FeS04   reduction  of  the
nitrogen dioxide.  Some  interference  by PAN was  observed  in tests with this
technique for  measuring  nitric  acid.   In clear  air masses  the  nitric acid
concentrations  often  were  below  the detection  limit  but, when measurable,
were  in  the 0.13  to 0.26  yg  m~3  range.  When  polluted air  reached the
site,  the nitric  acid concentrations  frequently were 0.5  yg m~3  or more
and values over 2.6 were measured occasionally.

Huebert (1980) and Huebert and Lazrus (1978, 1980a,b) measured nitric acid on
samples collected from aircraft or shipboard over remote areas of the Pacific
Ocean  and western North America.  Samples  were collected  using the  same sort
of  tandem  filter  technique discussed earlier.   Samples were collected from
aircraft as  part  of  project GAMETAG.   Surface concentrations of nitric acid
in  the equatorial Pacific  region  averaged 0.1  yg nr3  {Huebert  1980).   The
concentrations of nitric acid measured in the boundary layer  ranged  from less
than  0.03 to  2.22 yg  m-3, with  a  median range of 0.15  to 0.21  yg m-3
(Huebert and Lazrus  1980a).  The  free  troposphere nitric  acid concentrations
ranged from  less than  0.08  to  1.39   yg m-3  with  a  median  of  0.31  yg
m-3.   The nitric  acid concentrations in  the  boundary layer  in remote  areas
are  a  factor  of  5  to 10   lower  than  at rural   locations in eastern  North
America.
                                     5-44

-------
5.3.4  Peroxyacetyl Nitrates

Peroxyacetyl  nitrates  can be  determined  by electron capture  gas  chromatog-
raphy  down to  the 0.1  ppb  (0.5 yg  nr3)  concentration level  and  below.
This method can be used in urban, rural, or remote locations.   Long path FTIR
spectroscopy  has  been  used  to  measure  peroxyacetyl  nitrate  at  locations
within the Los Angeles Basin area.

5.3.4.1   Urban Concentration  Measurements--Peroxyacetyl  nitrate  concentra-
tions  have  been tabulated  when  obtained concurrently with  nitric acid  and
ammonia concentrations in Table 5-5.   Many other measurements  of peroxyacetyl
nitrate have been made in urban areas.

Additional average peroxyacetyl nitrate measurements made in  the Los  Angeles
Basin  area  are shown  in  Table 5-7.    The highest peroxyacetyl nitrate  con-
centrations have been reported from the sites in  the western  part  of  the Los
Angeles  Basin area.   In  the  eastern  part  of the  Los  Angeles Basin  area,
average peroxyacetyl nitrate concentrations usually have  been  measured in the
5 to 25 yg m-3 range.

Maximum  peroxyacetyl  nitrate  concentrations occur  late  in  the  morning  or
early  afternoon in  downtown  Los Angeles  (Mayrsohn  and Brooks  1965)  and
progressively later in the afternoon  passing from west to  east across  the Los
Angeles Basin area from downtown Los Angeles to Pasadena  (Hanst et al.  1975)
to  West  Covina  (Spicer   1977)  to  Claremont  (Tuazon  et  al.  1981a,b)  to
Riverside  (Pitts   and  Grosjeans  1979).   Pitts  and Grosjeans (1979)  also
reported  seasonal  variations  in  peroxyacetyl  nitrate  diurnal  peak  concen-
trations.  Two peaks were observed at the site in  Riverside, CA. The  earlier
peak was  associated  with  formation of peroxyacetyl nitrate from local  emis-
sions  while  the later  peak was  associated with  formation  of  peroxyacetyl
nitrate from  emissions  in  air masses traveling from west to  east  across the
Los Angeles Basin.  The peroxyacetyl nitrate concentrations were observed  to
decrease  at  night,  but  were still  present  at  significant  concentrations
(Spicer 1977;  Pitts and Grosjeans 1979;  Tuazon  et  al.  1981a,b).

The average peroxyacetyl  nitrate concentrations  reported  within  some  urban
and suburban  areas in the  United  States are  shown in  Table 5-8. The  average
peroxyacetyl  nitrate concentrations at a few sites have been within the  5  to
50  yg  m-3  range  (Lonneman  et  al.   1976).    However, at  other  urban  and
surburban  locations  the  average  peroxyacetyl  nitrate  concentrations  have
ranged from 1.5  to 4.5 yg m~3.   The  times  of maximum peroxyacetyl  nitrate
concentration   during  the  day usually  were  reported to  occur  during  the
afternoon hours in Houston, TX (Westberg et al.  1978a), St. Louis,  MO  (Spicer
1977),  and New  Brunswick,  NJ  (Brennen  1980).  At  sites  in  the Houston,  TX,
area peroxyacetyl  nitrate  concentrations usually  were  below  detectability
limits at night (Westberg  et al. 1978a), but  were present  at  measurable
concentrations at other sites (Spicer 1977, Brennen 1980,  Singh et  al.  1982).

Only a limited  number  of  measurements  of the next higher member of the  per-
oxyacetyl   nitrate  series,  peroxypropionyl  nitrate,  have  been  obtained  in
urban  areas (Darley  et al. 1963;  Lonneman et al. 1976;  Singh et al.  1979,
                                     5-45

-------
TABLE 5-7.  AVERAGE PEROXYACETYL NITRATE MEASUREMENTS
           FROM THE LOS ANGELES BASIN AREA
Site
Los Angeles



Pasadena
Claremont
Riverside
Year
1961
1965
1976
1979
1973
1980
1967-68
1975-76
1977
1980
1980
Concentration
ug nr^
100
155
40
25
150
65
19
18
8
6
24.5
Reference
Renzetti and Bryan 1961
Mayrsohn and Brooks 1965
Lonneman et al . 1976
Singh et al. 1981
Hanst et al. 1975
Grosjean 1981
Taylor 1969
Pitts and Grosjean 1979
Singh et al. 1979
Singh et al . 1982
Temple and Taylor 1983
                         5-46

-------
TABLE 5-8.  PEROXYACETYL NITRATE MEASUREMENTS FROM SEVERAL URBAN
             AND SUBURBAN AREAS IN THE UNITED STATES
Site
Hoboken, NJ
St. Louis, MO
Houston, TX
(Lange)
Houston, TX
(West Hollow)
(Aldine)
(Crawford)
(Fuqua)
(Jack Rabbit)
New Brunswick, NJ
San Jose, CA
Oakland, CA
Phoenix, AZ
Denver, CO
Houston, TX
Chicago, IL
Pittsburgh, PA
Staten Island, NY
Year
1970
1973
1976
1977
1978
1978-80
1978
1979
1979
1980
1980
1981
1981
1981
Concentration
yg nr3
18.5
31.5
2.0
3.0
4.5
3.0
3.0
4.0
2.5
4.5
2.0
4.0
2.0
2.0
2.0
1.5
3.5
Reference
Lonneman et
Lonneman et
West berg et
HAOS 1979
Martinez et
al. 1976
al. 1976
al. 1978a
al. 1982
Brennen 1980
Singh et al
Singh et al
Singh et al
Singh et al
Singh et al
Singh et al
Singh et al
Singh et al
. 1979
. 1981
. 1981
. 1982
. 1982
. 1982
. 1982
. 1982
                              5-47

-------
1981,  1982).   The  peroxypropionyl  nitrate  concentrations  measured usually
averaged 10 to 20 percent of peroxyacetyl  nitrate  concentrations.

The ratios of  average  peroxyacetyl  nitrate to nitric acid concentrations at
urban  sites can  vary widely.   For example, the ratio of average PAN to HN03
concentrations was about 3:1 during the  1978 study in Claremont, CA (Tuazon
et al. 1981a), but this ratio averaged  only 1:3  during the 1973  study at West
Covina, CA  (Spicer 1977).   The ratios of PAN to  HN03  concentrations also
can vary substantially  from day to day  at  the same site.

Nitrogen dioxide  and/or nitrogen oxide  (NO +  N02)  have  been measured con-
currently with PAN and  HN03  1n several studies.   The average ratios  of the
23-hr  average concentrations  of (PAN  +   HNOs)  to  (PAN  +   HNOs  + NOX)  in
West Covina, CA, and in St.  Louis, MO, were  0.1 (Spicer  1977).   The average
ratio  of  (PAN  + HNOs)  to (PAN  +  HNOs  + NO?)  concentrations  measured  in
Riverside,  CA,  was 0.2  (Tuazon  et  al.  1980).   Grosjean   (1983), using  a
commercial   chemiluminescent  analyzer,  found  PAN  and  HN03   to  interfere
quantitatively with the N02 measurements.   The  observed  concentrations  of
N0£  were  corrected  using  the  concurrent measurements  of  PAN and  HNO^.
The  ratios   of   (PAN   +  HN03)  to   (PAN  +  HNOa +   NOX   +   N03~)   in
Grosjean's   results  ranged  from 0.01   to  0.39  and  averaged  0.18.    On  the
average, the  results of these  several  studies  (Spicer  1977, Tuazon  et al.
1980,  Grosjean  1983) indicate  that (PAN +  HNOs)  accounts for from  10  to 20
percent of  the measured nitrogen species  in these  urban areas.

5.3.4.2  Nonurban Concentration Measurements—Concentration   measurements of
peroxyacetyl  nitrate and  peroxypropionyl  nitrate  at rural  and  remote loca-
tions  are  given  in Table  5-9.    Additional   measurements  of   peroxyacetyl
nitrate concentrations  are listed  in Table 5-6.  The average concentrations
of  peroxyacetyl  nitrate are in the range  of  0.5  to  5  yg  m~3 overlapping
the  range  of  average  PAN  concentrations  at urban  and suburban sites.  The
concentrations of PAN  at the remote sites, Reese  River,  NV,  Badger Pass, CA,
and Point Arena, CA, are about 0.5 yg nr3.

Lonneman et al. (1976)  observed two diurnal patterns of PAN concentrations at
the  site near Wilmington, OH.   One pattern  involved afternoon and evening
elevation in  PAN  and in ozone concentrations.  The  other pattern involved a
flat diurnal  profile for  the PAN concentrations,  but an  elevation in ozone
concentrations.   An afternoon  peaking of  the  PAN  concentrations  also was
observed at  the  Sheldon Wildlife Preserve, TX  (Westberg et  al. 1978b).   At
night, measurable concentrations of PAN were obtained at both of these rural
sites.

The concentrations of  peroxyacetyl  nitrate at rural sites were in about the
same concentration range  as  measured  for nitric  acid at rural  sites (Tables
5-6  and  5-9).  The concentrations  of  PAN at  remote locations  of about 0.5
yg  m-3 were  about the same  as  those  reported  for nitric  acid by Huebert
and Lazrus (1980a) at remote locations.
                                     5-48

-------
  TABLE 5-9.   PEROXACETYL NITRATE MEASUREMENTS AT RURAL  AND REMOTE SITES  IN THE UNITED  STATES
Site
Wilmington, OH

nuntington Lake,
IN
East Central
Missouri
Sheldon Wildlife
Preserve, TX
Jetmore, KA
Reese River, NV
Badger Pass, CA
Mill Valley, CA
Point Arena, CA
Nature of
site
Rural-continental

Rural -continental
Rural-continental
Rural-continental
Rural-continental
Remote- high
altitude
Remote- high
altitude
Rural -maritime
Remote-maritime
Period of
measurement
August 1974

April 1981
February 1981
October 1978
June 1978
May 1977
May 1977
January 1977
Aug. - Sept. 1973
Concentration, ug m-3
PAN PPN
Avg Max Avg Max
NA

2.5
3.5
4.0
1.25
0.55
0.65
1.50
0.40
20.5

NA
NA
15.0
2.5
1.3
1.10
4.15
1.40
ND

ND
ND
ND
ND
0.22
0.28
0.22
ND
ND

ND
ND
ND
ND
0.50
0.50
0.60
ND
Reference
Lonneman
1976
et

Spicer et al
Spicer et al
Westberg
1978a
Singh et
Singh et
Singh et
Singh et
Singh et
et
al.
al.
al.
al.
al.
al.

. 1983
. 1983
al .
1979
1979
1979
1979
1979
ND = Not determined.
NA = Not available.

-------
5.3.5  Ammonia

Unlike nitric acid and peroxyacetyl nitrate, which are formed through atmos-
pheric  reactions  involving   precursor   hydrocarbons  and  nitrogen  oxides,
ammonia  is  emitted directly  into  the atmosphere  from  near-surface sources
(Chapter A-2, Sections 2.2.2.7  to  2.2.2.10).   Consistent with ammonia being
emitted from ground-level sources,  ammonia concentrations have been found to
decrease with altitude (Georgii  and Muller 1974,  Hoell et al.  1983).  Ammonia
has a  significant  role in  neutralization of acid sulfate and nitric acid in
the  atmosphere   (Brosset  1978).   In  addition  ammonia,  when  it undergoes
deposition, can  participate significantly in chemical  reactions in soil.

Various techniques have been  used  to  sample and analyze ammonia.  Long path
FTIR spectroscopy  was  used at  several sites  in  the Los Angeles  Basin area
(Tuazon  et  al.   1978,  1980,   1981a,b;  Hanst  et al.  1982).    Dual  catalyst
chemiluminescent instrumentation was used in Los Angeles, St. Louis, and the
Dayton area (Spicer et al. 1976a,  Spicer 1977).   This procedure depended on
the fact that ammonia  is  oxidized to nitric oxide by high temperature but not
low temperature  catalysts while nitrogen dioxide is reduced by both high and
low  temperature  converters.   A tandem  filter technique involving  a Teflon
prefilter and two oxalic-acid-impregnated fiberglass filters has been used at
several locations  (Cadle  et  al. 1982).   Both positive  and  negative inter-
ferences can  occur.   A  similar tandem  filter technique with a  glass fiber
prefilter was employed by Appel et  al.  (1980).   Another method involved use
of  oxalic-acid-coated glass  tube   diffusion  denuders.    Another technique
involved collection  on  Chromosorb  T beads and  desorption   either  into  an
opto-acoustic detector or  a  chemiluminescent  analyzer (McClenny and Bennett
1980).  Harward  et al. (1982) also  used the acoustic detector.  The tungstic
acid technique was used  by McClenny et  al. (1982)  to measure ammonia.  Gas-
eous ammonia  and nitric acid  are  separated  from  particulate species  as a
result of their  more  rapid diffusion  to the walls  of a tungstic-acid-coated
Vycor tube.  The ammonia is  desorbed  into  a carrier gas and  readsorbed on a
second tungsten-oxide-coated  tube which  passes  nitric  acid  now in  the form of
nitrogen dioxide.  The ammonia  is  desorbed into a  chemiluminescent analyzer
as nitrogen dioxide.

5.3.5.1   Urban  Concentration  Measurements—The concentrations  of ammonia
measured at a number  of urban locations  are given in Table 5-5.   The highest
concentrations of ammonia in ambient air have been  measured at Riverside, CA
(Tuazon et al. 1978,  1980, 1981a).   These high  concentrations  were attributed
to ammonia emissions from feed  lots upwind of the  site  in  Riverside.  Nitric
acid  was observed to  decrease in  concentration with  increases  in ammonia
concentration at Riverside (Tuazon et  al. 1978,  1980)  due  to the ammonium
nitrate  equilibrium  relationship.   The  ammonia  concentrations  at  sites  in
Claremont, West Covina, and Los Angeles were substantially lower  than in  the
Riverside area  (Spicer 1977, Tuazon  et al. 1981a,b).   Such a  gradient  in
concentrations  of  ammonia is consistent  with   strong  localized  sources  of
ammonia  rather   than  more uniform basin-wide  emissions  of  ammonia.    The
ammonia  concentrations  measured in  St.  Louis  (Spicer  1977) were  not sub-
stantially  different  from those measured at  locations in  the  Los Angeles
Bavin  area  other than  the  Riverside area.  Concentrations of ammonia remain
                                     5-50

-------
 high  at  night  in Los  Angeles  and St. Louis  (Spicer 1977) consistent  with
 surface  emissions  of ammonia  into  the  shallower  mixing  layers  occurring
 during  the  nighttime  hours.

 5.3.5.2   Nonurban Concentration Measurements—Earlier measurements of ammonia
 concentrations  at nonurban locations  were in  the  range  from less  than  0.07
 yg  m-3  to  several  factors  of ten  times  greater  (Breeding  et al.  1973,
 1976; Lodge et  al.  1974).   Other measurements of ammonia that  were obtained
 concurrently with nitric acid concentration measurements are given  in  Table
 5-6.   Average  concentrations  range  from 0.35  to  2.1  yg  m-3 and  maximum
 concentrations  reported ranged  up to  11.9 yg  m-3.   However, this  latter
 concentration value observed at Huber Heights, OH,  is unusually  high compared
 to  the maximum concentration values at other suburban and rural  locations.

 Several additional studies have been reported at nonurban sites.  Ammonia was
 measured  at several sites on Cedar Island off  the coast  of  North  Carolina  in
 August  1978 (McClenny and  Bennett 1980).   The ammonia concentrations ranged
 from  2.1  to  2.4  yg  m-3.    The highest concentrations  were measured  imme-
 diately  above marsh  grass.   A few measurements also were  made  at Research
 Triangle  Park,  NC,  and these ammonia concentrations were in  the  2.8 to 4.2
 yg  m~3  range.    Measurements  of  ammonia  also  were made  nearby in  south-
 eastern Virginia at a  site  bordering  the  Great Dismal Swamp (Harward et al.
 1982).    The  ammonia  concentrations obtained  in August  and September  1979
 ranged  from  1.0   to  2.8  yg  m-3  and  averaged  1.9  yg m-3.  Measurements
 were made for comparison at Hampton,  VA.   The average ammonia  concentration
 was lower in air masses arriving  over water  than  over  land.   The  ammonia
 concentration also was lower during periods of rain.

 At  Hampton,  VA, the ammonia concentrations decreased from the  1.4  to 2.1  yg
 m-3  range  in late  summer  to  less  than  0.14 yg  m-3  in  the  early winter
 {Harward  et al. 1982).   A decrease in ammonia  concentrations  also was ob-
 served  at  Warren,  MI,  from  0.9 yg  m-3   in the  spring to  0.6  yg  nr3  in
 the winter  (Cadle et  al.  1982).   Although  such  seasonal  changes have  been
 associated  with changes  in soil  emissions and  fertilizer volatilization,
 higher  temperatures  also  could  explain the  shift  in the  ammonium  nitrate
 equilibrium  resulting in higher ambient air  ammonia  concentrations  (Cadle  et
 al. 1982).

 5.3.6  Particulate Nitrate

 Serious  difficulties have been experienced in obtaining accurate ambient air
measurements of participate nitrates.   During recent years substantial posi-
 tive and  negative artifacts have  occurred  during  the sampling of  nitrates
 from air.   The artifacts arise as  follows:

      (1)   Positive artifacts derived  from
           (a)  adsorption  of nitric acid by  filter medium,
           (b)  adsorption  of nitrogen  dioxide  by filter medium,
           (c)  loss of nitric  acid onto the  collected particulate
                matter on a filter  as  a  result  of chemical reactions
                with,  or adsorption by,  the  particulate matter.
                                     5-51

-------
     (2)  Negative artifacts  derived  from
          (a)  reactions of participate  nitrate  in the collected matter
               with strong  acids  in the  participate matter, resulting
               in release of  nitric acid;
          (b)  volatization of ammonium  nitrate  from the filter to form
               gaseous nitric acid and ammonia.

As a result of the artifact problems  given above the earlier nitrate measure-
ments  reported in  the  literature  are  likely  to be  questionable,  if  not
erroneous.

Most of the  early  measurements of  particulate nitrate  involved analysis for
nitrates in  samples collected  on glass  fiber filters  in high-volume  (HIVOL)
samplers (NAS 1977, U.S. EPA  1982).

A number  of  investigators  have observed in  measuring  particulate nitrate in
source emissions (Pierson  et  al. 1974)  and  in ambient  air studies (Witz and
MacPhee 1977; Stevens  et al.  1978; Spicer and  Schumacher 1977, 1979; Appel et
al.  1979,  1981a;  Witz and  Wendt  1981;  Shaw et  al. 1982;  Witz  et al. 1982)
that much  higher  particulate nitrate concentrations  were  measured  on glass
fiber  filters  than  on Teflon, quartz,  and  some other  filter  types.   Nitric
acid was  demonstrated  to be adsorbed on  glass  fiber  filters  in laboratory
studies (Okita et al.  1976,  Spicer and  Schumacher 1977, 1978, 1979, Appel et
al. 1979).  Nitrogen dioxide  also has been shown in laboratory studies to be
adsorbed  on  glass  fiber filters  (Spicer and Schumacher  1977,  1978, 1979;
Rohlach et al. 1979).   Appel  et al.   (1979) reported a positive artifact from
nitrogen dioxide at high ozone concentrations.  However, adsorption of nitric
acid rather  than  nitrogen  dioxide appears to be the dominant  source  of the
positive interference  (Appel  et al.  1979,  1981a).

Substantial  positive  nitrate artifacts have  been measured  on a  number of
other filter types including  Teflon-impregnated  fiber filters  (Pierson et al.
1980b),  silicone   resin  coated glass   fiber  filters  (Appel  et  al.  1979),
cellulose filters (Appel et al. 1979),  cellulose acetate filters  (Spicer and
Schumacher 1978,  1979,  Appel  et  al. 1979),  and  nylon  filters (Okita  et al.
1976, Spicer 1977, Spicer and Schumacher 1978, 1979).  Smaller but measurable
positive  artifacts have   been  reported  on  some  types  of  quartz   filters
including Gelman microquartz (Appel  et al. 1978, Spicer and Schumacher 1977,
1979)  and  Pall flex  Tissuquartz (Spicer  and Schumacher  1977,  Forrest  et al.
1980).

Negligible  positive artifacts  have been  obtained  on  Fluoropore  (Teflon)
filters  (Stevens  et al. 1978, Appel et al.  1979,  1980, 1981a,b; Pierson et
al. 1980b) on polycarbonate filters  (Spicer and Schumacher 1977), and on ADL
quartz  filters (Spicer  and  Schumacher  1978, 1979).    However,  atmospheric
particulate  matter  on  Teflon filters can retain  nitric acid  (Appel  et al.
1980).

Harker et  al.  (1977)  observed that  an  inverse relationship occurred  between
ambient air  sulfate and nitrate  concentrations  in samples collected at West
Covina, CA.  A group of controlled  photochemical experiments were  designed to
investigate  this  behavior.  When sulfuric  acid  was  generated and collected


                                     5-52

-------
concurrently with nitrates on Gelman Spectro Grade A glass  fiber filters, the
nitrate concentration was  lower  than in the absence  of  sulfuric acid.  The
researchers  concluded  that the  sulfuric acid  reacted  with  and caused the
release of  nitrate  probably  as nitric acid from  the  surface  of the aerosol
particles  (Marker et  al.  1977).    The  possibility of  a  negative artifact
effect on Fluoropore filters as  a  result  of reaction with sulfuric acid and
as a result of volatization of ammonium nitrate was discussed  by Appel  et al.
(1979).

Pierson et  al.  (1980a,b)  observed losses of  nitrate  off of  Fluoropore fil-
ters,  an  effect  associated   with   the  high  sulfuric  acid   concentrations
measured at  the  Allegheny Mountain  site.   Appel  et al.  (1981b) also  found
that particulate  nitrate collected on Teflon filters at  Lennox,  CA  decreased
with increasing amounts of ambient air sulfuric acid.   About half the nitrate
was  lost  at ambient air sulfuric  acid  concentrations of  10  yg m~^.    About
50  percent  of the  nitrate collected could be  lost from  Teflon filters  at
higher ambient temperatures,  29  to  35 C, and  about 30 percent RH  (Appel  et
al.  1981a).   No  losses  of nitrate appeared to  occur  from samples  collected
during  the  night and  morning  hours.    In  samples  collected  at  Research
Triangle Park, NC, large losses of particulate nitrate, up to  90 percent off
Teflon filters, occurred particularly during the day (Shaw  et  al. 1982).

Laboratory  experiments  were  carried  out  by Appel  et  al.  (1981b)  to  inves-
tigate the  losses of nitrate  off Teflon  filters  loaded  with submicron  (^
0.2  pm)   ammonium  nitrate particles.    With  equal   loadings  of  ammonium
nitrate and sulfuric acid on the  Teflon  filters,  over 90  percent  of the
nitrate was  lost  off the  filters after exposure to a clean air  stream  at 90
percent RH for six hours.  Volatization of  nitrate  under the  same conditions
in the absence of sulfuric acid  resulted  in 30  to 50  percent  losses of ammo-
nium nitrate.  Losses  of  about 90 percent  of the nitrate  occurred  when the
filters were  exposed to 17 to 23 ppb of  hydrochloric  acid.    Forrest   et al.
(1980) observed losses  of preloaded nitrate  from Pallflex Tissuquartz exposed
to  sulfuric  acid.   Particulate  nitrates  other  than ammonium   nitrate   can be
present in the atmosphere but they,  unlike  ammonium nitrate,  do  not volatize
readily.

The artifact  problems discussed  above appear  to have been dealt with  satis-
factorily by use  of diffusion-denuder tubes.  These tubes are  used to  remove
gaseous species and  to  pass  aerosols (Stevens et al. 1978).   This  technique
was proposed  for  use with  nitrate species by Shaw  et  al.  (1979) and   demon-
strated by  Appel et al.   (1981a) and by Shaw  et al. (1982).   Ambient air
measurements using this approach  are of  particular importance (Appel  et al.
1981a, Forrest et al.  1982, Shaw  et al.  1982, Spicer et  al. 1982a,   Tanner
1982).

5.3.6.1  Urban  Concentration  Measurements—As discussed above,  much   higher
ambient air nitrate concentrations have been measured on glass  fiber filters
than on Teflon and  other  inert  filters.    The  magnitude of  the actual net
positive artifact on ambient air  samples  cannot  be estimated.   Therefore, the
substantial  body  of ambient  air nitrate  concentrations obtained  on   glass
fiber  filters will  not  be considered (NAS  1977,  U.S. EPA  1982).   The same
problem probably  applies  to the  measurements  on cellulose filters  used  to


                                     5-53

-------
collect samples in the  Los  Angeles  Basin  during 1972 and 1973 (Appel  et al.
1978).   Appel  et al.  (1981a),  using  Gelman  A glass  fiber filters  in low
volume sampling over 2  to  8 hour periods, obtained reasonable agreement for
many of the samples between the nitrate values  on  glass  fiber  filters and a
total  inorganic  nitrate  (nitrate  particulate  plus  nitric acid)  sampling
system.  However, Shaw et al. (1982) did  not  observe  glass  fiber filters to
collect nitric acid  with  reproducible  efficiency  at  the subambient pressure
in their sampling assembly.  While Appel  et al. (1981a) concluded that glass
fiber filters give an approximation  of total  inorganic nitrate,  Shaw et al.
(1982) did not consider glass fiber filters to be  satisfactory collectors of
total inorganic nitrate.   Neither group  used  the  24-hr high volume sampling
procedure.  While it is clear  that  24-hr average  HIVOL  samples  are totally
inadequate for measurement  of  particulate nitrate, it  is  not  clear to what
extent such  sampling  might have  provided an adequate  measurement  of total
inorganic nitrate.

Because of  the large losses of  nitrate  off  Teflon and  quartz filters, the
ambient air measurements made with these  filters are also in question  (Spicer
1977,  Spicer  and Schumacher 1977, Appel  et  al. 1979,  Spicer  et al.  1979).
Although the measurements can be  considered lower limit estimates, the losses
of nitrate are so large as  to make such estimates of  little  value.

Nitrate  measurements also are  available  from  particle-size  distribution
studies made  using cascade  impactors  (Lee and Patterson 1969, Lundren  1970,
Moskowitz  1977,  Patterson  and Wagman  1977,  Appel  et al.  1978).   However,
these  cascade impactors and  the backup  filters  used with  them  have the
potential for similar types of  artifact problems discussed above.  Therefore,
it is  not  possible  to  know whether  such  nitrate  measurements are  of  value
either.

The  remaining  nitrate  measurements  are those made recently  using  gas dif-
fusion  denuders  to  remove nitric  acid.    Appel   et  al.  (1981a)   collected
inorganic  nitrate  on  a Teflon  prefilter followed  by a nylon  or NaCl/W41
backup filter.  Particulate nitrate  was collected with  the same tandem filter
system  after removing  the nitric  acid   with the  diffusion denuder.   This
arrangement  allows  nitric  acid  to  be determined  by difference.    Diurnal
nitrate concentration profiles  obtained with  this  system were plotted  for the
period  between 23  July and  27   July  1979  at Claremont,  CA  (Harvey  Mudd
College).   The particulate nitrate peaked in concentration during the late
morning  hours.   Particle  nitrate  concentrations  exceeded  nitric  acid con-
centrations  between  2200  and  1200  hours.    The  average  particle  nitrate
concentration  during this  period  was 25 pg  m-3.    The  average  particle
nitrate   concentration  moderately   exceeded  the   average   nitric  acid
concentration.

Forrest  et  al.  (1982), as  part  of  an intercomparison  study  (Spicer  et  al.
1982a) at  Harvey Mudd  College in Claremont,  CA,  measured  nitrates by  using
the  gas diffusion denuder technique.   Two assemblies,  each  with  a  Fluoropore
prefilter  followed by two  pairs of  NaCl impregnated  filters, were  used, with
one  assembly  at  the  exit of a diffusion  denuder.   Measurements of nitrates
were made with  this system between  27   August  and  3  September 1979.    The
particulate nitrate concentrations tended to  peak  in the morning hours.   The


                                     5-54

-------
particulate nitrate concentrations exceeded the  nitric acid  concentrations in
the evening and morning hours.   This diurnal  pattern was the  same as observed
at  this  site earlier  in  the summer by  Appel  et al.  (1981a).   The average
particulate  nitrate  concentration  was  13.4 yg  m-3.    This  concentration
moderately  exceeded  the average  nitric  acid concentration.   Lower nitrate
concentrations were  obtained  in August and  September than  were  measured in
July (Appel et al. 1981a).   The peak ozone concentrations also were somewhat
lower during  this  period  (Spicer et al.  1982b)  than in the  period  in July
(Appel et al. 1981a).   The  results  indicate  that the  later  period was  one of
lesser photochemical  activity.

5.3.6.2  Nonurban Concentration Measurements—Discussion earlier in this sec-
tion notes  that  the  nitrate concentrations  obtained  at  nonurban sites using
glass fiber filter HIVOL sampling are considered too unreliable to use.  The
Teflon  impregnated  HIVOL  filters employed  by  Mueller et  al.  (1980) have
similar problems associated  with  them  (Pierson  et  al. 1980b).   Even  with a
positive artifact associated with their nitrate measurements, Mueller  et al.
(1980) usually measured less than 1 yg  irr3  of nitrate at  rural  sites,  and
during the  spring and  summer months the nitrate concentrations reported were
at  or  below  0.5 yg  m-3.    Pierson et al.   (1980b)  sampled  with  Fluoropore
Teflon and  quartz filters  at  Allegheny  Mountain;  on  Fluoropore  filters an
average  nitrate  concentration  obtained  was  0.5  yg  m-39  but the  negative
artifacts likely to occur with  these filters  also may make these measurements
unreliable.

Shaw et al.  (1982) made measurements  of  nitrates,  using  a diffusion denuder
at  a  site  in Research  Triangle Park,  NC during 16 days  in June, July, and
August 1980.  The assembly  used contained a  cyclone  to  remove coarse  parti-
cles.  The  cyclones were shown to  pass nitric acid efficiently.  The cyclone
was followed  by  a  manifold  to  which were connected tandem  Teflon and Nylon
filter  holders,  one  of which  had  a  diffusion denuder  between  it  and  the
manifold.   The  particulate nitrate  concentrations  measured exceeded  the
nitric acid  concentrations  in  the late evening and  early morning hours, as
was observed  at  Claremont,  CA  (Appel  et al.  1981a,  Forrest et  al.   1982).
During the  late morning, afternoon, and early evening hours, the particulate
nitrate  concentrations  were  substantially   lower   than   the  nitric  acid
concentrations.   Averaging  the  entire  study  period,  the particulate nitrate
concentration was  1.0  yg  nr3  and  the  particulate   nitrate  was  37  percent
of  the  total inorganic nitrate.    The  average particulate  nitrate  concen-
tration at  this  nonurban  site was  4  percent  (Appel  et  al.  1981a)  and  7
percent  (Forrest   et  al.   1982)   of  the  average   particulate   nitrate
concentrations measured in Claremont, CA.

Tanner (1982) used the same diffusion denuder assembly arrangement as Forrest
et al. (1982) at  a site within  Brookhaven  National Laboratory on Long Island,
NY.  Measurements of  nitrates were made several  hours each day on 7, 8, and 9
November 1979.   The  average particulate nitrate  concentration  was  1.7  yg
m--3  and  constituted   about   one-third   of   the  total   inorganic   nitrate
measured.  As at  the  Research Triangle  Park,  NC  site, the particulate nitrate
concentration at  this  site  was only a  small  fraction  of  the  particulate
nitrate concentrations measured at Claremont, CA  (Appel et al. 1981a, Forrest
et al.  1982).
                                     5-55

-------
5.3.6.3   Concentration  Measurements  at  Remote Locations--Huebert (1980) and
Huebert and Lazrus (1978, 1980b) used a tamden filter assembly consisting of
a Teflon prefilter followed by  a base-impregnated  cellulose filter to collect
nitrates.   As  already  discussed, these  filters   have  positive  and  negative
artifacts.  In  combination  such types  of filters  are adequate for measuring
total inorganic nitrate but are questionable for the accurate measurement of
particulate nitrate and nitric acid individually  (Appel et al. 1981a, Spicer
and  Sverdrup 1981, Forrest et al. 1982).   Teflon  filters  alone  were  used to
collect  particulate  nitrate  at  remote  locations  (Huebert  and Lazrus 1980a),
but  these filters have  the negative  artifact  problems  already discussed.
Based on  such  measurements at   remote locations,  the  authors concluded that
particulate nitrate concentrations exceed  nitric  acid  concentrations  in the
marine boundary layer  (Huebert  1980), but  particulate  nitrate concentrations
are  much  lower than  nitric  acid  concentrations  in  the   free  troposhere
(Huebert and Lazrus 1978, 1980b).

5.3.7  Particle Size  Characteristics  of  Particulate  Nitrogen  Compounds

The  available  literature  on measurement  of particle size characteristics of
particulate nitrogen  compounds is based on  studies  done between  1966 and
1976.  Therefore,  the investigators could not  have been aware of the positive
and  particularly  the  negative  artifact  problems with  particulate  nitrate
sampling discussed earlier in this section.

The  last stage of  the cascade impactors  used consists of cellulose acetate or
glass  fiber  filters.   Because  of  losses  of  nitric  acid   on  such  filters
substantial overestimates  of the amount of  nitrate  on the  last  stage are
likely.  This would  result in the mass  median diameters  computed  being too
small.  However, losses of nitric acid and particulate  may occur  on the  upper
stages of the  impactors.   The  Lundgren  impactor  has substantial  wall losses
(Lundgren  1967, 1970).   The  impactor   stages  usually were  constructed of
stainless steel.  Shaw et al. (1982)  found at  least  88  percent of nitric acid
in air  passed  through  a stainless steel  cyclone.   This may  be an indication
that  nitric acid  is  unlikely to be lost  to other stainless  steel  surfaces,
but  no studies have been made.

The  situation   is  complicated  by the  use  of films  and coatings  over the
original  stainless  steel  surfaces.   Appel  et al.  (1978)  used  polyethylene
strips coated with a sticky  hydrocarbon  resin, while Moskowitz (1977) used a
thin  film of vaseline  on  stainless steel  strips.   No measurements  have been
made  on  losses of nitric acid or of nitrogen dioxide  to such surfaces.  If
losses did occur on the  upper  stages of  the  impactors  only,  the mass median
diameters  computed  would be too large.   It  is  impossible  to  estimate the
extent to  which artifact  problems may  shift the apparent size distributions
in these  impactors.  Nevertheless, some  qualitative  results of these  impactor
studies appear  reasonable, and  these  will be discussed.

The  larger mass  median diameters given  in Table  5-10 were  computed  from
measurements  at locations  near the ocean likely  to  be influenced  by air
masses moving off the  ocean.   As  can be  seen  from the mass  median diameters
of  particulate  nitrate from the  work of Appel  et al.  (1978), the diameters
tended to decrease from  sites  near the  ocean,  Dominguez Hills,  CA, to  those
                                     5-56

-------
TABLE 5-10.  MASS MEDIAN DIAMETERS REPORTED FOR NITRATE FROM PARTICLE
                     SIZING WITH CASCADE  IMPACTORS
Site
Cincinnati, OH
(CAMP Site)
Fairfax, OH
Riverside, CA
U. Cal . Campus
Secaucus, NJ
Dominquez Hills,
CA
West Covina, CA
Pomona, CA
Rubidoux, CA
Measurement
period
3/14-23/66
3/25-4/21/66
11/1-15/68
9/29-10/10/66
Background
Level A
Level B
Level C
10/4-5/73
0/10-11/73
7/23-24/73
7/26/73
8/16-17/73
9/5-6/73
9/18-19/73
Mass median
diameter in ym
Reference for nitrate
Lee and Patterson (1969)
Lee and Patterson (1969)
Lundgren (1970)
Patterson and Wagman
(1977)
Appel et al. (1978)
Appel et al. (1978)
Appel et al. (1978)
Appel et al. (1978)
0.23 (est)
0.59
0.8
0.20
2.6
0.38
0.37
1.64
0.72
1.13
0.62
0.68
0.33
0.34
                                 5-57

-------
well inland, Rubidoux, CA.   At Dominguez,  CA and to a lesser extent at West
Covina, CA  farther  inland,  a substantial coarse  mode  fraction of particles
greater than 2  ym were measured.

Moskowitz (1977)  observed the  same  sort of pattern of  particle  size dis-
tributions of particulate nitrate in  the South Coast Air  Basin.  The particle
size  distribution  of  nitrate  indicated two modes.   One mode  was located
between 0.05  and 1  ym,  while  the other mode was  between  2  and 8  ym  (8
urn was an arbitrary  upper cutoff).  At Hermosa Beach, CA, on the  coast, the
concentration of submicron nitrate was small with most of the nitrate  in the
2  to  8 ym  range.    At  Pasadena,  CA,  the  size  distribution  of particulate
nitrate was  bimodal  with significant amounts of nitrate in both size ranges.
At Chi no, CA, well  inland, a  large part of the particulate nitrate  was  in the
submicron range.  Coarse  mode nitrate was still present.   Chi no is a cattle-
feeding area with high ammonia concentrations available to react with  nitric
acid to form submicron ammonium nitrate.

Several studies  provide  results  bearing on  the  chemical  composition  of the
nitrates  in the  fine and coarse modes.   Grosjean and  Friedlander   (1975)
claimed  that ammonium  nitrate  accounted   for  95  percent  of  the measured
nitrate,  based  on  infrared  spectra  of  extracts  from samples  collected  on
water-washed Gelman type A glass  fiber filters in Pasadena,  CA during 1973.
O'Brien et al.  (1975) usually observed  the presence  of ammonium nitrate based
on infrared spectra and paper chromatograms  of samples collected on prewashed
Gelman type A  glass  fiber filters at several locations  in  California.   At
Santa Barbara,  CA,  a sample collected within a mile  of the ocean contained 16
percent nitrate, but no  ammonium ion  was  detected.   The authors suggested
that  the  nitrate was  sodium nitrate  formed  from the reaction  of nitrogen
dioxide with sodium  chloride.   Lundgren (1970),  in the samples collected at
Riverside, CA,  identified by  x-ray diffraction very  hygroscopic, crystalline-
like  particles  making up a  large part of  the 0.5  to 1.5 ym  size range as
ammonium nitrate.

High-resolution  mass  spectrometric  measurements  were   applied to  samples
collected during a  smog episode  at  West   Covina,  CA  (Cronn  et  al.   1977).
Ammonium  nitrate and  sodium  nitrate  were identified as  present in the size
range  below 3.5 ym.    The ammonium  nitrate  concentration   substantially
exceeded the sodium nitrate concentrations measured.

Kadowaki  (1977)  size-classified  particle nitrate using  an  Andersen sampler
with  a  type A  Gelman glass  fiber  backup filter  in  Nogoya,  Japan.   The size
distribution of  nitrate  was  bimodal.   The  submicron nitrate  was shown to be
ammonium nitrate and the coarse particles sodium  nitrate  based  on analysis by
paper chromatography.   Increases in coarse mode  nitrate were observed when
sea salt aerosols were transported to the sampling location.

5.4  OZONE

Ambient  air concentrations  of ozone  are of  interest with  regard to  acidic
deposition  for  several reasons.   Ozone can  contribute to adverse  effects on
field  crops,  forest trees,   and  other  forms of vegetation  (Chapter E-3,
Section 3.3.1).  Ozone in combination with  sulfur dioxide  can cause damage to


                                     5-58

-------
vegetation.   Ozone  also may interact with acidic  deposition  to cause damage
to vegetation.  However, the results of the several studies completed to date
are  preliminary and  inconclusive.   Transformations of sulfur  dioxide to sul-
fate in  aqueous droplets in  clouds, fogs, and acid mists  may  be contributed
to  significantly  by reactions  with ozone.  Therefore,  ozone  concentrations
both at  ground level and aloft, cloud heights, are of interest.

This presentation will  not include a discussion of  ozone  concentration mea-
surements within cities.   The  literature  on  ozone  measurements within cities
is  too  extensive  to consider  in  detail  here.   A  discussion  of  ambient air
ozone  concentration  levels  within cities can  be  found  in the  Air Quality
Criteria for Ozone (U.S. EPA 1978a).

Most of  the ozone measurements  made from  the early 1970's  to  the present at
ground  level  and from  aircraft have used chemiluminescent ozone analyzers.
Investigators using  these  instruments at  rural sites  and in aircraft believe
the method to be reliable,  specific, and precise  (Research Triangle Institute
1975, Decker et al. 1976).

Ozone is formed in the  atmosphere  from the reaction  of oxygen  molecules with
atomic oxygen.  The  atomic oxygen  is formed from  the photolysis of nitrogen
dioxide.  Ozone reacts  very  rapidly with  nitric  oxide.   Maintaining the pro-
duction  of ozone in  the atmosphere requires the presence  of  radical species
produced  from the  reactions of  nitrogen oxides  in  sunlight with  organic
vapors  (U.S.  EPA 1978a).    Peroxyacetyl  nitrates  and  nitric  acid  also are
formed in the atmosphere by  the reaction  of radical  species formed in  these
reactions with  nitrogen dioxide.   Hydroxyl  radicals, OH, are  particularly
important in their reactions with organic  vapors  to form other radicals, with
nitrogen dioxide to  form  nitric  acid,  and with  sulfur dioxide  to  form sul-
fates.   Therefore,  homogeneous  photochemical  reactions are important to the
formation of a number of the chemical  species discussed in this document.

Ozone is formed  in  the  stratosphere  and  can be transported into the tropo-
sphere  by tropospheric extrusion  events.    Aircraft  measurements  provide
evidence for the transport of ozone from stratospheric extrusions to within a
few kilometers of the surface  (Viezee  and Singh  1982).  Direct  evidence for
transport from the stratosphere,  free troposphere,  and through the planetary
boundary layer to rural  locations near sea level  is lacking (Viezee and  Singh
1982).  The air packets from the stratosphere have  been observed to level out
horizontally at a few kilometers above the surface.   Ozone previously trans-
ported to these  altitudes  eventually will be  transported  to  the surface by
vertical movements,  depending  on  the lifetime of  ozone under these circum-
stances.   A  number  of  reports  in the literature note stratospheric  ozone
contributing to ozone concentration levels at or  near the surface (Viezee and
Singh 1982).   If stratospheric ozone extrusions are an important  source of
ozone at rural locations,  a  spring  maximum and a fall minimum in  ozone con-
centrations would be expected.

Another  source of  ozone at  the surface could be  the reactions  of biogenic
hydrocarbons.   Because  background nitrogen oxide  concentrations are so low
(Section 5.3.2.5), biogenic  hydrocarbons,  if present  at  significant ambient
air concentrations,  would  have  to mix with anthropogenic  nitrogen  oxides to


                                     5-59

-------
react.  However, the  ambient  air concentrations of biogenic hydrocarbons in
urban and rural  locations outside of forest canopies are too low to generate
significant concentrations of  ozone  (Altshuller  1983).

Ozone formed  in homogeneous photochemical reactions  in  the atmosphere  from
anthropogenic precursors can be  present  at elevated concentration  levels at
rural locations as a  result of one  or more of the  following processes:   (1)
local synthesis, (2)  fumigation by a specific urban or industrial plume,  (3)
a high  pressure system near the  rural  location.   Ozone concentrations  gen-
erated  from  these  processes  are higher  in  the  warmer  than  in  the  cooler
months of the year.   If  homogeneous photochemical  reactions of anthropogenic
precursors are  the more  significant source,  the higher ozone  concentrations
would be expected to  occur in  the late  spring, summer months, and early fall.

5.4.1  Concentration  Measurements Within  the  Planetary Boundary Layer
       (PBL)

Average ozone concentrations in rural areas  have  been  reported as  low as 20
to  40  yg m-3,  at  night  and  during  the early morning  hours  (Martinez  and
Singh 1979, Research  Triangle  Institute  1975,  Decker et  al.  1976,  Evans et
al. 1982).  Maximum ozone concentrations  often are found  downwind of the  core
areas of large  cities.   Maximum  annual  one-hour ozone concentrations  in  the
ranges  of 800 to 1300 yg  m~3  have  been  observed  during  most years between
1964 and 1978 at several  locations in  the South  Coast  Air  Basin (Trijonis and
Mortimer 1982,  Hoggan  et al.  1982).  Well out  into the  eastern  part  of the
South Coast Air Basin at San Bernardino  and  Redlands maximum annual one-hour
ozone  concentrations  of  600   to 800  yg m-3  have been  measured  (Trijonis
and Mortimer 1982,  Hoggan et al.  1982).

A number of studies on urban plumes  of large  cities in the  United States  have
been reported.  The effects of these plumes  on  elevated  ozone  concentrations
have been shown to extend out to distances  as  far as several  hundred kilo-
meters  downwind.   Measurements have been  made  on the flow  of the  New  York
metropolitan  area  plume  into   southern  New England (Cleveland et  al. 1976,
1977, Siple  et al.  1977,  Spicer et al.  1979),  the Boston  plume  into  the
Atlantic  Ocean  (Spicer   et   al.   1982c),  the   Philadelphia-Camden  plume
(Cleveland and  Kleiner  1975),  the Chicago metropolitan area plume (Swinford
1980, Sexton  and Westberg  1980),  the St. Louis  plume  (White et  al . 1976,
1977;  Hester et  al.  1977, Spicer  et  al.  1982b),  and  the  Houston plume
(Westberg et al. 1978a,b).

The  concentrations  of ozone  measured within  these  urban plumes typically
ranged  up  to between  300  to  500 yg  m~3.   in  the case of  a  city the  size
of  St.  Louis,  MO, an urban plume  30  to  50  km wide was observed downwind
(White  et  al.  1977).   The ozone concentrations  within  the St.  Louis plume
were  about  twice  the concentrations  in  the  background  in  adjacent rural
areas.  A  definable  plume containing excess ozone  concentrations over rural
background also has been demonstrated to occur  shorter distances downwind of
small cities such as Springfield, IL (Spicer  et  al. 1982b).

Impacts of  urban  plumes  from large  or medium-sized cities  within  several
hundred  kilometers  on  elevated  ozone  concentration   levels  at   specific


                                     5-60

-------
 nonurban  sites have  been  reported.    Examples  of such  observations  include
 those made at  Research Triangle Park,  NC,  Duncan  Falls,  OH,  and Giles Co, TN
 (Martinez  and Singh  1979);  at Kisatchie  National Park,  LA and  Mark  Twain
 National  Park, MO  (Evans  et  al.  1982);  and  at a  rural  site  outside  of
 Glasgow,  IL  (Rasmussen et  al.  1977).   The  peak ozone  concentrations reported
 during  such episodes  at  these  nonurban  sites  ranged  from 140  to  260  yg
 m~3.

 Davis  et  al.  (1974)  reported measurement  of  excess  ozone  concentrations
 within power plant plumes.  Measurements of  ozone  in  four  power plant plumes
 in  the States  of  Washington, New  Mexico, and Texas by Hegg et al.  (1977)  did
 not  show  any excess  of ozone  in the  plumes over that  in surrounding  air  out
 to distance of 90 km.  Other measurements of power plant plumes in  the States
 of  New Mexico and  Texas by Tesche  et al.  (1977)  revealed ozone  depletion
 within the plumes in  the  vicinity of the stack  and  a gradual increase  in
 ozone  concentrations to  background  levels  far  downwind.    Gillani  et  al.
 (1978) observed a significant  ozone  excess in the Labadie  power plant plume
 190  km  and 9  hours  downwind during 9 July 1976.   The ozone  concentration
 within the plume  at this  distance  downwind was  220 yg  m-3,  about 100  yg
 m~3  above  the  rural  background.   Before 5  hours  downwind an  ozone  deficit
 was  observed.   During another day in  July  1976  a transition  from an  ozone
 deficit to an ozone excess was observed after only 2  hours.  On both days  the
 first  indication  of  ozone  production  was observed around  1400  hours.   There
 appears to be less likelihood of observing  excess ozone in  power plant plumes
 in  the western than  in the eastern United States.  This result may be  asso-
 ciated with the availability of more hydrocarbon  in rural  air  in the  eastern
 United States  to  diffuse  in and react with  excess   nitrogen  oxide  in  the
 plume.   Observations of the direct  effect of power  plant plumes   on  ground
 level ozone concentrations at rural locations are lacking.

 Several  studies  have been made of the effects  of high pressure  systems  on
 ozone concentrations over the midwestern and eastern  United  States  (Research
 Triangle  Institute 1975, Decker et al. 1976, Husar et al.  1977, Vukovich et
 al.  1977,  Wolff  et  al.  1977).    The  distribution of ozone  concentrations
 relative to  a  moving high  pressure system have been  represented for  several
 rural locations in Pennsylvania,  at Creston in southwestern Iowa, and  at Wolf
 Point in northeastern Montana (Decker  et al. 1976, Vukovich  et al.  1977).   A
 relative minimum in the maximum diurnal ozone concentration  occurs  somewhere
 in  the  region  between  the  initial  frontal  passage  and  the   high pressure
 center.   The  highest  ozone concentrations  diurnally occur  after the high
 pressure center  passes  the site  or  on  the  back  side  of  the  high pressure
 system.  The exception was at Wolf Point,  MT, where no substantial  variation
 in  the  ozone  concentrations  was  seen as  the  high  pressure   system  passed
 through that location.   Meteorological analysis  indicated  no reason  why  the
 average  downward  transport by  general  subsidence or  by  enhanced vertical
mixing should  increase the  ozone  concentration in the  backside of the high
 pressure  system.   The  aircraft  measurements  showed  no  indication  on  the
 average  that  the  vertical gradient   of  ozone  through  the  troposphere   is
 greater in  the eastern than  in  the western United States.   Therefore,  the
 elevated  ozone concentrations  measured  from  Iowa  eastward  could  not   be
 attributed to  downward  transport  of ozone.   It  was  concluded  that the most
                                     5-61

-------
appropriate  explanation  was   the   availability  of  sufficient  amounts  of
precursors reacting  to  form ozone  within  the high  pressure systems.   The
backside of  the  high pressure  systems is  the  region  where air parcels have
the highest residence times for precursors  to  react to  form ozone.

The peak ozone concentrations during the movement of the high pressure system
were  between 200  and  500  yg  m-3  at the Pennsylvania  sites,  150  yg  m-3
at  Creston,  IA  and  less than  100  yg nr3 at Wolf  Point,  MT.    Such  high
pressure systems were influencing the  sites much  of  the time in the July to
September period.  For  example,  at  one or another of  the  rural  sites where
measurements were being made in 1973, 1974, and 1975,  a high pressure center
or  ridge was  within  450 miles  of the site between 80  and  90 percent of the
time (Decker et al. 1976, Vukovich et al. 1977).

A study of factors responsible for  higher  ozone concentrations also was made
over the Gulf Coast area (Decker et  al. 1976).  Elevated ozone concentrations
of  160  yg m"3  or  more  were frequently measured  in  plumes  downwind  of
cities, major refineries, and  petrochemical installations.  Ozone concentra-
tions over the Gulf  of  Mexico  usually were lower than over land except when
the air parcels had previously  passed over  continental  sources of pollution.

Diurnal profiles of ozone concentrations averaged over  study  periods or quar-
ter of year  are  available from several  studies (Research Triangle  Institute
1975,  Decker  et al.  1976,  Vukovich et al.  1977,  Martinez  and  Singh 1979,
Evans  et al  1982)  at the rural  sites discussed and  additional  sites.   The
average  profiles are very similar,  with ozone concentrations  rising in the
morning  hours, peaking  in the  afternoon,  and   falling  after establishment of
the noctural inversion in the evening hours through the night to 0600  or 0700
hours.   From  a 1974  study made between  June 14  and August  31   (Research
Triangle  Institute  1975), the  average 0900 to 1600  ozone concentrations of
interest  in  crop  yield  studies  can  be   computed  for  the  rural   sites  as
follows:   Wilmington,  OH.  125 yg nr3;   McConnelsville,  OH,  117  yg  nr3;
Wooster,  OH,  119  yg  m-3,  McHenry, MD, 116  yg  nr3;  DuBois,  PA,  132
yg m~3.

In  some  of  the  studies  discussed above,  either sulfate measurements  or vis-
ibility measurements as  a  surrogate  for fine  particles are available  (Decker
et  al.  1976, Husar  et  al  1977).    The  sulfate  concentrations  (in yg nr3)
and the  sulfate  as a percentage of  total   suspended particulate from  west to
east  were  as follows:   Wolf  Point, MT, 1.8,  6.2;   Creston,  IA,  7.2, 9.2;
Bradford,  PA, 9.9,  29.0.   These  measurements  show  the  same directional
characteristics  from west to east as do  the  ozone concentrations.   Husar et
al.  (1977)  analyzed an  episode during  late June  1976, finding  that the
geographical  location  of high  ozone  concentrations  roughly corresponded to
areas  of low visibility  and   high  sulfate concentrations.   The air  quality
measurements  at  St. Louis  during  June through  August of  1975  showed that
ozone   concentrations   above   160   ug  m-3  roughly   coincided  with  light
extinction  coefficients above  5.   Therefore,  a  similar behavior occurs  for
ozone  and  for light  scattering aerosols such  as sulfate.
                                     5-62

-------
 5.4.2   Concentration Measurements at Higher Altitudes

 Ozone  measurements at  several  higher  altitude  mountainous sites  have  been
 compiled  by  Singh  et al.  (1978).   Hourly ozone concentrations  are as high as
 140  to 160  pg m-3  during the  spring  months,  and  as  low as 40  to 60  pg
 m-3  during the fall  months.    While  the seasonal patterns tend  to  be  con-
 sistent,  the absolute concentrations differ  from year to  year.   Relatively
 high  summer  ozone  concentrations  have been observed at  some  sites  (Singh et
 al.  1978).  Viezee and Singh (1982)  have assembled  results from  recent air-
 craft  observations.   Observations  between  the altitudes  of  1.5  and  4.5  km
 indicate  ozone concentrations  during May  in the 110 to  150  yg m-3  range
 and  during October in  the 70  to 90 pg  nr3  range.    A summary  of  aircraft
 observations  of  ozone  concentrations  during  stratospheric  air  extrusions
 results in a power curve  from which the  ozone concentration obtained is 140
 pg  m-3  at  3  km,  210  pg  m-3  at  5 km and  330 pg  m-3  at  7  km.    Based
 on these  aircraft  measurements  compared  to  the elevated  ozone  concentrations
 attributed to stratospheric ozone at sites between sea level and 3 km, Viezee
 and Singh (1982)  believe  that  reports  of ozone concentrations above  200  yg
 m~3  near  the  surface  attributed to stratospheric  air  extrusions  are  un-
 likely  and should  be reexamined.

 5.5  HYDROGEN PEROXIDE

 The oxidation of sulfur dioxide in  aqueous  droplets  by hydrogen peroxide may
 be the  most  important of the mechanisms  for  conversion  of sulfur dioxide to
 sulfuric  acid  (Chapter  A-4).    Therefore, the measurements of  hydrogen  per-
 oxide concentrations are of considerable interest.

 Several chemical methods  for  measuring  hydrogen peroxide  in ambient  air and
 in rainwater are in use.  Both  the  reaction of titanium  sulfate and  8-quino-
 linol  with hydrogen peroxide (Cohen  and Purcell  1967)  and the  reaction  of
 titanium  (IV)  tetrachloride with  hydrogen  peroxide  (Pilz  and Johann  1974)
 have been  used in  colorimetric  procedures for measuring  hydrogen  peroxide  in
 air.  The  chemiluminescent  oxidation  of luminol  by hydrogen peroxide  in the
 presence  of Cud I) catalyst is  the  basis  of a sensitive  automated system for
 continuous monitoring  of hydrogen  peroxide  in  the  atmosphere (Kok et  al.
 1978b).   Addition  of  a  known  amount of  scopoletin to  a  buffered  sample  con-
 taining hydrogen peroxide  followed  by addition of horseradish  peroxidase  to
 catalyze  the oxidation  by  scopoletin  results  in fluorescence decay  (Zika  et
 al. 1982).   The  amount  of hydrogen peroxide  is  determined by  difference  in
 the fluorescence before  and after  addition of  the horseradish peroxidase.

 The long-path Fourier transfer  infrared technique has not  proved applicable
 to measuring  hydrogen  peroxide  because  of  its  high  detectability  limit  of
 about 56 ug m-3 (Tuazon  et al. 1981a).

 Recent  studies  (Heikes  et al.  1982,  Zika  and  Saltzman  1982)   indicate  that
 hydrogen  peroxide  can  be produced  from other species within  aqueous  solu-
 tions.   These  results  suggest that methods  involving  collection  in  aqueous
 solutions  may not  provide  useful  measurements of  ambient air  hydrogen  per-
 oxide concentrations.   Both  groups found hydrogen peroxide to be generated
within   the  aqueous  collecting  solutions  when  ozone   in oxygen-nitrogen


                                    5-63

-------
mixtures  is  passed  through  aqueous  solutions  in  bubblers  or  impingers.
Heikes  et al.  (1982)  also  observed that  sulfur dioxide  vapor  acts  as  a
negative interferent by depleting hydrogen peroxide in  its aqueous  collection
or formation.

5.5.1  Urban Concentration Measurements

Ambient  concentrations  of hydrogen  peroxide up  to  56  yg nr3  in  Hoboken,
NJ  and  251  yg  m-3 in  Riverside,  CA were  measured  in  1970  by  Bufalini et
al. (1972) using Cohen and Purcell's (1967)  method.  Subsequent  measurements
of hydrogen peroxide in 1977  at sites in Claremont, CA  and Riverside,  CA  gave
hydrogen  peroxide  concentrations  typically ranging from 14  to  70  yg  m"3
with  a  maximum concentration  near 140  yg  m-3  (Kok  et  al.  1978a).   Three
chemical methods (Cohen and  Purcell  1967, Pilz and Johann 1974, Kok et al.
1978b) were  used  in intercomparisons.   The  hydrogen peroxide concentrations
measured by  the three methods  differed by  as  much  as  a factor  of  two to
three.   Substantial  ozone  concentrations  were  present  in  the  atmosphere
during most of the time hydrogen peroxide was being measured.

Subsequent measurements of hydrogen  peroxide were made  in 1979  and 1980 in
the Los  Angeles Basin  area at sites within  Los Angeles,  Claremont, and  Palo
Verde, CA  {Kok  1982).   In Los  Angeles  at  California  State  University,  the
hydrogen peroxide concentrations on 18 and 19 June 1980  ranged between about
0.7 and  3.5  yg m-3.   The  hydrogen peroxide  concentrations were  1 to  2
percent  of  the maximum  ozone concentrations.   At Claremont,  CA, hydrogen
peroxide measurements  were  reported  during  a  number  of days  in  June to
September 1979  and in September 1980.   In  June  and July 1979   the hydrogen
peroxide concentrations were  much higher than  reported  in August 1979  and
September 1979  and 1980.   Peak concentrations  exceeded 14  yg  nr3  in June
and July, while in August and September  the  hydrogen peroxide concentrations
were  only  a  few  ppb.   At Point  San  Vincente,  located in   the Palo Verde
peninsula, on 11 and  12  September 1980  the  hydrogen peroxide concentrations
peaked at  8  to  11 yg  m-3.    The maximum  hydrogen  peroxide  concentrations
compared to  the maximum ozone concentrations show no  distinct  relationship
(Kok 1982).

Heikes et  al. (1982)  obtained  about equal   amounts of hydrogen  peroxide in
each of three impingers in series sampling ambient air over a series of  days
in February  and March  1981  at  Boulder, CO.   If the  ambient  air hydrogen
peroxide was  collected  efficiently in  the  first  impinger, the  ambient  air
hydrogen peroxide  concentrations ranged  from  0.4 to  3.1 yg m-3.   Approx-
imately equivalent  amounts of hydrogen  peroxide  measured  in  the second  and
third  impingers  indicate  substantial  amounts  of  hydrogen  peroxide  were
generated in solution.

5.5.2   Nonurban Concentration Measurements

Measurements of hydrogen peroxide concentrations were obtained by the luminol
chemiluminescence  technique at a  rural  site east of Boulder,  CO in February
1978 (Kelly and Stedman 1979a).  The hydrogen peroxide concentrations ranged
from 0.4 to 4 yg m-3 during this period.
                                     5-64

-------
Hydrogen peroxide was  measured  in water condensate  by  the luminol chemilu-
minescence technique at rural  sites near Tucson,  AZ  (Farmer and Dawson 1982).
In more remote areas around Tucson the  hydrogen  peroxide concentrations were
about  1.4  yg  m-3,  while  at a  Thurber  Ranch  site  the  hydrogen peroxide
ranged  up  to 6  yg  m-3.   The hydrogen  peroxide  concentration  was observed
to drop off drastically when  high sulfur dioxide  concentrations  were mea-
sured.  With a correction for the interference  by sulfur dioxide, the authors
estimated that the hydrogen peroxide reached  10 yg nr3.

5.5.3  Concentration Measurements in Rainwater

Because the key interest in hydrogen peroxide is  with  respect  to  its behavior
in solution, available measurements of hydrogen peroxide in rainwater will be
discussed.

Hydrogen peroxide  in rainwater  collected  in Claremont, CA  during 1978 and
1979  was  analyzed  by  luminol  chemiluminescence  (Kok 1980).    The hydrogen
peroxide content  of the rainwater over  long sampling intervals dropped off
substantially  during  precipitation events.   The highest  hydrogen peroxide
concentration  obtained  was   1590  yg  jr1,  but  hydrogen  peroxide  concen-
trations  also  frequently  were  below  100   yg  £-1.   The lower  concentra*
tions  could  be accounted  for  by  the  absorption of  less  than  0.14  yg m"-5
of hydrogen peroxide from ambient air into  the  cloud water.

Measurements of  hydrogen peroxide  in  rainwater also were made in  Claremont,
CA during  1980  and  1981 (Kok 1982).   Hydrogen peroxide concentrations were
found  to  be extremely variable  in  rainwater samples  during  the course of  a
storm.  The results were interpreted as  suggesting  that hydrogen peroxide is
incorporated into the  rain at cloud  levels.  Most  of the hydrogen peroxide
concentrations in the rainwater samples were  at or below 500 yg  £-1.

Hydrogen peroxide was  measured  in rainwater samples  collected  in Miami, FL
and  the Bahama  Islands  (Zika et al. 1982).   The concentration of hydrogen
peroxide  in rainwater,  expressed  as  yg £-1,   ranged from  3.06  to  25.5   x
102  in Miami,  FL samples  and was  6.8  x 102  in a sample collected  in the
Bahama  Islands.   The variations of  hydrogen peroxide concentrations during
the  precipitation events  were   different  from  the  changes   in  sulfate and
nitrate concentrations.  The authors  believed that the results  for hydrogen
peroxide were  consistent with a substantial part  of the  hydrogen peroxide
being  present as a result of its being generated  within  the cloudwater  rather
than  being present as  a result of rainout  and  washout of gaseous hydrogen
peroxide.

5.6  CHLORINE COMPOUNDS

5.6.1   Introduction

Chlorine can exist in a number of gaseous and particulate forms  in  the  atmos-
sphere.  The  gases  can include hydrogen chloride, chlorine gas, and carbon-
containing  vapors  such as phosgene and halocarbons.   The particulate  forms
include sodium  chloride, usually as sea salt particles from  the bursting of
                                     5-65

-------
bubbles  at the  sea  surface (Junge 1963).   Ammonium chloride also has  been
reported  (Cronn et al. 1977).

The most  likely form for gaseous chloride is hydrogen chloride.   Chlorine gas
reacts  rapidly  with  hydrogen-containing  organic molecules  to  abstract  hydro-
gen  and  form  hydrogen  chloride  (Hanst  1981).   Phosgene  (Cl2c°)  nas  been
measured  in  the ppt  range  in the ambient  atmosphere  (Singh  et al. 1977b).
Numerous  chlorocarbons have been  measured in the ppt  to  ppb  range in  urban
atmospheres  (Singh  et al.  1982)  and in  the ppt range  at rural and  remote
sites (Singh et  al.  1977a,b).   Most chlorocarbons have long  residence  times
in the  atmosphere (Singh et al. 1981).   Their  inert  chemical  structure  tends
to limit  their rates of dry deposition and wet  scavenging  to very low values.
The  shorter-lived  chlorinated olefins   react  in  the  laboratory  to  form
chlorine-containing products such  as hydrogen chloride, phosgene, chlorinated
acetyl  chlorides,  and chlorinated peroxyacetyl nitrates  (Gay et al.  1976).
The chlorinated  acetyl chlorides  and chlorinated  peroxyacetyl nitrates  have
not been  detected in the  ambient atmosphere.

A  number  of the  same  type  of artifact  problems may  exist for particulate
chlorine  measurements as for particulate nitrate measurements because of  the
volatility  of  hydrogen  chloride.   However,  such   studies  of  sampling of
chlorides on filters are  not available.

5.6.2  Hydrogen Chloride

Junge  (1963)   reported  early  measurements  of gaseous  chlorine-containing
compounds  that  probably  were hydrogen chloride.  His  measurements at  three
sites gave  the following average  concentrations  in yg  m-3:   Florida—1.6,
Ipswich,  MA--4.4, and Hawaii--!.9.  Gaseous  chlorine compounds were measured
by the  same  technique  by Duce et  al.  (1965) on the island of  Hawaii.   The
concentrations of  qaseous  chlorine compounds  ranged from less  than  0.3 yg
nr3  to  218  yg nr3  although  the  gaseous  chlorine  concentrations  were at
or below  10  yg  m-3  -jn  m0st  samples.    The  halide  ion  analysis  does  not
permit identification of  the original  chemical  species  collected.

Although  hydrogen  chloride   has been measured  by  infrared techniques  in a
number  of studies  in  the  stratosphere, limited effort  has  gone  into  its
measurement in the troposphere.   Farmer  et  al. (1976)  reported  both  tropo-
spheric  and stratospheric measurements at the ground and from aircraft.   The
troposDheric mixing  ratio  at  ground  level  was  10-9  corresponding  to  1.5
yg m~3,   with  the mixing   ratio  decreasing to  10-1°  in  the upper  tropo-
sphere.    At ground level, the  tropospheric  levels  were essentially the  same
inland  in the  Mohave Desert,  CA,  as near  the coast  (Farmer et al. 1976).
Hydrogen cloride was not detected  by the  FTIR  system with  a 1 km pathlength
in measurements  at  Riverside and  Claremont, CA (Tuazon et al.  1981b).   The
established detection limit  was about 12  yg  m-3.

5.6.3  Particulate Chloride

Junge (1963) measured comparable  amounts of particulate chloride to gaseous
chlorine-containing compounds.  His  measurements  gave  the following average
concentrations   in yg  m-3;    Florida—1.5  and  Hawaii —5.    Duce  et   al.


                                     5-66

-------
(1965) measured particulate chloride on  a  four-stage cascade impactor.  The
total  chloride  concentrations ranged  from 0.5  to  137  yg  nr3.    Three of
the  nine  samples had  total  chloride  concentrations of 39,  95 and  137 yg
nr3; the remainder had concentrations below 10  yg m~3.

Particulate  chloride  concentration distribution  was measured  at  about 30
sites  in  the Houston-Galveston,  TX area  on  2 days  in  June and  2  days in
September 1975 (Laird and Miksad  1978).   The natural background of chloride
varied  from  0.2  to 6.6  yg  nr3 with  wind speed and direction.   The  higher
background concentrations corresponded  to the stronger inland penetration of
fresh maritime air from the Gulf  of Mexico.  Significant incremental concen-
trations  of  5 to  10  yg  nr3  above  background were observed,  particularly
in the industrialized  Pasadena-Houston  Ship Channel  area.

At  urban  and nonurban locations  somewhat  inland,  atmospheric chloride  con-
centrations  typically  average  1   yg m~3  and less  (Gartrell  and Friedlander
1975, Flocchini et al. 1976,  Paciga and  Jervis 1976, Crecelius et al. 1980,
Dzubay 1980).

5.6.4  Particle Size Characteristics of Particulate  Chlorine  Compounds

Junge (1963)  discussed the particle size characteristics  of chloride  parti-
cles.  The chloride particles  associated  with maritime air  are found in the 1
to  10 ym  range.   Measurements  at  a  rural coastal  site  50 miles  south of
Boston, MA  (Round  Hill),  support these  conclusions.   In contrast, chloride
particles  less  than  1  ym  were  associated with  processes  occurring  over
land.

Gladney et al. (1974)  reported measurements of chloride on cascade impactors
at several sites in the Boston, MA area.  The shapes  of the site distribution
curves for a  number of samples  indicated that the chloride present was  pre-
dominantly  marine  aerosol  and  that  there  also  was  a  strong  correlation
between sodium and  chloride  for   these samples.   The concentrations of  both
chloride  and  sodium  were usually low,  and the  size distributions flatter,
when the winds were from  inland.

The  size  distribution of chloride  particles  at  Secaucus,  NJ,   have  been
reported for varying visibility conditions (Patterson and Wagman 1977).  The
MMD  increased  from  the background  condition  of best visibility  of 0.17 ym
to  1.1 ym under the  poorest  visibility conditions  experienced.    The  size
distributions for chloride appeared to be trimodal.  Particles less than 0.5
ym  were  associated  with  lead aerosols from automobile  exhaust,  the parti-
cles  near  1 ym  with  the  contribution  from  sea  salt,  and  the  largest
particles with dredging operations.

The  particle size  distributions   of  chloride  particles  were  reported at
several sites in Toronto, Canada,  by Paciga and Jervis (1976).  The chloride
had a mass median diameter of 0.6 ym during the  summer  at this  inland site.
The sources of chlorides  were associated with lead aerosols from automobiles
and emissions from a power plant and an incinerator.  Winter samples showed a
10-fold increase  in chloride  concentration, and  an increase in  the  MMD of
                                     5-67

-------
chloride  to about  9 ym.   These  increases  were  attributed  to  salting  of
roadways.

Hardy et  al.  (1976)  reported  chloride size  distributions  at three sites  in
the  Miami,  FL  area.  Two  of  the sites were  2  km  from the seacoast and  the
third, 15 km inland.  The cascade impactor stages collecting  particles  larger
than  2  ym  contained most of  the  mass.    There was a  low concentration  of
chloride  on the  stages  collecting  particles between 0.25  and  1  ym, but  the
concentration increased  again  on the filter  used  to collect particles less
than  0.25  ym.    The  small-particle chloride  was  attributed  to chlorine
associated  with  lead aerosols emitted from gasoline-powered vehicles.   The
large particles were associated with  particles emitted  from the sea  surface.

Particle  size distributions of chloride  were measured  by  Lee and  Patterson
(1969) at sites  in  Philadelphia, PA, Cincinnati,  OH (Fairfax), and Chicago,
IL,  in the  summer and fall.  The MMD's obtained were all  near 0.85 ym.   Lee
and  Patterson  concluded that  the  chlorides  at these  sites  were  primarily
influenced  by   industrial  and   vehicular  emissions  rather   than  sea salt
aerosols.

5.7  METALLIC ELEMENTSi

The various interests and possible  concerns related to  metallic elements have
been discussed briefly in  the  introduction to this chapter.   Alkaline earth
elements  such as calcium and magnesium can help neutralize acidic materials
either during precipitation events  or as  a  result of dry deposition.   Manga-
nese and  iron are possibly of consequence  in  the chemical  transformations  of
sulfur dioxide to sulfur (Chapter A-4, Section  4.3.5).  Aluminum, manganese,
nickel,  zinc,  lead  and  mercury  are discussed elsewhere  in  this document
(Effects Chapters)  in relation  to possible adverse  effects  in  soil, lakes  and
streams, and indirect effects on  health.

5.7.1  Concentration Measurements and Particle Sizes  in  Urban Areas

An  extensive  literature base  on the air quality measurements  of metallic
elements in urban areas  is available.  It  is  not appropriate to discuss this
literature  in  great detail.     Concentrations  of  most  of the  elements   of
interest  here  have  been  reported  by Stevens et  al.  (1978) for six urban
areas.   These measurements along with particulate sulfur  concentrations  are
given in  Table  5-11 as  examples of  reasonably representative urban concen-
tration levels of these  elements.  This study is useful  in  also providing  the
percentages of these elements  in particles below and above 3.5  y m at these
urban  sites.    Sulfur is  the  most  abundant element,   followed  by calcium,
aluminum, iron  and lead.
lEditor's  note:    Although  several  public  reviewers objected  to  the  in-
 clusion of Section 5.7  on metallic  elements and Section 5.8 on visibility,
 during the November 1982 Technical Review Meeting, the reviewers and author
 viewed these discussions as useful and  their inclusion justified.
                                     5-68

-------
Lead concentration  measurements  have been  extensively reviewed in  the Air
Quality  Criteria  for  Lead  (U.S.  EPA  1977b).    In  urban communities  the
percentage  of  monitoring  sites  with measurements  falling within  selected
annual   average  lead  concentration  intervals during  1966  to  1974  were  as
follows:   less  than 500 ng  m-3, 8;  500 to 999 ng m-3, 38: 1000  to  1999  ng
m-3, 45;  2000  to 3999  ng  nr3,  8;  4000  to 53000  ng m-3,  1.   The  lead
concentrations at over 80  percent of these monitoring sites were in the 500
to  1999  ng m"3  range.   The average concentrations  of  lead  at  the  urban
sites given in Table 5-11 also fall within  this concentration range.

The National Academy  of Sciences (1975) review on  nickel  contains a compi-
lation of measurements of ambient air nickel  concentrations  from the National
Air Surveillance Networks.   The overall  average ambient air  concentrations of
nickel  at  urban  sites was  21  ng nr3.   Nickel,  as vanadium,  is associated
with the  type of  fuel  oils used in cities within the  northeastern United
States.  In such areas the  average nickel concentrations often are in the 100
to  300 ng  m-3 range  during the  first and  fourth  quarters.   The  nickel
concentration listed at a site in New York  City in Table 5-11 is at the lower
end of this range.

The  percentages  of  fine (less  than 3.5 pm)  compared to  coarse  particles
(greater than 3.5 pm)  in Table  5-11  indicate that  sulfur,  nickel,  zinc and
lead are most  often  associated with  fine particles.  Calcium, aluminum, and
iron are usually found in coarse particles.   Sulfur and  lead  show the least
variability in size distribution. As discussed earlier (Section 5.2.4), most
of the particle sulfur is present in  submicron particles.   Lead also is asso-
ciated mostly  with  submicron particles  in  urban areas  (Robinson and Ludwig
1967, Lee et al.  1968,  Lundgren  1970,  Gillette  and Winchester 1972, Martens
et al.  1973, Patterson  and  Wagman  1977).  Patterson and Wagman (1977) found
70 percent of the zinc measured in background air and  80 to  90 percent of the
zinc measured  in  more polluted air  on  particles less  than 1.5  pm with most
of the zinc associated with particles between 9.5 and  1.5pm.

Those elements present in coarse  particles  would be  expected to be subject to
rapid deposition near their areas of  emission.  Fine particles have small dry
deposition  velocities  (Chapter  A-7,  Section 7.4.2).    However, atmospheric
dispersion should tend to rapidly decrease the ambient air  concentrations of
both coarse and  fine  particles  associated  with  primary emissions from urban
sources.

Mercury occurs as  a  vapor  in the atmosphere but also can be associated with
particles.   Mercury  concentrations  have  been  measured  in ambient  air  in
several urban areas.   In Washington, DC  a mercury vapor concentration of 3.2
ng m-3  was  measured  during February  1972 (Foote 1972).   Dams  et al. (1970)
reported mercury  concentrations  of  4.8 ng  m-3  on particulate  matter col-
lected in East Chicago, IN.  In Los  Altos,  CA in the San Francisco Bay area,
mercury vapor  concentrations varied  from 1 to 25 ng  m-3  in winter and from
1.5  to 2 ng m-3  up to  50  ng m-3  in  summer  (Millisten  1968).   This area
has  Franciscan  sediments high in mercury,  100 to 200  ppb,  and two  mercury
mines  exist within  25 miles of  Los  Altos.  The  lowest  concentrations were
observed with strong westerlies bringing clear marine air ashore after rainy
weather (Williston 1968).


                                     5-69

-------
            TABLE 5-11.   CONCENTRATIONS AND PERCENTAGES OF ELEMENTS PRESENT AS FINE
                 PARTICLES IN  PARTICULATE MATTER AT SITES IN THE UNITED STATES
Site
Period of measurement Parameter
New York City, NYa
February 1977
Philadelphia, PAa
Feb. -March 1977
Charleston, W VAa
April-Aug. 1976
and January 1977
St. Louis, M0a
December 1975
Portland, ORa
February 1977
Glendora, CAa
March 1977
Smoky Mt., PA<*
July-Aug. 1977
Cone, ng
% Fineb
Cone, ng
% Fine
Cone, ng
% Fine

Cone, ng
% Fine
Cone, ng
% Fine
Cone, ng
% Fine
Cone, ng
% Fine
m-3

m-3

nr3


m~3

m-3

m-3

m-3

S
5936
93
3550
87
4119
92

3526
79
1679
83
1852
87
3948
95
Concentrations and
Ca Al Mn
1509
24
1104
15
924
10

2130
6
832
8
541
18
338
5
969
13
690
7
1372
19

-C
— C
1385
15
>331
NA
215
9
99
56
31
55
19
37

73
55
48
56
11
45
ND
NA
percentages, ng
Fe Ni
1340
29
904
24
788
21

1338
25
1123
17
484
26
146
19
75
76
37
81
1
67

25
60
52
81
17
82
2
50
m-3
Zn
458
81
186
80
50
60

221
67
91
67
61
74
<12
I75
Pb
1227
86
1115
85
757
82

1076
77
1040
83
706
87
114
85
aStevens et al.  1978.
Percentage of mass of element present as particles less than 3.5 ym.
concentrations  reported not  consistent with other Al  measurements  at  site.
dStevens et al.  1980.
NA = not available.
ND = not determined.

-------
5.7.2  Concentration Measurements  and  Particle  Sizes in Nonurban Areas

The  concentrations   of  the  metallic   elements of  interest  and   sulfur  in
particles are given at a number of  rural  and  remote  sites within  the United
States and  Canada  in Table 5-12.   Sulfur  in  particles  collected  at the two
sites in the eastern United States is  in large excess to  the other elements.
Calcium, aluminum,  and iron usually  are the next most  abundant elements.  The
three elements at the  Smoky Mountains, TN site,  as at  the  urban  sites, are
found to  a  large extent in  the  coarse particles (Table  5-12).   All of the
elements listed  except for sulfur and aluminum occur at  substantially lower
concentrations at the rural and remote sites than at  the  urban sites (Tables
5-11 and 5-12).  Lead concentrations at the three rural continental sites are
a factor of 10  to  20 below those at the urban sites.  At the Quillayute, WA
site, lead concentrations in Pacific maritime  air are a factor of 300 to 600
lower than at the urban sites.   Nickel  concentrations  at the rural and remote
sites show similar behavior compared to nickel  at urban sites.  However, zinc
does not  show concentration reductions as large at  rural compared to urban
sites as do lead and nickel.

Additional measurements of sulfur, zinc, and lead have been reported for the
period October 1979  to May 1980  and  from the  40-site Western Fine Particle
(WFP) Network, including the States of Arizona,  New Mexico, Utah, Colorado,
Wyoming, Montana, North  Dakota,   and  South Dakota  (Flocchini et  al. 1981).
Sulfur concentrations rarely exceeded 100 ng  nr3 and frequently  were below
500  ng nr3  on the  average at these sites.  Lead concentrations  were in the
30 to 80  ng nr3 range,  but on the average were below 50 ng nr3  at almost
all   of  the  sites.   The  overall  mean  concentration of  coarse  particles was
8000 ng m-3 with 60 percent associated  with  soil  elements  and  their asso-
ciated oxides.   The  percentage  of iron in fine particles (less than 2.5 ym)
for  the sites in the study area  ranged from 10 to 35  percent with the range
at most  sites between  15  and  25 percent.   These  percentages are  in good
agreement with those  for  fine  particle  iron  at  the  urban sites  and at the
Smoky Mountains site (Table 5-11).

Dams and  Dejonge (1976)  measured aerosol composition  from August  1973  to
April 1975  at Jungfraujoch  (3752  m  above sea  level)  in Switzerland and also
tabulated unpublished results by K.  A. Rahn obtained  at Lakely in marine air
at North Cape, Norway during the winter of 1971-72.  The concentrations in ng
m~3  of  the elements  considered  above were as follows:   Jungfrau,  Al, 51;
Mn,  1.5; Fe, 36;  Zn, 9.9; Pb, 4.4; Lakely, Al, 43;  Mn, 2.5;  Fe, 51; Zn, 8.9;
Pb,  5.6.  These concentrations  are not much different  than at Twin Georges in
the  Northwest Territory,  Canada.

A number  of  the rural   and  remote  sites  discussed  are  in  mountainous and
marine locations.  It is reasonable that  the concentrations of most  elements
would be  low.   In  particular,  sources of  soil-derived  elements  would  be
limited near such sites.   In areas with significant numbers of unpaved roads,
agricultural activities,  and other sources of  windblown  soils, the concentra-
tions of  soil-derived elements  should be  substantially  higher.    The much
higher concentrations  of  aluminum  at Chadron, NB  and Col strip,  MT (Table
5-12)  than  at  mountainous and marine  sites  are  consistent  with  this
expectation.


                                    5-71

-------
                    TABLE 5-12.   CONCENTRATIONS OF ELEMENTS IN PARTICULATE MATTER AT NONURBAN SITES
                                          IN THE UNITED STATES AND IN CANADA
en
Site
Period of measurement
Allegheny Mountain, PA
July-August 1977
Smoky Mountains, TN
September 1978
Chadron, NB 1973
Col strip, MT
May-September 1975
Quillayute, WA

April-November 1974a
December-May 1975a
Twin Georges, NW Terr.,
Canada
S
4690

3948

ND
550



ND
ND
ND

Ca
330

338

ND
390



ND
ND
ND

Al
70

215

535
930



ND
ND
66

Mn
9

ND

6
9



0.7
0.8
1.5

ng m-3
Fe
320

146

ND
410



25.3
13.1
71

Ni Zn
ND 20

2 <12

ND 16
0.6 6.5



0.1 4.2
0.1 11.3
ND 3.8

Cd Pb
3 90

ND 114

0.6 45
ND 14



ND 1.9
ND 1.8
ND ND

References
Pierson et al .
1980b
Stevens et al .
1980
Struempler 1975
Crecelius et al .
1980
Ludwick et al .
1977


Dams and Dejonge
1976
    aonly those days included with trajectories having marine histories for at least three days before arriving
     at the Quillayute, WA site.

    ND =  not determined.

-------
Ambient  air concentrations  of mercury  vapor  at  nonurban  sites  have  been
summarized  as  a function of  soil  conditions (U.S. Geological Survey  1970).
Over areas without mercury containing minerals,  ambient air concentrations of
mercury  vapor  were in  the  3  to  9 ng m-3.   Over  areas containing mercury
minerals, ambient air concentrations of mercury vapor  were  in the  7  to 53 ng
m~3  range,  while in the  vicinity  of known mercury mines  the mercury  vapor
concentrations  reached  the  24 to  108  ng  m-3 range.   Mercury concentrations
were  found  to  peak  at  midday and to  decrease rapidly  with altitude  (U.S.
Geological Survey 1970).

At  nonurban locations  on  the beach in  the  San Francisco  Bay area mercury
vapor  concentrations  of  3.1  ng  m-3  have  been  reported  (Foote  1972).
Willisten (1968) collected samples at 10,000  foot altitudes 20 miles  offshore
of the San Francisco Bay area and obtained concentrations of mercury  vapor of
0.6 to 0.7  ng  m-3.   At a rural  site,  Miles, MI,  a mercury  concentration  of
1.9 ng m-3  was measured in particulate matter  (Dams et  al. 1970).  Ambient
air  mercury vapor  concentrations  of  25  ng nr3  were  reported  in  samples
collected in Research Triangle Park, NC (Long et al.  1973).

5.8  RELATIONSHIP OF LIGHT EXTINCTION AND  VISUAL RANGE  MEASUREMENTS TO
     AEROSOL COMPOSITION

Visual range measurements can  be  influenced  by  a  number of  natural  and  man-
made factors.  Visual  range can be reduced substantially  on an episodic basis
by rain,  fog,  snow,  and by windblown dust and sand.   Rayleigh scattering  by
air molecules  contributes to  light extinction and  limited visual  range,  but
the contribution  is  small  except  in remote  areas.   Nitrogen dioxide  is  the
only other  gas in the atmosphere  with the  potential  to contribute  signifi-
cantly to light extinction,  but  its concentration  in the atmosphere usually
is too low  for it  to  contribute  substantially in  practice.  Particles  in  the
size  range  between   about  0.1  and  2 urn  are  effective  light  scattering
components of  the  atmosphere  while elemental  carbon particles are effective
absorbers of  light (Charlson et al. 1978b).   Most of  the emphasis  in  this
section will be  on  the  relationships between aerosol  composition  and  visual
range and light extinction.

Sulfates  and   nitrates  as  suspended  aerosol components of the  atmosphere
contribute to  visibility reduction through light scattering.  These  aerosols
also contribute  to acidic deposition  and its effects.   To  the substantial
extent that visual range and  light extinction are  accounted  for by  sulfates
and nitrate concentrations  in the atmosphere, these visibility measurements
can serve as surrogates for concentration measurements in geographical   areas
where concentration measurements are not  available.  Because aerosol concen-
trations  are related  to deposition rates, the  visibility  measurements  also
can be related to deposition or to  the  potential for deposition.

5.8.1    Fine   Particle  Concentration  and Light  Scattering  Coefficients—A
number ofinvestigatorshavedemonstratedaproportionalitybetweenFine
particle  concentration   and   light scattering  coefficient.    Sulfates  and
nitrates,  in   some  locations, are major  components   of the  fine  particle
concentration.
                                     5-73

-------
Waggoner and Weiss  (1980) obtained a  ratio  of  fine  particle  concentration  to
the  light  scattering  coefficient,   bsp,  of  0.36  g  nr2   (corrected  for
temperature)  from measurements  at five  urban and  rural  locations  in the
western  United  States.   In  Denver,  CO,  Groblicki et  al.  (1981)  obtained  a
ratio  of fine  particle  concentration to  bsp of 0.29 g nr2.   In Houston,
TX,  Dzubay  et  al.  (1982)  obtained a  very high  correlation coefficient  of
0.987  between   fine  particle concentration and  bsp and a  ratio of  0.28   g
m~2.    The  ratios  obtained  in  Denver  and  in   Houston  are  in  reasonable
agreement with  the results obtained by Waggoner and  Weiss (1980).

At a  site  in  the Shenandoah Valley,  VA, Weiss et al.  (1982)  obtained a  cor-
relation coefficient of  0.94 for  the  measurements  of fine  particle concen-
tration  as  related  to bsp  and  a ratio of  0.24  g nr2.  A  cyclone  was  used
to eliminate  particles  above 1 ym  from the measurement as   fine particles.
Ferman  et  al.  (1981)  made   measurements  at the  same  site  during  the  same
period.  These  workers obtained a correlation coefficient  of  0.91  for the
measurements of fine particle concentration  as  related to  bsn and  a ratio
of 0.14 g nr2.   However,  a  substantially  higher particulate   size cutoff was
used by Ferman et al. than by Weiss et al.

Although there  is variability in the ratio  of  fine  particle  concentration  to
bsp  from  site   to   site,  consistently  high  correlation  coefficients  are
obtained at  individual  sites.   The  variability in  ratio is related  to the
corresponding variability in  the ambient  air  aerosol composition  (White and
Roberts 1977,  Ferman et al.  1981).

5.8.2  Light Extinction or Light  Scattering Budgets  at  Urban  Locations

At several  locations in the  South Coast Air Basin concurrent measurements  of
light  scattering  and of  aerosol  composition  were  available from  the  1973
Aerosol  Characterization  Experiment   (ACHEX).    White  and   Roberts  (1977)
analyzed these  results to obtain relationships between light scattering and
aerosol  composition.   Sulfate,  nitrate  and  organic  aerosols  all  made   a
substantial  contribution  to  the  overall   aerosol  concentrations at  these
locations.   The average percentage  contribution  of aerosol   classes  to the
light  scattering  (based on  all  emission sources) was  as follows:   sulfate,
47; nitrate, 39; organics, 14.  Except at  high humidities, the  contribution,
on a unit mass basis, of sulfate  was  higher than that of nitrate.   A lack  of
dependence  on  humidity of  the contribution  of  sulfate to  light scattering was
found.  In  contrast Cass (1976),  from  similar measurements in the South Coast
Air Basin,  did  find a dependence on  humidity   of both the contributions  of
sulfates and nitrates  to  light scattering.   The  sum  of species  other  than
sulfates,  nitrates,  and  organics  was  found  to  have  about one-third  the
effectiveness  of  sulfate  on a  unit  mass  basis  in  contributing to  light
scattering  (White and Roberts 1977).

In Riverside,  CA  the average percentage  contributions  of aerosol  classes  to
the  light  scattering coefficient  were found  to be 70 to   75  percent for
sulfate and 20  to  25  percent for nitrate  on a  unit  mass basis  (Pitts and
Grosjean 1979).   No  statistical association  could  be found in  this  study
between light  scattering  with organic  carbon  or any  other   aerosol  species
measured.
                                     5-74

-------
In November  and  December 1978 at a  location  in Denver, concurrent measure-
ments were made of both light scattering and adsorption of nitrogen dioxide,
and  of  ammonium, sulfate,  nitrate,  organic carbon,  elemental  carbon   and
other species in the fine particle fraction (Groblicki et al. 1981).  Of the
chemical species measured the percentage contributions to the light extinct-
ion were as  follows:   sulfate as ammonium  sulfate,  20;  nitrate as ammonium
nitrate,  17; organic  carbon, 12;  elemental   carbon, 38  (scattering,  6.5,
adsorption,  31.2);  remainder  of  fine particle mass,  6.6;  nitrogen dioxide,
5.7.  Elemental  carbon  was  found to be the most effective species on a unit
mass basis  in  contributing  to light  extinction.   Both  sulfate and nitrate
were  found  to  have their  contributions  to  light scattering  dependent  on
relative humidity.   Sulfate was  a more  effective  scatterer  on a  unit mass
basis than nitrate or organic  carbon.   The  sum  of other  fine particle species
showed  a  much lower  effectiveness  on  a  unit mass  than the  other species
specifically considered above.

During  September  1980  in Houston,  TX  concurrent measurements  were  made  of
light scattering and light  extinction,  of  nitrogen dioxide,  and of sulfate,
nitrate, carbon-containing  compounds  and many other  species  (Dzubay  et al.
1982).   The percentage  contributions  of  the  chemical   species measured  to
light  extinction were  as  follows:    sulfate  and  associated  cations,  32;
nitrate, 0.5; carbon, 17 to 24 (scattering, 11,  adsorption,  6  to  13);  other
aerosol  components, 4;  water, 16;  nitrogen dioxide,  5;  Rayleigh  (air),  6.
The crustal  elements constituted 29  percent of the total mass  concentration
of  particulates,  but  only  2.9  percent of the  fine particle  mass.    As  a
consequence, the crustal elements only  contributed  2.6  percent of the light
extinction.   No functional  relationships  of  sulfate and  nitrate including
humidity were used.  Instead, the contribution of  water to light extinction
was computed separately.    If the  contribution of water  is  associated pre-
dominately with  sulfates, the sulfates  and  associated species  would account
for about one-half of the light extinction.

The  contribution of  light  extinction associated  with  nitrates  was  much
smaller in  Houston  than  in  Los Angeles and Denver  (White  and  Roberts  1977,
Groblicki  et al.  1981,  Dzubay  et  al.  1982).  Nitrates  were determined in both
Houston and Denver studies on Teflon filters,  so a negative nitrate artifact
would be  expected  in both  sets  of  measurements.   Therefore, at  least on  a
relative basis,  the  nitrate concentrations in Denver should  have  been much
higher  than  in Houston.   The  difference in season during which sampling was
done may in  part explain the  differences in nitrate concentration obtained.
In the measurements used by  White and Roberts (1977) glass fiber filters were
used, so overestimates  of nitrate  concentration  are  to  be expected.   Pitts
and Grosjean (1979) made measurements with tandem filters and concluded that
there  was   only  a  moderate,  11  percent  on  average,  nitrate  artifact
correction.

All of the studies at urban  locations discussed above  involved concurrent air
quality and  instrumental light scattering  absorption  or extinction measure-
ments.  Several  other studies  have used visibility measurements  combined with
HIVOL sampling results  obtained at  sites within the  same  urban area  (Trijonis
and Yuan 1978a,b; Leaderer et  al. 1979).  Aside from the  usual limitations in
regression models themselves,  these  studies are subject to a  number of other


                                     5-75

-------
possible sources of  error.   These sources of  error include some related to
airport visibility  measurements:   (1)  inadequate  sets of markers, (2) changes
in markers,  and (3)  changing  environment in  vicinity of   airports.   The
differences in the locations where  the visibility  and the  air quality mea-
surements are taken  can  also result  in differences in aerosol  concentration
and composition at these  locations.   The lack of compositional measurements
on some  significant  species  can  result  in overestimations  of  the contribu-
tions of measured  species.   Such  overestimations  can occur when  there are
good correlations between measured and unmeasured species.  The use of glass
fiber filters in the  HIVOL samplers means that  positive nitrate artifacts are
likely, as discussed  earlier  in this  chapter.

Despite  the  limitations  discussed  above,  the  airport  studies  do provide
results at a number  of urban locations at which  more acceptable studies are
not available.  The  estimated contributions of the  chemical species measured
to light extinction  budgets has been tabulated and  discussed elsewhere (U.S.
EPA 1979) and will be only  briefly discussed here.  On the average, for the
midwestern and northeastern  locations used (Trijonis and Yuan 1978b, Leaderer
et al.  1979)  the average percentages and ranges  of percentage contributions
of  chemical   species measured  to  the  light  extinction  were  as  follows:
sulfates 56, 27 to 81;  nitrates,  2,  0 to 14;  remainder of  TSP,  8,  0 to 44;
unaccounted for, 34, 19  to  73.   At southwestern  sites  (Trijonis  and Yuan
1978a)   the  nitrates   were  reported  to  make  a larger  contribution  to light
extinction than at the midwestern  and northeastern  locations considered.

5.8.3  Light Extinction or Light  Scattering Budgets at  Nonurban Locations

At  Allegheny  Mountains,  PA,  concurrent  light  scattering  and  air quality
measurements were made during the  latter part of July and  early August 1977
(Pierson et al. 1980a,b).  The  authors comment that the multiple regression
analyses showed bsp  to be remarkably insensitive to any aerosol constituent
but sulfate or  its associated  cations.  Sulfate alone accounted  for  94+7
percent  of the variability  in  bsp.   An  even better  correlation  was found
for  bsp  with  the product of  sul fate  and  humidity  than  with sul fate alone.
With  respect  to visual  range  the authors concluded that  "sulfate  may  be a
good index of visibility (and vice versa)  if humidity is  taken  into  account."
In the Shenandoah Valley/Blue Ridge Mountain area of  Virginia  several  groups
of investigators made  measurements  during  July to August of 1980  (Ferman et
al.  1981,  Stevens  et  al.  1982,  Weiss  et  al.  1982).   Ferman  et al.  (1981)
obtained light scattering and light absorption  measurements,  nitrogen  dioxide
concentrations, and  aerosol  composition  measurements.   The aerosol composi-
tion  of  the fine particle mass  was reported.    Based  on these results, the
observed light  extinction  on a  percentage basis could  be  accounted  for as
follows:   sulfate  (including water),  78;  carbon-containing compounds,  15.5
(scattering, 13, absorption, 2.5);  nitrogen dioxide, 0.3; Rayleigh  (air), 5.
For  the  periods in  the  upper  decile  of bsp values  the  sulfate (and  water)
accounted  for 4 percent of the light  extinction.  Weiss  et al.  (1982),  from
their measurements at  the  same site, also  concluded  that all  of the water at
70 percent RH was associated with  sulfate and  ammonium.   The sulfate  with
associated cations and water accounted on average for 70  percent of the light
                                     5-76

-------
 scattering.    This  result  Is in  reasonable  agreement with  the 78  percent
 obtained  by Ferman et  al.  (1981).  Stevens  et  al.  (1982) measured  aerosol
 composition, but not light extinction.  However, it is of interest to compare
 their  composition  results  for the fine particle mass with  those  obtained by
 Ferman et al.  (1981).   The percentage of  the fine  particle mass contributed
 by the various chemical  species  (do  not  add  up  to 100  percent)  from  the
 Ferman et  al. study  and  the  Stevens et  al.  study,  respectively were  as
 follows:   sulfate  as ammonium bisulfate, 55.4,  60.8; elemental  carbon,  5.4,
 5.7; organic  carbon (measured carbon x 1.2),  23.6, 4.1;  nitrate as ammonium
 nitrate,  0.6,  ND;  Pb-Br-Cl, 0.2,  0.3;  crustal  (estimated  from Si),  7.3,  1.1.
 The  higher percentage  for sulfates  and  the lower  percentage  for  organic
 carbon in the Stevens  et  al. (1982)  study  would result  in  an even  larger
 contribution  of sulfates  to  light extinction  than found  by Ferman et  al.
 (1981).

 At another location in  the eastern mountains  of the  United States,  Great
 Smoky  Mountains,  TN, aerosol composition, but no light extinction  measure-
 ments,  were made (Stevens et al.  1980).  The  percentage of  the  fine  particle
 mass  contributed  by  the  various chemical  species  (do not  add up  to  100
 percent)  were as  follows:   sulfate  as  ammonium  bisulfate, 56;  elemental
 carbon,  5;  organic carbon  (measured  carbon  x 1.2),  11;  Pb-Br-Cl,   0.5;
 crustal, 0.5.  The  percentages of sulfates and elemental carbon  at  the  Great
 Smoky  Mountains site were  nearly the same as  at the  Shenandoah  Valley  site.
 In contrast,   the  organic  carbon  and  the  crustal  elements  made up  a  sub-
 stantially  lower  percentage of  the fine  particle  mass at the Great  Smoky
 Mountains site (Stevens et al. 1980) than reported by  Ferman et al.  (1981)  at
 the Shenandoah Valley site.

 In the midwestern  United  States  at rural  sites  in Missouri and  in  the  Ozark
 Mountains, Weiss et al. (1977) concluded that  essentially all of the  aerosol
 light  scattering was  due to  sulfates.   Measurements  of sulfate as ammonium
 sulfate at  rural sites  in  the vicinity of St.  Louis  indicate that  45 to  50
 percent  of the fine  particle mass  was  ammonium sulfate  in  the  first  and
 fourth quarters of the  year and over 70 percent of the fine  particle mass  was
 ammonium  sulfate in the  third  quarter  of the year (Altshuller 1982).  As  in
 nonurban sites in the eastern United States,  the sulfates in  the midwest  are
 the major contributors  to the  fine particle mass.

 In  the southwestern United States  at nonurban locations,  concurrent mea-
 surements  of   light  extinction  and  of aerosol  composition  have been  made
 (Macias et  al. 1980).   From  samples obtained in flights  over the Southwest
 the average percentage contributions of chemical species to light scattering
 were as follows:    sulfate as  ammonium sulfate,  16;  silicon  dioxide,  16:
 other  fine mode particles, 8; coarse mode  particles,  4; Rayleigh (air), 44.
 In measurements at  a  nonurban site, Zilnez Mesa,  AZ, measurements  of light
 extinction  and aerosol  composition  were  made  (Macias  et  al.  1981).   The
 average percent contributions to  light extinction were  as  follows:   sulfate
 as ammonium sulfate, 18; organic  carbon, 33;  elemental  carbon, 12;  nitrate,
 2; other  fine  particles,  20;  coarse particles,  15.   In individual  measure-
ments  Rayleigh scattering  contributed from  16  to  54  percent.   The light
 extinction budgets at these western nonurban sites are clearly substantially
different than at eastern  nonurban sites.   Sulfates at these western nonurban


                                    5-77

-------
sites make  a  much  smaller  contribution to  the  light  extinction  than  at
eastern sites.  Carbon-containing particles, other fine mode species, coarse
mode  species,  and  Rayleigh  scattering  are  relatively  more  important  at
western than  eastern nonurban  sites.    However,  the  light   extinction  is
smaller and  the  visual  range  much  greater at  the western  nonurban sites
because the absolute amounts  of aerosol species are  so much  smaller.

The  contributions  of sulfates  compared  to  other chemical  species  to light
extinction at rural sites in the midwestern and eastern United States appear
more important than in western  urban  areas (White  and Roberts 1977, Pitts and
Grosjean 1979, Groblicki et al. 1981) and western nonurban  locations  (Macias
et al. 1980,  1981).  At eastern rural  sites  visibility  should be a good index
or surrogate  for sulfates (Pierson  et al. 1980a, Ferman et al. 1981, Weiss et
al.  1982).  It is  less  evident that  visibility  in the  western United States
can be used as a surrogate for  sulfates or for sulfates and  nitrates.

5.8.4  Trends  in Visibility as  Related to Sulfate  Concentrations

Several  investigations  have  indicated that  the  patterns  of historical visi-
bility at airport sites and  sulfate  trends  in  the eastern United States are
consistent with  each  other  (Trijonis and  Yuan  1978b;  Husar et  al. 1979;
Altshuller 1980;  Sloane 1982a,b).   The  improvements  in visibility  in  the
first and fourth quarters of  the year appear consistent with the decreases in
sulfate concentrations.   Similarly,  the  deterioration of visibility during
the  1960's into  the  1970's was consistent  with  the increase in sulfate con-
centrations.   Further deterioration in visibility during the 3rd quarter of
the year did not occur  later in  the  1970's, again consistent with the trends
in sulfate concentrations (Altshuller 1980,  Sloane 1982b).

5.9  CONCLUSIONS

The  following  statements summarize  the  discussion  in  this chapter  on  the
atmospheric concentrations and distributions  of chemical  substances.  Table
5-13 summarizes measurements of sulfur,  nitrogen, and  chlorine compounds in
rural areas.

 0   Sulfur dioxide  concentrations  have  been high  in  urban  areas  in  the
     eastern United States,  but decreased substantially during  the 1960's and
     into  the  1970's.   The   decreases  in  sulfur  dioxide   appear  to  be
     associated with  local reductions in the  sulfur content of fossil fuels
     (Section 5.2.2.1).

 0   In rural  areas sulfur dioxide concentrations are appreciably lower than
     in urban  areas.   The  differences  in  concentrations  between  urban  and
     rural areas  were not as  great  by the  late  1970's as  in earlier years.
     This change  primarily  is  the result of  the decreases in urban sulfur
     dioxide concentrations (Section  5.2.2.2).

 0   Measurements  of  sulfur  dioxide concentrations  at   nonurban  sites  are
     limited and  values are  often near  limits  of  detectability.   No clear
     trends in  nonurban sulfur dioxide  concentrations  with time are  evident
      (Section 5.2.2.2).


                                     5-78

-------
        TABLE 5-13.  CONCENTRATIONS OF SULFUR, NITROGEN, AND CHLORINE
         COMPOUNDS AT RURAL SITES IN  THE UNITED STATES IN THE 1970'S
                                                      Range of
                                          Average concentrations, yg m~3
Compound
Sulfur dioxide
Sulfur aerosols (as sulfate)
Nitrogen dioxide
Nitrate aerosols
Nitric acid
Peroxyacyl nitrates
Ammonia

Hydrogen chloride

Chloride aerosols
Maritime

Inland

East
10-20*
5-153
10-ZOb
1C
0.3-1.3
0.5-1C

0.5-2<1




1-lOC

1 1C
West
NA
1-33
12C
NA
1 lc
0.1-0.3C

0.5-2C

1-lOC


1-10C

11C
aAnnual average.
bSummer months:  August to December averages.
cLimited number of measurements.
NA=Not available.
                                     5-79

-------
Sulfate concentrations decreased in  eastern  cities  during  the 1960's and
into the  1970's  except during  the  third  quarter  of  the  year  (Section
5.2.3.1).

In rural areas in the  eastern United States  sulfate concentrations have
not increased  substantially  on  an annual  average  basis,  but  increased
significantly during the summer  months  (Section  5.2.3.3).

Sulfate concentrations  within rural  areas in the eastern United  States
by the  1970's  were  almost  as high as  in  adjacent  urban  areas  (Section
5.2.3.3).

Sulfate aerosols can contribute one-third to one-half the sulfur  budget
(sulfur dioxide plus  sulfate) in  rural  areas within the  eastern  United
States during the summer, but contribute relatively little to the  sulfur
budget in  the winter months (Section 5.2.3.3).

Sulfate aerosols are substantially higher in rural  areas  in the eastern
United States than in remote areas of the  western United States (Section
5.2.3.3).

Sulfate aerosols  occur predominately  in  the fine  particle  size  range
with much of the mass of sulfate aerosols  concentrated between  0.1 and  1
urn.  Particles in this  size  range deposit more slowly  than does  sulfur
dioxide,  so  they can  be  transported substantial  distances   (Section
5.2.4).

Sulfate aerosols  tend  to be more  acidic  in rural   areas  than   in  urban
areas (Sections 5.2.3.2 and 5.2.3.4).

Much  of the  sulfate  aerosol has  been reported to be in  the form of
strong acid  species at  locations  in  the eastern mountains of the  United
States during the summer months  (Section 5.2.3.4).

Sulfur dioxide and sulfate concentrations in remote areas are  between  a
factor of 10 and 100 lower than  the concentrations  in  rural  areas  in  the
eastern United States  and  adjacent areas  of  eastern  Canada  (Sections
5.2.2.2, 5.2.2.3 and 5.2.3).

Nitrogen  oxides  reach  about the  same concentration  range  as  sulfur
dioxide in  cities.   Their concentrations have  become more  significant
relative  to  sulfur  dioxide  with  the  decrease   in  sulfur   dioxide
emissions (Section 5.3.2.3).

Nitrogen oxides are substantially lower in concentration in  rural  areas
than in urban areas (Sections 5.3.2.3 and  5.3.2.4).

Nitrogen  dioxide  concentrations are  substantially  lower in  rural  areas
within  the   western  United States  than  in  the  eastern  United  States
(Section 5.3.2.4).
                                5-80

-------
At remote  locations  the  concentrations  of nitrogen oxides can be  10  to
100  times  lower  than  in  rural  areas  of  the  eastern  United   States
(Section 5.3.2.5).

The  average concentrations of nitric acid or of  peroxyacetyl nitrates
are  about  a  factor  of  ten lower  than  the  average  concentrations  of
nitrogen  dioxide  in both  urban  and rural  areas  (Sections 5.3.3.1  and
5.3.3.2).

The  average concentrations  of nitric acid  are  in  the  same  concentration
range  as the average  concentrations of  peroxyacetyl  nitrates in  rural
areas (Section 5.3.3.2).

The concentrations of nitric acid in the boundary  layer  in  remote  areas
are  a factor of 5 to 10  lower than  in  rural  areas in  the eastern  United
States (Section 5.3.3.3).

The  equilibrium between  ammonia,  nitric  acid,  and ammonium nitrate  can
be  important in  determining the  ambient air  concentrations of  these
chemical  substances (Section 5.3.5).

Several  positive  and negative nitrate  artifacts  on  filters  have  been
identified  and investigated.  Such  artifacts make most  of  the measure-
ments  on  single  or   tandem  filter  systems  for  particulate  nitrate
unreliable  (Section 5.3.6).

Measurements of particulate nitrate made  using  diffusion  denuders  appear
to be reliable.  At both urban sites in  Los Angeles and rural sites  in
the  eastern United  States  such  measurements indicate that particulate
nitrate concentrations can exceed nitric  acid concentrations in the late
evening and in the early morning  hours.  Conversely, nitric  acid concen-
trations are higher than  particulate nitrate concentrations in the  late
morning and afternoon hours (Sections 5.3.6.1 and  5.3.6.2).

Particle  size  distributions of  particulate  nitrates  are influenced  by
the  same  nitrate  artifact problems.  It  does  appear  that the particle
sizes of nitrates decrease  in  going  from  coastal  locations  inland  in
California.  The reason is related to the greater  abundance  of submicron
sodium nitrate aerosols  in maritime air reacted with nitrogen dioxide,
compared  to  the  submicron  ammonium   nitrate aerosols  found   inland
(Section 5.3.7).

The  concentrations  of   sulfate  aerosols  appear   to  be  several   times
greater than the concentrations  of  nitric  acid and particulate nitrate
at rural  sites in the  eastern  United  States (Sections 5.2.3.3, 5.3.3.2
and 5.3.7).

Ozone concentration  levels in rural  areas can result from one  or more  of
the  following processes:   (1) local synthesis, (2) fumigation by  urban
or industrial plumes,  (3)  high  pressure  systems  near rural   sites, and
(4) ozone formed in  the stratosphere or  free  troposphere  reaching ground
level (Section 5.4).
                                5-81

-------
Rural locations within urban plumes may experience ozone concentrations
in  the  range  of  300 to  500 yg  nr3.    Within high  pressure systems,
ozone concentrations  at  rural  locations  can  range from 150  to  250 pg
m-3 (Section 5.4.1).

At remote elevated sites,  hourly  ozone  concentrations  are as high as 140
to  160  ug  m-3 during  the spring  months and  as  low as  40  to  60 yg
m~3  in  the fall months.   Occasional  observations  of ozone  concentra-
tions  in  excess  of  200  yg  m-3  attributed  to   stratospheric  air
extrusions at remote  sites appear too high compared  to aircraft measure-
ments of ozone through the troposphere  (Section  5.4.2).

Ambient air measurements  of  hydrogen peroxide are  in doubt because of
recent demonstrations  of   in  situ generation  of  hydrogen  peroxide in
aqueous solutions (Section 5.5).

Hydrogen  peroxide  concentrations  measured in  rainwater  usually  cor=
respond to  those  resulting from   the absorption of  less  than 1  yg m"-5
of hydrogen peroxide  from  the ambient atmosphere (Section 5.5.3).

The variations in hydrogen peroxide concentrations measured in rainwater
during precipitation  events  are  consistent with a  substantial  part of
the hydrogen peroxide being generated within the cloudwater rather than
being present  as  a result of  rainout  and washout  of gaseous hydrogen
peroxide (Section 5.5.3).

The concentrations of particulate chloride compounds can  be important
near the ocean, but  not inland.   At inland sites particulate chlorides
tend to be submicron in  size and have been associated with  automotive
lead  aerosol   emissions   and  with  emissions   from  combustion  sources
(Section 5.6.4).

The concentrations of metallic elements  in most urban areas occur at 1
to 2  yg  m-3 and below.   The bulk  of  the calcium, aluminum, and iron
occurs in  coarse  particles,  while most of the  lead and  zinc occurs in
fine particles.  The  substantial  differences in  size distribution should
result  in  those elements  found   in  coarse particles usually being of
local origin, while the elements  in fine  particles are capable of  being
transported substantial  distances (Section 5.7.1).

Although lead  aerosols  are largely  submicron  in  size,  lead  concentra-
tions  drop off rapidly  from  urban  to  rural  to  remote  sites.    At
continental rural  sites  lead concentrations  are  a factor of 10 to 20
below  concentrations at  urban  locations.   At  remote  sites  the lead
concentrations  are  several  hundred times lower  than  at  urban  sites
(Section 5.7.2).

High correlations exist between fine particle mass and light  scattering
coefficients (Section 5.8.1).

At eastern rural  sites sulfate  accounts  for  a large part of the  fine
particle mass and the light extinction  (Section 5.8.3).


                                5-82

-------
At  western  locations  nitrate  and  carbon-containing  particles  make a
substantial  contribution to fine  particle  mass  and to light extinction
(Section 5.8.2).

At  rural  sites  in  the  eastern  United  States  visibility measurements
should be a good index or surrogate for particulate  sulfate concentra-
tions (Section 5.8.3).
                                5-83

-------
5.10  REFERENCES

Altshuller, A.  P.   1973.   Atmospheric  sulfur dioxide and sulfate distribution
of concentration at urban and  non-urban  sites in the United States.  Environ.
Sci. Technol. 7:709-712.

Altshuller, A.  P.   1976.   Regional   transport and  transformation  of sulfur
dioxide and sulfates in the U.S.  J. Air Pollut. Control Assoc. 26:318-324.

Altshuller,  A.  P.   1980.     Seasonal   and  episodic   trends   in  sulfate
concentrations  (1963-1978)  in  the eastern  United States.   Environ.  Sci.
Technol. 14:1337-1349.

Altshuller, A.  P.  1982.   Relationships involving particle mass  and sulfur
content at sites in and around St.  Louis, MO.  Atmos. Environ. 16:837-843.

Altshuller, A.  P.   1983.   Review:  Natural  volatile  organic  substances and
their effect on air  quality in the United States.  Atmos.  Environ.  17(11):
2131-2165.

Appel, B. R., E. M. Hoffer, U. Tokiwa,  and E. L.  Kothny.  1982.  Measurement
of  sulfuric  acid  and particulate  strong acidity  in the  Los  Angeles Basin.
Atmos. Environ. 16:589-593.

Appel,  B.  R.,  E.  L. Kothney, E.  M.  Hoffer, and J.  J.  Wesolowski.   1977.
Comparison of  wet  and instrumental methods for measuring airborne sulfate.
EPA   Report  600/7-77-128.     Environmental  Sciences  Research  Laboratory,
Research Triangle Park, NC.

Appel, B. R., E. L. Kothney, E. M.  Hoffer,  G.  M.  Hidy, and J.  J. Wesolowski.
1978.   Sulfate  and nitrate  data  from  the California aerosol characterization
experiment (ACHEX).  Environ.  Sci.  Technol. 12:418-425.

Appel,  B.  R.,  S.  M.  Wall,  Y.  Tokiwa, and  M.  Haik.   1979.   Interference
effects  in sampling  particulate nitrate  in ambient  air.   Atmos. Environ.
13:319-325.

Appel, B. R., S. M. Wall, Y. Tokiwa, and M. Haik.   1980.   Simultaneous nitric
acid,  particulate  nitrate and acidity measurements in  ambient  air.   Atmos.
Environ. 14:549-554.

Appel,  B.  R.,  S.  M.  Wall, Y.  Tokiwa,  and  M.  Haik.    1981a.   Sampling of
nitrates in ambient air.  Atmos.  Environ.  15:283-289.

Appel,  B.  R.,  S.  M.  Wall,  Y.  Tokiwa, and  M.  Haik.   1981b.   Atmospheric
particulate  nitrate  sampling errors  due  to  reactions  with  particulate and
gaseous strong  acids.  Atmos. Environ. 15:1087-1089.

Barrie, L.  A.,  H.  A. Wiebe, K.  Aulsuf, and P. Felliu.   1980.  The Canadian
Air  and Precipitation Monitoring Network APN.  Atmos.  Pollut.  8:355-360.
                                     5-84

-------
Barrie,  L.  A., K.  G.  Aulsuf, H.  A.  Wiebe, and  P.  Felliu.    1983.   Acidic
pollutants  in  air and precipitation at  selected  rural  locations in Canada.
In  Proceedings  of the American Chemist  Society Symposium  on  Acid Rain.  J.
Teasley, ed. Ann Arbor Science, Ann Arbor,  MI.

Bellinger,  M.   J.,  D.  D.  Parrish,  C.  Hahn,  D.   L.  Albritton,  and  F.  C.
Fehsenfeld.    1982.    NOX  measurements  in  clean  continental   air.    2nd
Symposium Composition of the  Nonurban Troposphere.   Williamsburg,  VA American
Meterological Society, 45 Beacon St. Boston, MA.

Bonsang,  B.,  B.  C.  Nzuyen,   A.  Gaudry and  G.  Lambert.    1980.   Sulfate
enrichment in marine aerosols  owing to biogenic  gaseous  sulfur  compounds.  J.
Geophys. Res. 85:7410-7416.

Breeding, R. J., J. P. Lodge,  Jr.,  J.  B. Pate,  D.  C.  Sheesley, H. B. Klonis,
B. Fogle, J. A. Anderson, T.  R. Englert, P.  L. Haagenson, R. B. McBeth,  A. L.
Morris,  R.  Pogue,  and  A.   F.  Wartburg.    1973.   Background  trace  gas
concentrations in the central  United States. J. Geophys. Res.  78:7057-7064.

Breeding, R. J., J. B. Klonis, J.  P. Lodge,  J. B.  Pate,  D.  C. Sheesley,  T. R.
Englert  and  D.  R. Sears.  1976.   Measurements  of atmospheric pollutants in
the St.  Louis area.  Atmos.  Environ. 10:181-194.

Brennen, E.  1980.  PAN concentrations in ambient  air  in New Brunswick,  N. J.
Final  Report   on   EPA   Grant  805827   to   Environmental   Sciences  Research
Laboratory, Research Triange  Park,  NC.

Brosset,  C.    1978.   Water-soluble  sulfur  compounds in  aerosols.   Atmos.
Environ.  12:25-38.

Bufalini, J.  J.,  B.  W.  Gay  and K.  L. Brubaker.    1972.   Hydrogen peroxide
formation  from  formaldehyde   photooxidation and  its   presence  in   urban
atmospheres.  Environ. Sci. Technol. 6:816-821.

Cadle, S. H., R. J. Countess and N.  A.  Kelly.  1982.   Nitric acid and ammonia
in urban and rural locations.   Atmos.  Environ. 16:2501-2506.

Cass,  G. R.    1976.    The   relationship between  sulfate air  quality and
visibility at Los Angeles, CA.  Inst.  Technol. Environ.  Quality Lab. Memo No.
18, Pasadena, CA.

Charlson, R. J.,  D. S.  Covert, T.  V.  Larson,  and A.  P.  Waggoner.   1978a.
Chemical  properties  of  tropospheric  sulfur  aerosols.    Atmos.  Environ.
12:39-53.

Charlson, R.  J.,  A.  P.  Waggoner,  and  J.   F. Thielke.   1978b.   Visibility
protection for Class I areas:  the technical basis.  Council on Environmental
Quality, Washington, DC.

Charlson, R. J.,  A.  H.  Vanderpol,  D.  S.  Covert,  A.  P.  Waggoner,  and  N.  C.
Ahlquist.       1974.       H2S04/(NH4)2S04    background    aerosol:   optical
detection in St. Louis region.  Atmos.  Environ.  8:1257-1267.


                                     5-85

-------
Cleveland, W. S. and  B.  Kleiner.   1975.  The transport of photochemical air
pollution  from  the  Camden  -  Philadelphia  urban complex.    Environ.  Sci.
Techno!. 9:869-872.

Cleveland,  W.   S.,  B.  Kleiner,  J.  E.  McRae  and  J.  L.  Warner.     1976.
Photochemical air  pollution:   Transport  from  the New  York  City  area  into
Connecticut and Massachusetts.   Science  191:179-181.

Cleveland, W.  S.,  B.  Kleiner,  J.  E. McRae and  R.  E. Pasceri.   1977.   The
analysis of  ground-level  ozone  data from  New Jersey,  New York, Connecticut
and  Massachusstts:    Data  quality  assessment and temporal  and geographical
properties, pp. 185-196. In Vol  1,  International Conference of  Photochemical
Oxidant  Pollution and Its Control.   B. Dimitriades, ed.  EPA-600/3-77-0012.
Environmental Sciences Research  Laboratory,  Research  Triangle  Park, NC.

Coburn,  W.  G.,  R.   B.  Husar, and J.  D. Husar.    1978.    Continuous   in-situ
monitoring of ambient particulate  sulfur  using  flame photometry and  thermal
analysis.  Atmos.  Environ.  12:89-98.

Cohen, I. R. and T. C.  Purcell.   1967.   Spectrophotometric determination  of
hydrogen peroxide with 8-quinolinol.   Anal.  Chem.  39:131-132.

Corn, M. and L. Demaio.   1965.   Particulate sulfates in Pittsburgh air.   J.
Air Pollut. Control  Assoc.  15:26-30.

Coutant, R.  W.   1977.   Effect  of  environmental variables  on collection  of
atmospheric sulfate.  Environ. Sci.  Technol.  11:873-878.

Cox,  R.  A.   1977.   Some  measurements of ground  levels  of NO, N02,   and 03.
Concentration at an unpolluted maritime  site.  Tellus 29:356-362.

Crecelius,  E.  A.,   E.  A.   Lepel, J.  C.  Laul,  L. A. Rancitelli,  and R.   L.
McKeever.    1980.     Background  air  particulate  chemistry  near  Colstrip,
Montana.  Environ.  Sci.  Technol.  14:422-428.

Cronn, D.  R.,  R.  J.  Charlson,  R.  L. Knights,  A.  L. Crittenden,  and B.  R.
Appel.   1977.   A  survey of the  molecular nature of  primary and secondary
components of  particles  in urban air by  high-resolution mass  spectrometry.
Atmos. Environ. 11:929-937.

Dams, R. and J. De  Jonge.   1976.  Chemical  Composition of  Swiss  aerosols from
the Jungfraujoch.   Atmos.  Environ.  10:1079-1084.

Dams,  R.,  J.   A.   Robbins,  K.  A.  Rahn  and   J.   W.   Winchester.     1970.
Nondestructive  neutron  activation  analysis  of  air  pollution  particulates.
Anal. Chem. 42:861-867.

Darley,  E.  F., K.  A.  Kettner,  and  E.  R.  Stephens.    1963.   Analysis  of
peroxybeyl  nitrates by  gas  chromatography  with electron  capture  detection.
Anal. Chem. 35:589-591.
                                     5-86

-------
Davis, D.  D.,  G.  Smith and G. Klauber.   1974.   Trace gas analysis of  power
plant  plumes  via  aircraft  measurement:    03,   NOx,   and   S02   chemistry.
Science 186:733-736.

Decker, C.  E.,  L.  A.  Ripperton,  J. J. B. Worth, F. M. Vukovich, W. D.  Bach,
J. B. Tommerdahl, F. Smith and D. E. Wagoner.   1976.   Formation  and transport
of oxidants  along  Gulf Coast  and in northern U. S.   EPA-45013-76-033 Office
of Air Quality Planning and Standards, Research  Park,  NC.

Department  of  Health,  Education, and  Welfare.   1966.   Air Quality Data  from
the  National Air  Quality  Sampling  Networks  and Contributing State and  Local
Networks,  1964-1965.  Robert A. Taft Sanitary  Eng.  Center, Cincinnati, OH.

Drummond,  J.  W.   1976.    Atmospheric  measurements  of nitric oxide  using  a
chemiluminescent  detector.    Ph.D.  Dissertation,  University  of Wyoming,
La ramie.

Drummond,  J. W.  and A.  Volz.    1982.   Simultaneous  measurements  of  NO and
N0£  in  the  troposphere.     2nd   Symposium  Composition  of   the nonurban
Troposphere.  Williamsburg, VA American Meterological  Society,  45  Beacon St.
Boston, MA.

Duce, R. A., J.  W. Winchester, and T. W. Van  Nahl.   1965.   Iodine, bromine
and  chlorine   in   the  Hawaiian  marine  atmosphere.    J.  Geophys.   Res.
70:1775-1799.

Dzubay, T.  G.   1980.   Chemical element balance  method applied to dichotomous
sampler data.  Annal.  New York Acad. Sci.  338:126-144.

Dzubay, T. G.,  R. K. Stevens,  C.  W. Lewis, D.  H. Hern, W. J. Courtney,  J.  W.
Tesch,  and  M.  A.  Mason.    1982.    Visibility  and   aerosol  composition  in
Houston, TX.  Environ.  Sci. Techno!. 16:514-525.

Evans, G.,  P. Finkelstein, B.  Martin,  N.  Possiel and  M.  Graves.   1982.   The
National   Air   Monitoring   Background   Network   1976-1980.     Environmental
Monitoring Systems Laboratory, U.S. Environmental  Protection  Agency, Research
Triangle Park,  NC.

Farmer, C.  B.,  0.  F. Raper, and R.  H.  Norton.  1976.   Spectroscopic detection
and  vertical   distribution of   HC1  in  the  troposphere  and   stratosphere.
Geophys. Res. Let. 3:13-16.

Farmer, J.  C.  and  G.  A.  Dawson.   1982.    Condensation  sampling  of soluble
atmospheric trace gases.   J.  Geophys.  Res. 87:8931-8942.

Ferek R. J., A.  L. Lazrus, P.  L.   Haagenson, and  J.  W.  Winchester.    1983.
Strong and  weak  acidity  of aerosols collected  over the  northeastern United
States.   Environ.   Sci. Technol.  17:315-324.

Ferman,  M.  A.,  G.  T. Wolff  and N. A. Kelly.   1981.   The nature and source  of
haze  in the  Shenandoah Valley/Blue Ridge Mountains area.   J. Air Pollution
Control  Assoc.  31:1074-1082.
                                     5-87

-------
Flocchini, R. G., T.  A.  Cahlll,  D.  J.  Shadoan,  S.  J.  Lange,  R.  A.  Eldred,  P.
J. Feeney, G. W. Wolfe, D. C. Si mine roth, and J. K. Suder.  1976.   Monitoring
California's  aerosols by  size  and elemental  composition.    Environ.  Sci.
Technol. 10:76-82.

Flocchini, R. G., T.  A.  Cahill,  M.  L.  Pitchford,  R.  A. Eldred, P. J.  Feeney
and  L.  L.  Ashbaugh.   1981.  Characterization of particles in the  arid  west.
Atmos. Environ. 15:2017-2030.

Foote, R. S.  1972.  Mercury vapor inside buildings.   Science 177:513-514.

Forrest, J.,  R.  L.  Tanner, D.  Spandau, T.  D'Ottavio,  and  L. Newman.   1980.
Determination of total inorganic  nitrate utilizing collection of nitric  acid
on NaCl-impregnated filters.  Atmos. Environ.  14:137-144.

Forrest,  J.,  D.  J.  Spandau,   R.   L.   Tanner,   and  L.   Newman.     1982.
Determination of  atmospheric nitrate and  nitric acid employing a diffusion
denuder with a filter pack.  Atmos.  Environ. 16:1473-1485.

Frank, N. H. and N.  C. Possiel,  Jr.   1976.  Seasonality and  regional trends
in  atmospheric  sulfates.    1976.    Presented  before   Div.  of  Environmental
Chemistry American Chemical Society  San Francisco,  CA.

Gartrell,  G., Jr.  and  S.  K.  Friedlander.    1975.    Relating particulate
pollution to  sources:   The  1972 California aerosol  characterization study.
Atmos. Environ.  9:279-299.

Gay,  B.  W.,  Jr., P.  L.  Hanst,  J.   J.  Bufalini,  and  R.  C.   Noonan.    1976.
Atmosperic  oxidation  of  chlorinated  ethylenes.    Environ.  Sci.  Technol.
10:58-67.

Georgii, H.  W.    1978.   Large  scale  spatial  and temporal  distribution  of
sulfur compounds.  Atmos. Environ. 12:681-690.

Georgii, H.  W. and F. X.  Meixner. 1980.  Measurement  of the  tropospheric and
stratospheric S02 distribution.  J.  Geophys. Res. 85:7433-7438.

Georgii, H. W. and  W. A. Muller.   1974.   On  the  distribution of  ammonia  in
the middle and lower troposphere. Tellus 26:180-184.

Gillani, N.  V.,  R.   B.  Husar,  J.  D.  Husar,  and  D.  E.   Patterson.    1978.
Project MISTT:   Kinetics of particulate sulfur formation in  a power  plant
plume out to 300 km.   Atmos. Environ. 12:589-598.

Gillette, D.  A   and J.  W. Winchester.    1972.   A  study of  aging  of  lead
aerosols.  Atmos. Environ.  6:443-450.

Gladney,  E.   S.,  W.   H.  Zoller,  A. G.  Jones,  and   G.  E.  Gordon.     1974.
Composition and  size  distributions  of atmospheric  matter  in  Boston area.
Environ. Sci.  Technol. 8:551-557.
                                     5-88

-------
Gravenhorst, G.   1978.   Maritime sulfur  over  the North  Atlantic.   Atmos.
Environ. 12:707-713.

Groblicki,  P.  J.,  G.  T. Wolff,  and R.  J.  Countess.   1981.    Visibility-
reducing  species   in  Denver  "brown  cloud"  -   I.  Relationships  between
extinction and chemical composition.   Atm.  Environ. 15:2473-2484.

Grosjean,  D.    1981.    Critical  evaluation and   comparison  of  measurement
methods  for nitrogeneous compounds  in  the  atmosphere.   ERT  Document  no.
P-A706-04.  Prepared for CAPA-19  Project  Group,  Coordinating  Research Council
219 Perimeter Center Parkway, Atlanta,  GA.   30346.

Grosjean, D.  1983.  Distribution of atmospheric nitrogenous pollutants at a
Los Angeles area smog receptor site.   Environ.  Sci. Technol.  17:13-19.

Grosjean, D. and S.  K.  Friedlander.   1975.   Gas-particle  distribution factors
for organic and  other  pollutants in  the  Los  Angeles  atmosphere.   J.  Air
Pollut. Control  Assoc.  25:1038-1044.

Hanst, P.  1981.  Report of  Research  under  Innovative  Research program.  U.S.
Environmental  Protection Agency,  Environmental Sciences  Research  Laboratory,
Research Triangle Park, NC.

Hanst,  P.  L.,  W. E.  Wilson, R.  K.  Patterson,  B. W.  Gay,  Jr.,  and  L.  W.
Chaney.  1975.  A spectroscopic study of California smog.  EPA-65014-75-006.
Research Triangle Park, NC.

Hanst, P. L.,  N. W.  Wong and J.  Bragin.   1982.  A long-path infra-red  study
of Los Angeles smog.  Atmos. Environ.  16:969-981.

Hardy,  K.  A.,  R.  Akselsson,  J.  W.  Nelson,  and  J.  W. Winchester.    1976.
Elemental  constituents  of   Miami  aerosol  as  function of  particle   size.
Environ. Sci.  Technol.  10:176-182.

Harker,  A.  B.,  L.  W.  Richards,  and W.  E.  Clark.   1977.  The  effects  of
atmospheric  S02  photooxidation   upon   observed  nitrate  concentrations  in
aerosols.  Atmos. Environ.  11:87-91.

Harward, C.  N.,  W.  A.  McClenny, J.  M.  Hoell,  J. A.  Williams,  and  B.  S.
Williams.   1982.    Ambient  ammonia  measurements  in  coastal   southeastern
Virginia. Atmos. Environ. 16:2497-2500.

Hegg,   D.,  P.  V.  Hobbs, L.  F.  Radke, and  H.  Harrison.   1977.   Ozone  and
nitrogen  oxides  in   power  plant   plumes,   pp.   173-183.     Jji  Vol.  1,
International  Conference on  Photochemical Oxidant  Pollution and its Control.
B. Dimitriades,  ed.   EPA-600/13-77-0012.    Environmental  Sciences Research
Laboratory, Research Triangle Park, NC.

Heikes,  B.  G.,  A.  L.  Lazrus, G.  L.  Kok, S. M.  Kunen,  B.  W. Gandrud, S.  N.
Gitlin and P.  D. Sperry.  1982.   Evidence of aqueous  phase hudrogen peroxide
synthesis in the troposphere.  J.  Geophys.  Res. 87:3045-3051.
                                     5-89

-------
 Helas,  G.  and P.  Warneck.    1981.   Background  NOX  mixing  ratios in  air
 masses over the North Atlantic Ocean.  J. Geophys. Res. 86:7283-7290.

 Hester,  N. E.,  R.  B.  Evans,  F.   G.  Johnson  and E.  L.  Martinez.    1977.
 Airborne  measurements  of primary  and  secondary  pollutant  concentrations  in
 the  St.  Louis  urban plume,  pp.257-274.  _I_n  Vol.  1,  International  Conference
 on  Photochemical  Oxidant Pollution and  its  Control.    B.  Dimitriades,  ed.
 EPA-600/3-77-0012   Environmental   Sciences  Research   Laboratory,   Research
 Triangle  Park, NC.

 Hidy,  G.  M.,  P.  K. Mueller,  and  E.  Y. long.   1978.   Spatial  and  temporal
 distributions  of  airborne sulfate  in  parts  of the United  States.    Atrnos.
 Environ.  12:735-752.

 Hilst,  G. R., P.  K. Mueller, G.  M.  Hidy, T.  F.  Lavery, and J. G.  Watson.
 1981.  EPRI Sulfate  Regional  Experiment:  Results and Implications.   Electric
 Power Research Institute, Palo Alto, California.  Report No. EA-2165-SY-LD.

 Hoell,  J. M., J.  S.  Levine,  T.  R.  Augustsson  and  C.  N.  Harward.   1983.
 Atmospheric ammonia:  measurements and modeling.  AIAAJ. In press.

 Hoggan, M., A. Davidson, D.  C. Shikiya and W. Lau.  1982.   Air Quality Trends
 in  California's  South Coast  Air.  Basic South  Coast  Air  Quality  Management
 District, 9150 East  Flair Drive, El Monte, CA.

 Houston Area Oxidant Study (HAOS).  1979.   Program Summary.   Houston Chamber
 of Commerce, 1100 Mi lain St.  Houston, TX.

 Huebert,  B.  J.  1980.   Nitric acid and aerosol  nitrate  measurements  in  the
 equatorial Pacific region.  Geophys. Res. Lett.  7:325-328.

 Huebert,  B. J. and A. L. Lazrus.   1978.   Global  tropospheric measurements  of
 nitric acid vapor and particulate nitrate.   Geophys.  Res.  Lett.  5:577-580.

 Huebert,  B. J. and A. L. Lazrus.  1980a.  Bulk composition of aerosols in  the
 remote troposphere.  J. Geophys. Res.  85:7337-7344.

 Huebert,  B. J.  and  A.   L.  Lazrus.    1980b.     Tropospheric  gas-phase  and
 particulate nitrate measurements.  J.  Geophys. Res. 85:7322-7328.

 Husar,  R. B.  and  D.  E.  Patterson.   1980.  Regional  Scale Air  Pollution:
 Sources and Effects.  New York Academy of Sciences.

 Husar,  R. B., D.  E.  Patterson,  J. M.  Holloway, W.  E.   Wilson,  and T.  G.
 Ellestad.   1979.   Trends of  eastern  U.S.  haziness since  1948,  pp.  249-256.
Jjn Proceedings of  the  Fourth Symposium on  Atmospheric  Turbulence,  Diffusion
 and Air Pollution, American  Meteorological  Society, Reno,  NV.
                                     5-90

-------
Husar, R. B.,  D.  E.  Patterson,  C.  C.  Paley and N. V. Gillani.  1977.  Ozone
in  hazy  air  masses.    pp.  275-282.    In  Proceedings of  the  International
Conference on  Photochemical  Oxidant  Pollution and  Its Control.   Vol  I.   B.
Dimitriades,   ed.   EPA-600/3-77-0012.      Environmental   Sciences   Researh
Laboratory.  U. S. Environmental Protection Agency.   Research Triangle Park,
NC.

Jacobs, M.  B.   1959a.   Concentration of  sulfur-containing  pollutants  in  a
major  urban  area, pp.  81-87.   _In Monograph  No.  3.   American  Geophysical
Union, Proceedings of a Symposium on  Atmospheric  Chemistry  of Chlorine and
Sulfur Compounds.   J. P. Lodge,  ed. Waverly Press,  Inc., Baltimore, MD.

Jacobs, M.  D.  1959b.   Techniques  for measurement  of  hydrogen sulfide  and
sulfur oxides, pp. 24-36.  _Iji Monograph  No. 3.  American Geophysical Union,
Proceedings of  a  Symposium  on Atmospheric  Chemistry of  Chlorine  and Sulfur
Compounds.  J. P.  Lodge, ed.   Waverly  Press, Inc.,  Baltimore,  MD.

Junge, C. E.   1954.   The chemical composition of  atmospheric aerosols.   I.
Measurements at the Round Hill Field Station, June-July, 1973.  J. Meteorol.
11:323-333.

Junge, C. E.    1956.   Recent   investigations  in   air  chemistry.   Tell us
8:127-139.

Junge, C. E.   1963.   Air Chemistry and Radioactivity.   Academic Press, New
York.

Kadowaki,   S.    1977.    Size   distribution  and   chemical   composition  of
atmospheric  particulate  nitrate  in   the   Nagoya   area.    Atmos.  Environ.
11:671-675.

Kelly, T. J.  and  D.  H.  Stedman,  1979a.    Measurements of  H202  in rura1
air.  Geophys. Res. Lett.  6:375-378.

Kelly, T. J.  and  D.   H. Stedman.   1979b.   Chemiluminescence measurements of
HNOa  in  air.   In current methods to  measure  atmospheric nitric  acid  and
nitrate artifacts. (  R. K. Stevens, ed.)    EPH-600-12-79-051.    Environmental
Sciences  Research  Laboratory,  US EPA Research  Triangle Pk, NC  27711.

Kelly, T. G.,  D.  H.  Stedman,  J. A.  Ritter,  and  R.  B.  Harvey.    1980.
Measurements of oxides of  nitrogen  and  nitric  acid  in clean air.  J. Geophys.
Res. 85:7417-7425.

Kelly, N.  A.,  G.  T.  Wolff,  and  M.  A.  Ferman.   1982.  Background pollution
measurements in air masses affecting the  eastern half on the United  States.
I. Air masses arriving from the  northwest.   Atmos.  Environ. 16:1077-1088.

Kley, D.,  J. W. Drummond, M.   McFarland, and S. C.  Liu.   1981.  Tropospheric
profiles  of NOX.  J.  Geophys.  Res.  86:3153-3161.
                                     5-91

-------
Kok, G.  L.  1980.   Measurements  of hydrogen peroxide in  rainwater.   Atmos.
Environ. 14:653-656.

Kok, G. L.  1982.  Measurements  of formaldehyde  in the California South Coast
Air  Basin.    Report on  EPA Grant  no.  CR-806629 to  Environmental  Sciences
Research Laboratory.  Research Triangle  Park, NC  27711.

Kok, G. L.,  K. R. Darnall,  A. M.  Winer,  J.  N.  Pitts, Jr., and B. W. Gay, Jr.
1978a.  Ambient air measurements  of hydrogen peroxide in the California South
Coast Air Basin.  Environ.  Sci.  Techno!.  12:1077-1080.

Kok,  G.  L.,  T.  Holler,  M.  Lopez,  H.  Nachtrieb,  and  M.  Yuan.    1978b.
Chemileminescent method for determination of hydrogen peroxide in the ambient
atmosphere.   Environ.  Sci.  Technol.  12:1072-1076.

Kumar, R., S.  A. Johnson,  and  P.  T. Cunningham.   1982.  Seasonal and diurnal
variations  in  the  chemistry   of  ambient  fine-particle  aerosols  in  the
northeastern  United  States.     In  2nd  Symposium   Composition  of  Nonurban
Troposphere.  Amer. Meteorol. Soc.,  45  Beacon St., Boston, MA, 02108.

Laird,  A. R.  and  R.  W.  Miksad.   1978.   Observations on  the particulate
chlorine  distribution  in  the   Houston-Gal veston   area.    Atmos.  Environ.
12:1537-1542.

Leaderer, B.  R.,  T. R.  Holford,  and J. A.  J.  Stolwijk.  1979.  Relationship
between  sulfate aerosol and  visibility.   J.  Air  Pollution  Control  Assoc.
29:154-157.

Lee,  R.  E.,   Jr.  and  P.  K.  Patterson.    1969.    Size  determination  of
atmospheric phosphate, nitrate,  chloride, and ammonium particulate in several
urban areas.  Atmos. Environ. 3:249-255.

Lee,  R.  E.,  Jr.,  R.  K.  Patterson,  and J. Wagman.    1968.    Particle-size
distribution  of  metal  components  in  urban air.    Environ.  Sci.  Technol.
2:288-290.

Lee,  R.  E.,  Jr.  and  J.  Wagman.    1966.    A  sampling  anomaly  in  the
determination  of  atmospheric  sulfate   concentration.    J.  Am.   Ind.  Hyg.
27:266-271.

Lewis,  C. W.  and E.  S.  Macias.   1980.   Composition  of  size-fractionated
aerosol in Charleston, West Virginia.   Atmos. Environ.  14:185-194.

Lioy,  P.  J.,  P.  J.  Samson,  R. L. Tanner, B. P.  Leaderer, T. Minnich, and W.
Lyons.  1980.  The distribution  and transport of sulfate "species" in the New
York Metropolitan area during  the 1977 summer aerosol  study.   Atmos. Environ.
14:1391-1407.

Lodge,  J. P.  and J. B. Pate.   1966.   Atmospheric  gases and particulates in
Panama.   Science  153:408-410.
                                     5-92

-------
Lodge, J. P., P. A. Machado, J. B. Pate, D.  C.  Sheesley, and A. F. Wartburg.
1974.   Atmospheric  trace chemistry in  the  American humid  tropics.   Tellus
26:250-253.

Long,  S.  J., D.  R. Scott, and  R. J.  Thompson.   1973.   Atomic absorption
determination of elemental  mercury collected  from ambient air on silver wool.
Anal. Chem. 45:2227-2233.

Lonneman, W.  A.,  J. J.  Bufalini,  and  R.  L.  Scila.   1976.   PAN and oxidant
measurement in ambient  atmosphere.  Env.  Sci.  Technol.   10:374-380.

Ludwick, J. D., T. D.  Fox,  and  S.  R.  Garcia.   1977.   Elemental concentrations
of  northern  hemispheric  air   at  Quillayute,  Washington.    Atmos.  Environ.
11:1083-1087.

Ludwig, F. L. and E. Robinson.  1968.   Variations  in the size distributions
of sulfur-containing compounds  in  urban  aerosols.  Atmos. Environ. 2:13-23.

Lundgren, D.  A.   1967.   An aerosol  sampler for  determination  of particle
concentration as a function of  size and time.  J. Air Pollut. Control Assoc.
17:225-229.

Lundgren, D. A.  1970.   Atmospheric aerosol composition and concentration as
a  function  of  particle  size  and  time.   J.  Air.  Pollution  Control  Assoc.
20:603-608.

Macias, E.  S.,  D.  L.  Blumenthal,  J.  A.  Anderson, and B. K. Cantrell.  1980.
Size and composition of  visibility-reducing  aerosols in southwestern plumes.
Ann. New York Acad.  Sci.  338:233-257.

Macias,  E.  S.,  J. 0.  Zwicker,  J.  R.  Ouimette,  S.  V. Hering,  S.  K.
Friedlander,  T.  A.  Cahill, G.  A.  Kuhlmey,  and  L.  W.  Richards.    1981.
Regional  haze  case studies in the  southwestern  U.S.    I.  Aerosol  chemical
composition.  Atm. Environ.  15:1971-1986.

Maroulis,  P.  J., A.  L.  Torres,  A.  B.  Goldberg,  and  A.  R.  Bandy.   1980.
Atmospheric   SO?  measurements  on   Project   Gametag.     J.  Geophys.  Res.
85:7345-7349.

Martens, C. S., J.  J.  Wesolowski,  R   Kaifer, and W. John.   1973.   Lead and
bromine size  distributions  in  the San  Francisco Bay  area.   Atmos. Environ.
7:905-914.

Martinez, J.  R. and  H.  B. Singh.    1979.    Survey  of the  Role  of  NOX  1n
Nonurban Ozone Formation.  SRI  Project 6780-8.   U.S. Environmental Protection
Agency, Research Triangle Park, NC.

Martinez, J. R., F. L. Ludwig, and C. Maxwell.   1982.   1978 Houston oxidant
modeling study.   Vol.  1:  Data evaluation and  analysis.   SRI  Project 7938.
Environmental  Sciences Laboratory.   U.  S.  Environmental  Protection Agency,
Research Triangle  Park,  NC.
                                     5-93

-------
Mayrsohn, H. and  C.  Brooks.   1965.  The analysis of PAN by electron  capture
gas  chromatography.    Presented  at  the  Western  Regional  Meeting  of the
American Chemical  Society, Nov.  18.

McClenny, W.  A.  and  C.  A.  Bennett,  Jr.   1980.   Integrative technique for
detection of atmospheric  ammonia.   Atmos.  Environ.  14:641-645.

McClenny, W.  A.,   P.  C.   Gailey,  R.  S.  Braman, and  T.  J.  Shelley.    1982.
Tungstic acid  technique  for monitoring  nitric acid and  ammonia  in  ambient
air.  Anal.  Chem.  54:365-369.

Meinert, D. L. and J. W. Winchester.   1977.   Chemical  relationships in the
North Atlantic marine aerosol.   J.  Geophys.  Res. 82:1778-1782.

Meserole, F.  B.,   K.  Schwitzgebel,  B. F. Jones,  C.  M.  Thompson,  and  F. G.
Mesich.  1976.   Sulfur dioxide interferences  in  the measurement  of  ambient
particulate sulfates.   Radian  Corp.  Report  (Project  262)  to Electric  Power
Research Inst., Palo Alto, CA.

Meszaros, E.   1978.   Concentration  of sulfur  compounds  in  remote  continental
and oceanic areas.  Atmos. Environ. 12:699-705.

Moskowitz,  A.  H.   1977.   Particle size distribution of nitrate aerosols in
the Los  Angeles Air  Basin.  EPA-600/3-77-U53.   U.S.  Environmental  Protection
Agency,  Research Triangle Park, NC.

Mueller,  P. K.,  G.  M.  Hidy,  K.  Warren, T.  F. Lavery, and  R.  L.  Baskett.
1980.    The occurence  of atmospheric  aerosols in the  north eastern  United
States.  Ann.  New  York Acad. Sci.  338:463-482.

National  Academy   of  Sciences.    1975.    Nickel.   Committee  on  Medical and
Biologic Effects   of  Environmental  Pollutants.   Div.  of  Medical  Sciences.
National Research  Council.  ISBN 0-309-02314-9.

National Academy  of Sciences.   1977.  Nitrogen oxides.  ISBNO-309-  02615-6.
Printing and   Publishing  Office,   National   Academy   of   Sciences,   2101
Constitution Ave., NW Washington, D.C.

Noxon, J. F.   1978.  Tropospheric N02.  J. Geophys. Res. 83:3051-3057.

O'Brien, R. J.,  J.  H.  Crabtree,  J. R.  Holmes,  M.  C.  Hoggan,  and  A. H.
Bockian.    1975.     Formation  of  photochemical  aerosol  from  hydrocarbons.
Atmospheric analysis.  Environ. Sci.  Technol. 9:577-582.

Okita,  T., S. Morinoto,  M.  Izawa,  and  S.  Konno.   1976.   Measurement of
gaseous  and   particulate  nitrates   in  the  atmosphere.    Atmos.   Environ.
10:1085-1089.

Ottar,  B.   1978.   An assessment of the OECD study on long  range transport of
air pollutants (LRTAP).   Atmos. Environ. 12:445-454.
                                     5-94

-------
Paciga, J. J. and R. E. Jervis.  1976.   Multielement  size  characterization  of
urban aerosols.  Environ. Sci.  Techno!.  10:1124-1128.

Patterson, R.  K.  and J. Wagman.   1977.   Mass and  composition  of an  urban
aerosol as a  function  of particulate size for several  visibility  levels.  J.
Aerosol Sci.  8:269-279.

Pierson, W. R., J.  W.  Butler,  and  D. A.  Trayser.   1974.  Nitrate  and  nitric
acid  emissions  from catalyst-equipped  automotive  systems.   Environ.  Letts.
7:267-272.

Pierson, W. R., W. H. Mammerle, and W.  W. Brachaczek.   1976.   Sulfate  formed
by  interaction  of   S02  with  filters  and  aerosol   deposits.    Anal.   Chem.
48:1808-1811.

Pierson,  W.  R.,  W.  W.  Brachaczek,  T.  J.  Truex,  J.  W.  Butler,  and  T.  J.
Korniski.  1980a.  Ambient  sulfate  measurements on  Allegheny Mountain and the
question of atmospheric sulfate in the northeastern United States.  Ann. New
York Acad. Sci. 338:145-173.

Pierson,  W.  R., W.  W.  Brachaczek, T.  J.  Korniski,  T. J.  Truex,  and  J.  W.
Butler.  1980b.   Artifact  formation  of  sulfate,  nitrate and hydrogen  ion  on
backup  filters:    Allegheny  mountain experiment.    J. Air  Pollut.  Control
Assoc. 30:30-34.

Pilz, W, and  Johann,  I.   1974.  Die bestirnmung  kleinster mengen  von  wasser
stoff peroxyd in luft.   Int.  J. Environ.  Anal.  Chem.  3:257-270.

Pitts,  J.  N.,  Jr.  and  D. Grosjean.    1979.    Detailed  characteristics  of
gaseous and size-resolved  particulate pollutants at  a  south coast air  basin
smog  receptor site.   PB-302,  294 National  Technical  Information Service,
Springfield,  VA.
Platt, U.  and  D.  Perner.   1980.   Direct measurements  of atmospheric
HN02,  03,   N02,  and  SO?  by  differential  optical  absorption  in  the near
UV. J. Geophys. Res.  85:7453-7458.

Prahm, L. P., U. Torp, and R.  M.  Stern.   1976.   Deposition and transformation
rates  of  sulfur  oxides  during  atmospheric  transport  over the  Atlantic.
Tellus 28:355-372.

Rasmussen,   R.  A.,  R.  Chatfield,  and M.  Holdren.   1977.   Hydrocarbon and
oxidant  chemistry  observed at a  site  near St.  Louis.   EPA-600/17-77-056.
Environmental Sciences Research Laboratory.  Research  Triangle Park,  NC.

Renzetti, N. A. and R. J.  Bryan.   1961.   Atmospheric sampling for  aldehydes
and eye  irritation in  Los  Angeles  Smog-1960.   J. Air Pollut. Control  Assoc.
11:421-424.

Research Triangle  Institute.  1975.   Investigation of  rural oxidant  levels as
related to  urban control  strategies.   EPA-450/3-75-036.   U.S. Environmental
Protection  Agency,  Research Triangle  Park, NC.


                                     5-95

-------
Ripperton, L. A., L. Kornreich, and J. J. B. Worth.  1970.  Nitrogen Dioxide
and nitric oxide in non-urban  air.  J. Air Pollut. Control Assoc. 20:589-592.

Robinson, E. and F.  L.  Ludwig.   1967.  Particle  size  distribution  of urban
lead aerosols.   J.  Air Pollut.  Control Assoc. 17:664-669.

Roesler,  J. F.,   H.  J.  R.  Stevenson,  and  J.  S.  Nader.    1965.    Size
distribution of sulfate  aerosols in the  ambient air.  J. Air Pollut. Control
Assoc. 15:576-579.

Rohlach,  L.  A., W.  C.  Hawn,   K.  R.  Williams,  and T.  P.  Parsons.   1979.
Nitrogen oxide  interferences  in the measurement  of atmospheric participate
nitrate.   EPRI EA-1031,  Research  Project 801-1.   Electric  Power  Research
Institute, 3412 Hillview Ave.,  Palo Alto, CA.

Schurr,  S.  H.,  B.  C.  Netschert,  V. F.  Eliasberg,  J. Lerner,  and  H.  H.
Landsberg.   1960.    Energy in the  American economy  1850-1975.    The John
Hopkins Press,  Baltimore, MD.

Sexton, K.  and  H.  Westberg.   1980.   Elevated  ozone concentrations measured
downwind of  the Chicago-Gary  Urban complex.   J.  Air  Pollut.  Control Assoc.
30:911-914.

Shaw,  R.  W. and R.  J.   Paur.    1983.    Measurements of sulfur  in  gases and
particles during sixteen months in the  Ohio River  Valley.   Atmos. Environ.
17:1431-1438.

Shaw, R. W., T. G.  Dzubay, and R. K.  Stevens.  1979.  The denuder difference
experiment.   Current methods to measure  atmospheric nitric  acid and nitrate
artifacts,  pp.  79-84.   Report  EPA-600/12-79-051.   R.  K. Stevens,  ed.   U.S.
Environmental Protection Agency, Research  Triangle Park, NC.

Shaw, R. W., R. J.  Paur, and T. Royal.   1981.   Ohio  River Valley Study  Sites,
methods, data summary for 1980.  Environmental  Sciences Research Laboratory,
Research Triangle Park,  NC.

Shaw, R. W., Jr., R. K.  Stevens, and  J.  Bowermaster.  J. W.  Tesch and E. Tew.
1982.   Measurements of  atmospheric  nitrate  and  nitric acid:   The denuder
difference experiment.  Atmos.  Environ.  16:845-853.

Singh,  H. B.,  F.  L. Ludwig, and W. B.  Johnson.   1978.   Tropospheric  ozone:
Concentrations  and  variabilities   in  clean  remote  atmospheres.     Atmos.
Environ. 12:2185-2196.

Singh,  H.   B.,  L.  J.  Salas,  and  L.  A.  Cavanaugh.   1977a.    Distribution,
Sources  and sinks  of  atmospheric  halogenated compounds.   J.  Air Pollut.
Control Assoc.  27:332-336.

Singh,  H.  B.,  L.  J.   Salas,  H.  Shigeishi,  and A.  Crawford.     1977b.
Urban-nonurban   relati9nships   of   halocarbons,   SF6,    NeO   and   other
atmospheric trace constituents. Atmos. Environ. 11:819-828.
                                     5-96

-------
 Singh,  H.  B.,  L.  J.  Salas,  H.  Shigeishi, A.  J. Smith, E. Scribner, and L. J.
 Cavanaugh.   1979.   Atmospheric  distributions,  sources,  and  sinks of selected
 holocarbons,   hydrocarbons,  SF,  and   N02.      EPA-600/3-79-107,   Research
 Triangle Park, NC.

 Singh,  H.   B.,   L.  J.  Salas,  A.  J.  Smith,  and  H.  Shigeishi.    1981.
 Measurements  of  some  potentially  hazardous  organic  chemicals  in  urban
 environments.  Atmos. Environ. 15:601-612.

 Singh,  H.  B., L.  J.  Salas,  R. Stiles, and  H.  Shipeishi.   1982.   Measurements
 of  hazardous organic chemicals  in the  ambient  atmosphere.   Report  to  EPA.
 Cooperatative Agreement 805990 to Environmental Sciences Research Laboratory.
 Research Triangle Park, NC.

 Siple,  G.  W., C.  K.  Fitzsimmons,  K.  F.  Zeller, and R. B. Evans.  1977.  Long
 range airborne measurements of ozone off the coast of the northeastern United
 States,  pp.249-258.    In  Vol.  1,  International  Conference on  Photochemical
 Oxidant  Pollution and its  Control.   B  Dimitriades, ed.   EPA-600/3-77-0012
 Environmental  Sciences  Research  Laboratory,  Research  Triangle  Park,  NC
 27711.

 Sloane, C.  S.   1982a.   Visibility trends  -  I. Methods of analysis.   Atmos.
 Environ. 16:41-51.

 Sloane, C. S.   1982b.   Visibility  trends -  II.    Mideastern  United  States
 1948-1978.  Atmos. Environ.   16:2309-2321.

 Spicer, C.  W.   1977.   The  fate  of  nitrogen  oxides  in  the atmosphere,  pp.
 163-261.   In Advances in Environmental  Science and Technology.   Vol  7.  J. N.
 Pitts and "RT L. Metcalf, eds.   J.  Wiley & Sons, New York.

 Spicer, C.  W.   1979.   Measurement  of gaseous  HN03  by electrochemistry  and
 chemiluminescence in current  method to  measure  atmospheric  nitric acid  and
 nitrate artifacts.  R.  K.  Stevens, ed.   EPA 600/12-79-051.

 Spicer, C.  W.  and D. F.  Miller.   1976.   Nitrogen  balance  in   smog  chamber
 studies.  J. Air Poll.  Control  Assoc. 26:45-50.

 Spicer,  C.  W.  and  P.  M.  Schumacher.     1977.    Interferences  in  sampling
 atmospheric particulate nitrate.  Atmos.  Environ.  11:873-876.

 Spicer, C. W. and P.  M.  Shumacher.  1978.  Studies in effect of  environmental
 variables  on the  collection of  atmospheric nitrate and the development  of a
 sampling   and   analytical   nitrate   method.      EPA-600/2-78-009.      U.S.
 Environmental Protection Agency, Research Triangle Park,  NC.

Spicer, C.  W. and P. M.  Schumacher.   1979.   Particulate nitrate;  laboratory
and  field  studies  of  major  sampling   interferences.    Atmos.   Environ.
 13:543-552.
                                     5-97

-------
Spicer, C. W. and G. W. Sverdrup.  1981.   Trace nitrogen chemistry  during  the
Philadelphia oxidant data enhancement  study.   1979.   Report on contract  no.
68-02-0338   to   Office  of   Air  Quality   Planning   and  Standards.  U.S.
Environmental Protection Agency.  Research Triangle Park,  NC.

Spicer, C.  W.,  J.  L.  Gemma, D.  W.  Joseph,  P. R.  Strickel,  and G. F.  Ward.
1976a.  The Transport  of Oxidant beyond Urban Areas.   EPA-600/3-76-018.  U.S.
Environmental Protection Agency, Research Triangle Park, NC.

Spicer, C.  W.,  J.  L.  Gemma, P.  M.  Schumacker  and G. F.  Ward.   1976b.   The
fate  of  nitrogen oxides in the  atmosphere.   Second year report by Battelle
Columbus Lab. to Coordinating  Research Council  (CAPA-9-71).

Spicer, C.  W.,  M.  W.  Holdren and  G.  W. Keigley.   1983.   The ubiquity  of
peroxyacetyl nitrate in the lower atmosphere.  Atmos.  Environ.  17:1055-1058.

Spicer, C.  W.,  J.  E.  Howes,  Jr., T.  A. Bishop,  L.  H. Arnold,  and R.  K.
Stevens.    1982a.   Nitric  acid measurement  methods:    An  intercomparisen.
Atmos. Environ. 16:1487-1500.

Spicer, C.  W.,  D.  W.  Joseph,  P. R. Sticksel,  and G.  F. Ward.  1979.   Ozone
sources and transport  in  the  northeastern  United  States.   Environ.  Sci.
Technol. 13:975-985.

Spicer, C. W., D. W. Joseph, and P. R. Sticksel.   1982b.  An investigation of
the ozone plume from a small city.   J. Air pollut. Control  Assoc. 32:278-281.

Spicer, C.  W.,  J.  R.  Koetz, G.  W.  Keigley,  G. M. Sverdrup, and G. F.  Ward.
1982c.   Nitrogen oxides reactions within  urban  plumes transported over  the
ocean.  Report  on contract  no.  68-02-2957 to Environmental  Sciences Research
Laboratory, U.S. Environmental Protection Agency, Research Triangle Park,  NC.

Stevens, R. K., T. G.  Dzubay,  D. T. Mage, R.  Burton,  G.  Russwurm, and  E.  Tew.
1978.  Comparison of  Hi-Vol and dichotomous sampler results on nitrates  and
sulfates.   Div.  of  Environ. Chem.  Amer. Chem. Soc.  176th  National Meeting,
Miami, FL.

Stevens, R.  K.,  T.  G.  Dzubay,  R. W.  Shaw, Jr., W. A. McClenny, C.  W.  Lewis,
and W. E. Wilson.  1980.  Characterization of the aerosol  in the Great  Smokey
Mountains.  Environ. Sci. Technol.  14:1491-1498.

Stevens, R. K., W. A.  McClenny, T.  G.  Dzubay, M.  A. Mason and W. J. Courtney.
1982.   Analytical  methods  to  measure  the carbonaceous content of aerosols.
In  Particulate  Carbon:   Atmospheric   Life  Cycle.   G.  T.  Wolff   and  R.  L.
TTTimisoh, eds.  Plenum Press,  New York.

Struempler,  A.  W.     1975.     Trace  element  composition  in atmospheric
particulates during 1973 and  the summer of  1974  at  Chadron, Neb.    Environ.
Sci.  Technol. 9:1164-1168.
                                     5-98

-------
Swinford,  R.   1980.   Vertical  ozone profile  in  the lower troposphere over
Chicago.  J. Air Pollut. Control  Assoc.  30:794-796.

Tanner, R.  L.  1982.   An ambient experimental  study  of phase equilibrium  in
the atmospheric system:  Aerosol  H*,  m$+,  S042',  N0a_  -  NH3(g),  HN03(g).
Atmos. Environ. 12:2935-2942.

Tanner  R.  L.  and  W.  H. Marlow.   1977.   Size discrimination  and chemical
composition of ambient  airborne  sulfate  particles  by  diffusion  sampling.
Atmos. Environ. 11:1143-1150.

Tanner, R.  L.,  R.  Cederwall,  R. Garber,  D.  Leahy,  W. Marlow,  R. Meyer,   M.
Phillips, and L. Newman.   1977.   Separation and analysis of aerosol  sulfate
species at ambient  concentrations.   Atmos.  Environ. 11:955-966.

Tanner, R.  L.,  W.  H.  Marlow,  and  L.  Newman.   1979.   Chemical  composition
correlations of size-fractionated sulfate  in  New York City Aerosol.  Environ.
Sci. Technol.  13:75-78.

Taylor,  0.  C.    1969.    Importance  of  peroxyacetyl   nitrate   (PAN)  as  a
phytotoxic air pollutant.   J.  Air Pollut.  Control  Assoc.  19:347-351.

Temple, P.  J.  and  0.  C. Taylor.   1983.  World-wide  ambient measurements  of
peroxyacetyl  nitrate  (PAN)  and  implications  for  plant  injury.    Atmos.
Environ. 17(8):1583-1587.

Tesche, T. W., J.  A.  Ogren  and  D. L.  Blumenthal.  1977.   Ozone concentrations
in  power  plant  plumes, pp. 157-171.   Ir± Comparison  of  Models  and Sampling
Data.   Vol. II,  International  Conference on Photochemical Oxidant  Pollution
and Its Control.  B. Dimitriades,  ed.    EPA-600/13-77-0012.    Environmental
Sciences Research  Laboratory,  Research  Triangle Park, NC.

Trijonis, J.    1978.    Empirical  relationship  between  atmospheric nitrogen
dioxide and its precursors. EPA-600/3-78-018.   U.S.  Environmental Protection
Agency, Research Triangle Park,  NC.

Trijonis, J.  and  S. Mortimer.   1982.    Validation of the  EKMA model using
historical  air  quality  data.    EPA-600/3-82-015.    Environmental  Sciences
Research Laboratory,  Research  Triangle  Park,  NC.

Trijonis,  J.  and  K.  Yuan.   1978a.    Visibility  in  the  southwest.     An
exploration of this  historical  data  base.    EPA   Report 600/3-7-039   to
Environmental  Sciences Research Laboratory, Research  Triangle Park,  NC.

Trijonis, J.  and K.  Yuan.   1978b.   Visibility in the northeast.   Long-term
visibility  trends   and  visibility/pollutant  relationships.     EPA  Report
600/3-78-075 to Environmental  Sciences  Research Laboratory,  Research Triangle
Park, NC.
                                     5-99

-------
Tuazon, E. C., R. A. Graham, A. M. Winer, R.  R.  Easton,  and J. N. Pitts, Jr.
1978.  A kilometer pathlength Fourier-transform  infrared  system for the study
of trace  pollutants  in ambient and  synthetic atmospheres.   Atmos. Environ.
12:865-875.

Tuazon, E.  C.,  A.  M.  Winer,  R.  A.  Graham,  and J.  N.  Pitts,  Jr.   1980.
Atmospheric measurements  of trace pollutants by kilometer-pathlength FT-IR
spectroscopy,   pp. 254-300.    In^  Advances  in   Environmental  Sciences  and
Technology, Vol.  10.   J.  N.  Pitts  and R. L. Metcalf,  eds.  J. Wiley & Sons,
New York.

Tuazon, E.  C., A. M.  Winer, R. A.  Graham,  and J.  N.  Pitts, Jr.   1981a.
Atmospheric measurements  of trace  pollutants:    Longpath  Fourier transform
infrared  spectroscopy.    EPA  Grant  No.  R-804546.     U.S.  Environmental
Protection  Agency,   Environmental   Sciences  Research Laboratory,  Research
Triangle Park, NC.

Tuazon, E. C., A. M.  Winer,  and J,  N.  Pitts, Jr.  1981b.   Trace pollutant
concentrations in a multiday  smog  episode in the California South Coast Air
Basin  by  long path  Fourier transform infrared  spectroscopy.   Environ. Sci.
Technol. 15:1232-1237.

U.S.   Environmental  Protection Agency.    1977a.    National  air  quality  and
emission  trends   report  1976.   EPA-450/1-77-002.    Office  of  Air  Quality
Planning and Standards, Research Triangle Park,  NC.

U.S.  Environmental Protection Agency. 1977b.  Air quality criteria for lead.
EPA-600/8-77-017.  Office of Research and Development, Washington, D. C.

U.S.  Environmental Protection Agency. 1978a.  Air quality criteria for ozone
and  other  photochemical  oxidants.   EPA-600/8-78-004.    Superintendent  of
Documents, U.S. Printing Office, Washington, D.  C.

U.S.   Environmental  Protection Agency.    1978b.    National  air  quality  and
emission  trends   report  1977.   EPA-450/2-78-052.    Office  of  Air  Quality
Planning and Standards, Research Triangle Park,  NC.

U.S.  Environmental Protection Agency.  1979.  Protecting visibility.  An EPA
report  to  Congress.   EPA-450/5-79-008.    Office  of Air Quality Planning and
Standards, Research  Triangle Park,  NC.

U.S.   Environmental   Protection  Agency.     1982.    Draft  final  air  quality
criteria for  oxides  of nitrogen.   EPA-600/8-82-026.  Environmental Criteria
and Assessment Office,  Research Triangle  Park, NC.

U.S.  Geological Survey.   1970. Mercury  in  the  environment, a  compilation of
papers on the abundance, distribution and testing of mercury in vocks, soils,
waters,  plants  and  the  atmosphere.   Professional   Paper  No.  713.    U.S.
Government Printing Office.  Washington,  D.  C.
                                     5-100

-------
 Viezee,  W. and H.  B.  Singh.   1982.  Contribution  of  Stratospheric  Ozone to
 Ground-Level  Ozone  Concentrations—A  Scientific  Review of  Existing Evidence.
 Report on Grant  CR809330010  to Environmental Sciences  Laboratory,  Research
 Triangle,  NC.

 Vukovich,  F.  M.,  W. D. Bach,  Jr.,  B.  W. Crissman, and W. J. King.   1977.  On
 the relationship between  high  ozone  in  the rural  surface  layer  and  high
 pressure  systems.   Atmos. Environ. 11:967-983.

 Waggoner,  A.  P.  and  R.  E.   Weiss.    1980.    Comparison  of  fine  particle
 concentration and  light  scattering  extinction  in  ambient aerosol.   Atmos.
 Environ.  14:623-626.
 Wagman,  J.,  R.  E.  Lee,  Jr.,  and C.  J. Axt.    1967.
 atmospheric  variables  on  the concentration  and particle
 sulfate  in urban air.  Atmos. Environ. 1:479-489.
                                               Influence  of some
                                             size distribution of
 Weiss,   R.   E.,
 rapid-response
 Virginia
      T.  V.  Larson,   and  A.   P.   Wa<
      measurement   of    H?SO/L/(NH4)2-
Environ. Sci.  and Technol.  r6:S25-532.
igoner,    1982.    In  situ
104    aerosols    in   rural
 Weiss,  R.  E., A.  P.  Waggoner,  R. J
 Sulfate  aerosol:   Its geographical
 United States.   Science 195:979-981.
                           Charlson,  and  N.  C.  Ahlquist.   1977.
                          extent  in  the mideastern  and southern
Westberg,  H.,  K. All wine,  and E.  Robinson.   1978a.   Measurement of  light
hydrocarbons  and oxidant transport-Houston  Study  1976.   EPA-600/3-78-  062.
Environmental Sciences  Research Laboratory,  Triangle  Research  Triangle  Park,
NC.

Westberg,  H.,  K. Sexton,  and M.  Holdren.   1978b.    Measurements  of  ambient
hydrocarbons and oxidant transport.  Vol. I. Houston Study.   Report on  EPA
Grant  No. 805343  to Environmental   Sciences,  Research Laboratory,  Research
Triangle  Park, NC.

Whelpdale, D.  M. and L.  A.  Barrie.  1982.   Atmospheric  monitoring  network
operations and results  in Canada.   Water, Air, and Soil Pollut.  18:7-23.
White,  W.  H.  and  P.  T.  Roberts.
visibility-reducing  aerosols  in  the
11:803-812.
                          1977.   On  the  nature
                           Los  Angeles  Basin.
            and origins  of
           Atmos.  Environ.
White,  W.  H.,  J.  A. Anderson, D.
J.  D.   Husar,  and W.  E.  Wilson,
secondary  air  pollutants:   ozone
Science 194:187-189.
                       L.  Blumenthal,  R.  B. Husar, N. V. Gillani,
                       Jr.   1976.    Formation  and  transport  of
                       and  aerosols  in the St. Louis urban plume.
White,  W.  H.,  D.  L.  Blumenthal,  J. A.  Anderson,  R.  B.  Husar,  and  W.  E.
Wilson, Jr.  1977.  Ozone formation in the St.  Louis urban  plume,  pp.237-247.
In Vol 1, International Conference of Photochemical  Oxidant Pollution and its
"Control.   B.  Dimitriades, ed.   EPA-600/3-77-0012,    Environmental Sciences
Research Laboratory, Research Triangle Park,  NC.
                                     5-101

-------
Mil listen,  S.
73:7051-7055.
               H.   1968.   Mercury  in the  atmosphere.   J.  Geophys.  Res.
Winer,  A.  M.,  J.  VI.  Peters,  J.  P.  Smith,  and J.  N.  Pitts,  Jr.   1974.
Response of  commercial  chemilum'nescent  NO-NO?  analyzers to other nitrogen-
containing compounds.  Environ.  Sci.  Techno!.  8:1188-1121.
Witz,  S.  and R.
filters on total
Air Pollut. Control
                  D.  MacPhee.   1977.   Effect of  different types  of glass
                 suspended particulates and  their  chemical  composition.   J.
                  "  Assoc. 27:239-241.
Witz, S. and J. G. Wendt.  1981.   Artifact  sulfate  and nitrate formation at
two  sites  in  the  South Coast Air  Basin.   A collaborative study between the
South Coast Air Quality Management District and the California Air Resources
Board.  Environ. Sci.  Technol. 15:79-83.
Witz, S., M.  Smith,  M.  Shu,  and A. B. Moore.
lead, sulfate, and nitrate concentrations  in  a
size-selective, and standard Hi-Vol samplers.
32:276-278.
                                               1982.   A comparison of mass,
                                              field study using dichotomous,
                                               J.  Air Pollut. Control  Assoc.
Wolff, G. T., P.  J.  Lioy,  G. D.  Wight,  R.  E. Meyers,  and  R.  T. Cederwall.
1977.    An  investigation   of  long-range   transport  of  ozone  across  the
midwestern and eastern United States.   Atmos.  Environ. 11:797-802.

Wolff, G. T., R.  R.  Monson,  and  M.  A. Ferman.  1979.   On  the  nature of the
diurnal variation of sulfates at rural  sites  in  the eastern United States.
Environ. Sci. Technol. 13:1271-1276.
Zika, R. G. and  E.  S. Saltzman.
peroxide in  water:    Implications
Res. Letts.  9:231-234.
1982.   Interaction
for  anlaysis  of
of
                                                          ozone  and
                                                          in air.
                                                                    hydrogen
                                                                    Geophys.
Zika, R.,  E.  Saltzman,  W.  L.  Chameides,  and  D.  D.  Davis.    1982.
levels in  rainwater  collected  in South Florida and  the  Bahama  Islands.
Geophys.  Res.  87:5015-5017.
                                                                          J.
                                     5-102

-------
               THE ACIDIC DEPOSITION PHENOMENON AND ITS EFFECTS

                   A-6.  PRECIPITATION SCAVENGING PROCESSES

                                 (J. M. Hales)

6.1  INTRODUCTION

Precipitation  scavenging  is  defined  generally  as the  composite process  by
which  airborne pollutant gases  and  particles  attach  to  precipitation  ele-
ments, and thus  deposit  to  the  Earth's  surface.1  This  process  constitutes
a critically important pathway for  atmospheric  cleansing,  and  is  a  "natural-
recovery"  phenomenon  which  is  absolutely  essential   for maintenance  of  a
liveable  global   atmosphere.   Conversely,  however, the   pollutant  delivery
resulting  from  precipitation  scavenging often  can  be  sufficiently  large  to
impose severe  impacts on a variety  of surface  receptors.   Growing cognizance
of  this   point  has  resulted  in  the  "acid-precipitation issue"  as  it  is
generally perceived today.

The  goal  of this  chapter  is to provide  the  dedicated (but not  necessarily
expert) reader with an overview  of  precipitation  scavenging, which  discusses
physical  processes in a qualitative manner while  at the  same time establish-
ing a  solid basis  of  understanding.   This is accomplished by  first  breaking
down  the  scavenging   process  into  a  number  of  discrete steps,  and  then
scrutinizing the associated physical  mechanisms or  individual  and collective
bases.  This summary of physical  processes emphasizes  the importance of storm
type and meteorological  behavior on scavenging pathways.  Relative  to  this,
the  subsequent  section addresses storm climatology and storm classification,
with emphasis on practical  applications.

The next two sections  deal  respectively with  past field  studies  of precipi-
tation scavenging  and  precipitation-scavenging  models.   A qualitative empha-
sis continues throughout the modeling section,  although  sufficient  equations
are  used  to  facilitate the  general  discussion.  The  chapter  concludes  with
 One  should  note  that this definition  pertains  to removal  from  the  gaseous
 medium  of  the  atmosphere  combined with  deposition  to the  ground.    An
 alternative  definition,  employed  often' throughout  the  open  literature,
 pertains  to the  simple  attachment  of airborne  pollutants  to liquid  water
 elements, without regard to whether the material  is subsequently  conveyed to
 the  Earth's surface.  Which  of  these definitions is used is  unimportant so
 long  as  the  precise   definition   is  understood.     The   definition   of
 "scavenging"  adopted here  will   be  utilized consistently  throughout  this
 text.   When specific reference  to  the  alternative  situation is made,  the
 terms  "attachment"  and  "capture"   will  be  employed  essentially   inter-
 changeably.


                                     6-1

-------
an examination of predictive uncertainties,  and the  scientific  advances  which
will be necessary to reduce these uncertainties to an acceptable  level.

6.2  STEPS IN THE SCAVENGING SEQUENCE

6.2.1  Introduction

The  precipitation  scavenging process  typically contains  many parallel  and
consecutive steps, and  as  an introduction to this section  it  is  appropriate
to  provide  a  brief  overview of these  intermeshing pathways.    In  a  very
general sense  there  are four major  events  in  which a  natural or  pollutant
molecule2  may  participate,  prior  to its wet removal  from the  atmosphere;
depicted pictorially in Figure 6-1, these are:

   1-2.   The pollutant and the  condensed atmospheric water (cloud,
          rain, snow, ...)  must  intermix within the  same airspace.

   2-3.   The pollutant must attach to the condensed-water  elements.

   3-4.   The pollutant may react physically and/or  chemically  within
          the aqueous phase.

   3-5.   The pollutant-laden water elements must  be  delivered  to  the
or(4-5.)  Earth's surface via the precipitation process.

The  interaction  diagram of Figure 6-2  gives a somewhat more  detailed  por-
trayal of  these four major  events.  Here the individual  steps are  represented
as transitions of the pollutant  between  various states in the atmosphere,  and
one can note that a multitude of reverse  processes are  also possible; thus a
particular pollutant  molecule  may experience numerous  cycles through  this
complex of pathways  prior  to deposition.   Indeed, Figure  6-2  indicates  that
this cycling process may continue even after "ultimate" deposition.  By  pol-
lutant off-gassing and  other resuspension processes, the  deposited material
can be re-emitted to the atmosphere,  with the possibility of participating in
yet another series of cycles throughout  the  scavenging sequence.

Another important feature  of  Figure  6-2   is  that,  while physicochemical
reaction within  the  aqueous-phase is  potentially an important  step  in  the
scavenging process,  it is  not  essential.   This  contrasts to  the  remaining
forward steps  that must take place  if scavenging is  to occur.  Despite  its
nonessential  nature, this  step  is  often of  utmost importance  in  influencing
scavenging rates,  owing to its   role  in modifying reverse processes  in  the
sequence.   An  example  of  this  effect,  already discussed in Chapter A-4, is
2Initial portions  of  this chapter will  treat  precipitation  scavenging in  a
 general  sense,  with  limited reference  to  specific  types  of  atmospheric
 material.  The reader should continue to note,  however,  that  the  "natural or
 pollutant molecules" of  primary  concern in the present context  are  species
 associated  with  acid-base   formation,  such  as  S02,  HN03,  NH3,  sulfate,
 chloride, metallic cations,  and  so forth.


                                     6-2

-------
     UNREACTED POLLUTANT
     REACTED POLLUTANT
Figure 6-1.   Steps in the scavenging sequence:   Pictorial  representation.

-------
    o
    o
    m
    oo
    oo
    m
    co
k.

RESUSPENSION

EVAPORATION |

*.
EVAPORATION, DESORPTIONJ |

— —



1

i
2
C
C
i
3
POLLUTANT
IN
CLEAN AIR
0 §
a. ^



O UJ
oo
t-l CM O OO
Z UJ
UJ UJ >— i O
0. 0_ X O
>- >- t-i o:
' 1— t— 2: 0-
POLLUTANT AND
ONDENSED WATER
INTERMIXED IN
OMMON AIRSPACE
'o z
•— ' i— i
11
=£ UJ
ff^ f^ •
ATTACHMENT
POLLUTANT
ATTACHED TO
CONDENSED WATER
ELEMENTS
t
i
•z.
o
s
UJ
a:
\
REACTION
4 ATTACHED POLLUTANT
MODIFIED BY
AQUEOUS PHASE
PHYSICOCHEMICAL REACTIONS
i
DEPOSITION
5 POLLUTANT DEPOSITED


ON
EARTH'S SURFACE
4 —


o
o
m
o
oo
*— i
— I
i — i
0
•z.
o
o
m
oo
00
m
00
                                                                   1
Figure 6-2.  Scavenging sequence:   Interaction diagram.


                                   6-4

-------
the  devolatilization  of dissolved sulfur dioxide  via  wet oxidation to  sul-
fate.   This  effectively  eliminates  gaseous  desorption  from  the  condensed
water and  thus  has  a strong tendency to enhance the overall scavenging  rate
as a result.

From Figure 6-2 one can note also  that  precipitation scavenging  of  pollutant
materials  from  the  atmosphere  is intimately  linked  with the  precipitation
scavenging of water.  If one were to replace the word  "pollutant" with  "water
vapor" in  each  of the steps, Figure 6-2 (with  the exception of  box 4) would
provide a  general description of the  natural precipitation process.  In  view
of  this  intimate  relationship,  it  is  not  surprising  that  pollutant  wet-
removal  behavior tends to mimic that  of precipitation.   Pollutant-scavenging
efficiencies  of  storms,  for example, are  often similar  to water-extraction
efficiencies.   This relationship  is  useful  in practically estimating scav-
enging rates and  will  reappear  continually  in  the   ensuing  discussion of
wet-removal behavior.

Figure 6-2  is  interesting  also  because  of  its  indication  that,  if  some
particular  step  in the diagram  occurs  particularly slowly  compared  to  the
others,  then  this step will  dominate  behavior  of the  overall  process.   This
is similar  to the  "rate-controlling  step"  concept in  chemical  kinetics,  and
has  been   applied  rather  extensively  in  practical  scavenging  calculations
(Slinn 1974a).   Finally,  it is important to note that Figure 6-2 presents  a
framework  for developing  and  evaluating  mathematical   models  of scavenging
behavior.   Successful scavenging models must emulate these steps effectively
and tend to reflect the structure of Figure  6-2 as  a result.  This point  will
be  recalled later  when  scavenging models  are examined  specifically.    The
following  subsections will   address   qualitative  aspects of  the scavenging
sequence in the  order of their  forward progress to  ultimate deposition.

6.2.2  Intermixing of Pollutant and Condensed Water (Step 1-2)

Upon  first  consideration,   one  often  is  inclined  to  dismiss pollutant-
condensed-water  intermixing  as  an  unimportant or at least trivial  step  in the
overall  scavenging  sequence.   It is neither.   In  a  statistical  sense it
usually  is  neither  cloudy nor  precipitating in  the immediate  locality  of  a
freshly-released pollutant  molecule;  typically this molecule  must exist in
the clear  atmosphere  for  several  hours, or  even  days, before  it encounters
condensed water  with which it may  commingle.  This  in itself establishes  step
1-2  as  a  potentially  important   rate-influencing  event.   Moreover,   this
extended  dry  period  typically   presents   the  pollutant with   significant
opportunities to react  and/or  deposit via  dry  processes; thus,  the chemical
makeup of  precipitation is  influenced profoundly by  this preceding chain of
events.

Significant  insights  to the behavior  of  step 1-2 can be gained  via  past
analyses of  storm  formation (Godske et al.  1957)  and the atmospheric water
cycle (Newell et  al.  1972).   Several  statistical  analyses of  precipitation
occurrence (Rodhe and Grandell  1972, 1981;  Gibbs and SI inn 1973;  Junge 1974;
Baker et al.  1979)  have been applied  as general interpretive descriptors of
this  step.   These  will  not be  examined  in  detail  here;  rather  we  shall
                                     6-5

-------
concentrate  upon  the mechanisms  by  which step  1-2  can occur,  from a more
pictorial viewpoint.

Two types of mixing processes exist whereby  pollutant and condensed water can
come to occupy common airspace;  these are

     1)  Relative movement of  the initially unmixed pollutant and condensed
         water,  in  a manner such  that  they  merge  into  a  common   general
         volume;  and

     2)  In situ phase change of water vapor, thus producing condensed water
         in the immediate vicinity of pollutant molecules.

The relative importance of Type-1 and Type-2 mixing  processes will depend to
some extent on the  pollutant.   _Tf a particular  pollutant is easily  scaveng-
able and j_f  precipitation  is occurring  at the pollutant's release location,
then Type-1 processes are  likely  to  contribute significantly.    If these two
conditions are not met,  the pollutant will usually mix intimately with makeup
water  vapor  for  some future  cloud,  and  Type-2  processes  will  predominate.
Based  upon  in-cloud vs  below-cloud  scavenging estimates  (SIinn  1983) it is
not unreasonable to estimate  that, as a  global  average,  roughly 90 percent of
all precipitation scavenging  occurs as the consequence of a Type-2 process.

As Figure 6-2 indicates, reverse processes can serve  to reseparate pollutant
and condensed water.  Evaporation,  for  example,  can reinject pollutant from
cloudy to clear air, and relative motion such  as  precipitation  "fall-through"
can remove hydrometeors from contact with elevated plumes.  Cloud formation-
reevaporation cycles  are  particularly significant  in  this  respect.   Junge
(1963), for  example,  estimates  that a  single  cloud  condensation nucleus is
likely to  experience on the  order of  ten  or more  evaporation-condensation
cycles before it is ultimately  delivered to  the  Earth's  surface with  precipi-
tation.    The  rate-influencing  effect  of  such  cycling  on  precipitation
scavenging is obvious.  Additional  types of  cycles will  be described  below in
conjunction with succeeding steps of  the scavenging  sequence.

6.2.3  Attachment of Pollutant to Condensed  Water Elements  (Step  2-3)

The microphysics of the  pollutant-attachment process  have been  the subject of
extensive  research,  and numerous  reviews of  this  area have  been   prepared
(Junge 1963, Davies 1966,  Dingle  and Lee 1973, Pruppacher  and  Klett 1978,
Hales 1984, Slinn 1983,  Slinn and Hales  1983).   In  the context  of Figure 6-1,
this process is  complicated  somewhat in  the  sense  that,  depending   upon the
particular attachment mechanism, Step 2-3 may occur  either simultaneously or
consecutively with Step  1-2.

Simultaneous commixing  and attachment occur  in  the  case  of  cloud-particle
nucleation.   This  is a  phase-transformation  (Type-2)  process  wherein water
molecules,  thermodynamically inclined  to  condense  from  the  vapor phase,
migrate to some suitable surface for this purpose.   Pollutant  aerosol parti-
cles provide such  surfaces within the air parcel, and  the  consequence is  a
                                     6-6

-------
cloud  of   droplets  (or   ice   crystals)3   containing  attached  pollutant
material.

Different  types  of  aerosol  particles  possess  different  capabilities  to
nucleate cloud  elements  and grow by  the  condensation  process.   As a conse-
uence, there is typically a competition for  water molecules among  the aerosol
and associated cloud particles.   Some will capture water  with high efficiency
and grow substantially  in  size.   Others will  acquire  only small   amounts of
water, and  still  others  remain  essentially  as "dry" elements.  In addition,
some particles  may  nucleate ice  crystals, while  others  will be  active only
for the formation of liquid water.   The nucleating capability of a particular
aerosol particle  is  determined  by  its  size,  its  morphological   character-
istics, and  its chemical composition.  Various  aspects  of  this  subject are
discussed  at   length  in   standard  cloud-physics  textbooks  (Mason   1971,
Pruppacher  and Klett 1978)  and in  the  periodical  literature  (Fitzgerald
1974).

An  additional  important aspect  of  the cloud-droplet  nucleation  and growth
process is  the fact  that once  initiated,  cloud-droplet growth does not
proceed instantaneously  to  some  sort of thennodynamic  equilibrium.  Because
of diffusional  constraints  on delivering water molecules  from the  surrounding
atmosphere, the growth in droplet diameter  slows  appreciably  as droplet size
increases   (Slinn  1983).    Superimposition  of this  lag  on  the   continually
fluctuating environment of  a  typical  cloud  results  in  a dynamic  and complex
physical  system.

Finally,  the  competitive nature of  the cloud-nucleation process   results in
significant  impacts  by the  pollutant  on  the  basic character of the  cloud
itself.   If the  local  aerosol  were populated  solely  by a  relatively  small
number of  large,  hygroscopic particles,  for example,  one would  expect any
corresponding  cloud  to  be  composed chiefly  of small  populations of  large
droplets.    If  on the  other hand the  local  aerosol were  composed of  large
numbers of  small, nonhygroscopic particles,  the corresponding  cloud should
contain larger numbers of smaller droplets.

This is precisely what is  observed  in practice.   Unpolluted marine atmos-
pheres, for example, contain large sea-salt particles as a primary component
of their aerosol burden.   Warm marine clouds are noted  for  their wide droplet
spectra containing large droplet sizes and  their  corresponding capability to
form precipitation easily.   Continental clouds,  on the otherhand, are  typi-
cally  composed of  larger  populations  of  smaller droplets.   Figure  6-3,
3At  this  point it  is important  to  note  that  aerosols can  participate in
 several  types of   phase  transitions  in  cloud  systems.     These  include
 vapor-liquid, vapor-solid,  and  liquid-solid transitions,   in  addition to a
 subset of interactions between  numerous  solid  phases.   Particles active as
 ice-formation nuclei are generally much  less abundant  than those active as
 droplet (or  "cloud-condensation") nuclei.   As will  be demonstrated  later,
 the  relative abundance  of  ice nuclei  can have  a  profound  effect upon
 precipitation-formation  processes and  related scavenging phenomena.


                                    6-7

-------
                                                 8-9
  id
   c
   -5
   fD

   CTv
   I
   CO
3=- O
Q. —'
Cu O
-a c
rt- Q.
fD
CX Q-
   -5
-h O
^^ "^3
O —'
3 fD

CO
_Q (/I
C TD
_i. fD
-s n
ro c+
t/> ~5
   Co
 CO
 3 -••
 CL 3

 —I O
 S O
 O 3
 3 <
 fD fD
<< O
 l—> <
 <£> fD
 CD
 O O
                   CLOUD DROPLET  CONCENTRATION  (droplets  cm'3)
    fD
    a.
     CO
     -s
     -••
     r-f-

     3
     a>
     3
     Q.

     O
     O
     3
     ro
     3
     t-t-
     CO
     £U

     -5

     3
     fD
     (Si
O
I —
O
         O
         §
         -a
         r~
         m
          a

-------
prepared on  the basis  of results  published  by Squires  and Twomey  (1960),
provides a  good  example  of  this  point.   Here,  measured  convective-cloud
droplet spectra  are  compared for  two different cloud  systems.   The conti-
nental  air-mass  cloud exhibits a  distinct tendency  toward  smaller  droplet
sizes and larger populations, as compared to its maritime  counterpart.   It  is
interesting also in this context to note Junge's (1963)  estimates  with regard
to  relative  amounts  of  aerosol  participating  in the  nucleation  process.
Junge  suggests  that  while  50  to  80 percent  of  the  mass  of  continental
aerosols can be expected to  participate as cloud nuclei, as much as  90 to 100
percent of maritime aerosols can become actively involved.

As  a concluding  note  in  the  context  of nucleating capability  and  water
competition, it should  be pointed out that acid-forming  particles,  by  their
very  nature,  are chemically  competitive  for  water vapor and  thus tend  to
participate actively  as cloud-condensation nuclei.  This attribute  tends  to
enhance their propensity to  become scavenged early  in  storm  systems  and  has a
significant effect on the nature of the acid precipitation formation process.

There  are  numerous mechanisms  by  which pollutants can  attach  to  cloud and
precipitation elements after the elements already exist,  and  thus  in a manner
consecutive with  Step 1-2.    These  mechanisms  are  itemized in the  following
paragraphs.  They are typically active for both  aerosols  and  gases, although
the  relative  importances  and magnitudes  vary  widely with  the  state of the
scavenged substance.

Diffusion^!  attachment,  as  its  name  implies,  results  from   diffusional
migration of the pollutant though the air to the water surface.  This process
may  be  effective  both  in the case of suspended cloud  elements and  falling
hydrometeors.   It  depends  chiefly  upon  the  magnitude  of  the  pollutant's
molecular (or Brownian) diffusivity; because diffusivity is  inversely related
to particle size, this mechanism becomes less  important as pollutant elements
become  large.   Diffusional attachment is  of utmost importance for scavenging
of  gases  and very small  aerosol  particles.   For  all  practical  purposes,  it
can be  ignored  for aerosol particle sizes above a few  tenths  of  a  micron.

In concordance with Pick's law (Bird et al. 1960),  diffusional transport to a
water surface also depends upon the pollutant's concentration gradient in the
vicinity of  this  surface.   Thus  if the cloud  or  precipitation element can
accommodate the  influx  of pollutant readily,  it will effectively  depopulate
the adjacent air, thus  making a  steep concentration gradient  and  encouraging
further  diffusion.    If  for some  reason (e.g.,  particle   "bounce off"  or
approach to  solute  saturation)  the element cannot accommodate the  pollutant
supply, then further diffusion will be discouraged.  If the cloud  or precipi-
tation element, through some sort of outgassing mechanism, supplies  pollutant
to  the  local  air,  then the  concentration gradient will be reversed and dif-
fusion will carry the pollutant away from the  element.

Mixing  processes  inside cloud  or precipitation elements play  an  important
role in determining the accommodation of gaseous species.   If mixing is  slow,
for example, it is likely that  the element's  outer layer will  saturate with
pollutant and thus inhibit further attachment  processes.   This is  quite  often
a  limiting  factor  in  cases  involving   gas  scavenging  by  ice  crystals.


                                     6-9

-------
Internal mixing occurs  as a consequence of  diffusion  and fluid  circulation
and has been analyzed by  Pruppacher  and his  coworkers (Pruppacher and Klett
1978).

In general, diffusional  attachment processes  are  sufficiently well  understood
to  allow  their  mathematical   description   with  reasonable   accuracy,  and
numerous references are available as guides for  this purpose (Pruppacher and
Klett 1978, Hales  1984,  Slinn 1983).
Inertia! attachment processes depend directly upon the size of the  scavenged
particle, and  thus are unimportant  for gaseous pollutants.   In a  somewhat
general sense this class of processes depends upon motions of pollution par-
ticles and scavenging elements  relative to  the surrounding air, which arise
because both have finite volume and mass.  The most important example of in-
ertia! attachment is the impaction  of aerosols  on falling  hydrometeors.  Here
the hydrometeor (because of its mass and volume)  falls by gravity,  sweeping
out a volume of space.   Some of the aerosol  particles  (because of their mass)
cannot move sufficiently rapidly with the flow  field to avoid the hydrometeor
and,  thus,  are impacted.   In  principle, impaction could  occur  even if the
aerosol particles were  point masses with zero volume.   Assigning a volume to
a particle further increases its chance of collision, simply on  the  basis of
geometric effects.  The inclusion  of aerosol volume in this context  has been
generally referred to in the past literature as interception.

The effectiveness  of impaction and  interception depends  upon  both  aerosol-
particle and hydrometeor size;  mathematical  formulae exist which can be used
conveniently to estimate the magnitudes  of these processes (e.g., Hales 1984,
Slinn  1983).   These effects generally  become  unimportant for  aerosols less
than a few microns in size.  In this context,  it is interesting  to  note that
a two-stage capture mechanism can exist, in  which a small  aerosol first grows
via  nucleation to  form a  larger  droplet,  which  then  can be  captured by
inertia! attachment in  a secondary process.   This two-stage process  has been
postulated as an important mechanism in below-cloud scavenging (Radke et al.
1978, Slinn 1983).  It  is also  an essential  factor  in  the  in-cloud generation
of precipitation and is generally referred to as accretion.

A second example of inertial  attachment  is turbulent collision.   In  this case
the particles and  scavenging elements subjected to a turbulent field collide
because of dissimilar dynamic responses  to velocity fluctuations  in  the local
air.  This capture mechanism is thought  to be of secondary importance and has
received  comparatively  little  attention  in  the  literature  although  past
theoretical  treatments  of turbulent coagulation processes (e.g., Saffman and
Turner 1955, Levich 1962, Fuchs 1964) indicate  that it may be significant for
specific dropsize-particle size ranges.

While the mechanisms of diffusional and  inertial attachment are efficient for
capturing very fine and very coarse  particles,  respectively, a region of low
efficiency should  exist  approximately in the 0.1 to  5.0  micron  range where
neither  mechanism  is effective.   This  effect  is  shown  schematically  for  a
given  drop  in  Figure 6-4.   Because  its importance to  scavenging  was  first
recognized  by   Greenfield  (1957),  it  has   become  known  generally as  the
"Greenfield  gap."     Depending   upon   circumstances,   several   additional


                                     6-10

-------
                                         II-9
IQ
-s
n>
CTl


-P>
                                   CAPTURE  EFFICIENCY
O c± (Si — 1
3-O -•• 3-
CU (M fD
-5 0 fD O
tr*> O • —S
fD 3 fD
c— 1- r-i-
CT C"
O -5 3a -"•
0 -•• Q. 0
3 cr cu eu
Q-C T3 — '
-••r+ rt-
rt- ->• fD CO
-••O O. O
O 3 CU
3 01 -h <
(S> -5 fD
CT O 3
*• — •"**< 3 ' °
oo -".
fD fD O 3
fD — ' C Id
fD -S
O O < fD
-S rt- fD -h
-•.-S to -h
IQ _1. _1.
-"•n ua o
3 CU -i. _•.
CU — i < fD
— • fD 3
fa 3 O
-S3 <<
fD Q. cr
-h "< O
fD "O ~h
-s 3- -a
fD O -S CU
3 -S C
O fD T3 -h
fD (fa CU
-J. CU — '
— hO O — '
O 3- ->•
-S fD fD 3
Q_ll| "* ""^
fD fD CU -s
r±n 3 CU
CU r+ Q. -••
-J.01 3
^ 7^ Q-
01 C — • -S
— '3 fD O
• O. c-fO
fD ct-
-s cu
.—X Ol
O I—1
3" i*Q CU
O ^J
to 00 -h
fD - — 5=
3 • 3
O
3" r~h
c o -••
3 CU 0
-J- to 3
CL 3-
-"• fD O
<-h Q. -h
^<
— ' CU
CU ->• fD
33-5
a. fD o
-s o
CU O — '
_i. O
3 -s -a
a. -s cu
-5 fD -S
O 01 rt
T3 -a -'•
1 O O
3 — '
a. ro


























-Q
>^
o
t — 1
c:


o
TI

3>
m

0
oo
o
r—

-a
•yo

t — i
o
r~
m

^— ^
•e
3






























                                 INERTIAL ATTACHMENT

-------
attachment mechanisms (including  the  two-stage  nucleation-impaction mechanism
mentioned earlier)  can serve to "fill" the Greenfield gap.  Some of the more
important of these  are itemized in the following  paragraphs.

Diffusiophoretic attachment  to a scavenging element  can occur whenever  the
element grows via  the condensation  of water vapor.   In  effect,  the flux  of
condensing  water  vapor  "sweeps"  the  surrounding  aerosol  particles  to  the
element's surface.   In a competitive cloud-element  system where some droplets
grow  while others  evaporate, diffusiophoresis  can  be  a  rather  important
secondary  attachment  mechanism.   This  is particularly  true  when the  cloud
contains mixtures of ice and liquid.   Under such  conditions,  the  ice crystals
have a pronounced tendency,  owing to  their  lower equilibrium  vapor  pressure,
to  gain  water  at  the  expense  of  the  droplets.    Known  as  the  Bergeron-
Findeisen  effect,  this process  is  important  in  precipitation formation  as
well as in diffusiophoretic enhancement.

Thermophoretic  attachment  results from a temperature gradient in the  direc-
tion  of  the  capturing element.   Here  the element  acts  essentially as  a
miniature  thermal precipitator.   Warmer  gas  molecules on the  outward  side of
the  aerosol  particle  impart a  proportionately  larger  amount of  momentum,
resulting  in a driving force toward the capturing element.*

Thermophoresis depends directly upon the temperature gradient in the vicinity
of the capturing element.   In cloud  and precipitation systems local  tempera-
ture  gradients  are  caused most  often  by  evaporation/condensation  effects;
thus,  thermophoresis  is usually  strongly  associated  with diffusiophoresis,
and in fact these two processes often tend to counteract each other.

Phoretic  processes  are unimportant in the case  of  gaseous pollutants, owing
to  the overwhelming  contributions of  molecular  diffusion.  At  present,  the
theory  of diffusiophoretic/thermophoretic particle attachment is at  a state
where  reasonably quantitative  assessments can be made for simple systems such
as  isolated droplets {Slinn  and  Hales  1971, Pruppacher  and  Klett  1978,  See
Figure 6-4).   Rough estimates are possible  for  more  complex  and interactive
cloud/precipitation  systems, but much  remains to  be done  to  make  our know-
ledge  of  this area  satisfactory.

Electrical  attachment of  aerosol particles  to cloud and precipitation ele-
ments  has been  the subject  of continuing  study  over the past three decades.
Understanding  of this  process is currently at  a  state  where relationships
between  aerosols  and isolated  droplets can  be quantified  with reasonable
accuracy   (Wang  and Pruppacher  1977).    In  general,  electrical  charging of
cloud  and/or precipitation  elements  must be moderately  high  for electrical
 40ne should  note  that  the  precise mechanisms  of  thermal  transport differ
  radically,  depending  upon particle  size  (cf., Cadle 1965).

 5As  noted  by  Slinn  and  Hales  (1971),  inappropriate   treatment   of  this
  relationship has caused  erroneous conclusions to  be  drawn in  some of the
  past literature.   The reader should be cognizant  of  this if more  detailed
  pursuit is  intended.


                                     6-12

-------
effects  to  become competitive  with other capture  phenomena, although  such
charging  is   certainly  possible   in  the   atmosphere,   particularly   in
convective-storm  situations.    Understanding  of  electrical   deposition  in
clouds of interacting drops is still relatively unsatisfactory.

While the mechanisms of  attachment  processes  have  been presented here on  an
individual basis,  they tend  in  actuality  to  proceed  in a simultaneous  and
competitive manner.  Insofar as atmospheric cleansing is concerned,  this  is a
fortunate circumstance,  because  some  mechanisms  tend to operate  in  physical
situations where  others  are  ineffective.   Figure  6-4 gives  an  excellent
illustration of this point.   Theoretical attachment  efficiencies  appropriate
to  a  0.31  mm  radius   raindrop  are  presented  for  various  electrical   and
relative-humidity  conditions,  demonstrating  the capability  of phoretic  and
electrical mechanisms  to  "bridge" the  Greenfield gap.   This  simultaneous and
competitive interaction  of mechanisms  serves  to  complicate  profoundly  the
mathematics of  the  scavenging  process, and  lends  an  additional degree  of
difficulty  to  the  problem of  scavenging  calculations.   This   aspect  will
continue  to emerge throughout this  chapter, especially  during the discussion
of scavenging models.

6.2.4  Aqueous-Phase Reactions (Step 3-4)

Aqueous-phase conversion phenomena  have  been  discussed in  some detail  In
Chapter  A-4  and  will  not  be examined  further here  except  to  note  their
general  importance  within the framework of the  overall  scavenging  sequence.
As noted  previously in the context of Figure 6-2, aqueous-phase reactions are
not  essential   to  the  scavenging  process.  Depending  upon  the  pollutant
material, however,  these  reactions  often  can  have the  effect of  stabilizing
the  captured  material  within  the  condensed phase and, thus, enhancing  the
scavenging  efficiency  appreciably.    Much needs  to  be learned  before  this
important topic is satisfactorily understood.

6.2.5  Deposition of Pollutant with  Precipitation (Steps 3-5  and 4-5)

Although  a  variety of mechanisms  exist (e.g.,  impaction  of fog on vegeta-
tion),  the  predominant means  for depositing pollutant-laden  condensed  water
to the  Earth's  surface is simply  gravitational  sedimentation.  Sedimentation
rates  depend  upon  hydrometeor  fall  velocities^whichdepend  in turn  upon
hydrometeor size.   Thus,  the processes by which  the  pollutant-laden  cloud
droplets  grow to  precipitation  elements emerge as major determining factors
in this  final  stage of the scavenging sequence.

Once attached to  condensed  water,  a pollutant molecule  has  several  alterna-
tive pathways for  action  (Figure  6-2).  If the  captured pollutant  possesses
some degree  of  volatility it may  desorb  back into  the gas  phase.   Reverse
chemical  reactions  may occur.   Evaporation  of  the  condensed water may,  in
effect,  "free"  the pollutant  to  the  surrounding  gaseous  atmosphere.   This
multitude of pathways  results in an active competition for  pollutant.  If the
precipitation stage  of the scavenging  sequence  is to  be effective, it  must
interact  successfully  within this competitive framework.
                                     6-13

-------
Besides competing actively  for  pollutants,  the above interactions produce a
vigorous competition for water.   This parallel  relationship  between pollutant
scavenging and water scavenging,  apparent  in  some  of  the  preceding discussion
regarding attachment  processes,  can  be  drawn  even  more  emphatically when
considering  precipitation  processes.   The  following  paragraphs  provide a
brief overview of some of the more  important  mechanisms in  this  regard.

Once initial  nucleation has occurred, cloud particles  may  grow further by
condensation of additional water vapor.   Net condensation  will  occur to  the
surface of  a  cloud  element whenever  water vapor  molecules can  find a more
favorable thermodynamic  state in  association  with  it;  and  because clouds
contain varieties of  makeup  elements having  different thermodynamic charac-
teristics, competition for water  vapor usually exists.  Such interactions  are
discussed at  length  in  standard  textbooks (Mason  1971,  Pruppacher and  Klett
1978).    SI inn (1983) has  developed a conceptual  scavenging model   in  which
condensational growth is an important rate-limiting step.

Thermodynamic  affinity  for  water-vapor  molecules depends upon  the  cloud-
element's size,  its  pollutant  burden, and  its physical  structure.   These
latter two  factors often influence  precipitation  characteristics  profoundly.
In particular, the favored thermodynamic  state of  a water molecule in associ-
ation  with  an ice  crystal  (as   compared  with  a  supercooled water  droplet)
results in  rapid  competitive  growth of ice particles in mixed-phase clouds.
This "Bergeron-Findeisen"  process  has  been mentioned already in  the context
of  diffusiophoretic   and  thermophoretic  transport.   Growth  of large  cloud
elements  via  this process is the  primary reason  that ice-containing  clouds
tend to be  so strongly effective as generators of  precipitation  water.

A further mechanism  by  which suspended cloud droplets can  grow  to form pre-
cipitation  elements is coagulation.  This process  occurs  via the collision of
two  or more cloud elements to form a new element containing the  total  mass
(and pollutant burden)^  of its  predecessors.   Coagulation  occurs  over  size-
distributed  systems  of cloud elements by a  variety of physical  mechanisms
and,  because  of  this,  is  a  rather  poorly  understood and mathematically
complex process.   Comprehensive  analyses of coagulation processes have been
performed by  Berry  and  Reinhardt (1974).  Coagulation  can be  considered an
important initiator  of  precipitation in  single-phase clouds  (water  or  ice).
In  mixed-phase clouds,  the Bergeron-Findeisen process  can  be expected  to
enhance the coagulation process  by widening the droplet size distribution, as
well as contributing  to precipitation growth  in a  direct sense.

Once a moderate  number  of precipitation-sized  elements  have  been  generated,
the  process of accretion rapidly begins to dominate as a  means for generating
precipitation  water.    As noted  previously,  this  process  occurs by  the
 "sweeping"  action of  large  hydrometeors  falling through  the field of smaller
elements, attaching  them on  the  way.  As was the  case with coagulation,  the
^Coagulation  is often  referred  to  as  autoconversion  in  the  cloud-physics
  literature.    It  is  interesting  to noticeTntFFTs  context  that,  while
  coagulation   tends  to  accumulate  nucleated   pollutants,   the   Bergeron-
  Findeisen  process  tends to re-liberate  nucleated pollutants to the air.


                                     6-14

-------
accretion process tends to  accumulate  the  pollutant burden of all collected
elements.

Accretion can occur  via  drop-drop,  drop-crystal, and crystal-crystal  inter-
actions.  Drop-crystal  interactions  are particularly important  in  mixed-phase
clouds;  when  supercooled  droplets are  accreted by falling ice crystals,  the
process is usually referred to as riming.

Although the  above  discussion has been confined primarily to deposition  in
conjunction with rain and  snow,  it  should  be emphasized that  fog deposition
often  is an  important  secondary process  for  conveying  pollutants  to  the
Earth's surface.  A "fog" is (rather pragmatically)  defined here  as any cloud
adjacent  to  the  Earth's  surface.    Classification of  fog-bound pollutant
deposition is problematic for two major reasons.  The first of these  is that
no  sharp demarcation  exists between  "fog  droplets"  and  "water-containing
aerosols;"  thus,  the  choice of  considering  fog  deposition  as  simply  the
dry-deposition of wet particles, or  the wet-deposition of  contaminated water
depends  primarily  on personal  preference.   Secondly,   no real   distinction
exists  between  fog  droplets and  precipitation.  Cloud physicists  often find
it convenient to categorize  condensed  atmospheric water  into  "precipitation"
and  "cloud"  classifications, with  the presumption  that cloud  water  has  a
negligible sedimentation velocity.   Such  a classification is of  limited  use
when  we consider  fog  deposition,   however,  because fog  droplets  do  have
significant gravitational fall speeds.   A  50-micron  diameter fog  droplet,  for
example, will  fall  at a rate  of about 10 cm  s~l.   This, combined with  the
fact that typical fogs and  clouds contain  droplet-size distributions  ranging
between 0 to  100 microns (Pruppacher and Klett 1978), suggests that  gravita-
tional  transport  of fog  droplets will indeed  be  a significant  pollution-
deposition pathway under appropriate circumstances.

In addition  to purely gravitational  transport, fog droplets  have a  strong
tendency to impact on projected surfaces.   The  rates of  fog  impaction depend
in a complex  fashion upon drop size, wind  velocity,  and  geometry  of  the pro-
jected  object.   The common  observations of  rime-ice accumulation on  alpine
forests  and on  power-transmission lines give direct testimony to  the  effec-
tiveness of this process.

Chemical deposition  by  fogs is directly proportional to  fog-bound pollutant
concentration,  and  this  fact often   acts to  enhance substantially the  path-
way's  overall effectiveness.   Owing to their  proximity  to the Earth's sur-
face,  fogs typically form  in conjunction with  high  pollutant  concentrations.
Attaching  particles  and gases  via   the variety of  mechanisms described  in
Section  6.2.3,  the  droplets  typically  accumulate  extremely high  burdens  of
material.  It is not difficult to find evidence to support this point.  Scott
and  Laulainen (1979),  for  example,  reported sulfate and  nitrate  concentra-
tions  approaching  500 ym  £~1 in  water obtained  near   the  bases of  clouds
over Michigan, while the SUNY group  has reported (Falconer and  Falconer 1980)
numerous similar concentrations (as  well  as extremely low pH measurements)  in
clouds  sampled at the Whiteface Mountain,  NY, observatory.

Recently, Waldman et al. (1982) have reported  nitrate and  sulfate concentra-
tions  in Los  Angeles  fogs  ranging  up to  and  beyond   5000  ym  £-1.    This


                                     6-15

-------
compares  with  typical  precipitation-borne  concentrations  of  about  35 ym
&-1 for the northeastern United States.

Lovett  et al.  (1982)   have  applied  a  simple  impaction  model  to  estimate
fog-bound  pollutant deposition  to subalpine  balsam fir  forests,  and  have
concluded  that  chemical  inputs via this  mechanism exceed those by  ordinary
precipitation by 50 to 300 percent.  This is undoubtedly an extreme  case, and
it would  be more  meaningful  to possess a regional  assessment  indicating the
general importance  of  fog deposition  on an areal  basis.   This  requires  sub-
stantial  effort,  however, involving  climatological  fogging analysis  (Court
1966)  as  well  as numerous  additional  factors, and  no really  satisfactory
evaluation of this  type is presently  available.   Regardless, it is  appropri-
ated to conclude that fog-deposition processes probably play an important, if
secondary  role  in pollutant delivery  on  a regional basis.   In the  future,
more effort should address this important research area.

6.2.6  Combined Processes and the Problem of Scavenging Calculations

The  preceding discussion  of  individual  steps in the  scavenging  sequence has
been intentionally  presented on a highly visual and  non-mathematical  basis,
with appropriate  references  given  for the reader interested in  more detailed
pursuit.   Despite the  qualitative  nature of this  presentation, however, it
should  be  obvious  that  the  most direct  and  expedient  approach  to  model
development is  first  to formulate mathematical  expressions  corresponding to
each of these steps, and  then  to combine them in some sort of a model  frame-
work that describes the composite  process.   This subject will  be examined in
greater   detail   in  Section   6.5,   which   specifically  addresses  scavenging
models.

6.3  STORM SYSTEMS AND  STORM CLIMATOLOGY

In  the present text  the  term  "storm"  is intended  to  denote any  system in
which  precipitation occurs.    This  definition  thus  encompasses all  occur-
rences,  ranging  from  mild precipitation  conditions  up  to and through the
major  and cataclysmic events.

6.3.1   Introduction

From the  preceding discussion,  it is easy to  imagine  that scavenging rates
and  pathways  will be dictated  to  a large extent by the basic  nature  of the
particular storm  causing  the wet  removal  to  occur.  Storms containing water
that is  predominantly  in the  ice  phase, for  example, will  provide  little
opportunity  for  attachment  mechanisms  associated with  droplet nucleation,
accretion, or  phoretic processes.   The  abundance of  liquid water and the
temperature  distribution  in  a  given  storm will  have  a direct bearing  on the
degree to which aqueous-phase  chemistry can occur.  Storms containing  no ice
phase  whatsoever will  be generally  ineffective as generators  of precipita-
tion,  and thus  will tend  to inhibit  the  scavenging process.  An interesting
indication of  the  importance  of  storm  type  in this regard is presented in
Figure 6-23   (see  Section   6.5.4),   which  presents  estimated  scavenging
efficiencies  which vary  extensively with  storm  classification.   Different
storm   types  differ  profoundly  with  regard  to  inflow,  internal  mixing,


                                      6-16

-------
vertical   development,  water  extraction  efficiency,  and  cloud   physics;
consequently it is  appropriate  at this point  to  consider briefly the major
classes and  climatologies of  storm systems occurring  over the  continental
United States.

Two major points  should  be  stressed at the outset  of  this discussion.   The
first  of  these is  the essential  fact that all  storms  are  initiated by  a
cool ing of  air,  which leads  to a  condensation  process.    Such  cooling  may
occur  by the transport of sensible heat,  such as when  a  comparatively warm,
moist air parcel  flows over a cold  land  surface.   The  dominant cooling mode
for most  storm systems,  however,  is  expansion,  which occurs  via  vertical
motion  of  the air  parcel   to  elevationsof  lower  pressure.   The  second
noteworthy point in  this  context is that the overwhelming majority  of storm
systems is  strongly  associated  with fronts between  one or more  air  masses.
The primary  reason  for this  associaton  is that  thermodynamic perturbations
and  discontinuities  associated  with  the  frontal  surfaces  provide   the
opportunity  for  vertical motion (and  thus  expansion  processes)  to  occur.
This relationship is an essential component of storm classification  systems,
and will  emerge repeatedly in the following discussion.

Overlaps  in  the  characteristics  of different  storm types  render  a  strict
classification largely impossible.   For practical  purposes,  however, it  is
convenient  to  segregate  midlatitude  continental  storms   into  two  classes,
which  are usually described  as  being "convective" and  "frontal."  These  two
major  categories then  can be  subdivided further as  deemed expedient for  the
purpose at hand, although it  should be noted that  significant overlap among
storm  types occurs  even  at  this  major  level  of  classification.    Frontal
storms, for example, often possess  significant convective  character  in their
basic composition, and true convective storms often  occur  as the  consequence
of  fronts.     Because  of  this,  the  following  discussion  will   use storm
classification primarily  as a descriptive aid and will  not belabor taxonomic
detail.

6.3.2  Frontal  Storm Systems

Much of what is understood today  regarding midlatitude  frontal-storm systems
stems  from  the pioneering work  of the Norwegian meteorologist Bjerknes,  who
conducted a systematic survey of large numbers  of storm systems  and  from  this
survey  developed  a  conceptual  model  of frontal-storm development  and  be-
havior.  Characterized schematically in Figure  6-5,  the Bjerknes  model  can be
understood most  easily by considering a  cool  northern air mass, separated
from a warm  southern air  mass by an east-west  front, as  indicated in  Figure
6-5a.  The  progression of figures represents a typical result of the  atmos-
phere's natural tendency to  exchange heat from  southern to northern latitudes
across this  front.    This is  often referred to  as a  "tongue"  of  warm  air
intruding into the cold air mass.   In  the northern  hemisphere this wave  will
tend to propagate in  an  easterly direction;  thus,  the  intrusion  is  bound by
two moving fronts—a warm front  followed by a cold  front—as shown in  Figure
6-5c.

Flows  associated with  the wave  system occur in a manner  such that a depres-
sion in atmospheric pressure occurs  at the vertex of the  warm-air intrusion;


                                     6-17

-------
                         HIGH
00
                                                             LOW
                                                                               HIGH
                         HIGH
                  Figure 6-5.   Cyclonic storm development according to Bjerknes's conceptual model.

-------
as a consequence a general  counterclockwise  or  "cyclonic"  circulation  pattern
emerges.   Because  of this  feature, Bjerknes's  conceptual  model  is  often
referred to as  the  "Bjerknes cyclone theory,"  and frontal storms  associated
with this pattern are termed "cyclonic" storms.  A typical feature  of storms
of  this  type  is the tendency for the cold  front to  overtake the warm  front
and ultimately annihilate the wave.   The  "occluded"  front created  as a con-
sequence of this behavior is shown schematically in Figure 6-5d.   In  view of
this  birth-death  sequence  of  the  Bjerknes  cyclone  model,  the  progression
depicted in Figure 6-5 often has been termed the "life  history"  of a cyclone.
Some idea of spatial scale and the  general cyclonic flow  pattern  of a mature
cyclone are given  in  Figure 6-6.   In viewing  these indicated  flow patterns,
however, the  reader should  note carefully that considerable vertical  struc-
ture  exists  in such  systems,  and marked deviations of the  wind  field with
elevation are  typical.   In particular,  one  should take care not  to  confuse
the indicated general circulation patterns with corresponding surface  winds.

Although created  from the  limited  observational base available during  the
early twentieth century, the fundamental precepts of  the Bjerknes theory have
proven  valid  even as more sophisticated observational   and analytical  facil-
ities have  become available.   Certainly  non-idealities and deviations from
this  model  occur;  but  its general  concepts  have  proven  to  be  immensely
valuable as a  conceptual basis  and  as an idealized standard for  the  assess-
ment  of actual  storm systems.   Comprehensive  descriptive  and  theoretical
material pertaining to  such systems is  available  in the  classic  text  by
Godske  et  al.  (1957),  and more elaborate and  modern extensions are given in
the periodical literature (e.g., Browning et al. 1973,  Hobbs  1978).

6.3.2.1   Warm-Front  Storms--It is  important  to note that  the plan  views
exhibited by  Figure 6-6  are gross  simplifications,  since they  do nothing to
characterize  the  three-dimensional   nature  of  the cyclonic  system.   If  one
were  to  construct a vertical cross  section  of  the warm front (A-A1  in Figure
6-6), then typically  one would  observe  an inclined frontal  surface as  shown
in  Figure  6-7.   (See Table  6-1  for definitions  of cloud  abbreviations.)   In
this  situation  the presence of  warm air aloft  creates a relatively stable
environment,  which  inhibits vertical  mixing   of  air  between  the  two  air
masses.   The warm, moist  air  moves up over the cold  air wedge,  expanding,
cooling, and ultimately  forming clouds and precipitation.   Typically the warm
air  supplying  moisture  for  this  purpose  has been advected  from  deep within
the  southern  air  mass,  carrying water vapor  and  pollutant over  extensive
distances.  This  transport  trajectory has been aptly compared  to a "conveyor
belt" for moisture  by Browning et al. (1973).  It is  appropriate to note that
this  moisture conveyor belt  is a conveyor belt for pollution as well.

Warm-front  storms  are often associated with long periods of continuous pre-
cipitation,  although significant structure  can exist within  such systems.
Important  structurally  in this  regard  are  the prefrontal rain bands,  which
take  the form  of  concentrated  areas  of precipitation embedded within  the
major storm  system.    At  present, the  factors contributing   to  rain-band
formation  are not  totally  understood, although  mechanisms  such  as  seeding
from  aloft by  ice crystals and nonlinearities of the associated thermodynamic
and flow processes  undoubtedly contribute to a major extent.
                                     6-19

-------
I
t\3
O
  Figure  6-6.   General  flow patterns  in  the  vicinity of an  idealized  cyclonic  storm  system.   Arrows  denote
               general  circulation  patterns  and should not  be  interpreted  as surface winds  (cf.  Figures
               6-7,  6-8,  and 6-9).

-------
CTl
I
ro
                      >///^7/A^7/A;5>7//-

                            400
                                 Km
      FLOATING  ICE  NEEDLES

      FALLING ICE NEEDLES
                                                               600

                                                                FALLING RAIN
                       FLOATING FOG DROPS


                       "ICE NUCLEI LEVEL"


                       FALLING  SNOW
                                       :::::::::  FALLING DRIZZLE


                                       	  0°C ISOTHERM


                                            =J  RELATIVE VELOCITY OF  WARM AIR

                                               RELATIVE VELOCITY OF COLD AIR
   Figure 6-7.  Vertical  cross  section  of a typical warn, front (Section A-A<  on Figure 6-6).  Adapted
from Godske et al. (1957).

-------
  TABLE 6-1.  SUMMARY OF CLOUD  TYPES APPEARING
          IN FIGURES 6-7 THROUGH 6-9
Type                        Abbreviation
Cirrus                                 Ci
Cirrostratus                           Cs
Cirrocumulus                           Cc

Altostratus                            As
Atlocumulus                            Ac

Stratus                                St
Stratocumulus                          Sc
Nimbostratus                           Ns

Cumulus                                Cu
Cumulonimbus                           Cb
                         6-22

-------
 Warm-front  storms usually  can  be expected  to  be rather effective  as  scav-
 engers of pollution  originating  from  within  the  warm air mass,  especially if
 temperatures  in the  feeder  region are sufficiently high to allow the presence
 of  liquid  water and the nucleation-accretion process.   Scavenging  of pollu-
 tants from  the underlying cold air mass will  usually be less effective,  owing
 to  the  relative scarcity of clouds  and  generally  less definitive  flows  in
 this  sector.   Scavenging   in  both  regions  will  of  course  depend  upon  the
 physiochemical  nature of  the pollutant  of  interest  and the  microphysical
 attributes  of the cloud system in general.  Methods for estimating scavenging
 rates in such circumstances  are discussed in Section 6.5.

 6.3.2.2  Cold-Front  Storms—A  typical  vertical cross section  (B-B1  in Figure
 6-6)  of a  cold-front  storm is shown in  Figure  6-8.   This differs  substan-
 tially  from the warm-front  situation in  the sense that, instead  of flowing
 over the frontal surface, the warm air is forced  ahead by the  moving cold air
 mass.   This  action  produces a more  steeply inclined frontal  surface  that,
 combined with the presence  of low-elevation warm air, creates  a  relatively
 unstable  situation  leading  to  convective   uplifting  and the  formation  of
 clouds and  precipitation.

 Although discussed here in  a frontal-storm context,  this precold-front  situ-
 ation composes  an  important class of convective  storms,  which  will  be  dis-
 cussed  in  some  detail later.   Scavenging rates   and  efficiencies  associated
 with such storm systems will again depend upon the pollutant and the physical
 attributes  of the particular cloud system involved.

 6.3.2.3  Occluded-Front Storms—Because occluded  fronts are  formed via merger
 of warm and cold fronts, it  seems reasonable  to expect that  storms associated
 with  occlusions  should share  characteristics of the  respective  elementary
 systems.  Figure 6-9,  which shows a typical  vertical  cross section  (Section
 C-C1 on Figure 6-6) of an occluded system, demonstrates this point.   Typical-
 ly the easterly flow of warm air aloft maintains  a relatively  stable  environ-
 ment to the east of the occlusion, and clouds and precipitation  occur in  this
 region largely as a consequence of ascending flow from  the south.  Much  more
 detailed accounts  of occluded systems  can   be found in standard  references
 such as the book by Godske  et al. (1957).

 6.3.3  Convective Storm Systems

 An idealized  cross section  of  a  typical  convective storm is  shown in Figure
 6-10.  Such storms depend upon atmospheric instabilities to  induce  the neces-
 sary vertical motions and concurrent cooling and  condensation processes;  and
 as such they  are most likely to  occur  under  warm, moist conditions where the
 energetics  are  most  conducive  to  this  process.   Often  convective storm
 systems occur as "clusters"  of cells, such as that  shown  in Figure 6-10,  and
 exhibit a marked  tendency  to exchange moisture  and pollutant between cells;
 thus, the flow dynamics and scavenging  characteristics of such systems  tend
 to be extremely  complex.

Typically the moisture  and  pollutant  input  to  a  convective  cell occurs
 primarily through  the storm's  updraft  region (cf.,  Figure  6-10),  although
entrainment  from upper regions  is possible as well.   Dynamics  of this process


                                    6-23

-------
cr>
i
ro
         ^^s^^^

         0              200              400              600             800
                                  *** *
                                      ** *
         B                                       Km                                       B-


                                         FLOATING  ICE  NEEDLES

                                         FALLING  ICE NEEDLES


                                         FLOATING FOG  DROPS

                                         MICE NUCLEI LEVEL"


                                         FALLING SNOW


                                         FALLING RAIN


                               iliHIiilii  FALLING DRIZZLE

                               	  0°C  ISOTHERM

                                <3=   RELATIVE VELOCITY OF WARM AIR

                                •*—•   RELATIVE VELOCITY OF COLD AIR


Figure 6-8.   Schematic vertical  cross section of  a  typical  cold front (Section B-B1  on Figure 6-6)
             Adapted from Godske et al.  (1957).

-------
ro
en

                     •*
                      f*|«|*#
                    1*1*1* *l*l*l*
                    ft

                                            FLOATING ICE NEEDLES

                                            FALLING ICE NEEDLES



                                            FLOATING FOG DROPS


                                            "ICE NUCLEI LEVEL"


                                            FALLING SNOW



                                            FALLING RAIN


                                            FALLING DRIZZLE

                                            0°C ISOTHERM

                                            RELATIVE VELOCITY OF WARM AIR

                                            RELATIVE VELOCITY OF COLD AIR

                                            RELATIVE VELOCITY OF COLDEST  AIR
  Figure  6-9.
Schematic vertical cross section of a typical  occluded  front (Section C-C1 on Figure 6-6)
Adapted from Godske et al. (1957).

-------
                                               92-9
crt

i—»
O
CL
(V
01
(SI
n

o
re
o
O


o
-h

Ol
3



O

o>

fl>
CL

n
o
CD
O
I/)
c-h
O
                                         HEIGHT ABOVE GROUND  (m)
                                        TEMPERATURE (°C)

-------
are such  that  violent updraft velocities often  occur;  these are capable  of
lifting entrained  air, water  vapor, and  pollution to  extremely high  ele-
vations (sometimes breaching the  stratosphere).   Along  this course,  entrained
pollutant  is  subjected to  a large  variety of  environments and  scavenging
mechanisms; as  will  be noted  in Section 6.5, convective  storms tend to  be
highly effective scavengers of air pollution.

As was  stated  earlier, convective storms  often  are associated  with  frontal
systems,  although  frontal   influence  is not absolutely  necessary for  their
presence.  An isolated air mass,  for example,  is  totally  capable of acquiring
sufficient energy  and  water vapor to induce a convective  disturbance  on its
own accord.  Perturbations  arising from  fronts,  however,  often  contribute  to
the creation of convective  activity—if  for no other reason than supplying a
"trigger" to initiate convection  in a conditionally unstable atmosphere.

6.3.4  Additional  Storm Types:  Nonideal Frontal  Storms,  Orographic
       Storms and Lake-Effect Storms

As noted  previously,  the  Bjerknes cyclone  model  represents something of  an
idealized  concept,  and numerous  features can  contribute to deviations  from
this  "textbook" behavior.   Orographic  effects are highly  important  in  this
regard.   Consider,  for  example,  a  cyclonic  disturbance approaching  the North
American  continent from across the Pacific  Ocean;  the  frontal  patterns typi-
cally lose much of their  original  identity after  impacting  with the  western
mountainous regions.    In  addition  to  the  physical  distortion  of  flow  pat-
terns,  the  lifting induced by the terrain  encourages  further  precipitation,
resulting  in  large spatial  variability  in rainfall patterns  and pronounced
local phenomena such as "rain shadows" and chinooks.  Precipitation-formation
and  precipitation-scavenging  processes  associated with  such systems  tend  to
be highly complex.

Frontal systems often tend  to reconstitute their structure after crossing the
Rocky  Mountains,  but  continental  effects  still  impart  a  marked impact  on
their  basic  makeup.   In  the midwest-northeast  region,  for example,  fronts
tend to orient themselves in an east-west direction and become  stationary for
extended  periods,  often punctuated by several  minor low-pressure areas.  Even
under  relatively  ideal  conditions  continental  frontal  storms tend to  possess
more convective flavor  in  their  basic makeup than  do  their oceanic counter-
parts.

As indicated above,  terrain-induced  or  "orographic"  effects are usually most
important in  augmenting  major  storm  systems,  although   relatively  isolated
orographic  storms  (such  as  oceanic "island-induced" storms)   certainly  do
occur.    Orographic  effects obviously  will  tend  to  be  most   pronounced  in
regions  where  radical  terrain changes  occur;  but even  the small  elevation
changes typical of the  Midwest  can contribute significantly at times.   Oro-
graphic effects also  are  suspected  to influence  storm  behavior over substan-
tial  downwind  distances.    Lee waves  from  the  Rocky Mountains,  for example,
have been  suggested to trigger thunderstorm formation at extended distances.

Lake-effect storms are  yet another example of a  somewhat  non-ideal  phenome-
non,  which often  is  superimposed with  more  major meteorological  patterns.


                                      6-27

-------
Typically such storms occur during fall and early winter,  when  land surfaces
tend to  be cooler  than  their adjoining  water bodies.   Considering an  air
parcel  moving  on  an easterly course  across  Lake Michigan, for  example,  one
can note that  the warm lake  surface  should supply both heat and  water  vapor
as  it  proceeds.    As  this  parcel  is  advected  across  the downwind  shore,
however, two  important  things will occur.   First,  the  cold  land  mass  will
extract heat from the air; second, the  orographic lifting  (on  the order of a
few tens of meters)  will  result  in  ascent,  expansion, and further cooling.
The  net result  is  a  lake-effect  storm.    Such  storms  can  induce  highly
variable precipitation  patterns  in  specific  areas  around the  Great  Lakes
region.   Although  confined  largely   to  this portion  of  the  United  States,
these  storms  account for  a  majority of  the snowfall  that  accumulates  in
specific cities such as  Muskegon, MI,  and Buffalo, NY.   Some appreciation for
the magnitude  of this effect can be gained  by  viewing  the  climatological
precipitation map given  in Figure 6-11.

6.3.5  Storm and Precipitation Climatology

The exceedingly complex  subject  of storm climatology will  be discussed  here
only to  the point  necessary to  describe  some  key  attributes  and  indicate
references for  more detailed pursuit.   Factors  especially important in  the
context of  precipitation  scavenging   are  temporal  and  spatial  precipitation
patterns,  storm-trajectory  behavior,  and  storm-duration  statistics.   These
will be discussed in the following paragraphs.

6.3.5.1  Precipitation Climatology—Figure 6-12 provides climatological  aver-
ages of monthly  precipitation   amounts  at various  stations  throughout  the
United  States.    This  figure,  taken  directly from  the U.S.  Climatological
Atlas  (1968),  requires little elaboration at this point.   It is  interesting
to note, however, that precipitation  amounts do not vary radically throughout
the year at most  northeastern U.S. stations;  this contrasts especially  with
the  arid western  stations,   whose  seasonal   variabilities  tend  to be  pro-
nounced.   It should be noted as  well  that actual  precipitation  amounts  for a
given  single  month  can  vary appreciably from  the  climatological  averages
presented here.

6.3.5.2   Storm  Tracks—Because  of  the  difficulties  noted previously  with
regard  to  precise  classification  or  definition  of  storms, a truly  concise
climatological summary of storm-pathway behavior  is  largely impossible.   Some
useful  information  can be generated,  however,  by  observing  the  tracks of the
cyclonic (low-pressure)  centers  associated with  major storm systems.   Klein
(1958), for example, has conducted a   systematic survey of cyclonic centers in
the northern hemisphere and  from this has constructed  monthly climatological
maps of  low-pressure tracks.  Figure 6-13,  taken  from the book  by Haurwitz
and Austin  (1944),  presents  the combined results of the  analyses  by several
previous authors.   On the  basis  of the previous  discussion it should be  re-
emphasized that, owing to the complex flow processes  associated  with cyclonic
systems, one should  not interpret the motion of these low pressure centers as
being  identical with feeder trajectories for the  storms themselves.  Success-
ful  interpretation   of  such   information  in   the  context of  source-receptor
analyses requires careful and skilled meteorological  guidance.
                                     6-28

-------
                     \\vv
Figure 6-11.
Average annual snowfall pattern (inches) over Lake Michigan
and environs.  Adapted from Changnon (1968).

              6-29

-------
                                        NORMAL MONTHLY TOTAL PRECIPITATION (Inches)
CTl
I
00
o
                     JW/i..,,*  ^V	r-f
                     fiji-lS  , V-^-  f        J ;

                     /    /   Illll.	   fc H        *'
 Figure  6-12.  Climatological  Summary of U.S.  Precipitation.  From  U.S. Cl imatological  Atlas (1968).

-------
Several additional  points should be emphasized in the  context of Figure  6-13.
Firstly, it should be noted that this presents  a long-term  composite  average
and that marked  deviations from this pattern can be  expected  to occur with
season.  Secondly, the  statistical  variability  of storm tracks  is  such that
substantial departures  from  the long-term averages can  be  expected  for  any
particular year.  Finally, substantial evidence  documents longer-term shifts
in  average   storm-track  distributions   (Zishka  and  Smith   1980);   thus,
presentations (such as  Figure 6-13)  that are based upon historical data  may
vary considerably  from   storm patterns  to be  observed  over the next twenty
years.  The implications of  this with regard to  long-term  acidic  deposition
forecasting are obvious.

Additional  features of  cyclonic storm climatology can  be  found in  standard
climatological textbooks  (e.g., Haurwitz and  Austin 1944).   Convective-storm
climatology,  which tends to  be much  more  region-specific,  can  be  evaluated
from  such  references  as  well,  although  more   recent  weather  modification
programs such as METROMEX, NHRE,  and HIPLEX  have  generated a  considerable
amount of new information in  this area.

6.3.5.3  Storm Duration  Statistics—In preparing  regional scavenging  models,
it  often  is  desirable  to create some sort  of  statistical  average of  storm
characteristics  so that  "average"  wet-removal  behavior  can  be  defined.
Although little  activity has  been  devoted to this area until  very  recently,
the usefulness  of such  an approach to  regional  model  development  suggests
accelerated effort during future years.

The analysis  by  Thorp   and  Scott  (1982)  provides an  example  of one  such
effort.  These  authors  compiled data from hourly precipitation  records from
northeastern   U.S.   stations   to   obtain    seasonally-stratified   duration
statistics, which were  expressed  in terms of probability  plots as shown  in
Figure 6-14.   As  can  be noted from  these  plots, "average"  storm  durations
during summertime are significantly  less than durations of  their  wintertime
counterparts,  reflecting  relative   influences  of  short-term  convective
behavior.   Some of the references given  in Section  6.5  suggest  potential
modeling applications  for these statistical  summaries.

6.4  SUMMARY OF PRECIPITATION-SCAVENGING  FIELD INVESTIGATIONS

For the  purposes of this  document "field investigations" of precipitation-
scavenging  mechanisms   will  be  differentiated   from  routine  precipitation-
chemistry network measurements, which are intended primarily for characteri-
zation purposes.  Of course a great deal of  overlap occurs  between  these  two
classes of measurements, and  significant  reciprocal benefit  is generated as a
consequence  of  each.     Some  essential   differences  exist  between  the  two,
however,  and  it  is  convenient   for  present  purposes  to  segregate  them
accordingly.

The primary  distinguishing feature of  a scavenging  field   investigation  is
that the  study usually  is designed around  the basis of some  sort  of con-
ceptual or interpretive  model(s)  of scavenging behavior,  which  is  tested on
the basis  of  the field  data.   If the model  predictions  and data  disagree,
then  some  basic  precepts of  the  model  must  be  invalid,  and  additional


                                     6-31

-------
Figure 6-13.
Major climatological storm tracks for the North American conti'
nent.  Adapted from Haurwitz and Austin (1944).  Dashed lines
denote tropical cyclone centers, and solid lines denote those
of extratropical cyclones.
                    6-32

-------
01
i
CO
CO
o

CO
o;
LU <
    Q_
    i—i
    O
    o;
    Q.
                                           LEGEND

                                       CUMULATIVE  PERCENT  FREQUENCY
                                       OF  STORM  DURATION
                                       CUMULATIVE  FRACTION OF TOTAL
                                       REGIONAL  PRECIPITATION
                                       THREE-POINT SMOOTHED AVERAGE
                                       FRACTION  OF TOTAL REGIONAL
                                       PRECIPITATION
                                                                                        SUMMER
                                                                                  (JUNE,  JULY,  AUGUST)
                                                                                       < 3 HR  DRY
                                                                    TOTAL NUMBER OF STORMS = 4516.
                                                                    TOTAL NUMBER OF EVENTS = 6198.
                                                                                                    -1-
                                                                   GR. AVE. PCPN RATE - 00088 in  hr
                                                                     MAXIMUM EVENT RATE - 1.49 in hr'1
                                                                    .  .   i  .  .   i  I   i  i	i   i  i  I   i  i
                                                            40
                                                                   50
                                                                                60
                                                                                         70
CO

§  o.
CO
   o.


   Oa


   0.


   0.


   0.
      10


      08


      06

      04


      02


      0
        0


Figure 6-14.
                                                                                   WINTER
                                                                         (DECEMBER, JANUARY, FEBRUARY) -
                                                                     TOTAL NUMBER OF STORMS - 3870
                                                                     TOTAL NUMBER OF EVENTS - 857l'

                                                                GR. AVE. PCPN RATE - 0.025 in hr"1-
                                                                 MAXIMUM EVENT RATE - 0.38 in hr'1-
                                                                                                70
1.0

0.8


0.6


0.4


0.2

0



1.0


0.8

0.6

0.4


0.2


0
                                                                                                       O
                                                                                                       UJ
                                                                                                       OL
                                                                                                       Q.
                                                                                                        ID
                                                                                                        UJ
                                                                                                       
-------
mechanistic  insights  must be  generated to  rectify the  situation.    In  the
event  that  predictions and data  agree,  then this may  be taken as  evidence
that the  precepts may be  correct.   Regardless of whether positive  or nega-
tive results are  obtained  (and  assuming that the field study  has  been well-
designed  and  well-interpreted),  an   advance   in   understanding   has  been
achieved.  The importance  of  such input cannot be  overemphasized.   Examples
exist  wherein  field  investigations have demonstrated then-accepted models to
be in  error by several orders of magnitude (e.g., Hales et al.  1971).   Field
studies have been essential in keeping the models "honest."

Field  studies  of precipitation scavenging were  begun  in earnest  during  the
early  1950's for  the primary  objective  of  radioactive-fall out  assessment.
Pioneering  studies  in this  area that pertained  to  radioactive  pollutant
releases from point  sources in  anticipation  of  reactor  accidents  and related
phenomena were performed in England by Chamberlain (1953).  These  constituted
the basis  for  the washout-coefficient  approach  to   scavenging  modeling  (see
Section 6.5).  Other studies focused primarily on nuclear-detonation fallout,
thus approaching  the scavenging problem from a more global point of view.

Following the  English lead, nuclear-oriented  studies  were  conducted  by  the
United  States,  Canada,  and the  Soviet Union.   These  included   studies  of
tracers as well  as  those  of the radionuclides themselves; and  although  some
of this material   still remains in the classified literature,  it may be stated
with certainty that most of what we know today regarding scavenging processes
has been generated as  a consequence  of  the nuclear  era.   The review "Scaven-
ging  in Perspective"  by  Fuquay (1970)  presents  a   comprehensive  account of
this early stage  of scavenging field studies.

During  the late 1960's field-experiment emphasis shifted to more conventional
pollutants, with  the general recognition of precipitation scavenging's impor-
tance  in preserving  atmospheric quality and  its  potential adverse  impacts of
deposition on the Earth's  ecosystem.  Since  that  time  a  variety of large and
small  field  studies  has  been  conducted.   These  are  summarized in  Table  6-2,
which  provides a logical   classification  in  terms of source  type, pollutant
type,  and geographical scale.

Although field studies have been  focused  strongly on quantitative  aspects of
precipitation  scavenging,  they  have provided  important  qualitative informa-
tion  regarding  acidic-precipitation  processes  as   well.    The ensemble  of
studies listed in Table  6-2 presents a  rather  cohesive base of evidence in
this  regard;  and although  some  conflicting results  and  uncertainties  do
exist,  a  generally  coherent picture can be  constructed  in  several important
areas.   Although there  is considerable overlap  of  source-receptor distance
scales  among these studies, they tend to group rather conveniently into three
classes of  area!  extent:   0  to 20 km, 0 to 200 km,  and 0 to  2000  km.   These
classes  shall  be termed loosely  as  "local," "intermediate,"  and  "regional"
scales  in the following discussion, where key qualitative features  are illus-
trated  by  considering the  fate of  specific  acidic-precipitation  precursors
(SOX,  NOX,  and  HC1)  as  they  are transported  over these increasing  scales
of time and distance.
                                     6-34

-------
                TABLE 6-2.  SUMMARIES  OF  SOME PRECIPITATION  SCAVENGING FIELD INVESTIGATIONS
     General  source type
     Specific source type
               Selected references
     Continuous Point
     Source
cr>
i
oo
en
Tower releases of aerosols
Tower releases of radioactive
gases and simulated tracers

Tower releases of S02

Tower releases of tritiated
water vapor

Tower releases of organic
vapors

Power-plant plumes
                            Smelter  piumes
Chamberlain (1953), Engelmann (1965), Dana
(1970)

Chamberlain (1953), Engelmann (1965)
                                                             Dana  et al.  (1972),  Hales  et al.  (1973)

                                                             Dana  et al.  (1978)


                                                             Lee and Hales  (1974)
Dana et al. (1973, 1976, 1982), Granat and Rodhe
(1973), Granat and Soderlund (1975),
Hales et al. (1973), Barrie and Kovalick (1978),
Hutcheson and Hall (1974), Enger and
Hogstrom (1979), Radke et al. (1978)

Kramer (1973), Larson et al. (1975)
Mill an et al. (1982), Chan et al.  (1982)
      "Instantaneous" and/
      or Moving Sources
Aircraft releases of rare-
earth tracers
Dingle et al. (1969), SI inn  (1973), Young et al.
(1976), Gatza (1977), Changnon et al.  (1981)
                            Rocket releases of radioactive
                            tracers
                                 Shopauskas et al. (1969), Burtsev et al. (1976),

-------
                                           TABLE  6-2.   CONTINUED
     General  Source Type
     Specific Source Type
               Selected References
en
CO
en
     Urban Sources
     General  and Regional
     Sources
Uppsalla, Sweden

St, Louis, MO
Los Angeles, CA
Regional pollution flowing
into lake-effect storms

General sources in western
Canada

Regional pollution in the
eastern U.S. and Canada

Regional aerosol loadings at
a specific receptor point
Hostrom (1974)

Hales and Dana (1979a)

Morgan and Liljestrand (1980)

Scott (1981)


Summers and Hi tenon (1973)


MAP3S/RAINE (1981), Easter (1982), Mosaic (1979)
                                                            Graedel  and  Franey (1977),  Davenport and Peters
                                                            (1978)
     Global  and Strato-     Cosmogenic radionuclides
     spheric Sources

                            Nuclear fallout
                                Young et al.  (1973)


                                Numerous studies;  see Fuquay (1970)
     aThe reference by Gatz provides a comprehensive  list  of  past  tracer studies  of  precipitation
      scavenging.

-------
On a local scale (0 to 20  km),  field  studies  have  generally  demonstrated  the
precipitation  scavenging  of  sulfur and  nitrogen  oxides  from  conventional
utility and  smelting  sources to be minimal.   The  virtual absence of  excess
nitrate or nitrite ion in precipitation samples collected beneath such  plumes
(Dana  et  al. 1976) provides  strong evidence  that direct uptake of primary
nitric oxide and  nitrogen  dioxide  by  precipitation  and  cloud elements is  a
negligibly slow process.

Nonreactive scavenging of  plume-borne  sulfur  dioxide  is  solubility dependent
and tends also to be a rather inefficient process, although  it  is definitely
detectable in field studies conducted  in relatively clean environments  (Hales
et al. 1973; Dana et al. 1973,  1976).   This  phenomenon, which  is suppressed
under  conditions  involving high rain  acidity, is  relatively  well understood
at present (Hales 1977, Drewes and  Hales 1982).

Nonreactive  scavenging  of  sulfate aerosol   can  be  an  efficient  removal
process.   The  preponderance of relevant  field tests in Table  6-2,  however,
has  demonstrated  that  wet deposition  of  sulfate  from local   power-plant  and
smelter plumes  occurs  rather slowly.    This is undoubtedly a consequence  of
the  small  amounts  of  primary  sulfate available   for  scavenging under  such
circumstances.

Field  tests  conducted under  situations wherein sulfur  trioxide was  inten-
tionally injected into the stack of a  coal-fired power plant  (Dana and  Glover
1975)  show correspondingly high sulfate  scavenging  rates,  and it  has  been
suggested that under certain operating conditions  some types  of power  plants
(especially  oil-fired units)   will  produce  sufficient  primary  sulfate  to
account for appreciable local deposition.   To  date, however,  no  really  strong
field  evidence has  supported  this point.    Hogstrom   (1974)  reported  the
observation  of substantial  sulfate scavenging from  the local  plume  of  an
oil-fired power plant in Sweden, but these  results are  rather dependent upon
the  interpretation of background contributions.  Granat  and  Soderlund  (1975)
performed  a  similar  investigation  in  the  vicinity of  a  second  Swedish
oil-fired plant and found a comparatively  small scavenging  rate.

Reactive scavenging of plume-borne  sulfur dioxide  to form rainborne sulfate
is difficult to differentiate  from  primary sulfate removal.  The previously
noted  findings  of  low excess sulfate  in  below-plume rain samples,  however,
suggest that  neither  process is particularly  effective  in near-source plume
depletion.

The scavenging of hydrochloric  acid to produce chloride  and  hydrogen ions  in
precipitation  will  most  certainly  be  a highly effective process, depending
upon the quantities of hydrochloric acid available.  Considerable theoretical
and laboratory work has been conducted in  this area  for  space-shuttle  impact
assessment, and limited  data  suggest  that hydrogen chloride  is  scavenged  in
measureable amounts from  power-plant plumes (Dana et al.  1982).

With the exception of studies conducted under  rather  clean  ambient conditions
(e.g., Dana et al. 1973,  1976),  the influence  of background contributions  has
                                    6-37

-------
made the interpretation of plume scavenging a difficult  task.   Typically  the
sulfate and nitrate concentrations in precipitation collected adjacent to  the
plume are quite variable, and subtracting this  influence to  determine source
contributions involves substantial  levels of uncertainty.  This difficulty of
"source attribution" at the local scale  is  compounded  appreciably  as  greater
scales of time and distance are considered.

On  a  more  intermediate scale  (0 to  200 km)  an  enhancement of sulfate  and
nitrate  precipitation  scavenging  seems  to  occur,  presumably  because  the
precursors  have  had  more  opportunity to  dilute  and  to  react under  these
circumstances.  Hogstrom (1974), using an extended network of samplers in  the
vicinity of Uppsala,  Sweden, reported  substantial  scavenging rates of sulfur
compounds.  Hales and Dana (1979a)  have observed summertime convective storms
to  remove  appreciable  fractions  of urban  NOx  and   SOx  burdens  in  the
vicinity of  St.  Louis, MO.  Although  both  of these studies  were  subject to
the  usual  uncertainties with  regard  to background  contributions there  is
little doubt  about  their general conclusions  of significant  scavenging under
such circumstances.

On  a regional  scale  (0  to 2000  km)  there  are  relatively  few  data  from
intensivefield  experiments.   Precipitation-chemistry  network  data   are  of
some  use  in this regard,  however,  and  several  analyses have  applied these
measurements  to  specific ends.   One  result  of these  analyses is the  sug-
gestion that, in the  northeastern quadrant  of the  United States, roughly  one
third  of the emitted NOX  and  SOX  are  removed by  wet  processes (Galloway
and Whelpdale 1980).   Network data for the  Northeast (MAP3S/RAINE  1982)  show
also  that  the  molar  wet  delivery  rates  of  NOX  and  SOX  are  roughly
equivalent.    Combining  this  result  with  regional   emission  inventories
suggests that nitrogen compounds begin  to  wet deposit  with  a significantly
enhanced efficiency as distance scales become regional  in extent.

The  above  changes  in  behavior with  increasing  scale seem  to be  a  logical
consequence  of  current understanding  regarding  the  atmospheric chemistry of
SOX  and  NOX.  On  local  scales neither  is  scavenged very effectively owing
to  the chemical makeup of the primary emissions.  On intermediate scales both
groups  have  had  some  opportunity  to  react  into  more  readily scavengable
substances.   Depending  upon  ambient  conditions,  nitrogen  oxides  will  have
participated  to some  extent in  initial  photolysis reactions  and proceeded to
form  scavengable  products  such as  nitric acid,  peroxyacetyl  nitrate,  and
nitrate  aerosol.   Sulfur dioxide also  will  have reacted  homogeneously to a
limited  extent;  more importantly,  however,   this compound  will   have  been
diluted  to  levels  where  limited  reactants  (and  possibly   catalysts)  will
facilitate  its  oxidation  in  the aqueous  phase.  On  a  regional  scale  this
progression continues  with  the relative  acceleration of  NOX  scavenging.

Present  field-study  indications  that  NOX  scavenging  may   occur primarily
through  the attachment of gas-phase  reaction  products,  while the scavenging
of  SOv  may depend much more heavily  upon  aqueous-phase oxidation processes,
are  also reflected in precipitation-chemistry  data.   A possible  consequence
of  this  difference in mechanisms is  illustrated in  Figure  6-15,  which  is a
time-series  of  daily precipitation-chemistry  measurements for a northeastern
                                  6-38

-------
      150
      100
       50
   o
   3.
                        TOTAL SULFUR
                     •  •
                                  ; •  * */-\ •
      150
   O
   CO
      100
       50
                                  NITRATE
              y   -A:-*:       .-:.;•
             0.5   1.0    1.5    2.0   2.5   3.0

                       YEARS SINCE JULY 1, 1976
                                     3.5    4.0
Figure 6-15.
Sulfate and nitrate concentration data for event
precipitation samples  collected at Penn State University,
PA.  Lines are least-squares of linear and periodic
functions (MAP3S/RAINE 1982).
                          6-39

-------
U.S. site.   The decidedly  periodic?  behavior of sulfate-ion  concentrations
in  contrast  to  the largely disorganized  behavior  of nitrate-ion  concentra-
tions  has been  suggested  to occur  as  a  consequence  of  an  aqueous-phase
oxidation  of sulfur  dioxide,  which  proceeds  more  rapidly  during  summer
months.   Whatever  the  cause,  it is readily  apparent  from this figure  that
scavenging mechanisms for these two species differ  appreciably.

As  noted  above,  most past  field  experiments have  experienced  difficulty  in
resolving  precisely  which  source(s)  of  pollution  has  been responsible  for
material  wet-deposited  at sampled receptor sites,  and  this  problem  is
typically amplified as time and distance scales  increase.  Source attribution
is  particularly uncertain on  a regional scale, and  the  basic  data  obtainable
from  standard  precipitation-chemistry  networks  are of  little help  in  this
regard.   Combined  with the lack  of data from  well-designed  regional  field
studies,  this problem  of  source  attribution  poses one  of  the  most important
and uncertain questions facing the acidic deposition issue at present.

As  a consequence of this need, a major regional  field experiment has recently
been  designed  and conducted  in  the northeastern United  States (MAP3S/RAINE
1981, Easter 1982).  Known as the Oxidation and Scavenging Characteristics of
April Rains  (OSCAR)  study,  this  field experiment was based upon the concept
of  characterizing,  as  completely  as  possible,   the   dynamic  and  chemical
features  of  major cyclonic storm  systems  as they  traverse  the  continent.
Specific  objectives were:

     1.   To  assess  spatial   and  temporal  variability  of  precipitation
          chemistry  in  cyclonic  storm  systems,  and to  test the adequacy of
          existing networks to characterize this variability;

     2.   To  provide   a   comprehensive,   high-resolution  data   base  for
          prognostic, regional deposition-model  development; and

     3.   To develop  increased understanding of  the transport, dynamic,  and
          physiochemical mechanisms that  combine to  make  up  the  composite
          wet-removal process, and  to  identify  source  areas responsible for
          deposition at receptor sites.
70ne  should note in  Figure  6-15  that the periodic functions are  fit to the
  total  data, whereas  the linear  regressions are  fit only  for  the period
  January  1,  1977-December 31, 1979; thus the cyclic functions are not exactly
  symmetric  about the  linear regression  curves.    Some idea of statistical
  improvement in  fit may be obtained using the expression
               2                     2
          ?2  = q  linear regression - g periodic fit
                     o2iinear regression

  where  the  a2's  pertain  to variances  of  the  data points  over  the  three
  and  one-half period.   For sulfate  in  Figure 6-15 f2 equals  0.22, indica-
  ting a  significant  reduction  in variance;  the  corresponding  r2  value for
  nitrate  is  0.01, suggesting that no significant annual periodicity exists in
  this case.
                                    6-40

-------
The data collected and assembled by the  OSCAR  project are  summarized  in  Table
6-3.   These are  being  made available  to the  general  user  community  in  a
computerized data base.

A  general  layout of  the  OSCAR precipitation-chemistry network  is shown  in
Figure 6-16.   The points and  triangles on this map represent locations  of
sequential  precipitation-chemistry  stations   on   an  "intermediate-density"
network; the open square overlapping  Indiana  and Ohio  depicts  a  concentrated
network  of  47   additional   sites.     Specific  design   criteria  for   this
configuration are discussed  in the supporting  literature MAP3S/RAINE  (1982).

The  OSCAR data  set  is  presently  under  intensive  investigation,  and  only
preliminary results are currently  available.   It is of interest  to  consider
some  of these  results  at  this point,  however,  to  evaluate  the potential
future  utility of  this material.    One  early result, presented by  Raynor
(1981), is  primarily  of qualitative  interest  and involves the  first-sample -
last-sample pH  data obtained  by the  sequential rain  samplers  for individual
storms,  typified  by  the   plots  shown   in Figures  6-17  and  6-18.  It  is
interesting to  note  that Figure 6-17 is  strongly  reminiscent of  annual-  or
multi-year-average plots for the northeastern  United States in the sense that
it shows  the  familiar  acid "core"  region  centered  upon  Pennsylvania.   The
final-sample distribution in  Figure 6-18  is  quite different.   Besides  indi-
cating a much  cleaner sample  set,  very  little structure exists in this final
distribution.    This  relative  cleanliness of late-storm  precipitation  is
consistent  with  the  general   OSCAR   finding  that  most of  the pollutant  is
scavenged  comparatively early  in  a storm's  life   cycle  (Easter and  Hales
1983a).

It should  be  noted in this context  that  field studies having higher spatial
resolution  (e.g., Semonin  1976,  Hales and  Dana  1979b)  indicate that sig-
nificant  fine  structure  typically exists  in  spatial pH distributions.   Much
of this fine structure can be expected to be  hidden within  the relatively
coarse  sampling mesh  shown  in Figures 6-17 and 6-18.

Substantial  source-receptor analysis is  presently  being conducted   in con-
junction  with  the  Indiana-Ohio concentrated  network.   One  early  analysis,
conducted  for  the 22-24 April  1981  storm  is  presented in Figure  6-19.   Back
trajectories of  this  type are  currently being combined in diagnostic scaven-
ging  models  with  aircraft and  surface  data  to   evaluate  source-receptor
relationships  in  greater detail (Easter and Hales 1983a,b).

6.5   PREDICTIVE AND INTERPRETIVE MODELS OF SCAVENGING

6.5.1   Introduction

A  precipitation-scavenging  model  can be defined as  any conceptualization of
the  individual  or composite  processes  of  Figure  6-2,  in  a manner  which
allows  their  expression in  mathematical form.Often such models  take the
form  of submodels or  modules" within a larger calculational  framework, such
as a  composite regional  pollution code.  When  considered in  a modular sense
the  lines connecting the boxes of Figure 6-2  can be  considered as  channels
for  information  exchange  within the overall  framework, whereas the boxes (or


                                  6-41

-------
         TABLE 6-3.  SUMMARY OF DATA COLLECTED  FOR  THE  OSCAR DATA  BASE
METEOROLOGICAL DATA

   0   North American standard 12-hour upper  air  observations
       (rawinsondes)

   o   OSCAR special  rawinsonde data

   °   North American 3-hour standard surface observations

   °   North American hourly precipitation amount data

   °   Trajectory forecast data (Limited Fine Mesh and Global  Spectral
       Models)

   0   Gridded forecast data (Limited Fine Mesh Model)

   0   Satellite observations

PRECIPITATION-CHEMISTRY DATA

   0   OSCAR  network:   Sequential measurements  of rainfall,  field  pH,  lab
       pH, conductivity, N03~, N02~,J.S042~, S032~>
       Cr, NH4+, Ca2+, Mg2+, K+, Na+, A13-P,
       P043~» total Pb

   o   Additional  networks:   Time-averaged  data  as available  from  sources
       such as NADP, CANSAP, CCIW, and APN
   0   Special rainborne H£02 measurements

AIRCRAFT DATA

       Trace gases:  03, NO/NOX, S02, HNOs,

   0   Aerosol parameters:  Scattering coefficient (bscat), Aitken
       nuclei, aerosol sulfur, sulfate size distribution, aerosol  size
       distribution, aerosol acidity

   0   Cloud water chemistry:  NOV" > N02~ , SO^2",
       S032-,,pH, NH4+, conductivity, CT, Ca*+, Mg2+,
       K  , Na+, total Pb.

   0   Meteorological parameters:  Temperature, humidity, liquid,  water
       content, wind speed and direction, cloud droplet size
       distribution

   0   Position parameters:  Latitude, longitude, altitude, time
                                    6-42

-------
                             TABLE 6-3.   CONTINUED
SURFACE AIR CHEMISTRY DATA

       OSCAR SAC site (Fort Wayne 40°49.8'N,  85°27.6'W):   H202,
       peroxyacetyl  nitrate, sulfur aerosol  size distribution, NH3,
       S02, 50^2-t  03, NO/NOX,  HN03, aerosol  composition
       vs particle  size, aerosol  acidity

   o   Selected air quality data  from specific surface monitoring sites
       throughout eastern North America

EMISSIONS

   0   MAP3S/RAINE  standard inventory
                                    6-43

-------
CTl
I
                                                                                      EXISTING
                                                                                    MAP3S SITES
                                                                                    SUPPLEMENTAL
                                                                                    REGIONAL SITES
                                                                                    NE INDIANA GRID
   Figure 6-16.  General  layout of OSCAR sequential precipitation chemistry network, showing hypothetical
                 "design-basis" cyclonic system.

-------
Figure 6-17.   pH distribution  for  initial precipitation sampled during OSCAR storm of 22-24 April 1981.

-------
CT>
I
CTi
                      J-                  f
                     r      /	T—'-4-
   Figure 6-18.   pH distribution for final precipitation sampled  during  OSCAR storm of 22-24 April 1981.

-------
Figure 6-19.
Loci of points contributing pollution to the high density
network near 1400 EST on 22 April 1981.  Contour intervals
3, 6, 9 represent travel times in hours from source regions.
The large arrow represents the likely path of air originating
from points 9 hours upwind of the receptors.

                    6-47

-------
clusters of  boxes)  can  be identified  with  the modules,  themselves.   This
modular  relationship  is  described in  somewhat more detail  in Chapter A-9,
where composite regional  models are discussed.

Scavenging models are currently in a rapidly-evolving state, and a  profusion
of associated  computer codes  and  computational  formulae is currently  avail-
able.   Indeed, one of the  major problems  in  precipitation-scavenging assess-
ment is determining precisely which model  to select from the large  number of
available candidates.  A major aim of the present  subsection is  to  guide the
reader in this pursuit.

There are a number of potential uses  for precipitation-scavenging models, and
the intended use will to a large  extent determine  just  which model  should be
employed.  Some of the more important potential  uses are itemized as follows:

     °  Predicting the impact on precipitation  chemistry of proposed
        new sources, source modifications, and  alternate emission-
        control strategies;

     °  Predicting long-range precipitation  chemistry trends;

     °  Estimating relative contributions  of specific sources to
        precipitation chemistry at a  chosen  receptor point;

     0  Estimating transport of acidic-precipitation precursors
        across political  borders;

     °  Estimating and predicting  air-quality improvements  occurring
        as a consequence of the scavenging process;

     0  Selecting sites for precipitation-chemistry network sampling
        stations;

     o  Designing field studies of precipitation scavenging;  and

     0  Elucidating mechanistic behavior of the scavenging  process  on
        the basis of field measurements.

In  selecting  an  appropriate  model, the  user  should   review his  intended
application  carefully  with regard to  the pollutant  materials  of  interest,
time  and  distance  scales,  processes   covered  in  Figure  6-2,  source con-
figuration,  precipitation  type,   and  mechanistic  detail  required.    The
question of  pollutant materials is particularly important  when precipitation
acidity  is  of interest.   Acidity  in  precipitation  is  determined  by the
presence  of  a multitude  of  chemical  species, and  in principle  one must
compute  (via  a  model)   the  scavenging of  each  species  and  then  estimate
acidity on the basis of an ion balance:

        [H+] = i Anions -  (E Cations  other than H+).                    [6-1]

Inorganic  ions usually  important  in  precipitation  chemistry are itemized  in
Table  6-4.   Organic species  play  a   secondary  role  in  the  acidification


                                     6-48

-------
 TABLE 6-4.  SOME INORGANIC IONS IMPORTANT
        IN PRECIPITATION CHEMISTRY3
Cations                               Anions


H+

NH4+                                   Cl~

Na+                                    N03-

K+                                     S032-

Ca2+                                   S042-

Mg2+                                   P043-

                                       C032-
aAll ions are presented here in their completely-
 dissociated states.  The reader  should note, however,
 that various states of partial dissociation are
 possible as well  (e.g., HS03", HC03").
                        6-49

-------
process, which  appears  to vary widely by  region.   Modeling of all  of  these
species  simultaneously   requires   substantial   effort,  and  all   "acidic-
precipitation" models to date have focused upon only one or just  a few of the
more important  species,  with  contributions of the others  estimated  empiric-
ally.   Currently,  newer models tend  to  accommodate larger numbers  of  these
species; but  complete  modeling coverage will  not be  achieved  in the  fore-
seeable future.

Mechanistic detail is another  important  feature  determining the  basic compo-
sition  of  a scavenging model.   A comprehensive mathematical  description  of
the scavenging  process can become rapidly  overwhelming,  and there is usually
a  need  to  represent these  relationships  in  a comparatively simple,  albeit
approximate, manner.   The process of consolidating complex behavior in  this
fashion is  often  referred to  as  lumping the system's  parameters.    The  re-
sulting simplified expressions are  termed parameter!'zations.  Consolidating
the effects of  non-modeled species  in empiricalform, described  in  the  pre-
ceding  paragraph,  is  one example of  lumping.   Numerous other examples  will
arise throughout the remainder of this section.

This section will  not attempt  to  provide the  reader  with a detailed  treatise
on  how models  should  be formulated and  applied.^   The  approach,  rather,
will be to develop a  basic  understanding  of the  fundamental  elements  of  a
scavenging model and then to  provide  a  systematic  procedure for  choosing and
locating appropriate models  from the literature.   The  following subsection
discusses  the  basic  conservation equations, which  constitute  the conceptual
bases for scavenging models in general.  This  discussion is followed in  turn
by  two  simple  applications  of these relationships,  which are  presented  to
illustrate usage and to define some terms commonly used in  scavenging models.
The  final  subsection  attacks  the problem  of model  selection, using a  flow-
chart approach  designed  to guide  the  user  to a valid  choice in  a systematic
manner  that  avoids many of  the  pitfalls  normally  encountered  in  such
endeavors.

6.5.2   Elements of a Scavenging Model

6.5.2.1   Material  Balances--In Figure 6-2 the various  arrows between  boxes
correspond  physically to  streams  of pollutant  and/or water.   From this it is
not difficult  to  realize that any characterization  of this system  must in-
clude material  balances,  which form the  underlying  structure  for all scaven-
ging models.To formulate  a material   balance,  one  simply visualizes  some
chosen  volume  of  atmosphere,  and sums  over all inputs and  outputs of the
substance in question.
     the  reader  interested  in  more detailed pursuit of this  area,  the works
 by  Hales (1984) and SI inn  (1983)  are recommended.  The  Hales  reference is
 something  of  a beginner's  primer,  while  SI inn's  treatment  delves  substan-
 tially  deeper  into  mechanistic  detail.   Together they  constitute  a  rea-
 sonable  starting  point  for  understanding  and modeling  basic  scavenging
 phenomena.
                                     6-50

-------
Two basic types of material  balance  are  possible:

     1.  "Microscopic" material  balances,  based upon  summation  over  a
         limiting small volume element of  atmosphere;  and

     2.  "Macroscopic" material  balances,  based upon  summation  over  a
         larger volume element of atmosphere (e.g.,  a complete  storm
         system).

Microscopic  material   balances  invariably  lead to  differential  equations,
which  must be  integrated  over  finite  limits  to  obtain  practical  results.
Macroscopic  balances  result  in  mixed,  integral,  or  algebraic equations.
Again  the  choice  of material-balance  type depends upon  the  specific modeling
purpose at hand.

An important general  form  of  the differential material  balance for a  chosen
pollutant  (denoted  by subscript  A)  is given  by  the equations9  (cf.,  Hales
1984)

        9C
          Ay  = -v.cAyVAy - WA + rAy  (gas phase)                        [6-2]
and
                ~
          Ax  -V.CAXVAX + WA + rAx  (aqueous phase).                     [6-3]
         9t

Here  CAy and  CAX  denote  concentrations  of  pollutant  in  the  gaseous  and
condensed-water phases, respectively.  The  time  rate  of  change  of these con-
concentrations  within  the  differential  volume element is  related  to  the sum
of  inputs by  1) flow through the walls  of  the  element,  2) interphase trans-
port  between  the  gaseous  and  condensed   phases,  and  3)  chemical  (and/or
physical)  reaction  within  the  element.    The  v   terms  in  Equations 6-2
and 6-3  denote  velocity vectors, while  v.   is the  standard vector divergence
operator.   The interphase  transport term  WA  accounts for  all  "attachment"
processes  (impaction,  phoresis, diffusion,  ...)  as well   as  any  reverse
phenomena such  as pollutant-gas  desorption, while the r terms denote chemical
conversion rates  in the usual sense.   To formulate a  usable model  from these
equations, one  needs to specify  values for the functions v, w, and r and then
solve differential  Equations  6-2 and 6-3 (subject to  appropriate initial and
boundary conditions)   to  obtain  the  desired  concentration  fields CAy  and
      A  simple  example  of this procedure is given in Section 6.5.4.
Equations  6-2 and  6-3 are  quite general  in  the  sense  that  the  velocity
  vectors denote velocity of  pollutant (rather than that of  the bulk medial
  and thus  provide  for  all  modes  of  transport (convective,  diffusive,  ...)
  without yet specifying how this transport is to  occur.  These equations are
  not yet time-smoothed; thus,  no closure  assumptions  have  been  applied  at
  this point.
                                     6-51

-------
6.5.2.2   Energy Balances—Many terms  in  Equations 6-2  and 6-3, especially
*Ax»  WA>   and   rAx»  depend  strongly  upon  the  amount,  state,   and  inter-
conversion  rates of  condensed  water;  and  it  is important  to  note  that
atmospheric water itself obeys  material-balance  expressions  of this form.  In
selecting  a  scavenging  model,  one  often  is confronted  with  the problem of
deciding whether to  estimate precipitation  attributes  and these related terms
independently on  the basis  of assumptions or  previous  information,  or to
attempt to compute the desired  entities directly by solving  appropriate forms
of Equations 6-2 and 6-3.

If  the  latter  of  these alternatives  is  chosen,  then  the inclusion  of an
energy-balance equation is mandatory.  This need arises  because  the evapora-
tion-condensation process  influences,  and  is  influenced by, a variety of
energy-related considerations.  These include  temperature influences on vapor
pressure and latent-heat effects,  and can be incorporated in the  model via an
energy balance  performed over the same  element  of  atmosphere  as  that of the
associated material  balances.  In microscopic  form, a general expression of
the energy balance (cf., Bird et  al .  1960),  is


     pCv 
-------
terms and solves the differential  equation subject to appropriate  initial  and
boundary conditions to  obtain  fields of the velocity  vector  v.   An  example
applying  Equation  6-5  for scavenging  modeling  purposes  is  given  by  Hane
(1978).

Incorporating energy  and momentum  balances,  Equations 6-4  and  6-5,  into  a
scavenging model  is a  rather  challenging exercise,  and  a relatively  small
number of  models  that apply these  equations  for this  purpose exists.    The
usual  tack  is simply to "pre-specify"  the required parameters  and  proceed
with material-balance calculations alone.   Numerous examples of both types of
models will be presented in Section 6.5.5.

6.5.3  Definitions of Scavenging Parameters

Four key  parameters often  arise in the context  of  scavenging  models, and it
is appropriate at this point to define these terms and indicate their  general
application.  Reference  to these  entities  as  "parameters"  is  consistent with
the  usage applied in  the previous section, in that they serve to "lump"  the
effects of a  number of  mechanistic  processes  in a simple  formulation.  These
will be discussed sequentially in the following paragraphs.

The  first parameter to be defined is the  attachment  efficiency.   Also  known
as  the  capture  efficiency, this  term can be  visualized most  easily  by con-
si dering~a~liyarrometeorrTal ling through a volume of  polluted air space,  as
shown in Figure 6-20.  This hydrometeor sweeps out a volume of air during its
passage,  and attachment efficiency is defined as  the  amount of  collected
pollutant divided by the amount initially  in this volume.   The efficiency can
exceed 1.0 if pollutant  from outside the swept volume becomes attached to the
drop.

From the  discussion in   Section 6.2.3,  it is  apparent  that attachment  effi-
ciency accounts for a multitude of processes.   Usually the efficiency  is less
than 1; but  mechanisms  such  as diffusion, electrical  effects, and  intercep-
tion can give rise to larger values, especially when the collecting  element s
fall velocity is small.  Efficiencies can be  negative   if  the  element  is
releasing  pollutant to   the surrounding atmosphere^  sucfi  as  in  the  case  of
pollutant-gas desorption.   Typical  efficiencies for  aerosol  particles col-
lected by raindrops are  shown in Figure 6-4.

Another important parameter  is the  scavenging  coefficient.   This entity is
basically an expression of the law of mass action, defined by the  form

     A = !|A_                                                            [6-6]
         cAy

where (in a  manner  consistent  with  Equations  6-2 and 6-3)  WA  1S  tne  rate °f
depletion of  pollutant A from  the gaseous phase by attachment to  the  aqueous
phase in a differential   volume element.  This  is similar to a rate of  expres-
sion  for  a   first-order,  irreversible  chemical  reaction,  and   as  such  it
applies strictly only to irreversible attachment processes (e.g.,  aerosols or
highly-soluble gases).   A  can  be related to  the attachment  efficiency E by
the  form (which assumes  spherical  hydrometeors)


                                     6-53

-------
                        ^ 2R
                       o
Figure 6-20.
Schematic of a scavenging hydrometeor falling through a
volume element.
                     6-54

-------
    A (a)  = - TrNTo/00R2vz(R)E(R,a)fR(R)dR ,                               [6-7]


where a and  R  denote  aerosol  and hydrometeor radii, respectively; vz  is  the
hydrometeor  fall   velocity;  and  NT  and  fR  are  the   total   number  and
probability-density functions for the  size-distributed  hydrometeors  residing
in the volume element of Figure 6-20 at any instant in  time.   From this,  one
can  note  that  A  essentially  extends   the  parameterization over the  total
spectrum of hydrometeor sizes.

Atmospheric aerosol particles  are  typically  distributed over  extensive size
ranges.   Because  of this it  is  often  desirable  to  possess some sort  of an
effective  scavenging  coefficient,  which represents  a  weighted average over
the aerosol  size  spectrum.   Figure  6-21  presents a family of curves  corre-
sponding to such averages,  which are based upon assumed log-normal particle-
size spectra, with different geometric  standard  deviations. From  these  curves
one can observe that  for the same  geometric  mean particle size, changes in
spread  of  the size  distribution  can result  in  dramatic changes  in  the
effective scavenging coefficient.

Inclusion  of reversible  attachment processes in  a  scavenging model  usually
involves using the mass- tram s f e r coe f f 1 c i en t .  This parameter  can be defined
in terms of the flux of pollutant moving from the  scavenging element  as


     Flux = - JV {cAy - h'cA)  .                                         [6-8]
              c

Here  Ky  is the  mass- transfer  coefficient   and  CA  is  the  concentration,
within  the  scavenging element, of  collected  pollutant; h1 is essentially  a
solubility  coefficient  which,  when  multiplied  by c^,  produces   a gas-phase
equilibrium  value.    c  is  the  molar concentration  of  air molecules,  which
appears in Equation 6-8 because of  the  manner in  which Ky  has been  defined.
Thus, the  flux can be  either to  the drop  or  away  from  it,  depending  upon the
relative magnitudes of  the  parenthetical  terms.   Equation 6-8 can  be  inte-
grated over  all drop  sizes  in a manner similar to that used  in  Equation 6-7
(cf., Hales 1972), to form the following expression for  WA:
     WA=   TrT   o/°VfR(R)Ky(R)  (CArh'cA)dR.                          [6-9]


The final scavenging parameter to be  described  here  is  the scavenging ratio.
This  entity  is usually  the resul t of  a model  calculation,  rather  than  an
input, and is defined by the form

     r =  ^A
          ^"                                                          [6-10]
                                     6-55

-------
en
en
                  FRONTAL RAIN SPECTRUM
        0.01 r-
       0.001 -
      0.0001
           0.001
     Figure 6-21.
                                         v
Computed effective scavenging coefficients for size-distributed aerosols.  Based on a log-
normal aerosol radius distribution with geometric means and standard deviations a  and a  .
A typical frontal-rain dropsize spectrum is assumed.  Adapted from Dana and Halesy(1976).

-------
where  C^  is  the  concentration  of  pollutant  contained   in  a  collected
precipitation  sample.   £  is  a  term  immediately  usable  for  a  number of
pragmatic  purposes  because  once its  numerical  value  is known,   it  can be
applied  directly  to compute  precipitation-chemistry concentrations  on the
basis of  air-quality measurements.   Tables of measured (Engelmann 1971) and
model-predicted (Scott 1978)  scavenging ratios have been  published, although
caution is advised  in  the application  of these values.   A  simple  example of
scavenging-ratio application  is given in  the following  section.

It is useful  for the sake of  visualization to discuss briefly the qualitative
features  of  the  scavenging parameters  noted above.  The  parameter E  is  easy
to visualize  in  the context of  Figure 6-20; it  is,  simply, the  collection
efficiency of an individual cloud or precipitation element and as such should
be expected  to  fall  numerically  in  the  approximate range  between zero and
one.    The   scavenging  coefficient  A  can  be  visualized as  a first-order
removal  rate, in much the same manner as that of  a  first-order  reaction-rate
coefficient.   As  such  it may be used roughly as  a  characteristic  time  scale
for  wet  removal.    A  = 1 hr~l,  for example, would imply  that the  scaven-
ging process will cleanse 100(l-l/e) percent of the  pollutant in one  hour if
conditions remain constant and  competitive processes do  not  occur.  From  this
one  can  note  that  1  hr-1  is  a  moderately large  scavenging  coefficient.
A'S  ranging  from  zero  to  1  hr~l  and  beyond  have been  reported  in the
literature (cf., Figure 6-21).

The  mass-transfer coefficient  Ky  is  essentially  a  normalized interfacial
flux of pollutant between the  atmosphere  and an individual  droplet.   Little
needs to  be  said  here regarding  magnitudes of  Ky,  except to note  that a
variety of different definitions of Ky exist, and  one  must be  cognizant of
these definitions when employing values  obtained  from  outside sources.  The
washout ratio,  £, is  essentially  a measure  of the concentrating  power of
precipitation in its extraction of pollutant from  the atmosphere.   As  will be
noted in the next section, precipitation  often has the ability to concentrate
airborne  pollution  by  a factor  of a million  or  more,    c's  ranging  from
below 100 up through 108 and  higher have  been reported  in  the literature.

The expected magnitudes and uncertainty levels associated  with the  scavenging
parameters listed in  this section depend  strongly  upon the substance  being
scavenged and  the environment  in which  the scavenging takes place.    Large
aerosol  particles in below-cloud environments, for example,  are  characterized
by scavenging  efficiencies in  the range of 1.0 (Figure  6-4),  which  can be
estimated  with relatively high precision.   Smaller  particles,   especially
those in the "Greenfield-Gap"  region, are much more  difficult to simulate and
associated errors in estimated  efficiencies may approach  an order of magni-
tude or more.   Errors  in these  efficiency  estimates will of course be  com-
pounded by uncertainties in raindrop size spectra,  if extended  to  scavenging
coefficients via  Equation 6-7.   In the case  of  gases,  the mass-transfer
coefficient usually  can be estimated to within a factor  of two or less;  again
this error can be expected to compound when  integrated over  assumed raindrop
size-spectra.

In the case of in-cloud scavenging of aerosols  our  capability for  estimating
transport  parameters  is  seriously impeded,   owing  to  the   profusion  of


                                     6-57

-------
mechanisms and  the  complex  environments involved.  Typical uncertainties  in
both  A and  E, can  be  expected  to  approach an  order of  magnitude  in  some
cases.   Some  appreciation  for  the  factors influencing in-cloud  scavenging
coefficients can be obtained from the  work  of SI inn  (1977), who attempts  to
evaluate  theoretical,   "storm-averaged"  values   for  A.    An  idea   of  the
magnitudes and uncertainties of £ is given in Figure 6-23.

In  all cases  involving  reactive   gases,  the  values of  E,  A,   and  5  are
heavily contingent  upon  the aqueous-phase chemical  processes  involved.   Much
remains to  be accomplished in  our  understanding of  aqueous-phase chemistry
before a meaningful assessment of associated uncertainties is possible.

As a final note in this context it should be emphasized that uncertainties in
scavenging parameters  dictate  uncertainties in scavenging calculations  in a
complex  fashion,  and  that  errors  associated with  the microscopic phenomena
can  be either amplified or attenuated by their  applications  in  macroscopic
models  to produce  practical results.   Uncertainties associated with  macro-
scopic  modeling  applications  will  be  discussed  at some  length  in  a  later
section.

6.5.4  Formulation of  Scavenging Models;  Simple Examples of Microscopic
       and Macroscopic ApproacheT

As noted  previously, the description given in this document  will  refrain in
general from deriving  and applying scavenging models explicitly.  This is too
broad  and complex a subject to be discussed in detail  here, and the reader is
referred  to the previously-cited literature for more detailed pursuit of this
subject.  For  purposes of illustration, however, it is worthwhile to  consider
two  very  simple examples of scavenging-model  formulation, which  demonstrate
the  microscopic and macroscopic  approaches  to  the problem.  The present sub-
section is addressed to  this task.

The  microscopic material balance approach will be considered first.  For this
example,  it is useful  to visualize  an  idealized  situation where rain of known
characteristics  is falling  through a  stagnant volume  of atmosphere,  which
contains  a well-mixed,  nonreactive pollutant with  concentration  cAy-   Tne
air  velocity  is known (v=0),  so solution of the momentum equation (Equation
6-5) is not required.   The raindrop size distribution  is  presumed to remain
constant;  thus, evaporation-condensation  and other  energy-related effects are
immaterial, and the energy  equation  (Equation  6-4)  may be  disregarded.

Because the  pollutant  is well-mixed, no concentration gradients occur; thus,
the  divergence term in  Equation 6-2  is  zero.   Because  of nonreactivity  the
reaction  term is  zero  as well.

 Now  presume   that  the  pollutant  is   an  aerosol  whose  attachment  can  be
characterized  in  terms  of  the  known  scavenging  coefficient A ,  using
 Equation  6-6.  The corresponding reduced  form  of Equation  6-2  is,  then,


      8CAy = - A cAy.                                                   C6-2a]
       3t
                                      6-58

-------
Given  some  initial   pollutant concentration  c^y0,  Equation  6-2a  can  be
integrated to obtain the form
     cAy {*> = cAyo exP (-At),                                        [6-11]

which expresses  the decrease  of  the  gas-phase pollutant concentration  with
time.  Counterpart expressions for rainborne concentrations  may be derived by
subjecting Equation 6-3 to a similar treatment.

The reader is cautioned to  consider this  treatment  as  an  example only and to
recognize that actual atmospheric conditions seldom conform  to the  idealiza-
tions  invoked above.   Gas-phase  concentrations  are  usually not  uniformly
distributed in space, raindrop characteristics  are usually nojt invariant with
time,  and  wind  fields   are   usually  not  well   characterized  by  v =0.  A
is usually  not  a time-independent constant, and many  pollutants are  usually
not well  characterized  by  the washout  coefficient  approximation.  The  pol-
lutant  often  is not  unreactive.   Examples  of existing  models  where  these
constraints are  relaxed in  various  ways are presented in the following  sub-
section.

Figure 6-22 illustrates the formulation of a macroscopic type of scavenging
model.  Here, in contrast to  the  differential-element  approach,  the material
balances are  formulated around a  large  volume  element, in this  case  a  total
storm.   If  one  denotes  concentrations and flow rates  of  water and  pollutant
as follows:

        • airborne concentration of pollutant
        = airborne concentration of water vapor into cloud
     CA = concentration of scavenged  pollutant  in  rainwater
     pw = density of condensed water
    win = flow rate of water vapor into  the storm
   wout = flow rate of water vapor out of the storm
    f-jn = flow rate of pollutant into  the storm
   fout = flow rate of pollutant out  of  the storm
      W = flow rate of precipitation  out of the storm
      F = flow rate of scavenged pollutant out  of  the  storm,

then extraction  efficiencies  for  water  vapor  and  pollutant can  be defined,
respectively,  as

     EP = JL .                                                         [6-12]
          "in


     e = TTT                                                           C6-131

If one further performs material  balances  over this storm system for  pollu-
tant  and  water  vapor,  and  then  combines  the two,  the following form  is
obtained:
                                     6-59

-------
                CONDENSATION,
         PRECIPITATION FORMATION,
            POLLUTANT ATTACHMENT
FLOW RATE OF WATER VAPOR OUT = w
                                out
FLOW RATE OF POLLUTANT OUT = f
FLOW RATE OF WATER VAPOR
FLOW RATE OF POLLUTANT IN =
                                                                             out
                             in
                                                FLOW RATE OF PRECIPITATION OUT = W
                                          FLOW RATE OF SCAVENGED POLLUTANT OUT = F
             DEFINITIONS OF EFFICIENCIES:
                      WATER REMOVAL
                              W
                              w
                               in
     POLLUTANT  REMOVAL
                                                       e =
             fii
       Figure 6-22.  Schematic of a typical  macroscopic material balance.
                                      6-60

-------
          cAy

where the scavenging ratio,  £,  is  as  defined earlier  in Section 6.5.3.

Equation 6-14 is an important  result  in  the  sense that it demonstrates once
again  the  strong linkage between  water-extraction  and pollutant-scavenging
processes.     If  both  occur  with   equal   efficiency10  (ep   =  e  )   for
example, then

     *  =£{fs 10-5. 10-6.                                             [6-15]


Experimentally-measured scavenging ratios often fall  in this range, although
wide variability is usually  observed.

Using a rather involved series of arguments pertaining to cloud-physics pro-
cesses and attachment mechanisms,  Scott (1978)  has created a family  of curves
expressing scavenging  ratio  as a function of  precipitation  rate.  Shown in
Figure 6-23, curves 1,  2,  and 3  pertain respectively  to convective storms,
nonconvective warm-rain process  storms,  and  cold storms where the  Bergeron-
Findeisen process is active.

A major  assumption  in  Scott's analysis  is that  storms ingest pollutants in
the  form of  aerosol particles  which  are  active as cloud condensation nuclei.
The  analysis also  assumes a  steady-state storm system and complete vertical
mixing  of  pollutant between  the  storm height and the  surface.   Under such
conditions Scott's  curves  can be  considered  reasonably  good  estimators of
actual  scavenging  behavior.    More  elaborate  systems,  involving reactive
pollutants,  gases,  and nonhomogeneous systems,  are  discussed  in  references
given in the following section.

6.5.5  Systematic Selection  of Scavenging Models:  A  Flow-Chart Approach

Hales  (1984)  has  suggested  a flow-chart approach  to  aid in  selecting  a
scavenging-model.   Presented  with  a decision tree in  Figure  6-24, the user
proceeds  by  answering  a  series  of  questions that  relate to  the model's
intended  use,  temporal and  geographical  scales,  pollutant characteristics,
choice  between macroscopic  and  microscopic  material  balances,  and type of
10There  is  no  direct  reason  to  expect  that ep  should  be  similar  to e
  in magnitude.   In  the  absurd circumstance where all the  pollutant  is  con-
  centrated  into one particle, for example, then scavenging of that pollutant
  by  a  very light  rainfall   would  yield  E=I.O»EP.   Conversely  a  large
  storm  processing  an insoluble gaseous  pollutant  (SFs, say)  would  provide
  e*0«ep.    For  practical   conditions   involving   acid-forming   aerosols,
  however,  the  scavenging of  water  vapor  and  pollutant appears to  be  suf-
  ficiently  related  to  allow  ep*e  to   be   employed   as  an  approximate
  rule-of-thumb.
                                     6-61

-------
                                                            Z9-9
    ro
    GO
CO C/1
 i   o
o  o
O  r+
— ' c+
Q_  -
    to
to
c+ to
O  O
-s  a>
3  <
to  fD
-.   3

 -". I
 to  O
 n>  o
 3  3

-a  ro
 -s  o
 O  rt-
 o  -"•
 fD  <
 to  fD
to  o
Q)  3
O  00
fD  I

-—^ o>
00 -5
o  3
o  >»
 to n
 ^j o
 00 3
**—^' ^^
 •   fD
    O
    fD
                                                                                          °co
m


T3
»—t

5
                  O
                  o
    O


    to

-------
<                                                                                      PLUME MODEL OR   \.___	—. —	-
                                                                                      MEASUREMENTS   f
CT»


CO
CHEMISTRY NETWORK
RESULTS



< PLUME MODEL OR \_
MEASUREMENTS P~
1
1
1 	

AMONG THOSE
MEASURED ROUTINELY'
T
ARE HISTORICAL
CLIMATOLOCICAL
DATA SUFFICIENT?
* 	 ll'
DOES THE SUBSTANCE 3
REACT IN THE AQUEOUS
PHASE TO FORM A NEW
NONVOLTILE MATERIAL?
F| Tl
USE SOLUBILITY RELATIONSHIP
TO DETERMINE AQUEOUS -
PHASE CONCENTRATION
AT GROUND LEVEL 1t
1
T
T

V GROUND-LEVEL GAS - /
PHASE CONCENTRATION /
INFORMATION /

/PRECIPITATION /
CHARACTERISTICS f- 	 — »
AND CONCENTRATION FIELD /
T
1
1


UTILIZE CONVENTIONAL CAS 17
SCAVENGING CALCULATIONS
TO DETERMINE AQUEOUS-PHASE
CONCENTRATIONS
1

1
IS THE SUBSTANCE
TRANSFERRED TO THE
AQUEOUS PHASE AS A CAS 7
* *
DOES VOLATILE CONSTITUENT
SATISFY EQUILIBRIUM
SCAVENGING CRITERION?
I'
13
IS UPTAKE RATE BY
AQUEOUS PHASE LIMITED
BY MASS TRANSFER?
F] |T
'
J

	 ( PLUME MODEL \*— —

1
1
I 	

APPLY IRREVERSIBLE
SCAVENGING
APPROXIMATION

tt
COMPUTE GASEOUS
AND AQUEOUS PHASE
CONCENTRATIONS
t
1
1

F ^


DOES THE SUBSTANCE 1
REACT IN THE AQUEOUS
PHASE TO FORM A NEW
VOLATILE MATERIAL?
*
PHASE LIMITED
BY MASS TRANSFER?
1
11
IS UPTAKE RATE BY
AQUEOUS PHASE LIMITED
BY CHEMICAL REACTION?
1
1
1
<
/ PRECIPITATION /
J CHARACTERISTICS /
7 AND CONCENTRATION FIELD /
/ OR SOURCE STRENGTH /
12
DOES THE NONVOLATILE
SUBSTANCE INTERACT
WITH THE CLOUD SYSTEM'

IS
COMPUTE WASHOUT
COEFFICIENTS
I
23
PARAMATERIZE AEROSOL
SIZE DISTRIBUTION
F
T
F
»-
i-
-.
-

I
FORMULATE SIMULTANEOUS »
MASS TRANSFER -
CHEMICAL REACTION
MODEL
COMPUTE
CONCENT
M
CASEOUS
DUS PHASE
RATIONS

DOES THE SUBSTANCE L,
INTERACT WITH THE F
T
SCAVENGING i
APPROXIMATION

APPLY RE ACTION- LIMITED
SCAVENGING
APPROXIMATION
I
i A
DOES THE NONVOLATILE -
SUBSTANCE INTERACT '
WITH THE CLOUD SYSTEM?
	 1
11
IS CONDENSATION OF WATER
ON PRIMARY AEROSOL •-
SIGNIFICANT
F| IT
1
COMPUTE SIZE DISTRIBUTION
OF SECONDARY AEROSOL
3D
UTILIZE SCAVENGING
RATIOS TO CALCULATE AQUEOUS
;
t 	
T
CAN THE STORM AND SOURCE
BE APPROXIMATED BY A
QUASISTEADY STATE'

/ PRECIPITATION /
*— — t CHARACTERISTICS /
/ AND CONCENTRATION FIELD /

X PLUME MODEL \
OR MEASUREMENTS f~
	 	 1 1
/ PRECIPITATION /
F / CHARACTERISTICS, L
"* / AND CONCENTRATION FIELD /*
/ OR SOURCE STRENGTH f
1
APPLY APPROPRIATE "
RATE-LIMITING AND
STEP-E LIMI NAT 1 ON
SIMPLIFICATIONS
L
UTILIZE INTEGRAL
SCAVENGING COEFFICIENTS
TO CALCULATE AQUEOUS-
PHASE CONCENTRATIONS
1
IS AN INTERCRAL "
1 , CHARACTERIZATION
" OF THE STORM
SUFFICIENT?


COMBINE WITH DETAILED
STORM MODEL TO CALCULATE
CASEOUS-AND AQUEOUS-
PHASE CONCENTRATIONS
1
IS
APPLY APPROPRIATE
-I RATE-LIMITING
SIMPLI CATIONS


DEFINE MICROSCOPIC
SCAVENGING RATES AS
A FUNCTION OF X, y, i AND t

                                  Figure  6-24.   Flow chart for scavenging calculations.

-------
conservation (i.e., material, energy, momentum) equations  involved.   Various
pathways through this decision tree are  discussed  in  the  original  reference.

Proceeding through Figure 6-24 in this manner, the user  can  arrive  at simple
or complex  end  points,  depending upon the nature of  his particular applica-
tion.    A  trivial  example  is  pathway  1-5-6,  which instructs  the  user  to
disregard  modeling and  rely  solely  upon  past  measurements.    The  simple
microscopic-balance example  of  Section  6.5.4 can be  traced through  pathway
1-2-7-8-21-23-15-16.

Table  6-5   itemizes  some currently-available  models, which can  be  related
directly to the pathways of  Figure 6-24.   This  provides  the  reader with a
rapid  and  efficient means of access  to  current  modeling literature,  while
minimizing  the  chance  of pitfall  encounters  that can arise from  the  inad-
vertent  use of inappropriate physical  constraints.   For  a more  definitive
description of  this model  selection process, the reader is referred to  the
original reference (Hales 1984).

6.6  PRACTICAL  ASPECTS OF  SCAVENGING  MODELS:   UNCERTAINTY  LEVELS  AND SOURCES
     OF ERROR

Quantitatively  assessing the  predictive  capability  of  present  wet-removal
models  is  a complex task,  well beyond the scope  of  this  document. There are,
however, a  number  of general  statements which  are highly useful  for focusing
in   on   this  question   and   for  providing   insights pertaining  to  model
reliability.  These are itemized sequentially below.

 o   The predictive capability of a scavenging model is  strongly contingent
     upon its desired application.

     As  noted  in  6.5.1,  a variety of different applications exist  for  scav-
     enging models, and  some  are much more difficult  to  fulfill  than others.
     One  can,   for example,  employ  existing  regional   models  to  reproduce
     distributions  of   annually-averaged,  wet-deposited,   sulfate  ion   in
     eastern North America with moderate success.   If one is charged with the
     task  of  relating specific  sources  to deposition at a chosen receptor
     site,  however,  our predictive capability  can  be expected  to  be  rela-
     tively imprecise.   Similarly,  if one  is  expected to forecast the change
     in  deposition that  would  occur in  response to some  future  change  in
     emissions,  then  the  associated  uncertainty level  would be  very  high
     indeed.   The question of nonlinear response is  of  paramount importance
     in  this last application.

     A  large  component of our  uncertainty in predicting  source  attribution
     and transient response  is  based simply on the  fact that we  do not have
     adequate  data bases  for testing model   performance for these applica-
     tions.  Our  present models may in actuality be better predictors in this
     respect  than  anticipated,   but  because  we  have  no  immediate way  of
     confirming this, our uncertainty level remains high (Section 6.4).
                                     6-64

-------
TABLE 6-5.  PERTINENT LITERATURE REFERENCES FOR WET-REMOVAL MODELS

1.
2.
3.
4.
T 5.
CT>
cn
6.
7.
8.
9.
Model
Classical Washout
Coefficient
Distributed Washout
Coefficient
"Two-Stage" Nuclea-
ti on- Accretion
Nonreactive Gas
Scavenging
Reactive Gas
Scavenging
In-Cloud Aerosol
Scavenging
In-Cloud Aerosol
Scavenging
In-Cloud Reactive
Gas and Aerosol
Scavenging
In-Cloud Reactive
Gas and Aerosol
Scavenging
Type of Balance
Equation! s)
Material
(Differential
Material
(Differential)
Material
(Differential)
Material
(Differential)
Material
(Differential)
Material
(Differential)
Material (Integral)
Material
(Differential)
Material (Integral)
Mechanl sm( s)
Irreversible Attachment
Irreversible Attachment
Irreversible Attachment
Reversible Attachment
Reversible Attachment
with Aqueous-Phase
Reaction
Irreversible Attachment
Irreversible or
Reversible Attachment
Transport, Reaction and
Deposition
Irreversible or
Reversible Attachment
with Chemical Reaction
Typical Application
Below-cloud scavenging
of aerosols and reactive
gases
Below-cloud scavenging of
size-distributed aerosols
Condensati on-enhanced
below-cloud scavenging of
aerosol s
Below-cloud scavenging of
nonreactive gases
Below-cloud scavenging of
reactive gases
Scavenging in storm systems
(nonreactive)
Scavenging in storm systems
Scoping studies
Interpretation of field
study data
Pertinent References
Chamberlain (1953), Engelmann (1968), Fisher
(1975), Scriven and Fisher (1975), Wangen and
Williams (1978)
Dana and Hales (1976), SUnn (1983)
Radke et al . (1978), Slinn (1983)
Hales et al. (1973, 1979), Slinn (1974b),
Barrle (1978)
Hill and Adamowicz (1977), Adamowicz (1979),
Overton et al. (1979), Durham et al. (1981),
Drewes and Hales (1982)
Junge (1963), Dingle and Lee (1973), Storebo
and Dingle (1974), Klett (1977), Lange and Knox
(1977), SUnn (1983)
Engelmann (1971), Gatz (1972), Scott (1978),
Hales and Dana (1979a), Slinn (1983)
Gravenhorst et al. (1978), Omstedt and Rodhe
(1978)
Scott (1982)

-------
                                                                  TABLE  6-5.   CONTINUED
                   Model
  Type of Balance
    Equation!s)
                                                              Mechanism! s)
                             Typical Application
                                                                                                                            Pertinent References
         10. Composite Analytical    Material
                                    (Differential)
                     Transport, Reaction and  Regional scale deposition
                     Deposition
                                                      Astarita et al.  (1979),  Fay and Rosenzweig
                                                      (1980)
 I
cr>
         11. Composite Trajectory   Material
                                    (Differential)
         12. Composite  Grid
         13. Composite
             Statistical

         14. Nonreactive
         15. Reactive
Material
(Differential)
Material
                     Transport, Reaction and  Regional  scale deposition
                     Deposition
Transport,  Reaction and  Regional scale deposition
Deposition
                     Transport, Reaction and  Scoping studies and
                     Deposition               life-time assessment
Bolin and Persson (1975).  Hales (1977),
Eliassen (1978),  Fisher (1975), Bass  (1980),
Heffter (1980),  Henmi  (1980),  Sampson (1980),
Bhumralkar et al. (1980),  Kleinman  et al.
(1980), Shannon  (1981), McNaughton  et al.
(1981), Patterson et al. (1981),  Voldner (1981)

Liu and Durran (1977),  Prahm and  Christensen
(1977), Mil ken ing and Ragland  (1980),  Lavery
(1980), Lee (1981),  Carmichael  and  Peters
(1981), Lamb (1981)

Rodhe and Grande!1  (1972,  1981)
                                   Material Energy and  Irreversible Attachment,  In-cloud scavenging analysis  Molenkamp (1974), Hane (1978),  Kreitzburg and
                                   Momentum             Nonreactive                                           Leach (1978)
                                   (Differential)
                                   Material and Energy  All modes of scavenging
                                   (Differential)       including chemical
                                                        reaction
                                             In-cloud scavenging  analysis   Hales  (1982)

-------
    Regardless of the above considerations It should be emphasized  strongly
    that the first step  in  scavenging  model  evaluation must be the  precise
    definition of the  intended uses of  the  model.   All  subsequent  efforts
    will be confounded in the  absence of  this focal  point.

o   The predictive capability of a  scavenging model depends  upon  the choice
    or mode I'.

    At  first sight this  appears to  be  a  self-evident and  trivial  statement.
    A profusion of scavenging models exists,  however,  and it is  not  at  all
    difficult  to  choose an  inappropriate  candidate   inadvertently.    Such
    inappropriate selections  have  on  occasion  resulted  in  reported  calcu-
    lations which have been in error by several  orders  of  magnitude  (Section
    6.5.1).

    This component of error may  of course be totally  eliminated  by  select-
    ing the most  appropriate  model  for the  intended application.   The  flow
    chart presented in Figure 6-24  is  a  useful  guide   for  this  purpose,  es-
    pecially for those only casually familiar with the  field.

°   The predictive  capability of a  scavenging  model   depends strongly  upon
    the processes modeled.

    As  noted in the context of Figure  6-2 a scavenging model may encompass
    one, several, or all  of the steps in  the composite  wet-removal sequence.
    If only a small  portion of this  sequence is being  considered, the model
    depends heavily upon information supplied from the  remaining components.
    This  information  may   originate   from  assumptions,   from  empirical
    measurements, or  from the output  of other models.    Assuming  that  all
    input information  is error-free,  then  it may be   stated generally  that
    the more  steps  in Figure  6-2 encompassed by  a  given  model, the greater
    will be  its  predictive uncertainty.    This  is  simply a consequence  of
    propagating errors and  must  be considered as a primary  factor  when one
    addresses the validation of wet-removal  calculations.

o   The predictive  capability of a scavenging model depends upon its area!
    range.

    This statement  is largely a  corollary of the  one  immediately above.  As
    a scavenging model is extended to, say,  a regional  scale it is forced to
    include  essentially   all  of the  components  of  Figure  6-2.   As  noted
    previously, this is  likely to increase uncertainty  levels appreciably.

°   The predictive  capability of a scavenging model is contingent  upon its
    temporal averaging time.

    Owing to  the  propensity of stochastic phenomena to average out to  mean
    values, the predictive capabilities  of  (especially regional)  scavenging
    models can be expected to improve somewhat as  averaging times  increase
    (see  Chapter  A-9).    This  improvement  is,  of  course, gained  at  the
    expense  of sacrificing  temporal  resolution,  and  a  value judgment  is
                                    6-67

-------
     necessary  (again   requiring  a  precise  definition  of  intended model
     application)  at this juncture.11

     This observation should  be  tempered by  the  fact  that,  in addition  to
     random errors, scavenging models can be expected to  possess  substantial
     systematic  biases.    In  general   these  biases  do  not  decrease  with
     averaging  time  and  in  fact many  lead  to cumulative  discrepancies  on
     occasion.   Examples  of  systematic errors  are  biases  in   trajectory
     calculations and artificial  offsets induced by  the superimposition  of
     random events  on  nonlinear processes.   Again the  seriousness of  such
     factors is heavily  contingent  on  the intended model  application (Section
     6.5.1).

In general summary, it may be  stated  that several  important factors  lead  to
widely varying  levels  of uncertainty in scavenging-model predictions.  One
may  predict,  for example, the  scavenging of  S02  from  a local  power-plant
plume by  using  existing models and expect to match measured  results within a
factor of two.  On the other hand, similar predictions  of,  say, the fraction
of sulfate  at a given  receptor which originated from some  particular source
can  be expected to  have orders-of-magnitude  associated uncertainty.   Both a
comprehensive model-evaluation effort and a substantially-improved data  base
will  be  required before  this  situation  can  be remedied to any  appreciable
extent (Section 6.4).

6.7  CONCLUSIONS

This chapter  has provided an  overview  of meteorological processes contrib-
uting to  wet  removal  of  pollutants and  has summarized  the  current  state  of
our  capability to  describe  these  complex phenomena  in mathematical  form.
Because  of  the magnitude of  this problem, it  has  been  necessary  to refrain
from detailed descriptions of models and modeling  techniques; rather, we have
chosen to describe  the  general  mathematical  basis for  wet-removal  modeling,
to give  two  simple  examples  of direct  application,  and then to  supply the
reader with a  means  for efficiently pursuing  the  available  literature for
specific  applications of  interest.

In conclusion  to  this discussion it  is  appropriate  to summarize the state  of
these calculational techniques by asking the following questions:

      ° Just  how accurate and valid are current wet-removal  modeling
         techniques as predictions of precipitation chemistry and wet
         deposition; that  is, how well do they fulfill  the needs
         itemized in Section 6.5.1?

      0  What must be accomplished before the present capabilities can
         be  improved?
        issue  is  especially  pertinent  in  view  of  the  contention,  often
   voiced  by some scientists within the acid-precipitation effects community,
   that temporally-averaged results (averaging times of a few months or more)
   are  totally adequate  for  assessment purposes.


                                     6-68

-------
The answers to these questions are somewhat mixed.  Certainly  the  techniques
discussed in  this  chapter,  if used  appropriately,  are  capable of  order-of-
magnitude determinations  in many  circumstances;  and under  restricted  con-
ditions they  can even  generate  predictions having factor-of-two accuracy or
better.   Moreover, there is  ample explanation in  existing  theories of  wet
removal to account easily for the spatial  and  temporal variabilities observed
in nature.

These capabilities, however, cannot be considered to be very satisfactory in
the  context  of current  needs.    The noted ability to explain  spatial  and
temporal variability on a semi quantitative basis has not  resulted  in a  large
competence in  predicting  such variability in  specific  instances.   Moreover,
we possess very little competence in  identifying specific  sources responsible
for wet deposition at a given receptor site.   Finally, the order-of-magnitude
predictive capability noted above hardly can be judged satisfactory for most
assessment purposes.

In reviewing  the discussions  of this chapter  against the backdrop  of  these
deficits, several  research  needs  become  apparent.   The  most important of
these are itemized in the following paragraphs:

    o   Much more definitive information  is needed  with regard  to the scaven-
        ging  efficiencies  of submicron  aerosols,  for  both rain  and  snow.
        Especially important in  this  regard is the effect of  condensational
        growth of such aerosols  in below-cloud environments (Section 6.5.3).

    0   We need to know  much more about aqueous-phase conversion  processes,
        which are potentially important as alternate mechanisms resulting in
        the presence of species  such  as  sulfate and nitrate in  precipitation.
        Because virtually nothing  is  known  presently  regarding the  chemical
        formation  of  such species in  clouds  and  precipitation, there  is a
        tendency to  lump these   effects with  physical  removal  processes in
        most modeling efforts, expressing them  in terms of pseudo  scavenging
        coefficients  or  collection  efficiencies.   Such   phenomena  must be
        resolved in finer mechanistic detail than this before  a  satisfactory
        treatment  is  possible,   and  this  requires  a  knowledge  of  chemical
        transformation processes  that is  much  more advanced  than   exists at
        present (Sections 6.2.4  and 6.5.3  and  Chapter A-4).

    o   Much  more   extensive  understanding  of  the competitive   nucleation
        capability of aerosols in in-cloud environments is needed,  especially
        for those  substances  that do  n.ot compete  particularly  well in  the
        nucleation process.   The influence of  aerosol-particle composition—
        especially  for  "internally-mixed"  aerosols (those containing  indi-
        vidual particles  composed of  mixed chemical  species)--is  particularly
        important in this regard (Section  6.2).

    «   Identifying specific sources  responsible for chemical deposition at a
        given receptor location  requires  that  we  possess  a much more accom-
        plished  capability   to   describe  long-range  pollution   transport.
        Progress in this  area during  recent years has  been encouraging,  but
                                     6-69

-------
        much more  remains  to  be  achieved  before  we  are  sufficiently pro-
        ficient for reliable  source-receptor analysis  (Section  6.4).

    0    We still  need  to enhance  our understanding  of  the detailed  micro-
        physical and dynamic  processes  that occur  in storm systems.  Besides
        providing  required   knowledge  of  basic   physical  phenomena, such
        research is  important in providing  valid  parameter!'zations of wet-
        removal  for  subsequent use  in  composite  regional  models  (Section
        6.4).

As a final note, it  is useful  to reflect once  again  on the  fact  that scaven-
ging modeling  research—as  treated  in   this  chapter—has  been  in a  rather
continuous state of  development over  the past 30 years.  While  progress  has
been  indeed significant during  this  period, a  number  of  important  and
unsolved problems still  exist.  Accordingly, one must use this  perspective in
assessing our  rate of advancement  during future  years.  Reasonable  progress
in resolving the above  items can be  expected  over the next decade; but  the
complexity of these  problems  demands  that a serious and sustained effort be
applied for this purpose.
                                     6-70

-------
6.8  REFERENCES

Adamowicz,  R.  F.    1979.    A  model  for  the  reversible  washout  of sulfur
dioxide,  ammonia,  and  carbon  dioxide  from  a  polluted atmosphere  and the
production of sulfate in raindrops.   Atmos. Environ.  13:105-121.

Astarita, G.,  J.  Wei, and 6.  lorio.   1979.   Theory of dispersion transfor-
mation  and  deposition  of  atmospheric  pollution   using  modified  Green's
functions.  Atmos. Environ. 13:239-246.

Baker,  M.  B.,  H. Harrison,  J.  Vinelli,  and  K. B.  Erickson.   1979.   Simple
stochastic  models  for the  sources  and sinks  of  two  aerosol  types. Tell us
31:1-39.

Barrie,  L.  A.   1978.  An  improved  model  of  reversible S0£ washout  by  rain.
Atmos.  Environ. 12:407-412.

Barrie,  L.  A.  and J. Kovalick.  1978.   A  wintertime investigation of  the
deposition  of  pollutants around an  isolated  power plant in northern  Alberta.
Atmospheric Environment Service, Environment Canada, REP ARQT-4-78.

Bass,  A.   1980.   Modeling  long-range transport and diffusion.   Proc. Second
Conf.  on App.  Air Pol. Meteorol.  AMS/APCA,  New Orleans.

Berry,  E. X. and R.  L. Reinhardt.  1974.  An  analysis of cloud drop growth by
collection. Part IV.  A new  parameterization.  J.  Atm. Sci. 31:2127-2135.

Bhumralkar, C. M., W. B. Johnson, R.  H. Mancusco, R.  H. Thuillier, and  D. E.
Wolf.    1980.    Interregional   exchanges  of  airborne  sulfur  pollution  and
deposition  in  eastern North America.   Proc. Second  Conf. on App.  Air  Pol.
Meteorol.   AMS/APCA,  New Orleans.

Bird,  R.  B., W. E. Stewart,  and E. N. Lightfoot.  1960.  Transport Phenomena.
John  E. Wiley, New York, NY.

Bolin, B.  and C. Persson.    1975.   Regional  dispersion  and deposition of
atmospheric pollutants  with particular application  to sulphur pollution over
western Europe.  Tellus 27:281-309.

Browning,  K. A., M.  E. Hardman, T. W.  Harrold,  and  C.  W. Pardoe.   1973.   The
structure  of  rainbands within  a midlatitude cyclonic  depression.  Quart. J.
Roy.  Meteorol. Soc.  99:215-231.

Burtsev,  I.  E., L. V.  Burtsevva,   and  S.   G. Malakhov.    1976.    Washout
characteristics  of   a  32p  aerosol   injected  into   a  cloud.    Atmospheric
scavenging of  radioisotopes.   Symp.  Proc. Palanga, USSR.

Cadle, R.  D.   1965.   Particle  Size.   Reinhold,  New York, 390  pp.
                                     6-71

-------
Carmichael, 6.  R.  and L.  K.  Peters.   1981.    Application  of  the  sulfur
transport Eulerian Model  (STEM)  to a SURE data set.  12th Int. Tech. Meeting
on Air Poll. Mod.  and its  App.   NATO,  Palo Alto.

Chamberlain, A. C.   1953.   Aspects of travel  and deposition of aerosols and
vapor clouds.   AERE Harwell, Report R1261, HMSO London.

Chan, W. H., M. A. Lusis, A. J.  S. Tang,  C. U.  Ro,  and R. J.  Vet.   1982.
Precipitation  scavenging  and  dry  deposition  of  pollutants  from  the  INCO
nickel smelter in  Sudbury.   Presented at Fourth   International  Conference on
Precipitation Scavenging, Dry  Deposition,  and Resuspension.   Santa Monica,
CA.

Changnon,  S. A.    1968.   Precipitation  scavenging  of Lake  Michigan  Basin.
Bull. 52.  Illinois State  Water Survey Report, Urbana,  IL.

Changnon,  S.   A.   A.  Auer,   R.  Brahm,  J.  Hales,  and R.  Semonin.   1981.
METROMEX:  A Review and Summary.   AMS Monograph,  Vol.  18,  Am. Meteorol. Soc.,
Boston, MA.

Court, A.   1966.    Fog Frequency in the United States.    Geog.  Rev. N.Y.,
56:543-550.

Dana, M. T.  1970.  Scavenging  of  soluble dye  particles by rain.  Jji Precipi-
tation Scavenging  1970.    R. J.  Engelmann  and  W.  6. N.  SI inn, eds.   AEC
Symposium Series.

Dana, M. T. and D. W. Glover.  1975.  Precipitation  scavenging of  power plant
effluents:    Rainwater concentrations  of sulfur   and  nitrogen  compounds and
evaluation  of  rain  samples  desorption  of  S02-    PLN  Annual Report  to  U.S.
AEC, BNWL-1950.

Dana, M. T.  and J. M. Hales.  1976.   Statistical  aspects of the washout of
polydisperse aerosols.  Atmos.  Environ.  10:45-50.

Dana, M. T., D. R. Drewes, D. W. Glover, and J.  M. Hales.   1976.  Precipita-
tion  scavenging of fossil fuel  effluents.  Battelle-Northwest Report  to EPA,
EPA-600/4-76-031.

Dana,  M.  T.,  J.  M.  Hales,   and M. A.  Wolf.   1972.   Natural   precipitation
washout of sulfur dioxide.  Battelle-Northwest Report  to EPA, BNW-389.

Dana,  M.  T.,  J.  M.  Hales,  W.  G.  N.  Slinn, and  M.  A. Wolf.  1973.   Natural
precipitation  washout  of  sulfur  compounds  from  plumes.    Battelie-Northwest
Report to EPA, EPA-R3-73-047.

Dana,  M.  T.,  A.  A.  N.  Patrinos,  E.  G. Chapman,  and J.  M.  Thorp.   1982.
Wintertime  precipitation chemistry  in North Georgia.   Proc. ACS  Symposium on
Acid  Rain, Las Vegas, NV.
                                     6-72

-------
Dana, M.  T.,  N.  A.  Wogman,  and M.  A. Wolf.   1978.   Rain  scavenging  of
tritiated water  (HTO):   A field experiment  and theoretical  considerations.
Atmos. Environ. 12:1523-1529.

Davenport, H.  M. and  L.  K.  Peters.    1978.   Field  studies  of atmospheric
particulate  concentration  changes  during  precipitation.   Atmos.  Environ.
12:997-1008.

Davies,  C. N.  1966.  Aerosol  Science.   Academic Press, New York.

Dingle,  A. N.  and Y. Lee.   1973.   An  analysis of in-cloud  scavenging.   J.
Appl. Meteorol. 12:1295-1302.

Dingle,  A. N., D. F.  Gatz, and J. W. Winchester.   1969.   A pilot experiment
using indium as tracer in  a convective  storm.  J. Appl. Meteorol. 8:236-240.

Drewes,   D.  R.  and  J.  M.  Hales.    1982.     SMICK:    A scavenging  model
incorporating chemical kinetics.   Atmos. Environ. 16:1717-1724.

Durham,  J. L., J. H.  Overton, and V. P. Aneja.   1981.  Influence of gaseous
nitric  acid  on  sulfate production  and acidity in  rain.   Atmos.  Environ.
15:1059-1068.

Easter,  R. C.   1982.   The OSCAR Experiment.   Proc. ACS  Symposium on Acid
Rain, Las Vegas, NV.

Easter,  R  C. and J. M. Hales.  1983a.   Interpretations of  the  OSCAR data for
reactive  gas scavenging.   Proc.  Fourth International Conference on Precipi-
tation  Scavenging, Dry Deposition and Resuspension, Santa Monica, CA.

Easter,  R. C.  and J.  M. Hales.   1983b.   Mechanistic  evaluation of precipita-
tion-scavenging data using a one-dimensional  reactive storm model.   Battelle-
Northwest report to EPRI,  EPRI RP-2022-1.

Eliassen,  A.    1978.    The  OECD  study  of  long-range  transport of  air
pollutants.  Atmos. Environ. 12:479-487.

Engelmann, R. J.  1965.  Rain scavenging of zinc sulphide  particles. J. Atm.
Sci. 22:719-724.

Engelmann, R.  J.   1968.   The calculation of  precipitation  scavenging.   Jji
Meteorology and Atomic Energy 1968.   D. Slade,  ed.   U.S.  AEC.

Engelmann,  R.  J.    1971.   Scavenging  prediction  using  ratios of air and
precipitation.  J. Appl. Meteorol. 10:493-497.

Enger,  L.  and  U. Hogstrom.   1979.   Dispersion  and wet deposition  of  sulfur
from a  power-plant  plume.  Atmos. Environ.  13:797-810.
                                     6-73

-------
Falconer, R.  E.  and P.  D.  Falconer.    1980.   Determination  of  Cloud Water
Acidity  at  a  Mountain  Observatory in  the  Adirondack Mountains  of  New York
State.  J. Geophys. Res.  85:7465-7470.

Fay, J.  A.  and J. J. Rosenzweig.  1980.   An  analytical  diffusion model  for
long-distance transport of air pollutants.   Atmos.  Environ. 14:355-365.

Fisher,  B. E. A.  1975.  The long range transport of sulfur dioxide.  Atmos.
Environ. 9:1063-1070.

Fitzgerald, J. W.  1974.   Effect of aerosol  composition of cloud-droplet size
distribution:  A numerical study. J.  Atm.  Sci. 31:1358-1367.

Fuchs, N. A.  1964.  The Mechanics of Aerosols.  Pergamon Press,  Oxford, 407
pp.

Fuquay,  J. J.  1970.  Scavenging in perspective.  ^Precipitation Scavenging
1970.  R. J. Engelmann and W. G. N.  SI inn,  eds.   AEC  Symposium Series  22.

Galloway, J. N. and D.  M. Whelpdale.   1980.  An atmospheric sulfur budget for
eastern  North America.   Atmos. Environ.  14:409-417.

Gatz,  D. F.   1972.  Washout ratios  in urban and non-urban areas.  Proc. AMS
Conf.  on Urban Environment.  Philadelphia,  PA.

Gatz,  D.  F.  1977.   A  review of chemical tracer  experiments on precipitation
systems.  Atmos. Environ. 11:945-953.

Gibbs,  A.  G.   and  W.  G.  N.  SI inn.    1973.    Fluctuations in  trace  gas
concentrations in  the troposphere. J. Geophys. Res.  78:574-576.

Godske,  C. L.,  T.  Bergeron,  J. Bjerkness,  and  R.  E.  Bundgaard.   1957.
Dynamic  Meteorology  and Weather Forecasting.  Am. Meteorol. Soc., Boston, MA.

Graedel, T.  E.  and  J. P.  Franey.   1977.   Field  measurements  of  submicron
aerosol  washout  by rain.   Jji Precipitation Scavenging 1974.    ERDA  Symposium
Series 41.

Granat,  L.  and H. Rodhe.   1973.  A Study of fallout by  precipitation around
an  oil-fired power plant.   Atmos. Environ.  7:781-792.

Granat,  L.  and R.  Soderlund.  1975.   Atmospheric deposition   due  to long and
short  distance sources with  special   reference to  wet and dry deposition  of
sulfphur compounds  around  an  oil-fired  power  plant.   MISU   Report A-32,
Stockholm University, Sweden.

Gravenhorst,  G., T. Janssen-Schmidt,  D.  H. Ehhalt, and  E.  P.  Roth.  1978.
The influence  of  clouds and  rain  on  the  vertical  distribution of  sulfur
dioxide  in a one dimensional steady-state model.   Atmos.  Environ. 12:691.
                                     6-74

-------
Greenfield, S. M.   1957.   Rain scavenging of radioactive participate matter
from the atmosphere.  J.  Meteorology  14:115-123.

Hales, J.  M.   1972.  Fundamentals of the theory of  gas  scavenging  by rain.
Atmos. Environ. 6:635-659.

Hales,  J.  M.    1977.    An  air  pollution  model   incorporating  nonlinear
chemistry,  variable  trajectories,  and  plume-segment diffusion.   Battelle-
Northwest Report to EPA,  EPA-450/3-77-012.

Hales, J. M.  1982.  Mechanistic  analysis  of precipitation scavenging using a
one-dimensional, time-variant model.   Atmos. Environ. 16(7):1775-1783.

Hales, J.  M.   1984.  Precipitation chemistry:  Its  behavior  and its calcu-
lation.   j£ Air Pollutants and  Their  Effects on  the Terrestrial Ecosystem.
S. V. Krupa and A. H. Legge, eds.  John  M. Wiley, New York, NY.

Hales, J.  M. and  M.  T.  Dana.   1979a.   Precipitation  scavenging  of urban
pollutants by convective  storm systems.  J.  Appl. Meteorol. 18:294-316.

Hales, J.  M.  and M. T.  Dana.   1979b.  Regional  scale deposition of sulfur
dioxide by precipitation  scavenging.   Atmos. Environ. 13:1121-1132.

Hales, J.  M.,  J.  M. Thorp, and  M. A.  Wolf.  1971.   Field investigation of
sulfur dioxide  washout from the plume  of a large  coal-fired  power plant by
natural  precipitation.    Battelle-Northwest  Final  Report  to  Environmental
Protection Agency, NTIS PB  203-129.

Hales, J.  M.,  D.  M. Miller, A. J. Alkezweeny, and R.  N. Lee.  1979.  Ozone
formation related to power  plant emissions.   Science  202:1186-1188.

Hales, J. M.,  M.  A. Wolf,  and M.  T.  Dana.   1973.    A linear  model  for
predicting  the  washout of  pollutant  gases from industrial plumes.  AICHE J.
19:292-297.

Hane, C.  E.   1978.   Scavenging of  urban pollutants by  thunderstorm rainfall:
Numerical experimentation.   J. Appl.  Meteorol. 17:699-710.

Haurwitz, B.  and J. M. Austin.   1944.   Climatology.    McGraw-Hill, New York,
NY.

Heffter, J. L.   1980.  Air resources laboratories atmospheric  transport and
dispersion model (ARL-ATAD).  NOAA Tech. Memo, ERL-81.

Henmi, J.   1980.   Long-range  transport  model of  SOg  and sulfate  and its
application to the eastern  United States.  J. Geophys.  Res. 85:4436-4442.

Hill, F.  B.  and R.  F.  Adamowicz.  1977.    A model  for  rain  composition and
the washout of sulfur dioxide.  Atmos.  Environ.  11:912-927.
                                     6-75

-------
Hobbs, P. V.   1978.   Organization and structure of clouds and precipitation
on the mesoscale and microscale in cyclonic storms.  Rev. Geophys. and Space
Sci. 16:741-755.

Hogstrom, U.  1974.  Wet fallout of sulfurous pollutants emitted from a city
during rain or snow.   Atmos.  Environ.  8:1291-1303.

Hutcheson, M. R.  and  F.  P.  Hall.   1974.   Sulfate  washout  from a coal-fired
power plant plume.  Atmos.  Environ. 8:23-28.

Junge, C.  E.   1963.  Air Chemistry and  Radioactivity.   Academic Press, New
York.

Junge, C.  E.   1974.  Residence  time  and  variability  of tropospheric trace
gases.  Tellus 26:477-488.

Klein, W. H.   1958.   The frequency of cyclones and anticyclones in relation
to the mean circulation.  J.  Meteorol.  15:98-102.

Kleinman, L. J., J. 6. Carney, and R.  E. Meyers.   1980.   Time dependence on
average regional sulfur  oxide concentrations.  Proc. Second Conf. on App. Air
Pol. Meteorol.  AMS/APCA, New Orleans.

Klett, J.  1977.   Precipitation scavenging in  rainout  assessment:   The ACRA
system and summaries of  simulation results. LASL Report  to ERDA, LA6763.

Kramer,  J.  R.    1973.   Atmospheric  composition  and  precipitation  of  the
Sudbury Region.  Alternatives 2:18-25.

Kreitzberg, C.  W. and  M.  J.  Leach.   1978.    Diagnosis and  prediction  of
tropospheric  trajectories and cleansing.    Proc. 85th National Meeting AICHE,
Philadelphia, PA.

Lamb, R.  G.   1981.  A regional scale  model of  photochemical air pollution.
Draft  Report,  Meteorology  and Assessment  Division,  U.S. EPA/ESRL, Research
Triangle Park, NC.

Lange, R.  and J.  B.  Knox.   1977.  Adaptation  of  a three-dimensional  atmos-
pheric  transport-diffusion   model  to  rainout  assessments.    Precipitation
Scavenging 1974.  R. S.  Semonin and R.  W. Beadle, eds.  ERDA  Symposium Series
41, CONF 741003.

Larson,  T. V.,  R. J.  Charlson,   E.  J.  Knudson,  G.  D. Shristian,  and  H.
Harrison.  1975.  The influence of a sulfur dioxide point source on the rain
chemistry  of  a single storm  in the  Puget Sound  Region.   Water,  Air, Soil
Pollut. 4:319-328.

Lavery,  T.  L.   1980.    Development and  validation  of a  regional  model  to
simulate  atmospheric concentrations of S02 and  sulfate.   Proc. Second Joint
Conf. on App. Air Pol. Meteorol.,  New  Orleans, LA.
                                     6-76

-------
Lee, H. N.  1981.  An Alternate  Pseudospectral Model for Pollutant Transport.
Diffusion, and Deposition in the Atmosphere.   Atmos. Environ. 15:1017-1024.

Lee, R.  N.  and  J.  M. Hales.   1974.   Precipitation  scavenging  of organic
contaminants.  Battelle-Northwest Final Report to U.S. Army Research Office,
Durham, NC.

Levich, V. G.  1962.  Physicochemical  Hydrodynamics. Prentice Hall,  Englewood
Cliffs, NJ,  700 pp.

Liu, M. K. and D. Durran.  1977.   The  development of a  regional air  pollution
model  and  its  application  to   the  northern  Great  Plains.    EPA Report,
EPA-908/1-77-001.

Lovett, G. M., W. A.  Reiners, and R.  K. Olson.  1982.  Cloud droplet deposi-
tion in  subalpine  Balsam  Fir forests:   Hydrological and  chemical inputs.
Science 218:1303-1304.

MAPS3S/RAINE.  1981.  Biennial Progress Report.  NTIS  PNL-4096, U.S. EPA/DOE.


MAP3S/RAINE. 1982.  The MAP3S/RAINE  precipitation chemistry  network:  Statis-
tical overview for the periods 1976-1980.  Atmos. Environ. 16:1603-1631.

Mason,  B. J.   1971.   The Physics  of Clouds.   Clarendon Press, Oxford, UK. p.
579.

McNaughton,  D., D. Powell,  and C.  Berkowitz.   1981.  A User's Guide to RAPT.
MAP3S/RAINE  Report, PNL-3390.

Millan, M.  M., S. C. Barton, N. D. Johnson,  B. Weisman,  M. Lusis,  W. Chan,
and R. Vet.   1982.   Rain  scavenging  from tall  stacks:   A new experimental
approach.  Atmos. Environ.  16:2709-2714.

Molenkamp, C.  R.   1974.   A one-dimensional  numerical  model of precipitation
scavenging  with  application  to rainout  of   radioactive  debris.    Lawrence
Livermore Laboratory, Report to  U.S. AEC,  UCRL-51627.

Morgan, J. J.  and H.  M.  Liljestrand.   1980.   Measurements and interpretation
of acid rainfall  in the  Los Angeles Basin.    Cal Tech Final Report AC-2-80,
Pasadena, CA.

Mosiac.   1979.   Acid From the  Sky.   Mosiac (National  Science Foundation)
10:35-40.

Newell, R.  E., J. W.  Kidson, D.  G.  Vincent, and  G.  J.  Baer.   1972.   The
General Circulation of the  Tropical Atmosphere.  Vols. 1  and 2.  MIT Press,
Cambridge, MA.
                                     6-77

-------
Omstedt, G.  and H. Rodhe.   1978.   Transformation and removal processes for
sulfur compounds as described  by  a one-dimensional time-dependent diffusion
model.  Atmos. Environ. 12:503-509.

Overton, J. H., V. P.  Aneja, and J. L.  Durham.   1979.  Production of sulfate
in rain and raindrops  in polluted  atmospheres.   Atmos.  Environ. 13:355-367.

Patterson, D.  E.,  R.  B. Husar, W.  E.  Wilson,  and L.F.  Smith.   1981. Monte
Carlo simulation of daily  regional  sulfur distribution.   J. Appl. Meteorol.
20:404-420.

Prahm, L.  P.  and  0.  Christensen.   1977.   Long-range transmission  of pol-
lutants simulated  by  a two-dimensional  pseudospectral  dispersion model.   J.
Appl. Meteorol. 16:896-910.

Pruppacher, H.  R.  and  J.  D. Klett.  1978.  Microphysics  of Clouds  and Pre-
cipitation.  D. Reidel  Pub. Co., Boston, MA.

Radke, L. F.,  M. W. Eltgroth,  and  P. V. Hobbs.   1978.  Precipitation scaven-
ging of aerosol particles.  Proc.   Cloud Physics  and Atmospheric Electricity.
Am. Meteorol. Soc., Boston, MA.

Raynor,  G.  S.   1981.   Design  and preliminary  results  of  the intermediate
density precipitation-chemistry experiment.  Report BNL  29992.   For presen-
tation at  Third Joint AMS/APCA Conference on  Applications  of Air Pollution
Meteorology,  January.   San Antonia, TX.

Rodhe, H. and  J. Grandell.  1972.    On  the removal  time of aerosol  particles
from the atmosphere by precipitation scavenging.  Tell us 24:442-454.

Rodhe, H.  and J.  Grandell.    1981.   Estimates  of characteristic  times  for
precipitation scavenging.   J. Atm.  Sci.  38:370-386.

Saffman,  P. G. and  J.  S. Turner.   1955.   On the collision of drops in turbu-
lent clouds.   J. Fluid Mech. 1:16-30.

Sampson,  P. J.   1980.   Trajectory  analysis of summertime sulfate concentra-
tions in the northeastern  United States.   J. Appl.  Meteorol.  19:1382-1394.

Scott, B. C.   1978.    Parameterization of  sulfate  removal  by precipitation.
J. Appl.  Meteorol.  17:1375-1389.

Scott, B.  C.   1981.    Sulfate washout ratios  in  winter storms.   J.  Appl.
Meteorol.  20:619-625.

Scott, B.  C.    1982.    Predictions  of  in-cloud  conversion  rates of  S02  to
$04  based  upon a simple  chemical  and  kinematic   storm  model.    Atmos.
Environ.  16:1735-1752.

Scott, B.  C.  and N. S.  Laulainen.   1979.  On the concentration of sulfate in
precipitation.  J.  Appl. Meteorol.  18:138-147.
                                    6-78

-------
Scriven,  R.  A. and  B.  E. A.  Fisher.    1975.   The  long  range transport  of
airborne material and its removal by deposition and washout.   Atmos.  Environ.
9:49-68.

Semonin,  R.  G.  1976.   The  variability of  pH  in convective  storms.   Proc.
First International Symposium on Acid Precipitation and the Forest Ecosystem.
USDA Tech. Rept. NE-23, pp. 349-361.

Shannon, J.   1981.   A regional  model  of long-term average sulfur  atmospheric
pollution,  surface  removal,  and   net   horizontal  flux.    Atmos.  Environ.
5:689-701.

Shopauskas,  K.,  B. Styra, and  E.  Verba.   1969.   Spreading  and  rainout  of
passive admixture injected into a cloud.  7th  Int.  Conf.  on Condensation and
Ice Nuclei, Vienna, Austria.

SI inn, W.  G.  N.   1973.   In-cloud scavenging studies.  Annual  Report to U.S.
AEC/DBER.  Battelle-Northwest, BNWL-1751 pt.  1.

Slinn, W.  G.  N.   1974a.   Rate  limiting aspects  of in-cloud  scavenging.   J.
Atm. Sci. 31:1172-1173.

Slinn, W. G. N.  1974b.  The redistribution  of  a  gas plume caused  by  reversi-
ble washout.  Atmos. Environ. 8:233-239.

Slinn, W.  G.  N.   1977.   Some approximations for the wet and  dry  removal  of
particles and gases from the  atmosphere.  Water,  Air,  Soil  Pollut.  7:513-543.

Slinn, W.  G.  N.    1983.   Precipitation  scavenging.  Jji Atmospheric  Sciences
and Power Production.  D. Randerson, ed.  U.S.  DOE.

Slinn, W. G. N. and J. M. Hales.  1971.   A revaluation of the role  of  ther-
mophoresis  as  a  mechanism of  in-  and  below-cloud  scavenging. J. Atm. Sci.
28:1465-1471.

Slinn, W.  G.  N.  and J. M. Hales.   1983.   Wet removal of  atmospheric parti-
cles.   EPA MONOGRAPH SERIES.

Squires,  P. and S. Twomey.   1960. The relation between cloud  droplet spectra
and  the  spectrum  of cloud  nuclei.    In Physics  of  Precipitation  NAS/NRC
Monograph No. 5.   Am. Geophys.  Union, Washington,  D.C.

Storebo,  P. B. and  A. N. Dingle.  1974.  Removal  of pollution by rain in a
shallow air flow.   J. Atm. Sci.  31:533-542.

Summers,  P.  W. and  B.  Hitchon.   1973.   Source  and budget  of  sulfate   in
precipitation  from  Central  Alberta, Canada.   J. Air  Pollut.  Contr. Assoc.
23:194-199.
                                    6-79

-------
Thorp, J.  M.  and B. C.  Scott.   1982.   Preliminary  calculations of average
storm duration and  seasonal  precipitation  rates  for  the northeast sector of
the United States.  Atmos.  Environ.  16:1763-1774.

U.S. Climatological  Atlas.   1968.  U.S.  Dept.  of Commerce, Washington, D.C.

Voldner, E.  C.,  K. Olson,  K.  Oikawa,  and  M. Loiselle.   1981.   Comparison
between measured and computed concentrations  of  sulfur  compounds in eastern
North America.  J. Geophys.  Res. 86(C6):5339-5346.

Waldman, J. M., J. W.  Munger, D. J.  Jacob,  R.  C. Flagan, J. J. Morgan, and M.
R. Hoffman.  1982.  Chemical  composition of acid fog.  Science 218:677-679.

Wang, P. K. and  H.  R.  Pruppacher.   1977.   An experimental  determination of
the efficiency which aerosol particles  are collected  by water drops in sub-
saturated air.  J. Atm.  Sci.  34:1664-1669.

Wangen, L.  E.  and M. D.  Williams.   1978.  Elemental  deposition downwind of a
coal-fired power plant.   Water, Air, and Soil  Poll. 10:33-44.

Wilkening,  K.  E.  and K.  W.  Rag!and.  1980.   Users Guide  for the University of
Wisconsin  Atmopsheric  Sulfur Computer  Model  (UWATM-SOX).   Report  to  EPA/
Duluth Research Laboratory.

Young, J. A., C. W. Thomas,  and N.  A. Wogman.  1973.  The use of natural and
man-made radionuclides  to  study  in-cloud scavenging  processes.   PNL Annual
Report for 1972 to U.S.  AEC/DBER, BNWL-1751  pt. 1.

Young, J.  A., T. M.  Tanner, C.  W.  Thomas, and N.  A.  Wogman.   1976.   The
entrainment of tracers near the sides  of convective clouds.  Annual Report to
ERDA/DBER.   Battelle-Northwest, BNWL-2000 pt.  3.

Zishka,  K.  M. and  P.  J.  Smith.    1980.   The  climatology of  cyclones and
anticyclones  over North  America  and surrounding ocean  environs  for January
and July, 1950-77.  Mon. Wea. Rev. 108:387-401.
                                     6-80

-------
            THE ACIDIC DEPOSITION  PHENOMENON  AND  ITS  EFFECTS

                    A-7.   DRY  DEPOSITION  PROCESSES

                              (B.  B.  Hicks)

7.1  INTRODUCTION (Eds.)

The presence  of acidic  and acidifying  substances  in  the atmosphere  is  a
result of  natural  and anthropogenic emissions,  atmospheric transformations,
and  transport.    Receptors  are  exposed to  these  substances  through   wet
deposition, the  processes  of  which were discussed  in the previous chapter.
These substances also  impact  on various  receptors  in the  form of dry depo-
sitions.   This  chapter addresses  many of  the questions associated with  the
dry deposition phenomenon.

The acidic and  acidifying  substances  associated  with dry deposition  include
the  gases,  S02,  NOy,  HC1,  and  NH3  and  the  particulate  aerosols  of
sulfate, nitrate, and ammonium salts.  Some  of the questions  addressed are:
How does  dry  deposition  differ from  wet deposition?   How is  dry  deposition
measured  in  the field,  in the laboratory?    What modeling  techniques   are
available  currently  for  predicting dry deposition for  specified  atmospheric
concentrations and  other  controlling  factors?   The  important issues addressed
begin with the identification  of the  various  chemical, physical,  and biologi-
cal factors that play an  important role in  the processes controlling the rate
of dry deposition as a function of time and space.  These processes take into
account the aerodynamics  near receptor surfaces, boundary layer effects,  and
other receptor surface phenomena.

The following  chapter of  the  document discusses monitoring  of dry  and  wet
deposition.  Wet deposition network  data are analyzed and interpreted so as
to provide maps  of  the U.S. and Canada with  sampling site locations, median
concentration data  for specified  sampling  periods  for sulfates, nitrates,
ammonium ion, calcium, chloride, and  pH.

7.2  FACTORS AFFECTING DRY DEPOSITION

7.2.1  Introduction

The rate  of  pollutant transfer between the air  and exposed surfaces  is con-
trolled by a wide range  of chemical,  physical,  and biological  factors which
vary in their relative importance  according to the  nature of the surface,  the
characteristics  of  the pollutant,  and the  state  of the atmosphere.    The
complexity of the  individual  processes involved  and the variety  of possible
interactions between them combine  to prohibit easy generalization; neverthe-
less,  a  "deposition  velocity",   v^,   analogous  to  a gravitational   falling
speed,  is of  considerable  use.     In practice,  knowledge   of  v^   enables


                                     7-1

-------
fluxes, F,  to be  estimated  from airborne concentrations,  C,  as the  simple
product vd*C.

Particles larger  than about  20 ym  diameter  will  be  deposited at  a rate
controlled by  Stokes's  law,  although with some  enhancement due to inertial
impaction of particles transported  near  the surface  in  turbulent  eddies.  The
settling of  submicron particles in air  is  sufficiently slow that turbulent
transfer tends to dominate,  but the net  flux  is often limited by  the presence
of a quasi-laminar layer adjacent to the  surface, which  presents a consider-
able barrier to all mass fluxes and especially to  gases with very low mole-
cular  diffusivity.    The  concept  of a  gravitational   settling  velocity  is
inappropriate in the  case of gases.  The  case  of particles  between  1  and 20
ym  diameter   is  especially  complicated,  because  all  of  these   various
mechanisms are likely to be  important.

Sehmel  (1980a) presents a tabulation of  factors  known  to influence the rate
of  pollutant deposition upon  exposed  surfaces.   Figure 7-1  has  been con-
structed on  the  basis of  Sehmel's  list  and  has  been  organized to emphasize
the greatly  dissimilar processes  affecting  the  fluxes of  gases  and large
particles.   Small, sub-micron  particles  are  affected  by  all  of the  factors
indicated in  the diagram;  thus, simplification  is  especially difficult for
deposition of  such particles.   In  reality, Figure  7-1 already represents a
considerable  simplification,   since it  omits  many   potentially  important
factors.   In  particular, the  diagram  emphasizes  properties of  the  medium
containing the pollutants in question;  a  similarly  complicated diagram could
be constructed to  illustrate the effects of  pollutant characteristics.  For
particles, critical  factors include  size, shape, mass, and wettability; for
gases,  concern  is  with molecular  weight and  polarization,  solubility,  and
chemical reactivity.   In this  context,  the  acidity of  a pollutant  that is
being transferred to  some receptor  surface by  dry processes is an especially
important quality  that may have  a strong  impact  on  the efficiency  of the
deposition process  itself.

Figure  7-2 summarizes particle size distributions on a number, surface area,
and volume basis.  In this way,  the  three major  modes  are brought clearly to
attention.   The number  distribution  emphasizes  the   transient  (or   Aitken)
nuclei   range,  0.005   to 0.05 urn diameter, for which diffusion  plays  a role
in  controlling  deposition.    The area  distribution draws  attention  to the
so-called accumulation  size  range formed  largely  from  gaseous precursors
(0.05  to  2  ym diameter, affected by  both  diffusion  and  gravity).   The
remaining mode  (2  to 50 ym diameter,  most  evident in the  volume distribu-
tion)  is the  mechanically-generated particle  range  for which gravity causes
most of the  deposition.   In most literature, the 2 urn diameter is used as a
convenient boundary between  "fine"  and "coarse" particles.

As  discussed  in  Chapter A-5,  atmospheric sulfates, nitrates,  and ammonium
compounds are primarily associated with the accumulation  size range.  Figure
7-3 demonstrates that very little acidic or acidifying material is likely to
be  associated with  the  coarse particle  fraction  in  background conditions.
However, the  larger  particles  include  soil-derived minerals,  some  of which
can react chemically with airborne  and deposited  acids.   Moreover,  it has
been suggested that some of  these larger particles may provide sites  for the


                                     7-2

-------
AIRBORNE SOURCE
LARGE
PARTICLES
GASES
 AERODYNAMIC
   FACTORS
 NEAR-SURFACE
   PHORETIC
   EFFECTS
QUASI-LAMINAR
    LAYER
   FACTORS
                      SETTLING
                     TURBULENCE
                                           TURBULENCE
   THERMOPHORESIS   j
               «
|   ELECTROPHORESIS
DIFFUSIOPHORESIS
anrl
dllu
STEFAN FLOW




                                                   -L
                                           STEFAN FLOW
                     IMPACTION
    INTERCEPTION
              _L
J
                 BROWNIAN  DIFFUSION  \-
                                       MOLECULAR DIFFUSION!
   SURFACE
  PROPERTIES



[ORIENTATION)
FLEXIBILITY}


| SMOOTHNESS!
i
| MOTION |

\



| STOMATA | | WETNESS |
i
| WAXINESS | CHE
i '
i
:MISTRY I

I VESTITURE | | EMISSIONS J
i
| EXUDATES |
/
RECEPTOR

 Figure 7-1.  A schematic representation of processes likely to  influence
              the rate of dry deposition of airborne gases and particles.
              Note that some factors affect both gaseous and particulate
              transfer, whereas others do not.
                                   7-3

-------
         2 15
          x
         ro
     co
      i
       Q.
      a
   600


   400
i


   200


 ^  0

%  40
 o


 A 30
        S-  20
        o»
        o
        3  10
            0.001      0.01       0.1         1

                                 DIAMETER (yin)
                                                10
                                                               (a)
                                                               (b)
100
Figure 7-2.  Diagrammatic representations of aerosol size distributions
             according to number concentration (a), surface area (b),
             and volume (c).  Data are for typical urban area.  Adapted
             from Whitby (1978).
                                  7-4

-------
-J

Ul
             CO
              I

              o

             CM
               Q.
              O
 O
3
                                                        LEGEND
                                                       ALL PARTICLES
                                         (NH4)2 S04

                                      D  H2S04

                                   	LOG  NORMAL  FIT  TO ACCUMULATION MODE
                                                 DIAMETER

   Figure 7-3.  Surface area distributions of sulfate aerosol  (and other) particles  in  background
                (oceanic) conditions, as determined by Whitby  (1978)  from the  data of Meszaros  and
                Vissy  (1974).

-------
catalytic oxidation of sulfur  dioxide  (e.g.,  when the particles are carbon;
Cofer et al. 1981;  Chang  et al.  1981).  Little  is  known  about the detailed
chemical composition of large particle agglomerates.  However it is accepted
that their residence time  is quite  short  (i.e., they are deposited relatively
rapidly), that there are substantial spatial and  temporal  variations in both
their concentrations and  their composition, and  that  their  contribution to
dry acidic deposition should not  be ignored.

To  evaluate deposition  rates,  several  different approaches  are  possible.
Average deposition  rates can be  deduced  from field experiments that monitor
changes over time  in some  system  of  receptors.   More intensive experiments
can measure  the  deposition of particular pollutants  in  some circumstances.
Neither approach is capable of monitoring the long-term,  spatial-average dry
deposition of pollutants.   To understand why, we must first consider in some
detail  the  processes  that influence pollutant fluxes  and then relate these
considerations to measurement  and  modeling  techniques currently being advo-
cated.  The logical sequence illustrated in Figure 7-1 will  be  used to guide
these discussions.

7.2.2  Aerodynamic Factors

Except for the obvious difference that particles  will  settle  slowly under the
influence of gravity, small particles  and trace  gases  behave  similarly in the
air.   Trace  gases  are an integral  part of  the  gas mixture that constitutes
air and, thus, will be moved with  all  of the turbulent motions  that normally
transport heat,  momentum,  and  water vapor.   However, particles have finite
inertia and can  fail  to  respond to rapid turbulent fluctuations.  Table 7-1
lists  some  relevant characteristics of spherical  particles in air (based on
data  tabulated by  Fuchs  1964,  Davies  1966,  and Friedlander 1977).  The  time
scales of most turbulent motions  in the air  are  considerably  greater than the
inertial  relaxation (or  stopping)  times listed  in the  table.   These  time
scales vary with height, but even  as  close  as 1  cm from  a smooth, flat  sur-
face, most  turbulence energy will  be  associated with  time scales longer  than
0.01  seconds, so that even  100 ym  diameter  particles  would follow most  tur-
bulent  fluctuations.   However, natural surfaces  are  normally neither smooth
nor flat, and  it is clear that  in many  circumstances the flux of particles
will  be limited by  their inability to  respond to rapid air motions.

Naturally-occurring aerosol  particles are  not always  spherical, although  it
seems  reasonable to assume  they  are in the  case  of hygroscopic particles  in
the  submicron  size  range.   Chamberlain  (1975)   documents the  ratio of the
terminal velocity  of  non-spherical  particles to that of  spherical particles
with  the same volume.  In all cases, the non-spherical particles have  a lower
terminal settling speed than do equivalent spheres. The  settling  speed ratio
is indicated by a "dynamical shape factor,"a , as listed  in Table  7-2.

Thus,  trace  gases  and small particles are  carried  by atmospheric  turbulence
as 1f they  were  integral  components of the  air itself, whereas large parti-
cles  are also affected by  gravitational  settling which   causes them  to  fall
through  the turbulent  eddies.   In  general,  however, the  distribution  of
pollutants  in  the  lower atmosphere is governed by the dynamic structure  of
the atmosphere as much as by pollutant properties.


                                     7-6

-------
          TABLE 7-1.  DYNAMIC CHARACTERISTICS OF UNIT DENSITY AEROSOL
                PARTICLES AT STANDARD TEMPERATURE AND PRESSURE,
                    CORRECTED FOR STOKES-CUNNINGHAM EFFECTS
           (DATA ARE FROM FUCHS 1964, DAVIES 1966, FRIEDLANDER 1977)
Particle radius    Dlffuslvlty         Stopping  time         Settling  speed
   (ym)             (cm2 s-l)                (s)                  (cm s-1)
   0.001           1.28 x 10~2          1.33 x 10"^            1.30  x  10"6-
   0.002           3.23 x 10"^          2.67 x 10"^            2.62  x  10~°
   0.005           5.24 x ID" J          6.76 x 10"-J            6.62  x  !Q-°
   0.01            1.35 x 10'J          1.40 x 10'°            1.37  x  10'*
   0.02            3.59 x 10"?          2.97 x 10"°            2.91  x  10"5
   0.05            6.82 x 10'°          8.81 x 10'°            8.63  x  10";
   0.1             2.21 x 10"°          2.28 x 10"4            2.23  x  10'7
   0.2             8.32 x 10";          6.87 x 10"'            6.73  x  10~,
   0.5             2.74 x 10~4          3.54 x 10"°            3.47  x  10"|
   1.0             1.27 x 10"'          1.31 x IQ'l            1.28  x  10";
   2.0             6.10 x ID"!          5.03 x 10"J            4.93  x  10"?
   5.0             2.38 x 10"°          3.08 x 10"*            3.02  x  10"1
  10.0             1.38 x 10~8          1.23 x 10~J            1.20  x  10U
                                     7-7

-------
  TABLE 7-2.  DYNAMIC SHAPE FACTORS, a,  BY WHICH  NON-SPHERICAL PARTICLES
       FALL MORE SLOWLY  THAN SPHERICAL PARTICLES  (CHAMBERLAIN 1975)
            Shape                         Ratio of axes
Ellipsoid                                      4                   1.28
Cylinder                                       1                   1.06
Cylinder                                       2                   1.14
Cylinder                                       3                   1.24
Cylinder                                       4                   1.32

Two spheres touching, vertically               2                   1.10
Two spheres touching, horizontally             2                   1.17
Three spheres touching, as triangle            -                   1.20
Three spheres touching, in line                3                1.34-1.40
Four spheres touching, in line                 4                1.56-1.58
                                     7-8

-------
In daytime, the lower  atmosphere  is usually well  mixed up to a height typi-
cally in the range 1 to 2 km, as a consequence of convection associated with
surface  heating  by  insolation.    Pollutants residing  anywhere  within this
mixed layer are effectively available for deposition through the many possi-
ble  mechanisms.   Atmospheric  transfer  does  not  usually limit  the rate of
delivery of pollutants to the surface boundary layer in which direct deposi-
tion  processes  are  active.    However,  at  night,  the  lower  atmosphere may
become  stably stratified and vertical  transfer  of non-sedimenting material
can  be  so  slow  that, at times, pollutants  at heights  as  low as 50 to 100 m
are  isolated from surface deposition processes.

The  fine  details  of turbulent transport  of pollutants remain somewhat con-
tentious.    Notable among  the  areas of  disagreement is  the  question  of
flux-gradient relationships in the  surface  boundary layer.   It  is now well
accepted that the eddy diffusivity of sensible heat and  water vapor exceeds
that for momentum in unstable  (i.e., daytime) but  not in stable  conditions
over fairly smooth surfaces (see Dyer 1974,  for example).   However, it  is not
clear that  the  well-accepted relations  governing heat or momentum transfer
are  fully  applicable to particles or trace gases;  some  disagreement exists
even in  the  case  of water vapor.   The  situation is  even more uncertain in
circumstances other than over large expanses of horizontally  uniform pasture.
When  vegetation  is  tall,  pollutant  sinks  are  distributed  throughout the
canopy  so that close similarity with  the  transfer of  any  more  familiar  quan-
tity such as heat or momentum is  effectively lost.   There is even consider-
able  uncertainty  about  how  to  interpret  profiles of  temperature,  humidity,
and  velocity above  forests (Garratt  1978, Hicks et  al. 1979, Raupach et al.
1979).

7.2.3  The Quasi-Laminar Layer

In the  immediate vicinity of any  receptor surface,  a  number  of factors  asso-
ciated with molecular diffusivity and inertia of  pollutants become  important.
Large particles carried  by turbulence can be impacted  on  the  surface as they
fail to  respond  to  rapid velocity changes.  The  physics  of  this  process is
similar to that of  sampling by inertial  collection.

Inertia!  impaction   is  a process  that augments  gravitational  settling for
particles  in  the size  range typically  between 2  and  20 ym (SIinn 1976b).
Larger  particles  tend to  bounce,  and capture is  therefore less  efficient,
while  smaller  particles experience  difficulty  in  penetrating  the   quasi-
laminar layer that envelops many receptor surfaces.   Figures  7-2  and 7-3  show
that many air-borne  materials exist  in  the  size  range likely to be affected
by  inertial  impaction.   However,  from  the viewpoint of acidic  particles,
inertia! impaction may not be important  to dry deposition  because most  acidic
species are associated with particles (see Figure 7-2)  which  are  not strongly
affected by this process.  But, because many of the chemical  constituents of
soil-derived particles are capable of neutralizing  deposited  acids, inertial
Impaction may have important indirect effects upon acidic  deposition.

To illustrate the role of  molecular or  Brownian diffusivity, it is informa-
tive to consider the  simple ideal  case of  a knife-edged thin smooth  plate,
mounted horizontally and with edge normal  to the wind  vector.  As  air  passes


                                     7-9

-------
over  (and  under)  the  plate,  a  laminar  layer develops,  of thickness  6  =
cfvx/u)1'2, where  v  is kinematic  viscosity, x  is  the  downwind  distance
from  the  edge of the  plate,  and u  is wind  speed.   According  to Batchelor
(1967), the value of  the numerical  constant  c  is 1.72.   Thus,  for a  5  cm
plate  in   a  wind speed of  1  m  s~l, we  should  imagine  a  boundary  layer
thickness  reaching about 1.5 mm thick  at  the trailing edge.

Over  non-ideal surfaces, the  internal  viscous boundary  layer  is frequently
neither laminar  nor constant with  time.  The  layer generates  slowly  as a
consequence of viscosity and surface drag as air moves across a surface.  The
Reynolds number, Re  ( = ux/v,  where u  is the  wind speed,  x is the downwind
dimension   of  the  obstacle,  and  v  is  kinematic  viscosity),  is an  index  of
the  likelihood  that a  truly laminar  layer will  occur.    For  large  Re, air
adjacent to  the  surface remains turbulent;  viscosity is  then  incapable  of
exerting its  influence.   In  many cases, it seems  that  the surface layer is
intermittently turbulent.  For  these  reasons, and because close similarity
between ideal  surfaces studied  in wind tunnels and natural  surfaces is rather
difficult to swallow,  the term  "quasi-laminar  layer"  is preferred.

Wind-tunnel studies  of the  transfer of particles  to  the  walls of pipes tend
to  support the  concept of a  limiting  diffusive  layer adjacent  to  smooth
receptor surfaces.   Transfer  across  such a  laminar layer  is conveniently
formulated  in  terms  of the Schmidt number, Sc  = v/D, where  v is viscosity
and D  is the  pollutant  diffusivity.   The conductance,  or  transfer velocity,
vi» across  the quasi-laminar layer  is proportional  to the friction velocity
u*:

     V! =  Au* Sca                                                      [7-1]

where  A and a are determined experimentally.   Most studies  agree  that the
exponent  a is about  -2/3,  as  is  evident  in the experimental  data  repre-
sented  in  Figure 7-4.   However,  a  survey by  Brutsaert  (1975a)  indicates
exponents  ranging from  -0.4  to -0.8.   The value of  the constant A is also
uncertain.  The line  drawn through  the data of Figure 7-4 corresponds  to A
 -  0.06,  yet  the  wind-water  tunnel  results  of  Moller  and  Schumann  (1970)
appear  to   require  A  - 0.6.    These values  span  the  value  of  A   - 0.2
recommended for the  case of sulfur dioxide flux to fibrous, vegetated sur-
faces  (Shepherd 1974,  Wesely and  Hicks 1977).

Laminar boundary layer  theory  imposes  the  expectation  that particle deposi-
tion  to exposed surfaces will  be strongly  influenced by  the  size of the par-
ticle, with smaller particles being more readily deposited by diffusion than
larger.    It  is  clear that many artificial  surfaces or structures  made of
mineral material  will   have  characteristics  for  which  the  laminar-layer
theories might be quite appropriate.   However  the relevance to vegetation can
be  questioned.   Microscale  surface  roughness   elements can  penetrate the
barrier presented  by  this  quasi-laminar layer  and  should  be  suspected as
sites  for enhanced deposition  of both  particles  and gases (Chamberlain 1967).
Figure 7-5  is a  photograph  of the surface of a mature corn leaf  (Zea mays),
showing the dense  blanket  of  leaf  hairs, or  trichomes,  which  covers the
surface.  These  hairs are  easily visible  to  the  naked  eye  and  provide an
                                     7-10

-------
10"
I     I   II  1114-
                    I  I  I  I 11
I
10  L-                 LEGEND
            O  HARRIOT and HAMILTON (1965)

            A  HUBBARD and LIGHTFOOT (1966)

            •  MIZUSHINA et al. (1971)
in-5|       |    j   |  ||  | | | I       |     I   I  I  I INI	I     I   I   I  I  I I I

   102                      103                       104                       105

                                          Sc
     Figure 7-4.  Laboratory verification of Schmidt-number scaling for
                  particle transfer to a smooth surface.  The quantity plotted
                  is BEV
-------
Figure 7-5.   A photograph  of  a  leaf  of  common  field corn  (Zea mays),
             highlighting  the leaf hairs  that  potentially provide  a mechanism
             for partially circumventing  the otherwise  limiting quasi-laminar
             layer in contact with the  surface.   (Photograph by R.  L. Hart,
             Argonne National Laboratory)
                                     7-12

-------
obvious example of a  case  in  which the limiting transfer characteristics of
the quasi-laminar layer next to  the surface might not be a critical issue.

7.2.4  Phoretic Effects and Stefan  Flow

Particles near a hot surface encounter a force that tends to drive them away
from the surface.   Thermophoresis  depends  on  the  local temperature gradient
in the air, on the thermal  properties  of the particle,  on the Knudsen number,
Kn =  A/r  (where  x is  the mean free  path of  air  molecules,  and r  is the
radius of  the  particle),  and  on the  nature  of the  interaction  between the
particle and air molecules (see Derjaguin and Yalamov  1972).  For very small
particles  (<  0.03  ym diameter, according  to  Davies  1967),  this  "thermo-
phoresis" can be visualized as the consequence of hotter, more energetic air
molecules impacting the side  of the particle facing the  hot surface.  As a
"rule of thumb", the  thermophoretic velocity of very small particles (< 0.03
vim  diameter)  is  likely  to be  about  0.03 cm  s"1  (estimated  from  values
quoted by  Davies  1967).   For  larger particles,  radiometric  forces  become
important (Cadle 1966).   In theory,  thermal  radiation  can cause temperature
gradients across particles that  are not good thermal conductors,  resulting in
a  mean  motion  of the  particle away  from a  hot  surface.    For  particles
exceeding 1 urn diameter,  the velocity  will  be  about  four times less.

Diffusiophoresis results  when particles reside in  a mixture of intermixing
gases.   In most natural  circumstances, the principle  concern  is with water
vapor.  Close to an evaporating surface, a particle will be impacted by more
water molecules on  the  nearer  side.    Because these water  molecules  are
lighter  than  air  molecules,  there  will be a  net  "diffusiophoresis"  towards
the evaporating surface.

Diffusiophoresis and thermophoresis both depend on the  size and  shape of the
particle of interest  and hence, neither can be  predicted with precision, nor
can  safe generalizations  be made.   These  subjects  are sufficiently compli-
cated that  they constitute specialties in their  own  right.   Excellent dis-
cussions have  been given  by  Fried!ander  (1977)  and  Twomey (1977).   These
phoretic forces  are  generally small,  and  their influence  on dry deposition
can usually be disregarded.

Many workers include  Stefan flow in general discussions of diffusiophoresis,
but  because of  the conceptual  difference  between  the mechanisms involved it
is  of  current relevance  to consider  them  separately.    Stefan  flow results
from the injection into the gaseous medium of new gas  molecules  at an evapo-
rating or subliming surface.  Every gram-molecule of substrate material that
becomes  a  gas displaces  22.41 liters  of air, at  STP.   Thus, for example, a
Stefan flow velocity  of 22.41 mm  s"1  will  result when 18  g  of water evapo-
rates from  a  1 m^ area every second.   Generalization  to other  temperatures
and  pressures  is  straightforward.   Daytime  evaporation rates  from  natural
vegetation  often  exceed  0.2  g nr2  s"1  for  considerable times during the
midday period,  resulting  in Stefan flow of more  than  0.2 mm  s~l  away from
the  surface.    Detailed  calculation   for  specific  circumstances is  quite
simple.   For  the  present, it  is  sufficient to  note  that  Stefan  flow is
capable  of  modifying  surface  deposition  rates  by  an  amount that is larger
                                     7-13

-------
than the  deposition  velocity  appropriate for many  small  particles to aero-
dynamically smooth surfaces.

Electrical forces have often been mentioned  as  possible mechanisms for pro-
moting  deposition  (as  well  as  retention;   see  Section  7.1.5)  of  small
particles, particularly  through  the  "viscous"  quasi-laminar  layer immediately
above receptor surfaces.   Wason et  al.  (1973) report exceedingly high rates
of deposition of  particles in  the  size  range 0.6 to 6 ym to  the  walls of
pipes when  a  space charge is present.   Chamberlain (I960) demonstrated the
importance  of  electrostatic  forces  in   modifying  deposition  velocities  of
small  particles,  when  fields  are  sufficiently  high.    Plates charged  to
produce local  field  strengths  of more  than  2000  V cm"1,  experienced con-
siderably  more  deposition  of  small  particles  than  uncharged plates,  by
factors  between   2  and  15.     However,  in  fair-weather  conditions,  field
strengths are typically less  than 10 V  cnr1, so the net effect on particle
transfer is likely to be small.  Further  studies of the ability of electro-
static forces to  assist the transfer of  particulate  pollutants  to vegetative
surfaces were conducted by Langer (1965) and Rosinski  and Nagamoto (1965).
According to Hidy (1973),  a series of experiments was conducted using  single
conifer needles and conifer trees.   "For  single needles  or leaves, electrical
charges on  ~ 2 ym-diameter  ZnS dust with  up to eight  units  of  charge had
no detectable  effect at  wind   speeds of 1.2  to 1.6  m  s"1.   The  average
collection efficiency was  found to  be  ~ 6 percent for edgewise cedar  or fir
needles, with broadside  values an order of magnitude  lower.  Bounce-off after
striking the collector was not  detected, but  re-entrainment could take place
above  -2ms"1 wind  speed.    Tests  on   branches of  cedar and  fir  by
Rosinski  and  Nagamoto (1965)  suggested  similar  results  as  for single nee-
dles."   It  should  be  noted,   however,   that the electrical  mobility  of  a
particle  is  a  strong negative  function  of particle  size, ranging  from 2 cm
S"1  per  V  cm"1  of  field strength  for  0.001  ym-diameter   particles,  to
0.0003 cm s"1 per V cm"1 for  0.1 ym  particles (Davies 1967).

7.2.5  Surface Adhesion

Most workers assume pollutants that  contact a surface will  be captured  by it.
For  some  gases,  this assumption is  clearly  adequate.    For example,  nitric
acid vapor  is  sufficiently reactive that most surfaces should  act as  nearly
perfect sinks.   Less reactive  chemicals  will be less efficiently captured.
The  case  of particles is  of  special  interest,  however, because of the pos-
sibility of bounce and resuspension.

The  role  of  electrostatic  attraction  in  binding  deposited particles to sub-
strate surfaces remains something of a mystery.  The process by which  parti-
cles become charged  and set  up mirror-charges on  the  underlying  surface is
fairly well  accepted.   For smaller  particles, the principle charging  mecha-
nism is thermal diffusion, leading  to a  Boltzman charge  distribution.  The
resulting van der Waals  forces are often  mentioned as the  major mechanism for
binding particles once  they are deposited.   For large, non-spherical  parti-
cles, dipole moments can  be set up  in  natural  electric fields and can help
promote  the adhesion at  surfaces.    These  matters have been conveniently
summarized  by  Billings  and  Gussman (1976),   who provide mathematical  rela-
tionships for evaluating the  electrical  energy of a  particle on the basis of


                                     7-14

-------
its  size,  shape,  dielectric constant,  and  the strength  of the surrounding
electrical  field.

Condensation of water reduces  the  effectiveness of  electrostatic adhesion
forces, because leakage  paths  are then set  up and charge differentials are
diminished.  However, the presence of liquid films at the interfaces between
particles and surfaces causes a capillary  adhesive force that compensates for
the loss of electrostatic attraction.  These "liquid-bridge" forces are most
effective  in high  humidities,  and for  coarse  particles  (>  20 ym, according
to Corn 1961).

Billings and Gussman  (1976)  draw  attention  to  the  effect  of microscale sur-
face roughness in  promoting  adhesion  of particles  to surfaces.  Much of the
experimental evidence is for particle diameters much greater than the height
of surface  irregularities (e.g., Bowden and Tabor 1950).  It is the opposite
case that is likely to be of greater  interest in the present context, as will
be discussed later.

7.2.6  Surface Biological Effects

The  efficiency with  which natural  surfaces  "capture"  impacting particles or
molecules will  be influenced considerably by the chemical  composition of the
surface as  well as  its physical  structure.   The "lead candle" technique for
detecting  atmospheric  sulfur dioxide  is  an  historically interesting example
of how chemical substrates can be selected to  affect the  deposition rates of
particular pollutants.

Uptake rates of many  trace  gases  by  vegetation are controlled by biological
factors such as  stomatal resistance.   In  daytime,  this is known  to  be the
case for sulfur dioxide (Spedding  1969,  Shepherd 1974, Wesely and Hicks 1977)
and  for  ozone  in  most  situations  (Wesely  et al.  1978).    The similarity
between sulfur and ozone is not complete, however, because  the  presence of
liquid water  on the  foliage will tend to  promote  S02  deposition,  and to
impede uptake of ozone;  the former gas is quite soluble  until  the solution
becomes too acidic, whereas the latter is  essentially insoluble (Brimblecombe
1978).

The role of leaf pubescence in the capture of  particles has received consid-
erable attention.   Chamberlain  (1967)  tested  the roles of  leaf  stickiness and
hairiness in his wind-tunnel tests.    He concluded  that  "with  the large par-
ticles (32  and 19 ym)  the velocity  of deposition  to  the sticky artificial
grass was  greater  than to the  real  grass, but with those of  5 ym and less,
it was  the other  way round, thus confirming  .  .   . that hairiness is more
important  than stickiness  for  the capture  of  the  smaller  particles."   The
importance  of leaf hairs appears  to  be verified by studies of the uptake of
210Pb  and  210Po particles  by  tobacco  leaves  (Martell  1974,  Fleischer and
Parungo 1974),  and by  the  wind tunnel work  of Wedding  et  al.  (1975), who
report increases  by  a  factor  of 10  in  deposition  rates  for  particles to
pubescent  leaves compared with smooth, waxy leaves.   It  remains to be seen
how greatly biological factors  of  this kind  influence the  rates of deposition
of airborne particles to other  kinds  of vegetation.
                                     7-15

-------
7.2.7  Deposition to Liquid Water  Surfaces

Trace gas  and  aerosol  deposition  on open  water  surfaces is of considerable
practical  interest, especially considering  concern with the acidification of
poorly buffered inland waters.   Air  blowing from land across a coastline will
slowly equilibrate with the new surface at a  rate strongly dependent on the
stability  regime  involved.   If the  water  is much warmer  than  upwind land,
dynamic instability over the water will cause relatively rapid adjustment of
the  air  to its new lower  boundary,  but if the water  is cooler, stratified
flow will  occur and adjustment will  be very  slow.   In  the former (unstable)
case, dry  deposition  rates  of  all  soluble  or chemically reactive pollutants
are likely to be much higher than in the latter.   Clearly, air  blowing over
small lakes will be less likely to adjust  to  the  water surface than will air
blowing over larger water  bodies.   During  much of the  summer,  inland water
surfaces will  tend to  be cooler than  the air,  and  hence may be protected from
dry deposition because  of  the  strongly stable stratification  that will then
prevail.  This phenomenon will  occur more  frequently over  small  water bodies
than larger ones (Hess and  Hicks 1975).

Following the guidance of chemical engineering gas-transfer studies, workers
such as Kanwisher (1963), Liss (1973), and Liss and Slater (1974), have con-
sidered the role of Henry's law constant and  chemical  reactivity in control-
ling the rate of trace  gas  exchange between the atmosphere and the ocean. In
general, acidic  and acidifying species  like  S02  are  readily  removed upon
contact with a water surface.  Thus, Hicks and Liss (1976) neglected liquid-
phase resistance and  derived  net  deposition  velocities  appropriate  for the
exchange of reactive  gases  across the  air-sea interface.  The work of Hicks
and  Liss is  intended  to apply  to water bodies of sufficient size  that the
bulk exchange  relationships of air-sea interaction research are applicable.
Their considerations indicate  that deposition  velocities  for  highly soluble
and  chemically  reactive gases such  as NH3,  HC1,  and  SO? are  likely to be
between 0.10 percent  and 0.15  percent of  the wind  speed measured  at 10 m
height.  The analysis  leading  to  this conclusion  assumes that the molecular
and  eddy diffusivities  can be combined by simple addition.  This assumption
has been shown  to  approximate  the transfer of water vapor and sensible heat
from water surfaces.  However,  for fluxes  of trace gases, Deacon (1977) and
Slinn  et  al.  (1978)  argue that  it  is better to introduce  molecular dif-
fusivity through  a term analogous  to  the Schmidt (or  Prandtl)  number  of
Equation 7-1,  with the  exponent  a  - -2/3.   (In contrast,  the  linear as-
sumption used by  Hicks  and Liss  implies a = -l.o.)    Hasse  and Liss  (1980)
discuss the matter from the viewpoint of surface-film behavior, with emphasis
on the  role  of capillary waves.  In view  of  the  uncertainties  mentioned in
discussion of  Equation  7-1, further  comment on the implications and ramifi-
cations of these alternative assumptions is not warranted.

In the  limiting case  of a  trace  gas of low  solubility,  the  deposition ve-
locity is  determined by the large liquid-phase resistance, which is directly
influenced by the Henry's law constant.

It  is  probable  that  breaking  waves  will   modify the  simple  gas  transfer
formulation derived from chemical  engineering pipe-flow and wind-tunnel work.
It  is  not  clear to  what   extent such  features   account  for  the  apparent


                                     7-16

-------
discrepancy  between  the  various  Schmidt  number  dependencies of  the kind
expressed by Equation 7-1.  However, the fractional power laws are basically
extensions of laboratory work,  whereas  the  unit-power,  additive-diffusivities
result is an  approximation  to  field data.    It  is to be  hoped that the two
approaches produce results that will  converge in due course.

Wind-tunnel   results  such as shown  in  Figure 7-6,  indicate exceedingly low
deposition velocities to  water  surfaces  for particles in  the  size range of
most acidic pollutants.   As in  the case of  gas exchange,  there  are  conceptual
difficulties in extending these results to  the open ocean.   The role of waves
in the transfer of small  particles  between  the  atmosphere  and  water surfaces
remains  essentially  unknown.    Not only does engulfment  by  breaking waves
provide an alternative path across the  quasi-laminar  sublayer where molecular
(or Brownian) diffusion normally controls the transfer, but also waves are  a
source of droplets which can scavenge particulate material  from the air [see,
however, the study of Alexander (1967)  which indicates  otherwise].  Hicks and
Williams  (1979)  have proposed  a  simple model  of air-sea particle exchange
that  extends  smooth-surface,  wind- and  water-tunnel  results  (as  in  Figure
7-6) to natural circumstances,  by permitting rapid transfer to  occur whenever
waves break.  This results in very  low deposition  velocities in light  winds,
but  rapidly  increasing  velocities when winds increase  above  about 5  m s"1.
SI inn and SI inn (1980) also suggest that particle transfer is more  rapid than
the  wind-tunnel  studies  of Figure  7-6 might indicate,  but they  present an
alternative  hypothesis  for  this more  rapid  transfer:  that hygroscopic par-
ticles grow  rapidly when  exposed  to  high humidities such  as are  found  in air
adjacent  to  a water surface,  resulting  in  increased gravitational settling
and  impaction to the water surface.

7.2.8  Deposition to Mineral and Metal  Surfaces

Acidic deposition is an obvious source of worry  to architects,  historians and
others concerned with the potentially accelerated deterioration of  structures
(see  Chapter E-7).    Many popular  building  materials  react chemically with
acidic  air  pollutants,  generating new chemical  species  that  can  contribute
directly  to the decay process even if they  are rapidly and efficiently  washed
off  by precipitation.  Furthermore, in some cases the  chemical  product  causes
a visual  degradation  that cannot  be easily  rectified,  such as  the  blackening
of metal  work  exposed to hydrogen sulfide.   Livingston and Baer (1983) sum-
marize  the  various  mechanisms  involved,  and relate them  to the  formulations
that have been developed  in laboratory studies.

The  presence of water at  the surface is known to be a  key factor  in promoting
the  fracturing  and  erosion  of  stone.  Water  penetrates  pores  and  cracks and
causes mechanical stresses  both by freezing and by hydration  and  subsequent
crystallization of  salts (see  Winkler and  Wilhelm 1970,  Fassina 1978, Gauri
1978).   The  earlier discussion of  surface  effects that influence dry depo-
sition  indicated  that surface  scratches  and fractures  will  cause  accelerated
dry  deposition rates  in localized  areas.    Moreover,  phoretic  effects are
likely to be more important than in the case of foliage (because  dry surfaces
exhibit  wider  temperature  extremes than  moist  vegetation).    Stefan flow
associated with dewfall  is  also probably more important  than  for  vegetation.
                                     7-17

-------
                                                                            DEPOSITION  VELOCITY  (cm  s"1)
00
                         3
                   3- o> n>
                   fD rt- V>
                      n> c
                   rt- -S —•
                   fD    rt-
                   1 to V>
                   3 c:
                   -i. -S O
                   3 -t, -h
                   ftl ftj
                   —•OS

                    QJ
T3 (/>(/)
n> 3- rt-
fD fD C
Q. Q. Q.
                      3
                      fD
                   tn
                         cu
                   IQ rt- -5
                      3- rt-
                   O fD -<•
                   3    O
                     I -5 —•

                      U3
                   TO 3" Q.
                   0> rt- fD
                   -5   T3
                   rt- -S O
                   -i. tt> 
                   O -O -••
                   —• -5 rt-
                   fD fD -"•
                   l/> (/> O
                   •  fD 3
                      3
                      rt- rt-
                      to o

-------
Some of the more Important considerations  can be  summarized as follows (after
Hicks 1982):

     1.  Particle fluxes will tend  to  be  greatest to  the  coolest parts of
         exposed surfaces.

     2.  Both particle and gas  fluxes will  be increased when condensation is
         taking place at the  surface, and  decreased when evaporation occurs.

     3.  If  the  surface  is  wet,  impinging  particles will  have  a  better
         chance of adhering, and soluble  trace  gases  will  be more readily
         "captured."

     4.  The chemical nature  of the surface is important; if reaction rates
         with  deposited pollutants  are  rapid,  then  surfaces  can  act as
         nearly perfect sinks.

     5.  Biological   factors  can  influence uptake  rates, by  modifying  the
         ability of the surface to  capture and  bind pollutants.

     6.  The  texture of  the surface  is   important.   Rough  surfaces  will
         provide better  deposition  substrates  than  smoother  surfaces,  and
         will permit easier  transport of  pollutants  across the  near-surface
         quasi-laminar layer.

     7.  Microscale  surface  roughness  features  probably result in greater
         deposition  velocities  for aerosols,   due  to  disruption  of  the
         quasi-laminar layer  that normally limits transfer  of particles to
         aerodynamically smooth surfaces.

The  importance  of these  factors  is emphasized  by  the results  of corrosion
tests conducted during  the 1960's  at  57 sites of the  National  Air Sampling
Network  (see Haynie and  Upham 1974).   The  data indicate  a  nonlinear time
dependence, such that the build-up  of corrosion  tends  to  reduce the rate of
further  deposition  of the trace  gases  and aerosols  causing  the  corrosion.
Correlation analyses indicate significant effects of surface moisture,  simi-
lar to what is outlined above,  but no  support is  provided for  the  expectation
that deposition rates will generally  be greater to  colder  parts of exposed
surfaces.   Statistical analyses of  the  kind used by  Haynie and  Upham provide
excellent information on  the general  features  of corrosion of exposed  metal
surfaces,  but  generally fail  to  yield  clear-cut evidence  as to which pro-
cesses are controlling the deposition  that causes the corrosion.   The subject
of  damage  to materials  surfaces is  dealt with elsewhere  in this document
(Chapter E-7).

7.2.9  Fog and Dewfall

The processes  that  cause aerosol particles to  nucleate,  coalesce, and grow
into  cloud droplets are  precisely the  same  as  those which assist  in the
generation of fog.  Whenever  surface air supersaturates,  fog droplets form on
whatever  hygroscopic nuclei   are  available.    These  small  droplets slowly
settle onto exposed surfaces, or  are deposited  by interception and impaction.


                                     7-19

-------
The characteristics  of the liquid  that is deposited  are  much the  same as
those of cloud liquid water (see  Chapter A-6).

Low-altitude surface fogs  form under  conditions  of strong  stratification in
which  vertical  turbulent  transport  is minimized.   The  frequency  of  fogs
varies widely with location and with time  of year.  The depth is also  highly
variable.   However,  it must  be assumed  that fogs constitute  a mechanism
whereby  the  lower atmosphere  (say the  bottom  hundred  meters or  so)  can be
cleansed of particulate and some  gaseous pollutants.

At higher elevations, fog  droplets are  precisely the same as  the cloud drop-
lets that in  other  circumstances would  grow and finally precipitate in  sub-
stantially  diluted  form.   The importance  of  cloud droplet interception has
recently been demonstrated by  Lovett et al. (1982), at an altitude of  1200 m
in New Hampshire, where most of  the  net  deposition  of acidic species is by
cloud droplet interception. The  presence  of liquid water on exposed  surfaces
helps  promote  the deposition  of  soluble gases and wettable particles.   This
surface water arises through the action of  several mechanisms other  than the
direct  effect of precipitation.   Some  plants  exude  fluid  from  foliage,
usually at the tips of leaves, by a process known as  guttation.  Moisture can
evaporate  from  the ground  and  recondense on  other  exposed  surfaces,  a
mechanism known  as  distillation.  However, these  mechanisms are  frequently
confused with dewfall, which   is properly the process  by  which water vapor
condenses on  surfaces directly from the air aloft.   In practice,  the  origin
of the surface moisture is immaterial  to pollutants that come  in contact  with
it.   However, dewfall  and distillation are processes  that assist pollutant
deposition  through  Stefan flow,  whereas  guttation does not.   According to
Monteith (1963),  the maximum   rate  of  dewfall  is of  the  order of  0.07 mm
hr'1,  so   that   the  maximum  Stefan   flow enhancement  of  the  nocturnal
deposition velocity is about 8 cm hr'1 (see Section 7.2.4).

7.2.10   Resuspension and Surface Emission

Deposited particles  can be resuspended into  the  air,  and subsequently re-
deposited.   The  mechanisms involved  are  much the same  as those  that cause
saltation of particles from the beds of streams and from eroding  soils. These
subjects are  of  great  practical  importance  in  their  own  right,  and have  been
studied  at  length.   Concern about resuspension of  radioactive particles  near
sites  of accidents  or  weapons  tests injected a note  of  some urgency  into
related  studies   during  the  1950's and 1960's,  as  evidenced  in  the large
number of  papers on the  subject included in  the  volume "Atmosphere-Surface
Exchange of Particulate and Gaseous Pollutants" (Engelmann  and Sehmel  1976).

The momemtum  transfer  between  the atmosphere  and the surface is the driving
force  that  causes surface  particles to creep,  bounce, and eventually  saltate.
There  is a minimum frictional  force that will  cause particles of any  particu-
lar  size to  rise from  the surface.  Bagnold  (1954) identifies the  momentum
flux,  u*2,  as a  controlling  parameter, so  that it is  the   few  occurrences
of strongest  winds that are the most important.  While  most thinking  seems  to
center on wide-spread  phenomena  like  dust  storms,  Sinclair (1976)  points out
that  dust  devils provide a  highly efficient light-wind  mechanism  for re-
suspending  surface  particles  and carrying them to considerable  altitudes.


                                     7-20

-------
Clearly,  very large particles  will  not be moved  frequently,  or far.   Very
small particles are bound to the surface by adhesive forces that have already
been  discussed,  and  tend  to  be  protected  in  crevices  or between  larger
particles.

Chamberlain  (1983) has  provided  a  theoretical  basis  for  linking saltation  of
sand  particles  and  snowflakes,  and  for  relating  these  phenomena  to  the
generation of salt spray at sea.

It is not clear how saltation and related phenomena affect acidic deposition.
Surface particles that are injected into the air by the action  of the wind  do
not  normally move far, nor  do  they offer  much  opportunity for  interaction
with  other  air  pollutants  (firstly,  because  they  are confined in  a  fairly
shallow layer near the surface,  and secondly,  because  they  have a very  short
residence time).  Their effects are largely local.

Many  smaller particles  (in  the  submicron size range)  are  generated  by  reac-
tions between  atmospheric  oxidants and organic  trace  gases emitted by  some
vegetation,  especially  conifers  (Arnts et  al. 1978).  Once  again, it  is not
obvious how these should best be considered in the present  context of  acidic
deposition.   This  is  but one of many  natural  surface-sources that provide a
conceptual mechanism for injecting  particles  and trace gases into the  lower
atmosphere.  The subject is dealt with in Chapter A-2.

7.2.11  The Resistance Analog

Discussing the relative importance of  the various  factors  that  contribute  to
the net flux of  some  particular atmospheric pollutant and  determining  which
process might  be limiting  in specific circumstances are simplified by  con-
sidering a resistance model  analogous  to  Ohm's law.    Figure 7-7  illustrates
the way in which  the concept is usually applied.   An aerodynamic resistance,
ra,  is  identified with  the  transfer  of  material   through  the air  to the
vicinity of  the  final receptor  surfaces.  This resistance  is defined as that
associated with the transfer of  momentum;  it is dependent on the roughness  of
the surface,  the  wind  speed, and the  prevailing atmospheric stability.  The
aerodynamic resistance can  be written as


     ra =  (Cfri - V*)/u*                                               [7-2]


where Cfn is the appropriate friction coefficient (the square root of the
familiar  drag  coefficient)  in  neutral  stability,  u*   is   the   friction
velocity  (a  scaling  quantity defined  as  the root  mean  covariance between
vertical and longitudinal wind  fluctuations),  k  is the von  Karman constant,
and fc  is a  stability correction  function  that  is  positive  in unstable,
negative in stable, and zero in  neutral stratifications (see Wesely and  Hicks
1977).  Equation 7-2 is obtained by straightforward  manipulation of  standard
micrometeorological relations,  as  given by  Wesely  and  Hicks,  for  example.
The value of  k  is usually  taken to  be about 0.4.    Table  7-3   lists typical
values of the friction  coefficient  for  a range of surfaces.
                                     7-21

-------
                                                                 bs.
                                                                 cs
Figure 7-7.   A diagrammatic illustration of the resistance model
             frequently used to help formulate the roles of processes
             like those given in Figure 7-1.   Here, ra is an aerodynamic
             resistance controlled by turbulence and strongly affected by
             atmospheric stability, r^f and rbs represent surface
             boundary layer resistances that are determined by molecular
             diffusivity and surface roughness, and rcf and rcs are the
             net residual  resistances required to quantify the overall
             deposition process, to the eventual sink.  The subscripts f
             and s are intended to indicate pathways to foliage and to
             soil respectively.  There are many other pathways that might
             be important; the diagram is not intended to be more than a
             simple visualization of some of the important factors.
                                  7-22

-------
    TABLE 7-3.  ESTIMATES OF ROUGHNESS CHARACTERISTICS TYPICAL OF NATURAL
          SURFACES.  VALUES OF THE FRICTION COEFFICIENT Cfn (= U*/u)
           ARE EVALUATED FOR NEUTRAL CONDITIONS, AT A HEIGHT 50 CM
                   ABOVE THE SURFACE OR TOP OF THE CANOPY
                          Approx. canopy     Roughness       Neutral  friction
    Surface                 height (m)       length (cm)      coefficient,  Cfn


Smooth ice                     0                0.003              0.042
Ocean                          0                0.005              0.045
Sandy Desert                   0                0.03               0.055
Tilled Soil                    0                0.10               0.066
Thin Grass                     0.1              0.70               0.095
Tall thin grass                0.5              5.                 0.16
Tall   thick    grass          0.5             10.                 0.21
Shrubs                         1.5             20.                 0.25
Corn                           2.3             30.                 0.29
Forest                        10.              50.                 0.23
Forest                        20.             100.                 0.24
                                   7-23

-------
The  surface  boundary  resistance,   r^   (separated   further   in  Figure  7-7
between components  r^f and rDS,  associated  with foliage  and soil,  respec-
tively), is that which accounts for the difference between momentum  transfer
(i.e.,  frictional drag)  at the surface  and  the passage  of  some  particular
pollutant  through   the  near-surface quasi-laminar  layer.   In  agricultural
meteorology  literature,  a  quantity  B*l   is  frequently  employed  for  this
purpose (Brutsaert  1975a).  The relationship between these quantities  can  be
clarified by relating both to the micrometeorological concept of a  roughness
length, z0  (the height  of apparent origin of  the  neutral logarithmic  wind
profile).   Then the total  atmospheric resistance,  R, between the surface  in
question and the height of measurement z can be  written  as

     R = (ku*)-l(in(z/zoc)  -*c)

       = (ku*)-lUn(z/z0) + £n(z0/zoc) -wc

       = ra + (ku*)-1 • £n(z0/zoc)                                      [7-3]

where  zoc  is a  roughness  length  scale appropriate  for the  transfer  of the
pollutant.  The residual  boundary-layer  resistance,  rb = R -  ra,  is then

     rb = (ku*)-l •  £n(z0/zoc),                                         [7-4]

which alternatively is written as

     rb=(u*B)-1.                                                      C7-5]

B  is,  therefore, a measure  of the  non-dimensionalized limiting  deposition
velocity for concentrations measured sufficiently close  to a  receptor surface
such that the resistance to momentum transfer  can be disregarded.

It should be noted  that  some  workers  refer to rb as the  aerodynamic resist-
ance and use the symbol ra for it, (e.g., O'Dell et al.  1977).

Shepherd (1974) recommends  using  a  constant  value kB"l =  £n(z0/zoc) = 2.0
for  transfer to vegetation,  on  the basis of  results  obtained over  rough,
vegetated  surfaces.  However, the  role of the  Schmidt number  in  accounting
for  diffusion  near  a  surface needs to  be taken into  account.    Wesely and
Hicks  (1977) advocate  using a Schmidt number  relationship like that of Equa-
tion 7-1, so that surface boundary layer resistance would then be written as


          rb = 5 Sc2/3/u*.                                              [7-6]

Equation 7-6 implies a value  of 0.2  for A in  the boundary layer relationship
given  by Equation 7-1, as  was mentioned earlier.

The  final  resistances in  the conceptual  chain  of  processes  represented dia-
gramatically by  Figure 7-7 are those  which permit  material  to be transferred
to the surface  itself.   For many  pollutants, it  is necessary only to consider
the  canopy foliage resistance,  rcf,  but  for  some  it  is also  necessary  to
consider  uptake at the  ground  by invoking a resistance  to  transfer to soil


                                     7-24

-------
(or  a  forest floor),  rcs.   In  concept,  it is  also  appropriate to  differ-
entiate  between  boundary  layer  resistances  rsf  and rps  for  transfer to
foliage  and  soil, respectively,  as is  shown  in  the diagram.   Many other
resistances can  be  identified  and  might often  need  to be  considered,  but
further  complication  of  Figure 7-7  is  unnecessary.   Its main  purpose is
illustrative.

Transfer of many trace gases to  foliage  occurs  by way  of stomatal  uptake,
which, because of stomatal physiology, imposes a strong diurnal cycle on the
overall  deposition  behavior.   Following initial  work by  Spedding  (1969),
studies  of foliar uptake  of sulfur  dioxide  have  repeatedly  confirmed the
controlling  role of  stomatal   resistance.    Chamberlain  (1980)  summarizes
results  of experiments by Belot  (1975)  and Garland and Branson (1977), who
compared surface conductances of  sulfur  dioxide  with those for water vapor,
over  a  broad  range  of stomatal  openings (which largely  govern   stomatal
resistance).    The conclusion  that  stomatal  resistance  is  the controlling
factor  when  stomata  are  open  appears to be well  founded.   However,   once
again, it is  necessary to apply  corrections  to account for  the  diffusivity of
the  trace  gas  in question;  the  higher the molecular  diffusivity of  the  gas,
the lower the stomatal resistance.

Fowler and Unsworth (1979) point  out  that S02  deposition to wheat continues
even when  stomata are  closed, at  a  rate  suggesting  significant deposition at
the leaf cuticle.  Thus, it  is  not  always sufficient to compute the  canopy-
foliage  resistance  rcf  on   the  assumption  that S02  uptake is via  stomata
alone (although this may indeed  be a sufficient approximation in most circum-
stances).  Instead,  it is  more  realistic to estimate  rcf  from its component
parts via
     rCf = (rst-l + rcut-M-Vd-AI)                                      [7-7]
(following  Chamberlain  1980),  where  rst  is  the  stomatal  resistance,  and
rcut  1S  tne cuticular resistance.  LAI  is the leaf  area  index (total area
of foliage per unit horizontal  surface area).   Note  that in most literature
the LAI  is assumed to be  the  single- sided  leaf area  index.  However, some-
times both sides of the leaves are counted.

The resistance analogy permits a closer look at the mechanisms  that  transfer
gaseous material into leaves.   Figure 7-8 illustrates the  pathways involved:
via stomatal openings and  into  the  interior of the leaf (involving  stomatal
and  mesophyllic  resistances,  rs^  and  rm)  or through  the  epidermis (in-
volving a cuticular resistance,  rcut)-

The resistance model  is somewhat limited by  the manner in which  it structures
the chain  of relevant processes, each being  represented  by a  resistance to
transfer that occupies a  prescribed location in a conceptual  network.  The
structure  of this network  is  sometimes  not clear;   furthermore,  there  are
important processes that do  not conveniently  fit  into the  resistance  model.
Mean  drift velocities (e.g.,  gravitational  settling of  particles)  are not
easily accommodated  in the  simple  resistance  picture,  and  it is  doubtful
whether some of  the  biological  factors  are relevant to the question of par-
ticle transfer.   Studies  of  leaves show  that stomata are typically slits


                                     7-25

-------
                                                                    EPIDERMIS
                                                                    SPONGY,
                                                                    MESOPHYLLIC
                                                                    CELLS
                                                                    PALISADE
                                                                    CELLS
     CX
                                            GUARD  CELLS
Figure 7-8.   An illustration of the roles of different resistances
             associated with trace gas uptake by a leaf.   Material is
             transferred along several possible pathways, of which two
             are shown.  These involve cuticular uptake via a resistance
rcut'
sub
                   anc* transfer through stomatal  pores (via rs^)  into
                stomatal  cavities,  with subsequent transfer to mesophyllic
             tissue (via  r,J.   The  way in which the various resistances
             are combined to provide the best visualization of the overall
             transfer process  in not clear-cut.
                                 7-26

-------
of the order  of  2  to 20 ym long.   For stomatal  uptake of particles to be a
controlling factor of deposition, we would need to hypothesize  spectacularly
good aim by the particles.

7.3  METHODS FOR STUDYING DRY  DEPOSITION

7.3.1  Direct Measurement

There Is little question that  the deposition  of large  particles  Is  accurately
measured by collection devices exposed  carefully above a surface of Interest.
Deposit  gauges and  dust buckets  have been  important weapons in the geo-
chemical  armory for  a  long  time.  They are  intended  to measure the  rate of
deposition  of particles  which  are  sufficiently  large  that  deposition  is
controlled by  gravity.   In  studies of  radioactive  fallout conducted in the
1950's and 1960's, these same devices were used.   In  the case of debris from
weapons  tests, the  major  local  fallout  was  of  so-called  hot radioactive
particles, originating  with the fragmentation of the weapon  casing  and its
supporting structures,  and  the  suspension of soil in the vicinity  of the
explosion.  These large particles fall  over  an area of rather limited extent
downwind of the explosion.   This area  of greatest  fallout was the major focus
of the work  on fallout dry  deposition.  It was largely in this context that
dustfall  buckets were  used to  obtain  an estimate  of how much radioactive
deposition occurred.   It was  recognized  that collection  vessels  failed to
reproduce the  microscale  roughness  features  of natural surfaces.   However,
this was  not seen  as a major  problem,  since the  emphasis  was  on  evaluating
the maximum rate of deposition that was likely to occur so that upper limits
could be  placed on  the extent  of  possible  hazards.   Nevertheless, efforts
were made to  "calibrate"  collection  vessels  in terms  of  fluxes to specific
types of vegetation,  soils,  etc. (Hardy and Harley 1958).

Much further downwind, most  of the deposition was  shown to  be associated with
precipitation, since the  effective  source of  the  radioactive  fallout being
deposited was  typically  in  the  upper  troposphere  or  the lower  stratosphere.
The acknowledged inadequacies of collection  buckets for  dry deposition were
then of only  little  concern,  since dry fallout composed  a small fraction of
the total  surface flux.

In the context of present concerns about acidic deposition, we must worry not
only about  large,  gravitatlonally-settling   particles, but also about small
"accumulation-size-range" particles that are  formed in the air from gaseous
precursors, and  about  trace gases themselves.  All  of these materials con-
tribute to the net flux of acidic and  acidifying substances by dry  processes.
It is known  that collection vessels do indeed provide a measure of the flux
of large  particles.    However,  accumulation-size-range particles,  typically
less  than  1  ym diameter,  do  not  deposit   by gravitational  settling  at a
significant  rate.    These  small  particles  are  transported by   turbulence
through  the   lower  atmosphere  and  are  deposited by diffusion  to  surface
roughness elements,  with  the  assistance of  a  wide  range  of surface-related
effects (e.g., phoretic processes, Stefan flow, etc.), many of  which will be
influenced by the detailed structure of the surface  involved.
                                     7-27

-------
Early work  on  the deposition of  radioactive  fallout made use of collection
vessels and surrogate surface techniques that  were  frequently  "calibrated"  in
terms of fluxes to specific types of vegetation,  soils,  etc.   Studies  of  this
kind were relatively easy, especially in the  case  of  radioactive pollutants,
because  very  small  quantities of many  important  species  could  be measured
accurately by straightforward techniques.  Most of the  radioactive materials
that  were  of   interest  do  not   exist  in  nature,  so  experimental   studies
benefited from  a zero  background against  which  to  compare  observed data.
Moreover,  major  emphasis  was  on  the   dose  of  radioactivity  to  specific
receptors, a quantity  strongly  influenced by contributions  of large, "hot"
particles in situations  of practical  interest.   Such circumstances included
deposition  of  bomb  debris,  fission products, and soil  particles  from the
radioactive cloud downwind of nuclear  explosions.    In  such cases,  highest
doses were incurred  near the source,  and were  due to  these  larger particles.

The applicability of collection vessels and surrogate surfaces in studies  of
the dry deposition of acidic pollutants is  in dispute (see also Chapter  A-8,
Section 8.2).   Principal  among  the conceptual  difficulties concerning their
use  is  their inability  to  reproduce  the  detailed  physical,  chemical, and
biological  characteristics of natural surfaces, which are known  to control,
or at least strongly influence,  pollutant uptake  in most instances.  Further-
more, the continued  exposure of  already-deposited materials to  airborne trace
gases and aerosol particles  undoubtedly causes some changes to occur, but  of
unpredictable magnitude and  unknown  significance.   A recent  intercomparison
between  different kinds  of surrogate  surfaces  and  collection  vessels has
indicated that  fluxes derived  from exposing dry  buckets are  greater  than
those obtained using small  dishes, which in  turn exceed  values  obtained using
rimless  flat plates  (Dolske and  Gatz  1982).   This  provides  a  tantalizing
tidbit of evidence  for an ordering of  performance characteristics according
to the total exposed  surface area per  unit horizontal  projection.    In  this
context,  the  similarity  with arguments  concerning  leaf area  index seems
especially  attractive.   Micrometeorological  data  obtained  during  the  same
experiment fall  between the  extremes represented by the  buckets and the  flat
plates.

Dasch (1982) reports on a comparison  between many different configurations  of
flat-plate collection surfaces,  pans, and buckets.  The  results indicate  that
glass surfaces  provide  the greatest flux estimates  for almost all  chemical
species considered,  and  teflon the  lowest.  Plastic bucket data generally
fall  midway in  the range.

Tracer techniques developed  in the radioecology era for  investigating fluxes
to natural  surfaces offer  some   promise.   A 3-emitting  isotope  of  sulfur,
S-35,  lends itself  to  use  in   studies  of  S02  uptake  by  crops   because
measurements of low  rates  of sulfur accumulation  are then possible.  Garland
et al.  (1973),  Owers  and Powell  (1974),   Garland  and   Branson  (1977),  and
Garland  (1977)  report the  results of   a  number  of studies  of 3^S02 uptake
by various vegetated surfaces ranging from pasture  to  pine  plantation, and  by
non-vegetated surfaces such as water.

In concept, it  is feasible to extend studies of this kind to  the deposition
of sulfurous particles,  but as  yet no  such  experiment  has  been reported.


                                     7-28

-------
However,  analogous  studies  of  particle  deposition  using  non-radioactive
aerosol  tracers have been  carried  out.   In wind-tunnel  experiments,  Wedding
et  al.   (1975)  employed  uranine  dye  particles  in  conjunction  with  lead
chloride particles  to  study the  influence of  leaf  microscale roughness  on
particle capture  characteristics;  uranine particles  are relatively  easy  to
measure by  fluorimetry, whereas measurements  of lead deposition require  far
more painstaking chemical  analysis  of the deposition surface.  The  particle
sizes used by Wedding et al.  were  in the range 3 to 7  ym diameter.

Considerably larger particles have  been used  in many studies.  In  detailed
wind-tunnel  studies,  Chamberlain   (1967)  used  lycopodium  spores  (-30 ym
aerodynamic diameter).   Workers  at  Brookhaven  National Laboratory  extended
these  wind-tunnel   techniques  to   real-world  circumstances by conducting  a
series of experiments employing  pollen in the  same general  size range (Raynor
et al. 1970, 1972a,b, 1974).

In  general, these  methods of tracer  measurement  have  not been  applied  to
natural  circumstances for the particle sizes of major interest in the present
context  of acidic deposition.   An  important  exception concerns  studies  of
deposition on snow surfaces.   The retention of deposited material at the  top
of  or within  a  snowpack has been studied in  some detail and  continues  to  be
an  intriguing area  of  research.   Particulate  materials such as sulfate  were
considered by Dovland  and  Eliassen  (1976), who  studied  the  accumulation  upon
snow surfaces during periods of  no precipitation and  found  average deposition
velocities  in the  range 0.1  to  0.7  cm  s-1, depending on the  assumption  made
regarding  the contribution  by  gaseous S02  deposition.   Similar  work  by
Barrie  and  Walmsley (1978) yielded  average  sulfur  dioxide deposition  velo-
cities  to   snow  in the  range 0.3  to  0.4  cm  s'1,   with   a  standard  error
equivalent to about a factor of  two.

Eaton et al.  (1978)  and Dillon  et al.  (1982)  present examples of the use  of
calibrated  watersheds  to  estimate  atmospheric deposition.   Dry deposition
fluxes are  estimated as a  residual  between measured  fluxes out of  a concep-
tually-closed  system,   assumed  to  be  in steady  state,  and  measured  wet
deposition  into  it.    Considerable  effort is  required to document  annual
chemical mass balances  for specific watersheds.  Once the effort is  made,  it
appears  possible  to   draw conclusions  regarding  dry  deposition,   although
obviously such estimates will be the result of  the difference  between fairly
large numbers.   According to Eaton et  al.,  the  annual dry deposition  flux
estimate obtained  at the  Hubbard  Brook Experimental  Forest in New  Hampshire
is  accurate to  about  +_ 35 percent  (one standard  error).   The data do  not
permit apportionment between gaseous and particulate sulfur inputs,  but  the
total  sulfur  flux  corresponded  to  a  deposition velocity of  about 0.6  cm
s"1.

7.3.2  Wind-Tunnel and  Chamber Studies

Figure 7-1  illustrates  the overall  complexity of the problem of dry deposi-
tion.   While it  is indisputable  that  no  indoor  experiment  can  provide  a
comprehensive evaluation of pollutant deposition that would be applicable  to
the natural countryside, laboratory  studies provide  the unique  attraction  of
controllable conditions.   It  is  feasible to  compare  the relative  importance


                                     7-29

-------
of various factors, as in Figure 7-1,  and especially  as  in  Figure 7-8, and to
formulate these processes in a logical manner.  In this general category, we
must include  the  extensive wind-tunnel  work  referred to  earlier,  the pipe-
flow and flat-plate studies conducted  in  experiments  more aligned to problems
of chemical  engineering, and the chamber experiments  favored by ecologists
and plant  physiologists.   Distinction among  these  kinds  of experiments is
often difficult.  Many exposure chambers  and  pipe-flow  studies have features
of wind tunnels.

The utility of chamber studies  is well illustrated by  the  series of results
reported by  Hill  (1971).   By comparing   the  rates of  deposition of various
trace  gases  to oat  and alfalfa  canopies exposed in   large  chambers,   Hill
concluded  that solubility  was  a critical  parameter in  determining uptake
rates of trace  gases  by vegetation.  The ordering  of  deposition velocities
was: hydrogen fluoride > sulfur dioxide > chlorine >  nitrogen dioxide > ozone
> carbon dioxide > nitric oxide > carbon  monoxide.   Furthermore, the chamber
studies indicated a wind speed  dependence of  the kind predicted by turbulent
transfer theory,  and demonstrated  a  physiological  effect  of  chlorine  and
ozone uptake on stomatal opening:   exposure to high  concentrations of either
quantity caused  partial  stomatal closure,  thus  limiting  the  fluxes  of all
trace gases that are stomatally controlled.

Experiments  conducted by  Judeikis  and  Wren  (1977,  1978)  yielded valuable
information on  the deposition of hydrogen sulfide,  dimethyl  sulfide, sulfur
dioxide, nitric oxide, and  nitrogen dioxide to non-vegetated surfaces (Table
7-4).  The values listed were derived from  initial deposition  rates obtained
before  surface  accumulation limited  uptake rates.   For comparison, surface
resistances derived from Hill's  (1971) studies of  trace  gas uptake by alfalfa
are also  listed.   On the  whole,  the  ordering of deposition velocities  sug-
gested by  Hill's  work appears to be  supported,  providing  some justification
for  extending  the   ordering  to  CO,   H2S,   and   ((^3)2$  in  the  manner
indicated  in  the  table.    Residual  surface  resistance  to  uptake  of soluble
gases  by  solid,  dry  surfaces appears to be  substantially greater than for
vegetation, which is as would be expected.

The values listed  in  Table  7-4  represent resistances to transport very  near
the  surface,  much  like  the  surface  boundary-layer  resistance  discussed
earlier to which  other  resistances must be  added   to  obtain  values repre-
sentative  of natural, out-door  conditions.  The reciprocals  of the tabulated
numbers provide upper limits of the  appropriate deposition  velocities.

Similarly, informative data  have  been obtained  about particle deposition on
surfaces that can  be  contained  in wind tunnels.   Studies of  this kind are an
obvious extension of  pipe-flow  investigations by workers such  as Friedlander
and Johnstone (1957) and Liu and Agarwal  (1974), which provide strong support
for theories  involving  particle  inertia  and  Schmidt number scaling.    Wind
tunnels provide a means  to extend  chamber and  pipe-flow investigations to
situations more closely approximating  natural  conditions.

Results obtained  in studies  of  particle  deposition  to dry gravel (Sehmel et
al.  1973a) are shown in  Figure 7-9.   Experiments on  the  deposition to wet
gravel  were  also  conducted.  These indicated deposition  velocities some 30


                                     7-30

-------
      TABLE 7-4. RESISTANCES TO DEPOSITION (S  CNT1)  OF  SELECTED  TRACE
      GASES, MEASURED FOR SOLID SURFACES IN A  CYLINDRICAL  FLOW REACTOR
      (JUDEIKIS AND STEWART 1976) AND FOR ALFALFA IN A  GROWTH CHAMBER
                                (HILL 1971)a
                                      Substrate Surface
Pollutant             Adobe Clay            Sandy  Loam              Alfalfa
CO
HoS
(CH3)?S
NO J *
C0?
Oo
NDo
C1J
SO?
HF

62.0
3.6
7.7
-
-
1.3
_
1.1
*™

67.0
16.0
5.3
_
-
1.7
_
1.7
—
oo
io"o
3.3
0.7
0.5
0.5
0.4
0.3
aSolid-surface data  are  derived from  Table  2 of  Judeikis  and Wren  (1978).
 The alfalfa values are obtained from  Table 1  of  Hill  (1971).
                                    7-31

-------
                        OBTAINED  AT ABOUT 2.4 m  s
                     O OBTAINED  AT ABOUT 16  m s
       0.01
0.1              1.0

   PARTICLE  DIAMETER (ym)
Figure 7-9.   Results  of wind-tunnel  studies  of particle  deposition  to
             1.6 cm diameter dry gravel.   Adapted  from Sehmel  et  al.
             (1973a).
                                 7-32

-------
percent less than the values evident  in Figure  7-9  (for  particles  in  the  0.2
to 1.0 urn size range), as  might be expected  from  considerations  of  Stefan
flow and  diffusiophoresis.  When  surface  roughness  was  increased,  deposition
velocities also increased.  The  wind  speed effect  evident  in these data  is
fairly typical  and applies also in the case of vegetation (Figure 7-10).

Chamberlain  (1967)  extended his  earlier  (1966) wind tunnel  studies  of  gas
transfer  to  "grass  and  grass-like surfaces" by considering particle  deposi-
tion to rough  surfaces.   Sehmel  (1970) conducted  similar wind-tunnel  experi-
ments,  employing  monodisperse  particles  ranging  from   about  0.5  to  20 y m
diameter.  Figure 7-10 combines results from Chamberlain  (1967)  and Sehmel  et
al. (1973b).  The Chamberlain  data refer  to live grass, but  the  Sehmel  et  al.
data were obtained  using 0.7  cm high artificial grass.   Moreover, the  two
sets of data were obtained at  different wind speeds   (Chamberlain,  u*  = 70 cm
s  ,  Sehmel  et  al.,  u*  -   19  cm   s"^).    Further   tests  conducted   by
Chamberlain  (1967)  indicated   that deposition  velocities  to  natural   grass
exceeded  those to  artificial   grass by a factor  of about two  for particles
smaller than about  5 ym.  This appears contrary to the  indication of  Figure
7-10,  where VA  (natural)  of   Chamberlain  is seen  to be about half  the  v
-------
*-J

co
                             UD
                             c
                             -~j
                              i
ID ~s n>
3" 01 t/>
3 i/> c
n> (f> —'

      1/1
ID Q)
e-t-  O
      -h
OJ -S
—• ID £
.  -a -••
   O 3
—--s a.
i—• d- i
vo n> r+
^J Q. C
OJ    3
cr cr 3
^^**< fD

 I  O
   3" t/l
O Qi C+
c 3 c
-5 cr a.

n> -s n>

   cu
   -j. o
                          (7* rl-

                          •—'O

                           I  fO

                          a. a.
                          o n>

                          ISI O
                          0> rt
                          3 -J.
                          Q- O
                             3
                                                                              DEPOSITION VELOCITY (cm  s"1)
m

o
                                       m

-------
      Fc = PW'C'                                                         [7-8]

where  P is  the  air  density  and the  primes  denote  deviations  from mean
values.  The over-bar indicates a time average.   This  is  an extremely  demand-
ing task and constitutes a specialized field of  micrometeorology.   Details  of
experimental procedures  are given,  for  example, by Dyer  and Maher  (1965),
Kaimal (1975),  and Kanemasu et al. (1979).

Figure 7-11 shows examples of sensor output  signals fundamental to the  eddy-
correlation technique.   Fast-response  sensors of any pollutant concentration
can  be  used;  the  trace  shown for  C02 in  the  diagram  is  an  interesting
example  of considerable  agricultural  relevance.   As a  basic requirement,
sensors suitable  for eddy-correlation applications should have response  times
shorter than one  second  for operation at convenient heights  on towers.  For
application aboard aircraft (Bean et al. 1972,  Lenschow  et al.  1980)  consid-
erably faster response is required.

Eddy-correlation  methods  have  been used in  field experiments  addressing the
fluxes of  ozone  (Eastman and  Stedman  1977), sulfur (Galbally et al.  1979,
Hicks and  Wesely  1980),  nitric oxides (Wesely et al.  1982b),  carbon  dioxide
(Desjardins  and  Lemon  1974,  Jones  and  Smith   1977),   and  small particles
(Wesely et al.  1977).

Rates of  transfer through  the  lower atmosphere  are  governed by  turbulence
generated by both mechanical mixing  and  convection.   In this  context,  three
atmospheric  quantities  cannot be  separated:  the vertical  flux of material,
the  local   concentration  gradient  (aC/9z),  and  its  corresponding  eddy
diffusivity (K).    Knowledge of any  two  of these quantities will  permit the
third to be evaluated.   Often,  when  sensors suitable for  direct  measurement
of pollutant fluxes  are not available,  assumptions  regarding the eddy dif-
fusivity are made to provide a method for estimating fluxes from measurements
of vertical concentration gradients:

     Fc = pK(3C/8z).                                                     [7-9]

Hicks and Wesely (1978)  and Droppo (1980)  have  summarized  a number of criti-
cal  considerations.   In  particular, with  a typical   value of u* =  40   cm
s"1  and  neutral   stability, the  concentration   difference between adjacent
levels differing  in height by a factor of two is about 9 percent,  for a 1  cm
s"1  deposition  velocity  (VH).    In  unstable (daytime)  conditions,   smaller
gradients would  be expected!  for the  same  v^;   in stable conditions,  they
would be greater.

The  demands  for  high resolution  by  the concentration measurement technique
are obvious.  Nevertheless,  a  substantial  quantity of excellent  information
has  been obtained,  especially concerning fluxes  of  SO?  (Whelpdale and Shaw
1974, Garland 1977, Fowler 1978).

It should  be  emphasized that  the  stringent  site  uniformity requirements
mentioned above for the case of eddy-correlation approaches are also relevant
for  gradient studies.    Detecting   a  statistically  significant  difference


                                     7-35

-------
                         330



              C02  (ppm)   319



                         308
                         0.8i
OJ
en
                        27.On
              T (°C)    25.5-
                        24.0J
                                         12:35
                                                12:36

                                            TIME (hrrrnin)
12:37
   Figure 7-11.
An example of atmospheric turbulence near the  surface.   These  traces  of C02  concentration,
vertical velocity (w), wind speed  (u),  and temperature  (T)  were  obtained over  a  corn
canopy by workers at Cornell  University at a  few meters  above  the  surface.

-------
between  concentrations  at  two heights  is  not necessarily  evidence  of a
vertical  flux and can only  be  interpreted  as such  after extremely demanding
siting criteria have been  satisfied.

Gradients of  particle concentration present special  problems  because it is
often not possible  to derive  internally-consistent results from alternative
measurements.   Droppo (1980)  concludes  that "(t)he  particulate  source  and
sink processes over  natural  surfaces  cannot be considered  as  a simple uni-
directional  single-rate flux."  Thus,  the  proper interpretation of gradient
data in terms of fluxes might not  be possible for airborne  particles,  even in
the  best of  siting circumstances,  because  of  the  role  of the  surface in
emitting and resuspending particles.  In this case, eddy-correlation  methods
will still provide an accurate determination of  the flux through a particular
level, but this flux will  be made  up of a downward flux of  airborne material
and an upward  flux  of similar  material of  surface  origin.   Disentangling  the
two is likely to present  a considerable  problem.

None of the various micrometeorological methods  has yet been developed to  the
extent necessary for routine application.   Rather, they are research  methods
that can  be  used in  specific  circumstances,  requiring  considerable  experi-
mental  care,  the  use  of  sensitive equipment,  and  fairly complicated data
analysis.  They  are more  suitable for investigating the processes that con-
trol dry deposition than  for monitoring  the flux itself.

Nevertheless, some new techniques  for  dry deposition measurement are present-
ly under development.  A "modified Bowen ratio" method is  being developed in
the  hope that it might permit an accurate determination of vertical  fluxes
without  the  need for very rapid response or great resolution   (Hicks  et  al.
1981).   High-frequency variance methods  are also being advocated but have  yet
to be  fully  investigated;  for  these,  sensors having very rapid response  are
required.   An eddy-accumulation method  that bypasses the need  for rapid  re-
sponse of the pollutant sensor is  of long-standing  interest (e.g., Desjardins
1977)  but has yet to be applied  to the  pollutant  flux problem  with signifi-
cant success.

7.4  FIELD INVESTIGATIONS OF DRY DEPOSITION

7.4.1  Gaseous Pollutants

Table 7-5 summarizes a number  of  recent  field experiments  on trace gas depo-
sition  to natural  surfaces.  The listing  is drawn  from a  variety of  sources
(especially  Sehmel  1979,  1980a; Garland 1979;  and Chamberlain 1980);  it is
not  meant to be exhaustive, but  is intended to  demonstrate that many of  the
available  data on surface  fluxes of  trace  gases  are biased toward  daytime
conditions,  when "canopy"  resistances  are usually the controlling  factors.
Extrapolation  of  these  deposition  velocities   to nighttime  conditions is
dangerous  on  two  grounds:  first,  because of  the  large  changes that might
accompany stomatal  closure and, second,  because of the much greater  influence
of  aerodynamic resistance in nighttime,  stable conditions.

Figure  7-12  illustrates  the large  diurnal  cycle typical  of the dry  deposi-
tion  rates of most pollutants.  These  observations were  made over   a  pine


                                     7-37

-------
              TABLE 7-5.   RECENT EXPERIENCE ON TRACE GAS DEPOSITION TO NATURAL  SURFACES
Worker
S02
Hill (1971)
Garland et al .
(1973)
Owers and Powell
(1974)
w Shepherd (1974)

Whelpdale and Shaw
(1974)
Garland (1977)

Fowler (1978)

Dannevik et al .
/ -t *» "1 y~ \
Method

35S02 with stable S02 carrier
over alfalfa
3 S02 over pasture
oc
S02 over pasture
S02 gradients over grass

S02 gradients over snow, water, and
grass
S02 gradients, calcareous soils

S02 gradients, over - wheat
- soybean
S02 gradients over wheat


vd
vd
vd
vd

vd
vd

vd
vd
vd
Results and Comments

* 2.3 cm s"1 (daytime)
Implies rc - 0.4 s cm'1
= 1.2 cm s1 (daytime)
rc - 0.6 s cm
= 1.3 cm s" (daytime)
- 1.3 cm s'1 (daytime)
- 0.3 cm s~ (autumn)
- 1 cm s (daytime for
grass, water, and snow)
- 1.2 cm s"1
rc~ 0.01 s cm"1
- 0.4 cm s"1
- 1.3 cm s"1
^ 0.4 cm s'1
Garland and Branson
  (1977)
                           35
S02 over a pine plantation
0.1 - 0.6 cm s"1

-------
                                                 TABLE 7-5.  CONTINUED
            Worker
         Method
     Results and Comments
         Belot  (1975)  (as
            summarized  by
            Chamberlain 1980)
         Galbally  et al.  (1979)
         Dovland and Eliassen
            (1976)
         Barrie and  Walmsley
            (1978)
04
  S02 over a pine plantation

Eddy correlation over pine forest
Accumulation to snow
Accumulation to snow
   < 1 cm s"1

   - 0.2 cm s"1
   - 0.1 cm s"1
   - 0.2 cm s
             -1
I
10
vo
      NO,
         Wesely et  al.  (1982b)
Eddy correlation
  -soybeans
vd ^ 0.6 cm s"1 (daytime)
rc - 1.3 s cm"1 (daytime)
      =  15 s cm'1 (night)
          Galbally  and  Roy
            (1980)
         Wesely et  al.  (1978,
            1982b)
Gradients over wheat
Eddy correlation over a range of
  natural surfaces
/d = 0.7 cm s"
      Implies rc - 1.4 s cm
rc - 0.8 s cm"1 (daytime)
     *  1.8 s cm"1 (night)
                                                                                                        -1

-------
                                     SULFUR DEPOSITION (ug  rri2  s1)
                                                                                   SENSIBLE  HEAT  (W rii2)      FRICTION  VELOCITY  (cm  s'1)
                      fD
                                                                            o
                                                                            en
                                                                                             ro             *»
                                                                                             o             o
                                                                                             o             o
                                                                                                                               en
                                                                                                                               O
o
o
 i
-F*
O
    O rh
    O O
    3 r+
    c* to
    c. sz
    o c.
    3 -5
    (/> *
   in -a rt- < pa
   c o o fD fD
   —" -s    —' o

   c1 ^.    o -s
   -S O T3 -"• CL

   O tn 3 <<
   O    fD    O
   CO    ct- -h
   —' -hT3 3"
   O-    —' -J m
      r+ cu o c

   O fD c-t-ia -h
   rl-    CU 3- C
      m r+    -S
       —i fD
       Cu
       Q.
       fD
"O

 O
 -h

CO
 CU

 fD
 O

 in

 cu

 a.

T3
 CU
 -s
 rt
 _j
 O


 CU

 fD
   —• O    -h
   -h 3 Q. —•
CL C    Cu C
fD -S ^-^ X

fD O. -••
n CU o O tn
rt rt- 7^- -h fD
fD CU tn    3
Q.       cu in
.   _l. QJ ^ ^J.
   33    cr
   CX CL -•• —•
3s-1'    3 fD
ct O S r-h
   CU (D fD 3"
CU rt- VI 3 fD
—> (T) fD in CU
—•    —' -•• r+
  -O << <
r+ fD    (D -h
-'•-51-'    —•
3 -"• vo tn c
fD O CO rt X
tn Q. o C »
>.   tn -— CL
      •  << CU
c+ S.       3
3- 3-    O Q.
fD fD —I -h

CL    fD Q. -S
Cu to    -5 —*•
c+ Cu CL^ O
Cu in cu    <-<•
   fD -S CL -••
-S O X- fD O
fD C fD T3 3
-h tn -s O
fD       in
-S       -••
                      to

                      ^     -
                                   ro
                                           00

-------
plantation in North Carolina,  using eddy  correlation  to measure each quantity
(Hicks and  Wesely 1980).   The  eddy fluxes  of  total sulfur  demonstrate a
diurnal cycle that appears to be as  strong  as  for  the meteorological  proper-
ties, a result which is not surprising when  it is remembered  that many of  the
causative  factors are  common  (e.g.,  vertical  turbulent exchange).   Some
caution must be associated with interpreting the negative (upward) fluxes of
sulfur evident on two periods as  evidence  of emission or resuspension from
the canopy.   Similarly  large  diurnal cycles of  S02  deposition are reported
by  Fowler (1978),  who introduces  the   further  complexity   of  enhanced  SO?
deposition to wheat covered with dewfall.  Using the  notation of Figures  7-7
and 7-8,  Fowler finds typical  daytime values to be

           ra = 0.25 s cm-1
           rjj = 0.25 s cm"1
           rst =  1.0 s cm-1

           rcut = 2.5 s cnr1.

For deposition to dry soil,  Fowler  suggests using  rcs  = 10.0  s  cm-1,  and
rcs = 0 when the  soil is wet.

Aerodynamic  resistance,  ra,  influences   the  deposition  of all non-sediment-
ting pollutants.   It  is not possible  for any  trace gas  to have a  deposition
velocity  greater than  l/ra,  i.e.,  about  4  cm  s~*  in  the daytime condi-
tions  of  Fowler's experiment.   Because of stability effects,  the  maximum
possible  deposition  velocity  at  night would be considerably lower.   Many of
the exceedingly large  deposition velocities  reported in  the open  literature
appear to exceed  the limits  imposed by  our knowledge  of  the aerodynamic
resistance.   Thus,  several  of the results  included  in  the   exhaustive  tabu-
lation presented  by  Sehmel (1980a)  should  be  viewed more as indications of
experimental  error than as determinations of a physical quantity.

Figure 7-13  addresses  the question of the  time  variation of the  deposition
velocity  vj.   Values plotted are  the maximum deposition velocity permitted
by the prevailing aerodynamic  resistance, evaluated directly  from eddy fluxes
of  heat  and  momentum determined  during the pine plantation  experiment of
Figure 7-12.   In daytime, deposition velocities could  be as much as 20 cm
s"1  if   the  surface  resistance  is  zero,   implying  ra   -  0.05   s  cnr1
during midday  periods.    At night, however, v^  can  decrease to 0.1  cm  s'1
on  infrequent occasions but often  is less  than 2.0 cm s"1.   Fowler's recom-
mendations are probably representative of the long-term average.

The importance of diurnal  cycles in pollutant deposition and the close  rela-
tionship  with other meteorological  quantities  is   further  illustrated  by
Figure 7-14,  which  provides  examples of the  trend  from  nighttime,  through
dawn,  and into  the  afternoon  of the  residual  canopy resistance,  rc,  for
ozone and water vapor determined using eddy-correlation  (Wesely et  al. 1978).
These  data  were  obtained over  corn (Zea  mays)   in  July 1976.    The  upper
sequence  shows good matching between  rp  for  ozone  and water vapor, with  the
former exceeding  the latter by a  small   amount, on the  average.   As  the  day
                                     7-41

-------
I
-p»
ro
        to
        o
        0.
              100.0
               10.0 —
               10.0
                1.0
                0.1
                      18

                      _L
                      19

                     J_
 20

J_
 21

_L
                               00
                               00
          00
                                                                                   00
                                                     HOUR  (est.)
    Figure 7-13.
Values of the maximum possible deposition  velocity of trace gases, determined as the

inverse of the aerodynamic  resistance, ra, for the pine plantation experiment of Hicks

and Wesely (1980).

-------
-J

co
          'E
           O
           u
                   0
                                                                           o
                                                                           •
                                           10          12          14

                                                   HOUR  (CST)
                                                              16
18
   Figure 7-14.
Evaluations of the residual  "canopy resistance" rc, to the transfer of ozone and water
vapor, based on eddy fluxes  measured above mature corn in central Illinois on 29 July
1976 (upper sequence) and 30 July 1976 (lower sequence).  Data are from Wesely et al.
(1978).

-------
progresses,  rc  increases  gradually,  presumably as a consequence of  increas-
ing water stress and eventual stomatal closure.  The lower data  sequence  has
two features  of considerable interest.   First, the gradual  initial  decrease
of rc  for Oo corresponded to  a period of evaporation  of dewfall  (note  the
relatively  Tow  value  of  rc for  H20 during  the same period),  suggesting
that the  presence  of liquid  water on the  leaf surfaces might inhibit  ozone
deposition  (much as  might be expected on the  basis of  ozone insolubility in
water).  This would  not  be   the  case  for  S02  deposition   (Fowler 1978).
Second,  the peak in both  evaluations  of rc at about  1000  hr is  associated
with the passage  of clouds, which  caused a  rapid  and  strong  decrease  in
incoming radiation and lasted for about an hour.   The peak is seen  as further
evidence for stomatal control, because some stomatal closure  would  be expect-
ed with reduced insolation.

The preceding discussion  of both S02  and 03 deposition  confirms the gen-
eralization made by Chamberlain (1980)  that the deposition of such  quantities
might  be modeled after the  case of  water vapor  transfer with  considerable
confidence.

Recently, Wesely et  al.  (1982b) have  reported a  field study  in which both
03  and  N02  fluxes  were  measured.    For a  soybean  canopy,  bulk canopy
resistances to ozone uptake  exceeded water vapor  values by  about 0.5 s cm
during daytime, with rc for N02 still  greater  by a  similar amount.

7.4.2  Particulate Pollutants

No technique  for measuring particle  fluxes has been developed to  the extent
necessary to provide universally accepted data. Use of  gradient  methods,  for
example, is limited  by the inability to  resolve concentration  differences  of
the order of  1 percent.    Turbulence  methods require  rapid-response,  yet
sensitive chemical  sensors which are  not often available.   In  both cases,
practical application is hindered by  the need for a  site meeting stringent
micrometeorological  criteria.   Nevertheless,   results  from  several  applica-
tions   of micrometeorological  flux-measuring   methods  have  been  published.
Table  7-6  provides  a  list  that  illustrates  the  narrow  range of available
information.   The  evidence  points  to  a difference between  the  deposition
characteristics  of  small  particles and sulfate;  the  latter  seems to   be
transferred with deposition velocities  somewhat greater  than  the  value of  0.1
cm s-1  that has been  assumed in most assessment  studies,  and greater than
the values  appropriate for small  particles, on the average.   At this  time,
the possibility  that sulfate fluxes are  promoted by the  strong  effect of a
few large particles cannot be dismissed.

As must  be  expected, taller  canopies  are associated  with higher  values  of
Vd, on the  average.   Figure  7-15  shows  how small  particle  fluxes varied
with time of day over a pine  plantation  in North Carolina  during  1977  (Wesely
and Hicks 1979). These eddy-correlation  results display a run-to-run  smooth-
ness that engenders considerable confidence;  moreover,  they are  supported  by
the  finding  that  simultaneous eddy  fluxes  of momentum and  heat   closely
satisfied the usual  surface  roughness  and energy  balance constraints.  There
is little  doubt that the  surface under  scrutiny  (or at least the air  below
the sensor) did indeed represent a source of particles rather than  a  sink  for


                                     7-44

-------
 TABLE 7-6.  FIELD EXPERIMENTAL EVALUATIONS  OF  THE  DEPOSITION VELOCITY
                     OF SUBMICRON DIAMETER PARTICLES
Surface
  Size and Method
  Results and Comments
Snow

 Dovland and Eliassen
   (1976)
 Wesely and Hicks
   (1979)

Open Water
 Si even'ng et al.
   (1979)
 Williams et al.
   (1978)

Bare Soil
 Wesely and Hicks
   (1979)
Grass
 Sehmel et al
   (1973b)
 Chamberlain (1960)
Lead aerosol, surface
sampling
0.05-0.1 urn parti-
cles eddy correlation
0.2-1.0 ym parti-
cles, gradients
0.05-0.1 ym parti-
cles, eddy
correlation

0.05-0.1 ym parti-
cles, eddy correla-
tion
Polydispersed
rhodamine-B particles
with mass median
diameter 0.7 ym,
deposited to
artificial  grass
exposed outdoors

Radon daughters
deposited to natural
grass.  Work attri-
buted to Megaw and
Chadwick
0.16 cm s-1 in
stable stratification,
greater  values  in  neutral
All light-wind data.
Net fluxes small  but
upwards; v^ too small
be determined.
                      to
Gradients  highly  variable.
Range  of  vj  typically  0.2
-  1.0  cm  s~l  in  magnitude.
Including reversed gradients
in long-term average reduces
average  v^  to  near  zero.
(See Hicks and Williams
1979).

Preliminary indications
only: V(j very small, 95%
certainty < 0.05 cm s"1.

Surface frequently a
source: vj very low on
the average, but often
large for short periods.
Av
 Average vd =^0.2 cm
   -  0.20 cm s'1
                                     7-45

-------
                           TABLE 7-6.   CONTINUED
Surface
  Size and Method
   Results and Comments
 Hudson and Squires
   (1978)
 Davidson and
   Fried!ander (1978)
 Wesely et al.  (1977)
Cloud condensation
nuclei fluxes
measured by gradient
methods over
sagebrush and grass.
Particle size prob-
ably 0.002-0.04 urn

- 0.03 ym particles
gradients over wild
oats

0.05-0.1 ym parti-
cles, eddy correla-
tion
 Everett et al.  (1979) Particulate lead and
                       sulfur,  gradients
v
-------
                             TABLE 7-6.   CONTINUED
Surface
  Size and Method
  Results and Comments
Trees

 Hicks and Wesely
   (1978, 1980)
 Wesely and Hicks
  (1979)
 Lindberg et al
   (1979)
Sulfate particles,
eddy correlation,
Loblolly pine
0.05-0.1 pm parti-
cles, eddy correla-
tion
Strong diurnal  variability
but less marked than for small
particles;  average  v 0.1 cm s'1 for all
tides foliar washing  quantities on the average
 Wesely et al. (1982a) Sulfate particles,
                       eddy-correlation
                       VH not significantly
                       different from zero for a
                       winter deciduous forest
                                     7-47

-------
                     fD
                                                                                         DEPOSITION VELOCITY  (cm  s"1)
 I
-P»
CO
     _.. fl> ,_.
     3 X 10
     Q- rt "^J
     —'• fD VO
     O 3 —•
     O> Q. •
     rt fD
     fD Q.
     Q.    Z
       T3 O
     CT fD rt
     << -5 fD

     rt O rt
     3-0-3-
     fD  fD

     3 O (O
     fD -*» rt
     IQ    -J
     D> fD O
     rt 3 3
  = O  C
Q. 3  -S
fD     3
TJ  -S  Q»
O  0»  —'
to  rl-

f-*- fD "<
-••-SO
O     —'
3  r* fD

<  Q>
fo  3  s:
—i    _j.
O  O. rt-
O  fD  3-
-J.T3
rt O  -h
-J.   <-»-J3
  = -<• C
	O  fD
•   3  3
       rt
          O CT O
          a» "<  fD

          O fD  O

          -j. a. -j.
          3 <<  c+

             o  o
          -J. O  3
          3 -S
             -s  <
             fD  fD

             o*  o
             rt O
             -J. —J.
          '—O  rt
          m 3 <<
          _i.
          O Q)  O
          7^- cr -t,
          I/I O
             <  l/>
          O) fD  3
          23     P*
          Q. CD  —*
                    •a
               (/)  3  Q)
               fD  fD  -5
               — «    rh
              »< T3  -'•
                  — < O
               I— • Q<  — J
               U3 3  fD
               •^ r+  W
               co a>
               s o  o
               fD  3  •
               to     >-•
               fD  -••
               — i 3  T±
              *<     3
                  Z-^
               Ol  O
               3  -5
               Q. rt
                   -
                     D>
                              m
                              co
        fl)    _i.
              O     -S
              7T    fD
              (/>     Q.

-------
substantial  periods (Arnts et al. 1978).  A basic question then arises about
the meaning of the measured deposition rates, since these probably represent
a net result of continuing but varying surface emission and a deposition flux
that is also  varying  with time.   In particular, it  is  not obvious  how to
relate such results to the common situation in which we wish to evaluate the
atmospheric deposition rate of some  particulate pollutant that is not emitted
or resuspended from the surface.

Figure 7-12 identifies periods of the 1977  pine plantation study during which
no gaseous  sulfur  was detectable.   These  occasions  were used by  Hicks and
Wesely (1978) to evaluate  residual  canopy resistances for particulate sulfur
that averaged about 1.5  s cm'1  (with a standard error margin of  about  +_ 15
percent)  for 17 July,  and about 1.1  s cnr*  {+_ 25  percent) for 18 July.

Two  tests  of  sulfate gradient equipment  over  arid  grassland,  reported by
Droppo (1980),  yielded  values  of  0.10  and  0.27  cm  s-1  for  vd,  in  very
light winds  (~ 1  m  s'1).  The  residual  surface resistances evaluated from
his data are  7.7  and  3.3 s cm-1, respectively.  These  values  are consider-
ably higher  than  the  pine plantation results quoted above, but might not be
wholly discordant  when the nature  of the surface  present  in  the gradient
studies is taken into account.

Results of  an extensive series of  eddy-correlation measurements of particu-
late sulfur fluxes to a variety of vegetated  surfaces  have been summarized by
Wesely et al. (1982a).  In daytime conditions,  deposition velocities to grass
range from  about  0.2  to  0.5  cm  s'1.    Values  for  a  deciduous  forest in
winter (few  leaves) are  not significantly different from zero.  In general,
somewhat lower values are appropriate at night.  In  almost  all  of the cases
summarized  by  Wesely  et al.,  normalization of surface transfer conductances
by  u* appears to  reduce  the  residual variance.   Hicks et  al.  (1982)  pre-
sent supporting data  from  another study of the same  series,  also over grass-
land.

Considerable controversy  remains  concerning  the  value of v
-------
these features of the collecting surface are not easily reproduced by common-
ly  used  artificial  collecting  devices.   Monitoring  the  accumulation  of
particles  in  collection  vessels  continues to be a wide-spread  practice  (See
Chapter A-8); however, relating the data obtained to  natural  circumstances  is
difficult  (Hicks et al. 1981).  In a special  category of its  own,  however,  is
the method of foliar washing, as used by  Lindberg et  al.  (1979).   As applied
in careful studies of particle dry  deposition  at the Walker  Branch Watershed
in Tennessee,  this method  of removing and  analyzing material  deposited  on
vegetation  has  succeeded  in demonstrating  long-term average  values of  vd
larger than the usually accepted values for several elements.

7.4.3  Routine Handling in Networks

The discussion given  in  this chapter is  intended to focus on  the  processes
that  cause dry deposition,  and  on methods by  which these processes can  be
investigated.  Discussion of network monitoring of dry deposition  is left for
Chapter A-8.   However,  for the sake of completeness  a  brief summary of  pre-
sent capabilities to monitor dry deposition should be given here.

It is important to recognize  dry  deposition  for what it is:   a highly vari-
able exchange of trace gases  and  aerosols between the atmosphere  and exposed
surfaces.  In some special circumstances,  natural surfaces are  such  that the
accumulation of deposited  material  can be measured  directly, such  as in the
case of some  icefields,  snowpacks,  stone, and  metals.   However,   in  general
there is  no  "monitor"  that will  give a clear-cut measurement of  dry deposi-
tion  rates to natural  surfaces.    Work  on  developing such  a  monitor  must
continue,  but should  be  conducted with the realization that  science has yet
failed to develop such a device for monitoring  the surface fluxes  of meteor-
ological quantities such  as  sensible heat, moisture, and momentum.   Even  in
these cases,  micrometeorological  methods  such  as eddy  correlation  and  gra-
dient interpretation remain  research  tools that are  applied with  great  care
in intensive case studies.  These  field studies are intended  to  formulate the
atmosphere/surface exchange  in  a  manner  that  can then be extended  to other
situations.  Laboratory and modeling studies provide  the  basic  understanding
necessary  for developing  the  techniques  for  interpolating between  infrequent
direct measurements (by any available method)  and for extending  them to other
situations.

It appears unlikely that  collection-vessel or  surrogate-surface methods  will
be capable of providing direct measurements of dry deposition fluxes of trace
gases  and  aerosols   to   natural  surfaces.    Likewise,  micrometeorological
methods seem  unable  to address  the case  of  particles that  fall  under  the
influence  of  gravity, and  a  micrometeorologically-based deposition "monitor"
does not seem an  immediate possibility.   Thus,  any network for evaluating dry
deposition should concentrate on providing data from  which surface  fluxes can
be evaluated, by applying the rapidly expanding  understanding of dry deposi-
tion processes that is  presently  being developed.   The minimum requirements
would be for data on  atmospheric concentrations of the relevant  trace gas and
aerosol  species,  and for sufficient meteorological data to enable  appropriate
deposition velocities  to be calculated for specified  surface  characteristics
and for the species of interest.  Surrogate  surface  devices  might  be used  to
evaluate fluxes of particles falling under the  influence of gravity.


                                     7-50

-------
These matters are discussed  at  greater length in Chapter A-8.  A summary of
methods  for  measuring dry deposition,  with  emphasis on  the  suitability of
various  techniques  as deposition  "monitors"  has been presented  by  Hicks et
al. (1981).

7.5  MICROMETEOROLOGICAL MODELS  OF THE DRY  DEPOSITION PROCESS

7.5.1  Gases

Almost all models of  dry deposition of trace  gases have as their foundation
either the resistance analogy  illustrated in  Figures  7-7 and  7-8  or  some
equivalent to it.   The convenience of this approach is obvious; it permits
separate  processes  to be  formulated  and  combined in   a  manner  that mimics
nature, while providing a clear-cut mechanism  for determining  which processes
can be omitted  from consideration  in  specific circumstances.   The relevance
of the resistance  approach to the  matter  of  particle   deposition is  not so
obvious,  especially when gravitational  settling  must  be  considered.

A  useful   start  is  to  identify  the properties of interest and possible  pro-
cesses that control  the uptake of  various gases:

S02:     Uptake by plants is largely via stomata  during daytime, with about 25
        percent  apparently via  the epidermis  of  leaves  (Fowler  1978). At
        night, stomatal resistance  will  increase substantially,  but cuticu-
        lar resistance should be  unchanged.   When  moisture condenses on the
        depositing   surface,  associated  resistances  to  transfer should be
        allowed  to  decrease  to  near zero  (Murphy  1976,  Fowler  1978).   To a
        water surface, water  vapor appears to provide  an acceptable analogy
        to S02 flux.

03:     Behavior is like S02  but  with significant  cuticular uptake at night
        (rcut  ~  2  to 2.5  s cm~l  at  night;  see  rc   quoted  by Wesely et
        al. 1982b)  and with  surface moisture effectively minimizing uptake.
        Deposition  to water surfaces,  in general, is  very  slow.

N02:     Similar  to  03  in  overall  deposition  characteristics,  but with  a
        significant additional  resistance  (possibly  mesophyllic; see Wesely
        et al.  1982a)  of  about 0.5 s cm"1.   Even  though  N02  is insoluble
        in water in low concentrations (see Chapter A-4),  deposition to water
        surfaces might be quite   efficient.     Chamber  studies  (Table  7-4)
        indicate similar overall surface resistances  for S02 and  N02.

NO:     Typical  canopy resistances  are in  the range 5  to 20  s  cm-1,  as in-
        dicated by  chamber studies  (Table 7-4) and field  experiments (Wesely
        et al.  1982a).    NO  appears   to be   emitted by  surfaces  at  times,
        possibly as  a  consequence of  N02  deposition   and of  the  intimate
        linkage with ozone  concentrations (Galbally and  Roy  1980).

HN03:    Little direct information  is available;  however, on the basis of its
        high   solubility  and  chemical   reactivity,  substantial  similarity to
                                     7-51

-------
        HF  should  be  expected.   Consequently, the use  of rc = 0 appears  to
        be a reasonable first approximation.

NH3:    Again, no  direct  measurements  are available but  in  this case  simi-
        larity with S02  appears likely.   Natural  surfaces  may be emitters
        of NH? because of a  number  of  biological  processes occurring in and
        on soil.

Variations in aerodynamic resistance must be expected to modulate all of the
behavior patterns summarized above.   In many circumstances, deposition  rates
at night will be  nearly zero solely  because atmospheric  stability is so  great
that material cannot be transferred  through the lower atmosphere.  The evalu-
ations given  in  Figure 7-12  are especially  informative, because even over a
pine  forest  whose surface  roughness  operates to  maximize  v
-------
stability conditions  and  for different seasons,  they  produced a map of  S02
deposition velocities  suitable  for use  in numerical models  and for inter-
preting  concentration  data (see  Chapter  A-8).    This  approach has not been
extended to other pollutants.

7.5.2  Particles

Modeling of particle deposition is  complicated  by three major factors:    (1)
gravitational   settling,  which causes  particles to  fall  through the atmos-
pheric turbulence that provides the conceptual basis for conventional micro-
meteorological  models (Yudine  1959);  (2)  particle inertia,  which  permits
particles to  be projected through  the near-surface  laminar layer by turbu-
lence,  but  also  prohibits  particles  from  responding  to the  high-frequency
turbulent motions  that transport material  near receptor  surfaces;  and  (3)
uncertainty regarding  the processes  that control  particle  capture.    These
three factors  are interrelated in  such  a manner  that clearcut  differentiation
of their separate consequences is  not  possible.

The  problem   has  attracted the  attention  of  many  theoreticians,  and many
numerical models have been developed.   Each model represents  a selected com-
bination of processes, chosen for  consideration  on the  basis of the modeler's
understanding  of the problem.  Without adequate consideration of all  of  the
mechanisms involved, none of these models  can  be considered as a  simulator of
natural   behavior.   This   is  not  to question  the worth  of  such  models,  but
rather  to  emphasize that  each  should  be applied with  caution,  and only  to
those situations commensurate with its  own assumptions.

The many numerical models can be classified in  several   different ways.  Some
extend  chemical  engineering  results to surface geometries that  are intended
to represent  plant communities.  Others extend  agrometeorological air-canopy
interaction models  by  including  critical  aspects of aerosol  physics.  Both
approaches have benefits,  and the final  solution  will   probably   include
aspects of each.

An  excellent  review  of  model  assumptions has  been given by Davidson  and
Friedlander (1978).  They trace  the evolution  of models  from the  1957 work  of
Fried!ander and Johnstone (which  concentrated  on  the  mechanism  of inertial
impaction and  assumed that particles shared the  eddy diffusivity  of momentum)
to the canopy filtration models of  SI inn  (1974)  and  Hidy and  Heisler  (1978).
Early work concerned deposition  to flat surfaces and made various assumptions
about  the surface  collection process.    Friedlander   and  Johnstone  (1957)
permitted particles  to be carried  by  turbulence to within one  free-flight
distance  of  the  surface,  upon  which  they were  assumed to  be  impacted  by
inertial penetration of  the  quasi-laminar "viscous" sublayer.   Beal  (1970)
introduced viscous  effects  to limit the  transfer  of small particles, while
retaining inertial  impaction  of larger  particles.   Sehmel (1970)  assumed that
all particles  that contact the surface will  be captured and  used empirical
evidence obtained  in  his wind-tunnel   studies  to determine the  overall  re-
sistance to transfer,  assumed to  apply  at a distance of one  particle radius
from  the  surface.   Sehmel's  work has been  updated recently  to  provide  an
estimate of deposition velocities  to  canopies  of  a range of geometries  in
different meteorological  conditions (Sehmel  1980b).


                                    7-53

-------
The above models  are  based  largely on observations and theory regarding the
deposition of particles to  smooth  surfaces,  usually of pipes.   More micro-
meteorologically-oriented  models   have  been  presented by  workers  such  as
Chamberlain  (1967),  who  extended  the  familiar  meteorological   concepts  of
roughness length and zero plane displacement to the case of particle  fluxes.
Much of this work was considered as an extension of models developed  for the
case  of gaseous  deposition  to  vegetation, which  in  turn were  based on an
extensive background of agricultural  and forest meteorology, especially con-
cerning evapotranspiration.   A recent development  of this genre is  the canopy
model  of  Lewellen and  Sheng (1980),  which uses  recent techniques  in turbu-
lence modeling to reproduce  the main  features of  subcanopy  flow  and combines
these with particle deposition formulations like those represented in Figure
7-4.   Lewellen  and  Sheng  emphasize their model's omission of several poten-
tially  critical  mechanisms,  especially electrical  migration,   coagulation,
evolution of  particle  size  distributions,  diffusiophoresis, and  thermophore-
sis.  To this list we can  add a  number of other  factors about which little is
known  at  this time,  such  as subcanopy chemical  reactions,  interactions with
emissions, and the effect of microscale roughness  elements.

Although outwardly simpler than the case of particle deposition  to a  canopy,
deposition to a water  surface has  given  rise  to  a similar  variety  of  models.
Once again, however, different models focus on different mechanisms.  That of
Sehmel and Sutter (1974) is based on  their wind  tunnel  observations and  lacks
a component that can be identified with wave  effects.   SI inn and  SI inn (1980)
invoke the rapid growth of hygroscopic aerosol particles in very  humid air to
propose rather  rapid  deposition  to open water;  deposition velocities on the
order  of  0.5  cm s-1  appear possible in this case.  On the  other hand,  Hicks
and Williams (1979)  propose negligible  fluxes  unless  the  surface quasi-
laminar layer is interrupted by breaking  waves.   At present, none of  these
models has strong experimental evidence to support it.   However,  experimental
and theoretical  studies are proceeding, and  a  resolution of the  matter can
certainly be expected.

7.6  SUMMARY

All of the many  processes  that  combine to permit  airborne  materials  to be
deposited  at the surface have  aspects  that are  strongly  surface  dependent.
While  broad  generalities  can be made  about  the velocities of deposition of
specific  chemical  species   in  particular  circumstances,   wide   temporal  and
spatial  variabilities  occur in most of  the  controlling  properties.    The
detailed  nature of the vegetation covering  the surface is often  a  critical
consideration.   If depositional  inputs to a special  sensitive  area  need to be
estimated,  then  this  can  only be accomplished if characteristics specific  to
the  vegetation  cover  of the  area  in question  are   adequately taken  into
account.

Recent field studies  investigating  the  fluxes  of small  particles have con-
firmed  wind-tunnel  results  that point to  a  surface limitation.    Studies of
the  rate  of  deposition of  particles to the internal  walls of pipes  and  in-
vestigations  of  fluxes to surfaces more characteristic of  nature,  exposed  in
wind  tunnels, tend to  confirm theoretical expectations that surface uptake is
controlled  by the ability  of particles to  penetrate a quasi-laminar  layer


                                     7-54

-------
adjacent to the  surface  in  question.   The mechanisms that limit the  rate of
transfer of particles involve their finite  mass.   Particles fail to  respond
to  the  high  frequency  turbulent fluctuations  that  cause  transfer  to  take
place in  the  immediate  vicinity of  a  surface.   However,  the  momentum of
particles  also   causes  an  inertia!   deposition phenomenon  that  serves to
enhance  the rate of deposition of particles  in  the 10  to 20 ym size range.

The general features  of  particle deposition to  aerodynamically  smooth  sur-
faces are  fairly  well  understood.   Studies  conducted  so  far  support the
theoretical expectation  that  particles  smaller  than  about 0.1 ym  in diam-
eter will be deposited at a rate  largely determined by Brownian  diffusivity.
In  this  instance,  the limiting  factor  is  the transfer by  Brownian motion
across the quasi-laminar layer referred to above.  On the  other  hand, parti-
ticles larger than about  20 ym  in diameter  are effectively transferred via
gravitational  settling,  at rates  determined  by  the familiar Stokes-Cunningham
formulation.  Particles in the intermediate-size ranges are transferred  very
slowly.   The  minimum value of the  "well"  of the deposition velocity versus
particle size  curve is approximately 0.001 cm s~l.

However, natural  surfaces are rarely aerodynamically smooth.   Wind-tunnel
studies  have shown that the  "well" in  the  deposition velocity curve is filled
in as the surface becomes rougher.  Although studies have been conducted, in
wind  tunnels, of deposition fluxes to  surfaces such  as  gravel, grass, and
foliage, the  situation  involving natural  vegetation  such as corn,  or  even
pasture, remains  uncertain.   It  is well  known that  many plant  species  have
foliage  with  exceedingly  complicated  microscale surface roughness features.
In  particular, leaf  hairs increase  the  rate of particle deposition;  studies
of  other  factors, such  as  electrical  charges associated  with  foliage and
stickiness of the surface, indicate that a natural canopy might  be consider-
ably different from  a  simplified  surface  that  is suitable for investigation
in the laboratory and wind tunnel.

Caution   should  be exercised  in  extending laboratory  studies  using  artifi-
cially-produced   aerosol  particles to  the  situation  of  the deposition of
acidic quantities.  Special  concern  is associated with  the  hygroscopic nature
of many acidic  species.   Their growth  as  they enter into  a  region  of  high
humidity and  their  liquid  nature  when  they  strike  the  surface  are  both
potentially important  factors  that might work to increase  otherwise small
deposition   velocities.    Moreover,  there  is evidence that  acidic  species,
especially   sulfates,  might  be  carried by  larger particles;  the  rates of
deposition  of such  complicated  particle structures  are essentially  unknown.
However, the shape of particles  can have a considerable influence upon their
gravitational  settling speed and  probably on their impaction characteristics
as well.

It is not clear to what extent special considerations appropriate for acidic
species, such as those mentioned above, contribute to  the finding of unex-
pectedly high  deposition velocities for atmospheric  sulfate particles (some-
times exceeding 0.5  cm  s'1),  as  reported in  some  recent  North  American
studies.   European  work  has been  fairly  uniform  in  producing velocities
closer to  0.1 cm s~l, while North  American experience has generated larger
values.
                                     7-55

-------
It Is informative to  consider  the flux of any airborne quantity to the sur-
face underneath in  terms  of an electrical  analog,  the so-called resistance
model developed initially in studies of agrometeorology.  In this model, the
flux of the atmospheric property  in question  is  identified  with the flow of
current in an electrical circuit; individual resistances can then be associ-
ated with readily  identifiable  atmospheric  and  surface  properties.  While the
electrical analogy has obvious  shortcomings,  it permits an easy  visualization
of many  contributing  processes and enables  a comparison of their relative
importance.   Micrometeorological  studies of the  fluxes of atmospheric heat
and  momentum  show  that  the aerodynamic  resistance  to transfer  (i.e., the
resistance to transfer  between  some convenient level  in the air and a  level
immediately above the quasi-laminar layer)  ranges  from between 0.1  s cm"1
in  strongly  unstable, daytime  conditions,  to more  than  10 s  cm'1  in many
nocturnal cases.

There are several  resistance  paths  that permit  gaseous   pollutants  to  be
transferred into  the  interior  of leaves.   An obvious  pathway is directly
through  the  epidermis  of  leaves,  involving  a   cuticular  resistance.   An
alternative route,  known  to be of  significantly  greater  importance in many
cases,  is via  the  pores of leaves,   involving  a stomatal  resi stance that
controls  transfer to  within stomatal  cavities, and a  subsequent mesophyllic
resistance that parameterizes   transfer  from  substomatal  cavitiesto leaf
tissue.   Comparison among  resistances to transfer  for  water  vapor,  ozone,
sulfur  dioxide,  and  gases  that  are  similarly  soluble  and/or  chemically
reactive,  shows that  in general  such  quantities  are transferred  via the
stomatal  route, whenever  stomata  are  open.  Otherwise, cuticular resistance
appears  to play  a  significant  role.    Cuticular  uptake  of   ozone  and  of
quantities like NO  and  N02 appears  to be  quite significant,  whereas for
S02  this pathway  appears to be  less  important.   When leaves  are wet, such
as  after heavy dewfall,  uptake  of sulfur  dioxide  is exceedingly  efficient
until  the pH of  the  surface water becomes  sufficiently  acidic to impose  a
chemical  limit  on  the rate of  absorption  of gaseous  S02-   However, the
insolubility of ozone causes dewfall to inhibit ozone dry  deposition.

The  same conceptual model  can be applied  to  the case of  particle  transfer
with  considerable utility.   While the  roles of factors  such as  stomatal
opening  become less clear when  particles are being considered,  the concept of
a residual surface resistance to particle uptake  appears to  be  rather  useful.
Studies  of the  transfer of sulfate particles to a pine forest have  shown that
this  residual  surface resistance  is  of  the  order of 1  to 2  s cm'1.  It
appears   probable  that  substantially  larger  values   for  residual  surface
resistance will be  appropriate for non-vegetated surfaces,  especially snow,
for  which the  values are  more  likely to  be approximately 15  s  cm"1.  At
this time, an exceedingly limited quantity of field information is  available;
however,  it appears that in North American conditions the  surface  resistance
to uptake of sulfate particles will be in the range 1.5 to 15 s cm"1.

While  sulfate  particles have  received  most  of  the  recent  emphasis,  the
general  question  of acidic deposition requires that  equal  attention  be paid
to  nitrate and ammonium  particles.   There  is no information  regarding  the
deposition  velocities of  these  particles, but likewise  there is  no  strong
indication that they are  different from the case of sulfate.


                                     7-56

-------
Regarding  trace  gas  uptake,  sulfur  dioxide  has  received  the  majority of
recent attention.  Chamber studies and  some  recent field work indicate  that
highly reactive  materials  such as hydrogen  fluoride  (and presumably iodine
vapor, nitric acid vapor, etc.)  are readily taken up by a  vegetative surface,
whereas  a  second  set  of pollutants,  including  S02,  N02,  and  03,   seems
to be  easily transferred via  stomata,  and  a  third category  of  relatively
unreactive trace gases is poorly taken up.

Transfer  to water surfaces  presents special problems,  especially when the
surface concerned is  snow.  As  mentioned above,  surface resistances to parti-
cle uptake  by  snow appear  to be of the order 15 s cm"1.   Soluble  gases  will
be readily  absorbed  by all water surfaces, so equivalence to  transfer of
water vapor might be expected.   An important exception occurs  in the case of
SOp,   in  which case  absorbed  S02  can increase the  acidity of  the surface
moisture  layer  to  the extent  that further  S02  transfer   is cut  off.    Trace
gas transfer to  liquid  water  surfaces is  influenced  by  the  Henry's Law
constant.

Wind-tunnel  studies of particle  transfer  to  water  surfaces all show exceed-
ingly  small  deposition velocities  of  particles  in the  0.1   to  1 urn   size
range.  Several workers have suggested mechanisms by which larger  deposition
velocities might exist in  natural  circumstances;  for  example,  the growth of
hygroscopic particles in  highly-humid, near-surface  air can cause accelerated
deposition of such particles, and breaking waves might provide a route  that
bypasses the otherwise limiting quasi-laminar layer in contact with the  sur-
face.   Once again field observations  are lacking.

While large deposition velocities of  soluble trace  gases to open water  sur-
faces  might appear quite  likely, water bodies are frequently  sufficiently
small  that an air-surface thermal equilibrium cannot be achieved.  Air  blow-
ing from  warm  land across  a  small, cool lake,  for example, will not rapidly
equilibrate with the  smooth, cooler  surface.   Flow will   then  be stable and
largely laminar, with  the  consequence that very small deposition  velocities
will  apply for all  atmospheric  quantities.   In many  circumstances,  especially
in daytime  summer occasions, deposition velocities are likely  to be so  small
as to be  disregarded for all practical  purposes.   On  the other hand, during
winter when the land surface  is frequently  cooler than the water, the result-
ing convective activity over small water bodies will  induce  the air to  come
into  fairly rapid equilibrium  with   the water, and  rather  high  deposition
velocities  (in  agreement  with  the  open  water surface   expectations)   will
probably be attained.

An associated special  case  concerns the effect of dewfall, which can acceler-
ate the net transfer of trace gases and particles in some circumstances. The
velocities of deposition involved are small; however, they permit  an accumu-
lation of  material  at the  surface in  conditions  in which  the atmospheric
considerations are likely to  predict  minimal  rates  of exchange  (i.e., limited
by stability  to an extreme  extent).    When  surface fog   exists,  the highly
humid  conditions will  permit  airborne hygroscopic  particles to nucleate and
grow  rapidly.   This  process provides a mechanism  for  cleansing  the  lower
layers of the atmosphere of most  airborne  acidic  particles.   The small fog
droplets that are formed  around the hygroscopic  acidic nuclei are transferred


                                     7-57

-------
by the  classical  process  of fog interception,  to  foliage and other surface
roughness elements.

Recent workshops  (e.g.,  Hicks  et  al.  1981)  have  concluded  that it  is not
possible to  measure  the dry deposition  of acidic  atmospheric  materials by
using exposed collection vessels  because  they fail  to collect trace gases and
small  particles  in  a  manner  that can  be  related in  a direct  fashion to
natural  circumstances.   However,  surrogate surface  methods  appear  to be
useful in indicating space  and  time  variations of deposition in some cases,
and may  provide  reasonable estimates  of fluxes to individual  leaves  under
some  conditions.    It  is   possible  to  measure the  flux of  some  airborne
quantities by micrometeorological means,  without interfering with the natural
processes involved.  These studies,  and  laboratory and wind-tunnel  investi-
gations, provide evidence  that the controlling properties in the deposition
of many trace  gases and  aerosols are  associated with  surface structure,
rather  than  with  atmospheric properties.   The  exception to  this generali-
zation  is  the nocturnal case,  in  which atmospheric  stability  may  often be
sufficient to  impose a severe  restriction  on   the  rate of delivery  of all
airborne substances to the  surface  below.

7.7  CONCLUSIONS

The conclusions presented above can be  summarized as follows:

    °   Dry  deposition  of  small  aerosol  particles and  trace  gases  is a
        consequence  of  many  atmospheric,   surface,  and  pollutant-related
        processes, any  one  of  which  may dominate  under some set of condi-
        tions.  The complexity of each individual process makes it unlikely
        that a comprehensive simulation will  be developed  in  the near future
        (Section 7.2).

    o   The convenient  simplicity  afforded  by  the  concept of  a deposition
        velocity (or its inverse, the total  resistance  to  transfer) makes it
        possible  to  incorporate  dry  deposition processes in  models  in a
        manner adequate  for modeling  and assessment purposes.   The simpli-
        city of the deposition velocity  approach imposes  limitations on  its
        application.   For  example, using  average  deposition velocities is
        inappropriate  when  time-or  space-resolved  details  of deposition
        fluxes are needed (Section  7.2.1).

    0   Sufficient information is  known  about  the  processes  controlling the
        deposition of trace gases that in many instances  deposition veloci-
        ties  can  be  considered to  be known functions of  properties such as
        wind speed, atmospheric stability,  surface  roughness, and biological
        factors such as stomatal  aperture.   Important  exceptions concern the
        case of insoluble  (or poorly soluble)  gases, and  deposition to  non-
        simple surfaces such as forests in  rough terrain (Section 7.2).

    °   The  deposition  of  particles  larger than  about 20 ym  diameter is
        controlled by  gravity  and  can be  evaluated using the  straightfor-
        ward  Stokes-Cunningham relationship.    Smaller  particles  are  also
                                     7-58

-------
influenced by gravity,  and many will  contribute to the deposition of
acidic and acidifying substances (Sections  7.2.2 and 7.2.3).

The deposition of small  particles  remains  an  issue  of considerable
disagreement.  On  the  whole, model  predictions agree  with the re-
sults  of  laboratory and wind-tunnel  studies,  at  least  for test
surfaces that  are  usually smoother than pasture,  but field experi-
ments provide data  that indicate greater deposition velocities.  The
reasons for  the  apparent disagreement  are  not  yet  clear  (Sections
7.3, 7.4.2, and 7.5.2).

Over water surfaces, there are  almost  no field  data  on the deposi-
tion of small  particles.  Different models have  been put forward,
predicting a  wide  range  of  deposition velocities.    At  this time,
there is little evidence  that would permit us to choose among them.
The  situation  for  trace  gases  like  sulfur dioxide and  ammonia is
much better.   On the  whole, models  agree  with the available field
data, although there is disagreement  among  the models on how factors
such as molecular diffusivity  should be  handled (Sections 7.2.7 and
7.5.2).

Dry deposition to the surfaces  of materials used in the construction
of buildings, monuments,  etc., can be measured in many instances by
taking  sequential  samples of  the  surface over  extended periods.
However, many  of the  drawbacks of  surrogate-surface  sampling are
also of concern here (Section 7.2.8).

Particulate material  at the surface can creep, bounce, and  eventual-
ly resuspend under the influence of wind gusts.  The large  particles
entrained in this way  can cause a  local modification  of the  acidic
deposition  phenomenon   that  is  associated with  accumulation-size
aerosol particles and  trace  gases  of  more  distant  origin (Section
7.2.10).

For both case-study  measurement  purposes and for long-term monitor-
ing, accurate measurements of pollutant air concentrations are nec-
essary.  For monitoring  purposes,  measurement of airborne  pollutant
concentrations in a  manner carefully designed  to  permit evaluation
of  dry  deposition  rates  by applying time-varying deposition  veloc-
ities specific to the  pollutant  and  site in question  appears to be
the most attractive option (Section 7.3).

Micrometeorological  methods for measuring dry deposition fluxes have
been developed from  the  techniques conventionally used to  determine
fluxes of  sensible heat,  moisture, and  momentum.  These methods are
technologically demanding, and  their use in routine monitoring ap-
plications is not yet possible (Section 7.3.3).
                             7-59

-------
7.8  REFERENCES

Alexander,  L.  T.   1967.   Does salt water  spray  strontium-90 from the  air?
U.S.A.E.G.  Health  and  Safety  Laboratory  Quarterly  Summary  Report  HASL-181,
1-21 - 1-24.

Arnts, R. R.,  R.  L.  Seila,  R.  L.  Kuntz,  F.  L. Mowry,  K.  R. Knoerr,  and A.  C.
Dudgeon.  1978.  Measurement of a-pinene  fluxes from a loblolly  pine  forest.
Proc. Fourth  Joint Conference on  Sensing  of Environmental  Pollutants.   Am.
Chem. Soc. 829-833.

Bagnold, R. A.  1954.   The  Physics of  Blown Sand  and Desert  Dunes.   Methuen
and Company, London.

Barrie, L.  A.  and J. L. Walmsley.   1978.   A study of sulphur dioxide depo-
sition velocities to  snow in northern Canada.  Atmos.  Environ. 12:2321-2332.

Batchelor,  6.  K.  1967.   An  Introduction  to  Fluid  Mechanics.    Cambridge
University Press, Cambridge, MA.   615 pp.

Beal, S.  K.   1970.  Deposition of  particles  in turbulent  flow on channel  or
pipe walls.  Nucl. Sci. Eng. 40:1-11.

Bean, B. R., R. F. Gilmer, R.  L.  Grossman, and R. McGavin.   1972.   An analy-
sis of airborne   measurements  of  vertical  water  vapor  flux during BOMEX.  J.
Atmos. Sci. 29:860-869.

Belot, Y.  1975.   Etude de la captation des pollutants  atmospheriques  par les
vegetaux.  C.E.A., Fontenay-aux-Roses,  France.

Billings, C. E., and R. A. Gussman.  1976.  Dynamic behavior of  aerosols, pp.
40-65.  _In Handbook on Aerosols.   R. E. Dennis, ed.  U.S. ERDA TIC-26608, 142
pp.

Bowden, F.  P., and D.  Tabor.   1950.  The Friction and  Lubrication of Solids.
Clarendon Press,  Oxford.

Brimblecombe,  P.   1978.   Dew as  a sink  for sulphur dioxide.   Tellus 30:
151-157.

Brutsaert,  W.  H.   1975a.   The  roughness  length  for  water  vapor,  sensible
heat, and other sea lars.  J. Atmos. Sci.  32:2028-2031.

Brutsaert, W.  H.   1975b.   A theory for local evaporation  (or heat  transfer)
from  rough  and   smooth surfaces  at  ground  level.   Water Resources  Res.
11:543-550.

Cadle, R. D.  1966.  Particles in  the  Atmosphere  and  Space.   Reinhold Press,
New York.  226 pp.
                                  7-60

-------
Chamberlain, A. C.  1960.  Aspects of the deposition of radioactive and other
gases and particles.   Int. J. Air Pollut. 3:63-88.

Chamberlain, A. C.  1966.  Transport of gases to and from grass and grasslike
surfaces.  Proc. Roy.  Soc. London, A, 290:236-265.

Chamberlain,  A.  C.   1967.    Transport  of lycopodium spores  and  other small
particles to rough surfaces.  Proc. Roy. Soc. London, A, 296:45-70.

Chamberlain, A.  C.   1975.   The  movement  of particles  in  plant communities,
pp.  155-203.   In  Vegetation  and  the Atmosphere, Volume  1, Principles.  J. L.
Monteith, ed. "TTcademic Press, London.

Chamberlain, A. C.  1980.  Dry deposition of sulfur dioxide, pp. 185-197.  Jji
Atmospheric  Sulfur Deposition.   D.  S.  Shriner,  C.  R. Richmond, and  S.  E.
Lindberg, eds.  Ann Arbor Science, Ann Arbor, MI.

Chamberlain, A. C.  1983.  Roughness length of sea, sand and snow.  Boundary-
Layer Meteorol.  25:405-409.

Chang, S. G., R.  Toossi,  and T.  Novakov.   1981.  The importance of soot par-
ticles and  nitrous acid  in  oxidizing S02  in atmospheric  aqueous droplets.
Atmos. Environ. 15:1287-1292.

Cofer, W.  R.,  D. R.  Schryer, and R. S.  Rogowski.  1981.   The oxidation of
S02  on  carbon  particles  in  the  presence  of  Op,  NO?  and N?0.  Atmos.
Environ. 15:1281-1286.

Corn,  M.   1961.   The  adhesion  of solid  particles  to  solid  surfaces,  I.  A
review.  J. Air Pollut. Contr. Assoc. 11:523-528.

Dannevik,  W.  P.,  S.  Frisella,  L.  Granat,  and R.  B.   Husar.    1976.   S02
deposition measurements  in the  St.  Louis regions, pp.  506-511.  In Proceed-
ings, Third  Symposium  on  Atmospheric  Turbulence, Diffusion, and  ATr Quality
Raleigh, NC (American Meteorological Society).

Dasch, J.  M.   1982.   A comparison  of surrogate surfaces  for  dry deposition
collection.     Proc.   Fourth   International   Conference   on   Precipitation
Scavenging, Dry Deposition, and  Resuspension.   Santa  Monica,  CA,  29 November
- 3 December.

Davidson, C.  I. and S. K. Friedlander.  1978.  A filtration model  for aerosol
dry deposition:   Application to  trace metal deposition  from  the  atmosphere.
J. Geophy. Res. 83, C5:2343-2352.

Davies, C.  N.    1966.   Deposition  from moving  aerosols,   pp.  393-446.   ln_
Aerosol  Science.   C.  N. Davies, eds.  Academic Press, New York.  468 pp.

Davies, C. N.   1967.   Aerosol  properties  related to surface  contamination,
pp. 1-5.  _I_n Surface Contamination.   B.  R. Fish, ed.   Pergamon  Press,  New
York.  415 pp.
                                  7-61

-------
Deacon,  E.  L.    1977.   Gas  transfer  to and  across  an air-water  interface,
Tellus 29:363-374.

Derjaguin, B. V., and  Yu.  I. Yalamov.  1972.   The theory of  thermophoresis
and diffusiophoresis of aerosol  particles  and  their experimental  testing,  pp.
1-200.  In Topics in Current  Aerosol  Research, Part 2.  G. M.  Hidy and  J.  R.
Brock, edT. Pergamon Press, New York.  384  pp.

Desjardins,  R.  L.   1977.   Energy budget  by  an  eddy correlation method.  J.
Appl. Meteorol.  16:248-250.

Desjardins, R.  L. and  E.  R.  Lemon.   1974.   Limitations of an eddy-correla-
tion technique  for the determination  of the carbon dioxide and sensible heat
fluxes.  Boundary-Layer Meteorol.  5:475-488.

Dillon, P. J.,  D. S. Jeffries,  and W.  A.  Scheider.  1982.   The  use  of  cali-
brated  lakes and  watersheds for  estimating  atmospheric  deposition near  a
large point source.   Water, Air, and  Soil  Pollut.  18:241-258.

Dolske, D. A. and D. F. Gatz.   1982.    A field intercomparison  of sulfate  dry
deposition monitoring  and  measurement methods:   Preliminary results.   Proc.
American Chemical Society Acid Rain Symposium, Las Vegas,  NV,  30 March 1982.

Dovland,  H.  and A.  Eliassen.   1976.   Dry  deposition  on a  snow  surface.
Atmos. Environ.  10:783-785.

Droppo,  J.  G.    1980.   Experimental   techniques  for  dry deposition  measure-
ments, pp. 209-221.  _I_n Atmospheric  Sulfur  Deposition,  D.  S.  Shriner,  C.  R.
Richmond and S.  E.  Lindberg,  eds.   Ann Arbor Press, Ann  Arbor,  MI.  568  pp.

Dyer, A.  J.   1974.   A review of  flux-profile relationships.   Boundary-Layer
Meteorol. 7:363-372.

Dyer, A. J., and F.  J.  Maher.  1965.   The  "Evapotron".   An  Instrument for  the
Measurement  of   Eddy   Fluxes in  the  Lower  Atmosphere,   CSIRO  (Australia)
Division of Meteorological  Physics Technical  Paper Number  15,  31 pp.

Eastman, J.  A.  and  D.  H.  Stedman.  1977.   A  fast response sensor for  ozone
eddy-correlation flux measurements.   Atmos.  Environ.  11:1209-1211.

Eaton, J.  S., G.  E. Likens,  and F. H. Bormann.   1978.  The  input  of gaseous
and particulate  sulfur to a forest ecosystem.   Tellus 30:546-551.

Engelmann, R. J., and  G.  A.  Sehmel.    1976.   Atmosphere-Surface  Exchange  of
Particulate and  Gaseous Pollutants.   U.S.  ERDA CONF-7409Z1,  988 pp.

Everett,  R.  G., B.  B. Hicks,  W.  W.   Berg, and  J.  W. Winchester.  1979.   An
analysis  of  particulate  sulfur and lead  gradient  data  collected at  Argonne
National Laboratory.  Atmos.  Environ.  13:931-9434
                                  7-62

-------
Fassina,  V.  1978.   A survey of air  pollution  and  deterioration  of stonework
in Venice.  Atmos. Environ. 12:2205-2211.

Fleischer,  R.  L.  and  F.  P. Parungo.   1974.   Aerosol  particles  on  tobacco
trichomes.  Nature 250:158-159.

Fowler,  D.    1978.    Dry  deposition  of  $03 on  agricultural  crops.   Atmos.
Environ.  12:369-373.

Fowler, D. and M. H.  Unsworth.  1979.  Turbulent transfer  of  sulphur  dioxide
to a wheat crop.  Q. J. Roy. Meteorol. Soc.  105:767-783.

Friedlander, S. K.   1977.   Smoke, Dust, and Haze.  John Wiley and Sons,  New
York.  317 pp.

Friedlander,  S.  K.  and  H. F.  Johnstone.    1957.    Deposition  of  suspended
particles from turbulent gas streams.  Indust.  Engr. Chem.  49:1151-1156.

Fuchs,  N.  A.   1964.   The  Mechanics  of Aerosols.   Pergamon Press,  New York.
408 pp.

Galbally,  I.  E. and  E.  R. Roy.   1980.   Ozone and  nitrogen oxides  in  the
southern  hemisphere  troposphere,   pp  431-438.    In Proceedings,  Quadrennial
International Ozone Symposium.   August 4-9, 1980,~¥oulder,  CO  (available  from
IAMAP).

Galbally,  I.  E.,  J.  A. Garland, and M.  J.  G.  Wilson.  1979.   Sulfur uptake
from the atmosphere by forest  and  farmland.  Nature 280:49-50.

Garland,  J.  A.   1977.   The dry  deposition  of sulphur dioxide  to land  and
water surfaces.  Proc. Roy. Soc. London A 354:245-268.

Garland,  J.  A.   1978.  Dry and wet  removal  of sulphur from  the  atmosphere.
Atmos. Environ. 12:349-362.

Garland, J. A.  1979.  Dry deposition of  gaseous pollutants, pp 95-103 of  WMO
- No. 538:  Papers presented at the WMO Symposium on the Long-Range Transport
of Pollutants  and its  Relation to  General  Circulation  Including  Stratos-
pheric/Tropospheric Exchange Processes, Sofia,  1-5  October.

Garland, J. A.  and  J. R.  Branson.  1977.  The deposition of  sulphur  dioxide
to a  pine forest assessed  by a  radioactive  tracer method.   Tellus  29:445-454.

Garland, J. A., W.  S. Clough,  and D.  Fowler.   1973.   Deposition  of  sulphur
dioxide on grass.   Nature  242:256-257.

Garratt,  J.  R.   1978.   Flux-profile relationships  above  tall   vegetation.
Quart. J. Roy.  Meteorol.  Soc.  104:199-211.
                                  7-63

-------
Gauri, K. L. 1978.  The  preservation of  stone.   Scientific  American 238:126-
128, 131, 134-136.

Hardy, E. P., Jr., and J. H.  Harley,  Eds.  1958.   Environmental  contamination
from weapons tests.  U.S. AEC Health  and  Safety  Laboratory Report,  HASL-42A.

Harriott, P. and  R.  M.  Hamilton.   1965.  Solid-liquid mass transfer in tur-
bulent pipe flow.   Chem.  Eng. Sci.  20:1073-1078.

Haynie, F. H. and  J.  B.  Upham.  1974.  Correlation between corrosion behavior
of steel and atmospheric pollution data, pp. 33-51.  _I_n  Corrosion  in Natural
Environments, ASTM STP 558, American  Society for Testing  and Materials.

Hasse, L., and P.  S.  Liss.   1980.   Gas  exchange  across  the air-sea  interface.
Tell us 32:470-481.

Hess,  G.  D.  and  B.  B.  Hicks.  1975.   The  influence  of surface effects on
pollutant  deposition  rates  over   the  Great  Lakes,  pp.  238-247.    Ini  The
Proceedings of the Second Federal Conference on  the Great Lakes,  ICMSE.

Hicks, B.  B.   1982.   Wet  and dry surface  deposition  of air pollutants and
their modeling, pp. 183-196.   In Conservation of historic stone  buildings and
monuments.    National  MateriaTs   Advisory  Board,  National  Academy  Press,
Washington, 365 pp.

Hicks, B.  B.  and   P.  S.  Liss.   1976.    Transfer of SOg and  other reactive
gases across the air-sea  interface.  Tellus  28:348-354.

Hicks, B.  B. and  M.  L.  Wesely.  1978.    An  examination of some  micrometeoro-
logical  methods  for  measuring  dry deposition.   U.S.  EPA  Report EPA-6000/
7-78-116.  27 pp.

Hicks, B.  B.  and  M.  L.  Wesely.   1980.   Turbulent  transfer  processes  to a
surface and interaction with vegetation, pp. 199-207.  _Iri Atmospheric Sulfur
Deposition.  D. S. Shriner, C. R. Richmond, and S. E.  Lindburg, eds.   Ann
Arbor Press, Ann Arbor,  MI.  568 pp.

Hicks, B. B. and R. M.  Williams.  1979.   Transfer and Deposition of Particles
to Water  Surfaces, pp.   237-266.   _I_n Atmospheric Sulfur Deposition.   D. S.
Shriner, C.  R.  Richmond, and S. E.  Lindberg,  eds.   Ann Arbor  Science, Ann
Arbor, MI.  568 pp.

Hicks, B. B., G.  D. Hess,  and M. L.  Wesely.  1979.   Analysis  of flux-profile
relationships above  tall  vegetation, an alternative view.    Quart.  J.  Roy.
Meteorol. Soc.  105:1074-1077.

Hicks, B. B., M.  L. Wesely,  and J. L.  Durham.   1981.  Critique  of  methods to
measure dry deposition;  concise summary of workshop.    Presented at the 1981
National  ACS Meeting,  Atlanta, GA,  Ann  Arbor Science.
                                  7-64

-------
Hicks, B. B.,  M.  L.  Wesely, R. L. Coulter,  R.  L.  Hart, J. L. Durham, R.  E.
Speer, and D. H. Stedman.  1982.  An experimental  study  of  sulfur  deposition
to  grassland.    Proc.  Fourth  International   Conference   on  Precipitation
Scavenging, Dry Deposition, and Resuspension.   Santa Monica, CA, 29  November
- 3 December.

Hidy, G. M.   1973.   Removal  processes  of  gaseous  and  particulate pollutants,
pp.  121-176.   _Iri  Chemistry of  the  Lower  Atmosphere.    S.  I.  Rasool, ed.
Plenum Press, New York.  335 pp.

Hidy, G.  M.  and S. L. Heisler.   1978.   Transport and deposition  of  flowing
aerosols.  J_n Recent Developments in  Aerosol  Science.   D.  Shaw,  ed.   John
Wiley and Sons, New York.

Hill, A.  C.   1971.  Vegetation.   A sink for atmospheric pollutants.   J. Air
Pollut.  Contr. Assoc. 21:341-346.

Hubbard, D.  W.  and E. N.  Lightfoot.   1966.    Correlation  of  heat and  mass
transfer  data for  high  Schmidt  and   Reynolds  numbers.   I/Ec Fundamentals
5:370-379.

Hudson,  J. G. and P.  Squires.  1978.   Continental  surface measurements on CCN
flux.  J. Atmos. Sci. 35:1289-1295.

Jones, E.  P. and  S.  D.  Smith.   1977.   A  first  measurement  of sea-air CO?
flux by eddy correlation.  J. Geophys.  Res.  82:5990-5992.

Judeikis, H.  S.  and T.  B.  Stewart.    1976.    Laboratory measurement  of S02
deposition  velocities on   selected  building  materials  and  soils.   Atmos.
Environ. 10:769-776.

Judeikis, H.  S. and  A.  G.  Wren.   1977.    Deposition  of  H2$ and  dimethyl
sulfide on selected soil  materials.   Atmos.  Environ. 11:1221-1224.

Judeikis, H.  S.  and  A. G.  Wren.  1978.   Laboratory  measurements  of  SO? and
N02  depositions  onto solid  and  cement surfaces.   Atmos.  Environ.  12:2315-
2319.

Kaimal,  C. J.   1975.  Sensors and techniques  for the direct  measurement  of
turbulent fluxes and  profiles  in  the  atmospheric surface layer. Atmospheric
Technology 7:7-14.

Kanemasu, E.  T.,  M.  L.  Wesely,  B.  B.  Hicks,  and  J.  L. Heilman.    1979.
Techniques  for  calculating  energy  and  mass  fluxes,  pp.   156-182.    J_n
Modification of the Aerial   Environment of  Plants.   B. J. Barfield and J.  F.
Gerber  eds.   ASAE  Monograph No.  2.

Kanwisher, J.  1963.   On  the exchange of gases  between the atmosphere  and the
sea.  Deep-Sea Res. 10:195-207.
                                  7-65

-------
Langer, G.  1965.  Particle deposition on and re-entrainment from coniferous
trees, Part II.   Kolloid Z.  204  119-124.

Lenschow,  D.  H.,  A. C.  Delany,  B. B.  Stankov,  and D.  H. Stedman.   1980.
Airborne measurements of  the  vertical  flux of ozone  in the boundary layer.
Boundary-Layer Meteorol. 19:249-265.

Lewellen, W. S. and Y.  P. Sheng.   1980.   Modeling  and  dry deposition of S02
and sulfate aerosols.   Electric  Power  Research  Institute Report EA-1452, 46
pp. (Available from EPRI, Research Reports  Center,  Box 50490,  Palo Alto, CA
94303, USA).

Lindberg,  S. E.,  R.  C.  Harris,  R.  R.  Turner,  D.  S. Shriner, and D. D. Huff.
1979.    Mechanisms  and  Rates  of Atmospheric  Deposition  of  Selected Trace
Elements  and  Sulfate to  a  Deciduous  Forest  Watershed,  Oak  Ridge National
Laboratory Report ORNL/TM-6674,  514 pp.

Liss,  P.  S.   1973.   Processes of gas exchange across  an air-sea  interface.
Deep-Sea Res.  20:221-238.

Liss,   P.  S. and  P. G.  Slater.    1974.    Flux  of gases  across the  air-sea
interface.  Nature 247:181-184.

Liu, B. Y. H.  and J. K.  Agarwal.  1974.   Experimental  observation of  aerosol
deposition in turbulent  flow.   Aerosol  Science 5:145-155.

Livingston, R.  A.,  and  N.  S.  Baer.    1983.   Mechanisms  of  air pollution-
induced damage to stone.  Proc. Sixth World Congress on Air Quality,  IUAPPA,
Paris, France, 16-20 May,  1983.   In press.

Lovett, G.  M., W. A.  Reiners, and  R. K.  Olson.   1982.  Cloud droplet depo-
sition in  subalpine  balsam  fir forests:   Hydrological and chemical  inputs.
Science 218:1303-1304.

Mantel 1,  E. A.   1974.   Radioactivity  of tobacco trichomes  and  insoluble
cigarette smoke particles.  Nature  249:215-217.

Meszaros,  A., and K.  Vissy.  1974.   Concentrations,   size distributions and
chemical nature of atmospheric aerosol  particles  in remote oceanic areas.  J.
Aerosol Sci. 5:101-109.

Mizushina   T., F. Ogino,  Y. Oka, and H.  Fukuda.   1971.   Turbulent heat and
mass  transfer  between  wall  and  fluid  streams of  large Prandtl and  Schmidt
numbers.   Inter. J.  Heat and Mass Transfer 14  1705-1716.

Moller, U.  and G. Schumann.   1970.   Mechanisms  of transport from the atmos-
phere to the earth's surface.   J. Geophys. Res. 75:3013-3019.

Monteith,  J.  L.   1963.   Dew, facts and  fallacies,  pp. 37-56.   Jji  The Water
Relations of Plants.  A. J.  Rutter  and F. H. Whitehead, eds.   John  Wiley and
Sons, New  York.
                                  7-66

-------
Murphy, B. D.   1976.   The  influence of ground cover  on the dry deposition
rate of  gaseous materials.   Oak  Ridge  National   Laboratory  Report UCCCND/
CSD-19, 33 pp.

O'Dell, R. A.,  M.  Taheri, and  R.  L. Kabel.   1977.   A  model  for uptake of
pollutants by vegetation.  J.  Air Pollut.  Contr. Assoc. 27:1104-1109.

Owers,  M.  J.  and A.  W.  Powell.   1974.  Deposition  velocity  of  sulphur dioxide
on land and water surfaces  using a  35S  method.  Atmos. Environ. 8:63-67.

Raupach,  M. R., J. B. Stewart,  and A.  S.  Thorn.  1979.  Comments on  "Analysis
of flux-profile relationships above tall  vegetation, an alternative view" by
Hicks et  al.   Q. J.  Roy. Meteorol.  Soc. 105:1077-1078.

Raynor, 6. S.,  J. V.  Hayes,  and E. C.  Ogden.  1974.  Particulate dispersion
into and  within a forest.  Boundary-Layer  Meteorol.  7:429-456.

Raynor, 6. S.,  E. C.  Ogden,  and J.  V.  Hayes.   1970.   Dispersion and depo-
sition of ragweed  pollen  from  experimental  sources.   J.  Appl.  Meteorol.
9:885-895.

Raynor, G. S.,  E. C.  Ogden,  and J.  V. Hayes.   1972a.   Dispersion and depo-
sition  of Timothy   pollen  from  experimental   sources.     Agric.  Meteorol.
9:347-366.

Raynor, G. S.,  E. C.  Ogden,  and J.  V. Hayes.   1972b.   Dispersion and depo-
sition of corn pollen from  experimental sources.   Agronomy  J.  64:420-427.

Rosinski,   J.  and   C.   T.  Nagamoto.    1965.    Particle  deposition on  and
reentrainment from coniferous trees,  Part  I.  Kolloid  Z.  204:111-119.

Roth,  R.     1975.    Der vertikale  transport  von  Luftbeimengungen in  der
Prandtl-Schicht und  die  deposition-velocity.  Meteorol. Resch. 28:65-71.

Sehmel, G.   A.    1970.    Particle deposition  from turbulent  airflow.   J.
Geophys.  Res. 75 1766-1781.

Sehmel, G.  A.    1979.    Deposition and  Resuspension  Processes.    Battelle,
Pacific Northwest Laboratory  Publication PNL-SA-6746.

Sehmel, G. A.   1980a.   Particle and gas dry deposition:   A review.  Atmos.
Environ.  14:983-1012.

Sehmel, G. A.   1980b.   Model   predictions and  a   summary  of  dry deposition
velocity   data,   pp.  223-235.   _In  Atmospheric Sulfur  Deposition.    D.  S.
Shriner,   C.  R.  Richmond,  and   S.  F. Lindberg, eds.   Ann Arbor Press,  Ann
Arbor,  MI.  568 pp.

Sehmel, G. A. and S. L.  Sutter.   1974.  Particle  deposition rates on a water
surface as  a   function  of  particle  diameter  and  air  velocity.    J.  de
Recherches Atmospheriques 3:911-920.
                                  7-67

-------
Sehmel, G.  A.,  W.  H.  Hodgson, and  S.  L. Sutter.   1973a.   Dry deposition
of particles.     Battelle  Pacific  Northwest  Laboratory  Report  BNWL-1850
3:157-162.

Sehmel, G.  A.,  S.  L. Sutter,  and  M.  T. Dana.   1973b.   Dry deposition pro-
cesses.  Battelle  Pacific Northwest  Laboratories  Report  BNWL-1751 1:43-49.

Sheih,  C.  M.,  M. L.  Wesely,  and B.  B.  Hicks.   1979.   Estimated dry deposition
velocities of sulfur over the  Eastern United States and  surrounding  regions.
Atmos.  Environ.  13 1361  1368.

Shepherd,  J.  G.  1974.   Measurements  of  the direct deposition  of  sulphur
dioxide onto grass and water by the  profile method.   Atmos. Environ.  8:69-74.

Shreffler, J.  H.  1976.   A  model  for  the transfer of gaseous pollutants to a
vegetational surface. J. Appl. Meteorol.  15:744-746.

Sievering,  H.    1982.    Profile  measurements  of  particle dry deposition
velocity at  an air-land  interface.   Atmos.  Environ.  16:301-306.

Sievering, H., M.  Dave,  D.  A.  Dolske, R. L.  Hughes,  and  P. McCoy.  1979.  An
experimental study of loading by aerosol transport and dry deposition in the
southern  Lake  Michigan   basin.    U.S. Environmental Agency  Report   EPA-905/
4-79-016.

Sinclair,  P.  C.   1976.    Vertical  transport  of  desert  particulates  by dust
devils  and  clear  thermals,  pp. 497-527.   _Iji Atmosphere-Surface Exchange of
Particulate  and Gaseous  Pollutants.   R. J   Engelmann and G.  A.  Sehmel, eds.
U.S. ERDA CONF-7409Z1.   989  pp.

Slinn,  W. G.  N.   1974.   Analytical investigations of inertial  deposition of
small aerosol  particles  from laminar flows  into  large obstacles  - Parts A and
B,  PNL  Ann. Report  to  the  USAEC,  DBER,  1973;  BNWL-1850,  Pt.  3,  Battelle-
Northwest, Richland, WA; available from  NTIS, Springfield,  WA.

Slinn,  W.  G. N.   1976a.   Formulation  and  a solution of the diffusion, depo-
sition, resuspension problem.   Atmos.  Environ. 10:763-768.

Slinn,  W.  G. N.   1976b.   Dry deposition  and resuspension  of aerosol particles
- a  new look at some old problems;  pp.  1-4U of  Atmospheric-Surface  Exchange
of  Particulate  and   Gaseous  Pollutants  -  1974,  R.   J.  Englemann and  G.  A.
Sehmel, (Coord.),  available  as  ERDA CONF-740921  from NTIS,  Springfield, VA.

Slinn,  S.  A. and W.  G.  N. Slinn.  1980.  Predictions for particle deposition
on natural waters.  Atmos. Environ.  14:1013-1016.

Slinn,  W. G.  N.,  L. Hasse,  B.  B.  Hicks, A. W. Hogan, D. Lai, P. S.  Liss, K.
0.  Munnich, G.  A.  Sehmel,  and  0.  Vittori.    1978.    Some  aspects  of  the
transfer of  atmospheric  trace  constituents past   the  air-sea  interface.
Atmos.  Environ.  12:2055-2087.
                                  7-68

-------
Spedding, D.  J.   1969.   Uptake of sulphur  dioxide  by  barley leaves at  low
sulphur dioxide concentrations.  Nature 224:1229-1231.

Twomey,  S.    1977.    Atmospheric  Aerosols.   Elsevier  Scientific  Publishing
Company, Amsterdam.  302 pp.

Wason,  D.  T., S.  K.  Wood,  R. Davies,  and A.  Lieberman.    1973.    Aerosol
transport.    Particle  charges  and  re-entrainment  effects.     J.   Colloid
Interface Sci. 43:144-149.

Wedding,  J.   B.,  R.  W.  Carlson,  J. J.  Stiekel, and  F.  A.  Bazzaz.    1975.
Aerosol deposition on plant leaves.  Env. Sci.  and Technol.  9:151-153.

Wesely,  M.   L.  and  B.  B.  Hicks.   1977.    Some factors  that  affect  the
deposition rates  of  sulfur dioxide and similar gases on vegetation.  J.  Air
Pollut. Contr. Assoc. 27:1110-1116.

Wesely, M. L. and  B.  B.  Hicks.   1979.   Dry deposition  and emission of  small
particles at the  surface of  the  earth.    Proceedings  Fourth  Symposium  on
Turbulence,   Diffusion  and Air  Quality  (Reno   NV,  15-18  January).    Am.
Meteorol. Soc., Boston, MA. pp. 510-513.

Wesely, M.  L., D. R.  Cook, R.  L. Hart,  B.  B.  Hicks, J.  L.  Durham, R.  E.
Speer, D. H.  Stedman, and R. J. Trapp.   1982a.   Eddy-correlation measurements
of  dry deposition  of  particulate  sulfur  and  submicron particles.    Proc.
Fourth  International  Conference on  Precipitation  Scavenging, Dry  Deposition,
and Resuspension.  Santa Monica, CA, 29 November  - 3  December,  in  press.

Wesely, M. L., J.  A.  Eastman,  D.  R. Cook, and B. B. Hicks.  1978.    Daytime
variation of ozone eddy  fluxes to maize.   Boundary-Layer Meteorology  15:
361-373.

Wesely, M. L., J.  A.  Eastman,  D.  H. Stedman,  and  E. D.  Yalvac.   1982b.   An
eddy correlation  measurement  of N0£  flux  to  vegetation and  comparison  to
03 flux.  Atmos.  Environ. 16:815-820.

Wesely, M. L.,  B. B.  Hicks, W.  P. Dannevik,  S.  Frisella,  and R. B. Husar.
1977.   An eddy  correlation  measurement  of  particulate  deposition  from  the
atmosphere.   Atmos. Environ. 11:561-563.

Whelpdale, D. M.  and  R. W. Shaw.  1974.  Sulphur  dioxide  removal by turbulent
transfer over grass,  snow, and  other surfaces.  Tellus 26:196-204.

Whitby, K.  T.   1978.    The  physical  characteristics  of  sulfur aerosols.
Atmos. Environ. 12:135-139.

Williams,  R.  M.,  M.   L.  Wesely, and B.  B. Hicks.   1978.   Preliminary eddy
correlation   measurements  of momentum,  heat,  and particle  fluxes to  Lake
Michigan.      Argonne   National  Laboratory   Radiological  and  Environmental
Research Division Annual Report.  Jan.  -  Dec.  1978.   ANAL-7865, Part III,  pp.
82-87.
                                  7-69

-------
Winkler, E. M., and E. J. Wilhelm.  1970.   Salt  burst by  hydration pressures
in architectural  stone  in  urban atmosphere.   Geol.  Soc.  America  Bull. 81:
567-572.

Yudine, M.  I.   1959.    Physical  considerations  on heavy-particle diffusion,
pp. 185-191.   In  Advances  in Geophysics, Volume 6.  H.  E.  Landsberg and J.
van Mieghem, eds.   Academic  Press,  New York.
                                  7-70

-------
               THE ACIDIC DEPOSITION PHENOMENON AND ITS EFFECTS

                          A-8.  DEPOSITION MONITORING
8.1  INTRODUCTION (G. J. Stensland)

The  previous  two chapters have  discussed  the deposition processes  by  which
acidic  and  acidifying  substances in the atmosphere impact on  various  recep-
tors.   Wet  deposition  in the form  of  rain,  fog,  and  snow and  dry deposition
of gases and particulate matter have been addressed.

This  chapter considers  both wet   deposition  monitoring during  periods  of
precipitation and  dry  deposition monitoring during periods of  no precipita-
tion.   Techniques  are  discussed  for collecting deposition data on  a routine
basis to determine the broad spacial patterns of deposition and their changes
over time.  Most of  the  techniques  are also  applicable  for measuring deposi-
tion  over smaller space and time  scales,  such as in  research  projects  to
study transformation and scavenging processes (Chapters A-4,  A-6,  and  A-7).
The  first section of this chapter  will  discuss techniques and  data bases for
wet  deposition  networks.   The  next section  will  emphasize  dry  deposition
techniques.

The  second  major  purpose  of this  chapter  is to  present  and discuss  data
available from routine, long-term networks.   Such data for dry  deposition are
limited and therefore are combined  with  the  techniques  discussion in Section
8.3.  Section 8.4 will  discuss wet deposition data. Section 8.5 will examine
the  data  record  from  glacier  studies.   Glaciochemical  investigations  are
given as a tool  in historical delineation of acid precipitation  problems such
as a bench mark on the natural background void of anthropogenic  pollution and
contamination.

Wet  deposition  monitoring techniques  vary  with the  chemical   species  being
investigated.   This  wet deposition  discussion will be limited to  the  major
soluble species in  precipitation   which  account  for  most of  the  measured
conductance of  the  samples.   This list would  include the following  ions:
hydrogen,   bicarbonate,  calcium,  magnesium,   sodium,   potassium,   sulfate,
nitrate, chloride, and ammonium.   Experience has shown that measurements  of
the last eight  ions  in the list  allow  one  to  calculate a pH  value  which  is
usually in  good agreement with  the measured pH value.    Samples  from  remote
locations  can be strongly affected  by organic acids and  are thus one group  of
exceptions (Galloway et  al.  1982).   The fact  that  we can often successfully
calculate  the pH of precipitation  samples  indicates  that  the  rather  small
list of measured  ions  is probably  sufficient for studies of wet deposition
emphasizing  the  acid  precipitation  phenomena.


                                    8-1

-------
This  chapter  will  present information  relevant  to the following  questions:
How  good  are the  current  network data?   Are the  networks  adequately  dis-
tributed  and operated  to provide  a  good evaluation  of the  temporal  and
spatial variations relative to pH and the acidic  and acidifying  substances of
interest?   Which  measurements need improvement,  what  are the nature of  the
improvements, and the reasons for them?  Are surrogate  types  of  air and  water
quality measurements available for trend analysis?

The  next chapter discusses the deposition  models  used  to  predict  exposure of
receptors to concentrations  of  specific pollutants.  Such models  are  needed
to predict deposition over prescribed periods  and with  required  resolution.

8.2  WET DEPOSITION NETWORKS  (G.  J.  Stensland)

8.2.1  Introduction and Historical Background

The measurement of chemicals  in  precipitation  is  not just a  recent endeavor.
In 1872, for example, Smith  discussed  the  relationship  between  air pollution
and  rainwater chemistry  in his  remarkable  book  entitled  Air and  Rain:   The
Beginnings  of  Chemical  Climatology.    Gorham (1958a)   reported  that  hydro-
chloric acid should be considered in assessing the  causes of rain acidity in
urban  areas.   Junge  (1963)   discussed the  role of  sea   salt  particles  in
producing rain from clouds.  A valuable historical perspective on  the subject
of acid precipitation has recently been published by Cowling  (1982).

There  are several  recent  reports describing wet  deposition  networks  and  the
data  generated  by  them;   the Acid Rain  Information Book,  prepared by  GCA
Corporation  in  1980  for  the U.S.  Department  of Energy   (GCA  1980);  the
Battelle Northwest Laboratories  (Dana  1980) report  for  the American Electric
Power  Service  Corporation;  and   the  Environmental  Research and  Technology
Incorporated report  for the  Utilities Air Regulatory  Group (Hansen et  al.
1981)  are but three examples.

Networks to monitor wet deposition can be physically characterized by:

     1.  Space scale--the total  area covered by the sampling  network.

     2.  Space density—the area represented by each site  in  the network,
         i.e., network area divided by the number of sites.

     3.  Time scale--the  time span during  which data were  collected in
         the network.

     4.  Time density—the frequency  of sample collection (the  sampling
         interval).

Networks have been of all  spatial and  temporal  scales  and densities, ranging
from  1  site operated  for only a  few days  to  more than 50 sites  distributed
over  several European countries  and operated for over  30 years,  to the cur-
rent rapidly growing NADP network with 115 sites as of mid-1983.
                                    8-2

-------
The  time and  space configurations  of networks  are  dictated by  scientific
objectives and available  financial  resources.   Networks  are  often  classified
either  as research networks  or as  monitoring  networks.  Research  networks
usually  have  smaller  space and time dimensions than  do  monitoring networks.
However,  the  data  generated  by all types  of monitoring networks are  even-
tually  used  for  research  purposes,  and the  data  from  single site  research
networks  are  frequently used  to monitor  the changes in time  of wet deposi-
tion.  Therefore, characterizing networks according to monitoring or  research
purposes  does not produce  a unique distinction.

8.2.2  Definitions

Some  widely  used technical  terms  that relate  to deposition  monitoring  are
defined as follows:

   -  For typical  rain  and melted  snow solutions  the  pH ranges  from 3.0  to
  .0. The  pH indicates the  acidity, i.e., the free hydrogen-ion concentration,
and  mathematically pH  =  -logiQ[H+].    Each  unit  of  decrease  on  the  pH
scale  represents  a 10-fold increase of acidity.   Chemically a pH of 7.0  is
approximately neutral  (for T = 20 C); a pH of less than  7.0  is acidic,  and a
pH of more than  7.0 is alkaline.   Therefore, rainwater with  a pH less than
7.0  is  acidic.   However,   pure  water in equilibrium with atmospheric  carbon
dioxide  has a pH  of about 5.6.  Therefore,  in practice many  scientists  adopt
5.6  as  the  reference  value, with samples of  rain  and melted  snow having  pH
less  than 5.6  referred  to as  acidic precipitation.   This pH = 5.6 reference
point will be  adopted  for this chapter.  (Values  varying  from 5.60 to 5.70
are  quoted  as the  reference  value  by  other authors.)   Discussion to  follow
(Section  8.4.2) will  indicate that natural  rain  in  areas  unaffected by man
can  have pH values of  5.0 or less and therefore the  value of  5.6 is more
arbitrary than natural.   This point is  also  discussed in Chapter A-2, Section
2.2.5.

A  more  rigorous  chemical  discussion  of pH  is  provided   in Chapter  E-4,
Sections 4.2.2 and 4.4.3.1.

Weighted-mean concentration -  The mean  concentration of  a precipitation con-
stituent  such as sulfate for five samples  would be simply the sum of  the five
concentration values divided by five.  The volume-weighted-mean concentration
for  five  samples  for sulfate  is  the sum of  five products (each sample volume
x the sulfate concentration in that sample volume) divided by  the  sum of the
five volumes.   The precipitation-weighted-mean concentration  is calculated  in
the  same  way  except the precipitation  amount from a  standard rain  gauge  is
used instead of the volume from  the precipitation  chemistry  sampling device.
For the ions generally  considered  to be conservative when samples are  mixed
together  (sulfate,  nitrate,  ammonium,  chloride, calcium, magnesium, sodium,
and  potassium),  the weighted-mean  concentration   for  five   samples  is  con-
ceptually the  same  as the  single  value that would be  measured  if  all  five
samples had been poured into  one large container.   This is  not conceptually
true  for non-conservative  ions  (such as  hydrogen   and bicarbonate ions).
However, if all  the precipitation samples  are  in equilibrium  with atmospheric
carbon  dioxide  and have   pH  values less than  about  5.0,  then  bicarbonate
concentrations are  relatively  small  and the  hydrogen ion would be conserved


                                    8-3

-------
 in  the mixing process.  The pH  calculated  for the volume- or precipitation-
 weighted-mean  hydrogen  concentration  will  be referred to  in  this  chapter as
 the weighted pH.

 Precipitation  - The  term wil be  used  to denote aqueous material  in liquid or
 solid  form, derived from the atmosphere.  Dew, frost, and fog are technically
 included but in practice poorly measured, except by special instruments.

 Acid  rain  - A popular term with  many  meanings, generally used  to describe
 precipitation  with a pH less than 5.6.

 Acid  precipitation -  Water from the atmosphere  in  the form  of  rain,  sleet,
 snow,  and hail, with a pH less than 5.6.

 Wet deposition -  A term that refers to:  (1)  the  amount of material  removed
 from  the  atmosphere  and  delivered to  the  ground  by rain,  snow, or  other
 precipitation  forms; and (2) the process of transferring gases,  liquids,  and
 solids from the atmosphere to the ground during a precipitation event.

 Dry deposition -  A term for  (1) the amount of  material deposited from  the
 atmosphere to  the ground in the absence of precipitation; and (2)  the  process
 of such deposition.

 Total   atmospheric deposition - Transfer from the atmosphere to the ground of
 gases,  aerosol  particles,  and precipitation,  i.e., the  sum  of wet and  dry
 deposition. Atmospheric  deposition  includes many  different  types  of  sub-
 stances, non-acidic as well as acidic.

 Acidic deposition - The transfer from the atmosphere  to  the ground of acidic
 substances, via wet or dry  deposition.

 Quality control  and  quality assurance  -  Each person  involved in  producing
 precipitation  chemistry  measurements,  from site  operators  through  central
 laboratory chemists, must  carry  out certain tests to  continuously  determine
 that his procedures and his equipment are "in control."  Without  such  tests a
 technician might, for  example, continue to  measure pH with a malfunctioning
 electrode, an "out of control"  electrode.

At  the next higher  level,  the  technician's  supervisor  must  assure  himself
 that his technician is producing high quality  data, within  the quoted  limits
 of  precision  and bias.   The supervisor  gains this  assurance,  in part,  by
 submitting check samples of specified chemical  concentration.   The  technician
 is not told the "known"  values and may not  even recognize  that  a  particular
 sample is designed to  check his work;  in  this case  the sample  is a  blind,
 unknown quality assurance sample.

The scientific  data  user will  likely  want to examine  the  same data set used
by the supervisor to  be assured that the data were  of  high  quality.

The terms quality  control  and quality  assurance  are defined  differently  by
various authors.  The definitions for  this chapter  are the  following:
                                    8-4

-------
Quality control  - A system of activities that accomplishes  two objectives:

     1.  To continuously control  the quality of measurements within
         established tolerances;  and

     2.  To provide data from tests to determine the  precision and bias  being
         achieved.

Quality assurance  -  A system of  activities that verifies and maintains  the
quality of the measurements.  The  quality control activities of the  analysts
are one complement of the quality assurance  system.

8.2.3 Methods, Procedures, and Equipment for Viet Deposition Networks

For ideal  data comparability, all  wet deposition networks  should use  the same
equipment and procedures.   In reality, this  rarely  happens.   The following
discussion outlines procedures and  equipment  which vary among networks,  past
and present, and indicates how the data used  should be checked for data  com-
parability.

Site selection - The selection of monitoring sites is based on criteria  which
should be described in the network documentation.  The siting criteria depend
on the objectives of the network.

Sample containers  -  The containers for collecting and storing precipitation
vary, depending on the  chemicals  to be measured.  Reuseable plastic collec-
tion containers  are  currently used in  most acidic wet deposition networks.
However, they  are unacceptable for monitoring pesticides in precipitation.
Glass collection  containers are  considered  less desirable than plastic  ones
(Galloway  and Likens  1979).   Frequent  quality  control  blank  checks  are
necessary to  monitor procedures  for  cleaning  containers,  and  great care  is
necessary to maintain acceptably  low  blank  levels.  Acid  washing procedures
can  potentially  produce  precipitation pH  levels that  are  too  low,   while
detergent washing may have  the opposite effect.  Several   networks now  avoid
both of these washing procedures.

Sampling mode  -  There are  three  sampling modes.  In bulk sampling  the  col-
lection container is continuously  exposed to  the atmosphere for sampling  and
thus collects  a  mixture of  both  wet  and  dry deposition.   Bulk sampling  has
been used frequently  in the past and  is still  often used for economic  rea-
sons.   For  studies of  total  deposition, wet plus dry, bulk sampling may  be
suitable.   A problem is  that exactly  what component of  dry  deposition  is
sampled by  open  containers is unknown.  The  continuously-exposed containers
are subject to  varying  amounts of  evaporation  unless  equipped  with a  vapor
barrier.  For studies to  determine the acidity of rain and snow samples (the
wet deposition component), bulk data pH must be used  with  great caution  (only
in  conjunction  with comprehensive  system blank data which demonstrate  that
dry deposition did not  significantly  bias  the results).    For wet deposition
sites  that  will  be  operated  for  a  long   time  (more than one  year),   site
operation and central laboratory  expenses are large  enough that wet-only  or
wet-dry samplers should  be used  instead of bulk samplers  to  maximize  the
scientific output from the project.


                                    8-5

-------
In  wet-only  sampling,  dry  deposition is  excluded  from  the  precipitation
samples by automatic devices that uncover  the sampling containers  only  during
precipitation  events.    Three  types  of   automatic   wet-only  samplers  were
evaluated  for event collection  in  a  Pennsylvania State  University  study,
which  found  differences  in both the reliability  of  the instruments and the
chemical  concentrations  in the  samples  (dePena et al.  1980).   In wet-dry
lampling,  the automatic  collecting device includes one container to capture
wet deposition  and a  second container  to  capture  dry deposition  where  a
precipitation sensor activates a motor which moves a cover  from one  container
to the other.  As with bulk sampling,  the dry container of a  wet-dry sampler
collects a not-well-defined fraction of the total  dry  deposition.

For both  wet-only  and  wet-dry sampling,  the automatic device  has been some-
times replaced by an observer making manual  container  changes, an  undesirable
alternative except in very  special  situations.  Generally, projects  have not
collected  and reported system blank data  to prove that the manual  procedure
prevented  bias due to dry deposition.

In sequent!'a! sampling, a  series of containers  are exposed to the atmosphere
to  collect wet  deposition samples, with  consecutive advances  to  new  con-
tainers  being triggered on a  time basis,  a  collected volume  basis,  or  a
combination.  Sequential samplers  can be rather complicated  and  are usually
operated  only  for short  time   periods  during  specific  research  projects.
Again  an  observer  sometimes replaces the automatic device to  provide  manual
sequential sampling.

Field measurements - Conductivity,  pH, sample weight  or volume,  and rainfall
amount  are frequently measured  at  field laboratories.   Making  these addi-
tional  measurements requires  that  site operators  have more training and work
longer  periods for each sample than  operators at sites where  samples are only
collected  and forwarded to a  central analytical  laboratory.   Rainfall  amount
determined with a  standard  rain  gauge  is  necessary as it provides an assess-
ment  of  the  fraction of   the  precipitation captured  by  the  precipitation
chemistry  sampler, and thus,  is  useful to  ascertain,  after the fact, that an
automatic  sampler  has not malfunctioned.

Sample  handling  -  Chemical changes  with time in  the  sample  are decreased by
refrigeration,  aliquoting,  filtering,  and  the  addition of preservatives to
prevent biological change.  Peden  and Skowron  (1978)  have  reported  that fil-
tering  is  more effective than refrigeration  for stabilizing Illinois samples.
When the  filtering procedure  is used,  it  is important to  obtain and evaluate
frequent  filter blank samples, because the chemistry of relatively clean rain
samples can be easily altered.

The  chemical  changes with  time seem to generally  increase  the measured pH.
Central laboratory pH seems to be generally  higher than the field pH measured
at an  earlier time but  this has not  been  carefully  documented and reported
very often due, in part, to the problem in quality assuring the field data.

Analytical methods - Appropriate analytical  methods are available to measure
the major ions  found  in  precipitation, but special precautions are  necessary
                                    8-6

-------
because the  concentrations  are low;  thus,  the  samples  are easily contami-
nated.  Although pH is  deceptively  easy to  determine with modern equipment,
achieving accurate  results  requires  special  care because of  the low  ionic
strength of rain and snow samples.  Frequent checks with low ionic  strength
reference solutions are  required  to  avoid  the  frequent problem of malfunc-
tioning pH electrodes.

Data  screening - Network  data  are in effect  screened  out if technicians  in
the  field or at  the central  laboratory discard  samples  because they  look
"unduly contaminated."   After samples are analyzed the  data can  be flagged  or
removed because samples were not collected in the field according  to standard
protocol  or  because the  data are statistical outliers.   The data screening
procedures should be documented  and updated at regular intervals during the
projects.

Quality  control  and  quality assurance  reports  -  For most  wet deposition
networks, too  few  quality  control  checks are  performed  routinely,  too few
procedures and results undergo continuous evaluation,  and  too few  results are
summarized into formal  written quality assurance and quality control reports.
This  is  even more  true  for past  network operations.   The reports  that are
available are  often  analytical  laboratory reports that document  the methods
used  to measure chemical  parameters and the bias and  precision of the  ana-
lytical methods.   However,  for  wet  deposition  monitoring networks,  a  much
greater effort should be made to develop a quality program that addresses all
of  the  steps leading  to the data base.   While  quality assurance  and quality
control reports can be relatively easily produced for the  analytical methods,
some  of the  greatest uncertainties  in comparing  data  from different networks
involve  estimating  the  bias and precision  resulting from  differences  in
sampling  mode, sample handling, and possibly data handling.

Thorough  quality assurance programs are costly.   Therefore, a network must  be
quite large  and be planned to run for a  long  time to  warrant implementing  an
elaborate quality assurance program.   A research  project  that  operates  five
sites for one year, for example,  generally  cannot afford  to  produce an  array
of  written documents to  describe  in  detail  all  aspects of the  quality  assur-
ance  program.

Because different networks  collect  daily samples, weekly  samples or monthly
samples,  the data  user  is  often faced  with deciding whether  two  different
data  sets are comparable.   Thus, quality  control  reports for the  separate
networks  should contain information to assess data bias and precision  for the
particular network and also for comparing results to other accepted  networks.
The  use of  colocated  sites for  various networks  is  one  of the  most  direct
ways  to  assess  network  design  differences.    Several   colocated  sites  at
locations  having different meteorological   and  pollution environments  are
necessary to evaluate  network  data  differences.   The  operation of  colocated
sites should be continuous rather than a one-time endeavor.

8.2.4  Wet Deposition Network Data Bases

The  wet deposition  data  bases  available for North America  have  been  summa-
rized  by many  authors (e.g.,  Eriksson  1952,  Niemann et  al.  1979,  Miller


                                    8-7

-------
1981, Wlsniewskl and  Kinsman  1982).   Miller  points  out that the history of
precipitation chemistry measurements  in North America has been very erratic,
with networks  being  established and disbanded  without  thought of long-term
considerations.  Miller suggested one possible time grouping of network data:

     1.  1875-1955, the  period when agricultural  researchers measured
         nutrients  in  precipitation to  determine  the input to the soil
         system;

     2.  1955-1975, the period when atmospheric  chemists were measuring
         the major ions  in  precipitation to  better  understand chemical
         cycles in  the atmosphere;  and

     3.  1975-present, the period  when  network  measurements  were often
         primarily  to  evaluate ecological  effects.

Table  8-1   (Miller  1981) summarizes  the  "agricultural  data  bases"   (taken
largely from the review by Eriksson 1952).

Table 8-2  summarizes  some regional-  and  national-scale  wet  deposition net-
works in Canada  and the  United States  that have begun operation since 1955.
These networks  were generally not  established  to monitor  acidic precipita-
tion.   The first two  are no  longer in  operation.   The  PHS/NCAR and EML-DOE
networks include sites influenced  by large  urban  areas  and  thus  are  not as
useful   in  addressing   acidic  precipitation  issues on  larger scales  as  are
other networks.  All   the  networks  followed  the  pattern  of  the Junge network
in measuring  major inorganic  ions that account for much  of sample conduc-
tance.   Sulfate was measured in all the networks;  pH was not measured  in the
Junge network.

In addition to regional- and  national-scale  wet  deposition  networks, local
networks also exist.   These  local networks:

     1.  may consist  of  only  one site  (e.g., Larson and Hettick 1956),
         or many sites concentrated  in  a rather small   area (e.g., 85
         sites in Gatz 1980);

     2.  may have operated for a year (e.g.,  the central  Illinois study,
         Larson  and  Hettick  1956), or much  longer  (e.g., the Hubbard
         Brook data base, Likens 1976);  and

     3.  may have  studied a  particular  pollution  source (e.g.,  the St.
         Louis area,  Gatz 1980) or  the  plume from  power  plants  (e.g., Li
         and Landsberg 1975, Dana et al. 1975).

Some of the local  network data have been very useful  in  interpreting time
trends  of chemical  concentrations in precipitation.

Wisniewski  and Kinsman (1982)  have prepared a detailed tabulation of nation-
al, regional,  and  state  or  province  networks currently  in operation  in the
United  States, Canada, and Mexico.  A total  of 71  networks  are described.
                                    8-8

-------
             TABLE 8-1.   AGRICULTURAL DATA BASES (1875-1955)
 Period
Number of studies
Locations of sites
1875 -1895

1895 - 1915



1915 - 1935



1935 - 1955
        3

        7



        8
Missouri, Kansas, Utah

Ottawa, Iowa, Tennessee,
Wisconsin, Illinois,  New  York
Kansas

Kentucky, Oklahoma, New York,
Illinois, Texas, Virginia,
Tennessee

Alabama, Georgia, Indiana,
Minnesota, Mississippi,
Tennessee, Massachusetts
                                  8-9

-------
                        TABLE 8-2.  SOME NORTH AMERICAN WET DEPOSITION DATA BASES (1955-PRESENT)
CO
I
APPROXIMATE

NETWORK
National
Junge
PHS/NCARb
WMO/EPA/NOAAC
CANSAPd
NADPe

PERIOD
1955-1956
1959-1966
1972-Present
1977-Present
1978-Present
NUMBER OF
SITES
60
35
17
54
115
SAMPLING
MODE3
W-M
W
W
W
W-D
SAMPLING
INTERVAL
Daily (with monthly
compositing)
Monthly
Monthly (weekly after joini
NADP in 1980)
Daily (with monthly composi
ing) (monthly before 1980)
Weekly



ng
t-

            Regional
            US Geological
            Survey Eastern
            (USGS)
            Canadian Centre
            for Inland
            Waters (CCIW)
            Tennessee Valley
            Authority (TVA)
            MAP3Sf
1964-Present       18

1969-Present       15

1971-Present        9
1976-Present        9
 W

W-D

 W
Monthly

Monthly

Biweekly
Daily

-------
                                                  TABLE 8-2. CONTINUED
00
I
NETWORK
Canadian APN9
EML-DOEh
EPRI-rSURE1
UAPSJ
U.S. EPAk
Great Lakes
PERIOD
1978-Present
1977-Present
1978-1981
1981-Present
1977-Present
NUMBER OF
SITES
8
7
9
20
30
SAMPLING
MODE*
W
B, W-D
W
W
B, W
SAMPLING
INTERVAL
Daily
Monthly
Daily
Daily
Monthly and Weekly
 B for bulk, W for wet-only with  automatically opening device, W-M for wet-only via manual
 operation, W-D for wet-dry with  automatic  device.

bU.S. Public Health Service/National Center for Atmospheric Research.

 World Meteorological  Organization/U.S. Environmental Protection Agency/National and Oceanic
 and Atmospheric Administration.  These sites are now part of NADP.

"Canadian Network for  Sampling Acid Precipitation.

eNational Atmospheric  Deposition  Program.   There were 115 operating sites on 1 July 1983 and
 the network was growing rapidly.  In  1983.  many of the NADP sites were also named as sites for
 inclusion in the National  Trends Network (NTN).

fMultistate Atmospheric Power Production Pollution Study.
^Canadian Air and Precipitation Network.

 Electric Power Research Institute-Sulfate  Regional Experiment.

 Environmental  Measurements Laboratory of the U.S. Department of Energy.

 Utility Acid Precipitation Study.  This was preceded at some of the same sites and with the
 same central laboratory by the 9 site, wet-only, daily sampling EPRI/SURE network.
b
 United States Environmental Protection Agency.

-------
 Whelpdale  (1979)  has  prepared a  tabulation  of  seven  major wet  deposition
 networks and  programs  in the world.   These  include CANSAP, MAP3S,  and  NADP
 (which  have  been  included  in Table  8-2);  the  Organization  for  Economic
 Cooperation and Development  (OECD)  network to  study the  long-range  transport
 of  air  pollutants which operated  from 1972 to  1975;  and the three  currently
 operating  networks  summarized  in  Tables  8-3  through 8-5.  Most of  the  World
 Meteorological Organization (WMO)  sites (see Table 8-3)  in Canada,  the United
 States,  and  Europe are  sites  operated  as  part of  the  CANSAP,  NADP,  or
 Economic Commission for Europe (ECE)  networks.   The ECE  network  (see  Table
 8-4)  is noteworthy in  that  (1)   only pH  and  sulfate  are required  to  be
 measured in  the precipitation samples (for  many sites  other major  ions are
 also  measured),  (2)  aerosol  sulfate  and  gaseous  sulfur  dioxide  must  be
 measured,  (3) each  participating country has  one or  more laboratories  to
 perform chemical analysis on  samples collected in that  country, and  (4)  the
 sample  collection  period is  24  hours.   The  European Atmospheric  Chemistry,
 Network (EACN) (see Table 8-5) is noteworthy in  that  its early  data  provided
 evidence that Scandanavian precipitation  is acidic.   Over  the last  20 years,
 these data have been central  to discussions of  why Scandanavian  precipitation
 is  so acidic  and  what  adverse  effects  are linked to this acidity.   Whelpdale
 (1979) and Wallen (1981) discuss  the European  and world  networks and  provide
 maps of site  locations.

 8.3  MONITORING CAPABILITIES FOR DRY DEPOSITION (B. B. Hicks)

 8.3.1  Introduction

 Dry deposition  delivers materials to  the  surface in both  solid and  gaseous
 phases, and  sometimes  in liquid  (e.g.,  when  the humidity  is  so  great  that
 "solid" hygroscopic particles are, in  fact, wet),  without  the convenience  of
 a natural  process (precipitation)   to organize  and concentrate its  delivery.
 Rainfall delivers pollutants in irregular but comparatively intense  doses,  in
 a manner that permits  relatively  simple sampling.   Dry  processes  are  far
 slower yet more  continuous.   Nevertheless, assessments  such as by  Galloway
 and Whelpdale (1980) and by Shannon  (1981)  suggest that wet and dry  deposi-
 tion processes are  of  roughly  equal  importance in  the average deposition  of
 specific chemical  species.

 As is explained at length in  Chapter A-7,  dry deposition  rates are influenced
 strongly by the nature  of the surface and by  the configuration of appropriate
 sources.  Surface emissions  are held in close contact  with  the ground  consid-
 erably more than are emissions released at greater  altitudes, so that in the
 former case rates of dry deposition would be expected to  be greater.   As  a
 direct consequence,  dry deposition fluxes must  be expected  to be highest  near
 sources, whereas the highest rates  of  wet deposition of the same pollutants
may be found  much farther downstream.  Thus, a  network designed specifically
 to study dry  deposition will  not  be the  same  as  one  designed  only to  study
wet.   Nevertheless, the  intent of most  networks is to obtain  the  maximum
amount of information on the deposition of pollutants by all processes;  con-
 sequently,  networks such as that of the U.S. National Atmospheric Deposition
Program (NADP) have emphasized the  importance  of obtaining data on both wet
and dry deposition rates and  amounts.
                                    8-12

-------
     TABLE 8-3.  CHARACTERISTICS OF THE  WORLD METEOROLOGICAL ORGANIZATION
                 (WMO)  AIR POLLUTION NETWORK (WHELPDALE 1979)
Program name:  WMO BACKGROUND  AIR POLLUTION  NETWORK.

Orgam' zati on/Country/Agency:   World Meteorological  Organization

Purpose:  to obtain, on a global  and regional  basis, background concentration
levels of atmospheric constituents, their variability and possible long-term
changes, from which the  influence  of  human  activities  on the composition of
the atmosphere can be judged.

Number of stations:   approximately 110.

Location:  in 72 countries throughout the  world.

Period of program:  from 1970  continuing indefinitely.

Collector type:   recommended  procedure is to use either open buckets during
periods of precipitation  only, or automatic precipitation  collectors with a
tight  seal.    Some  baseline  stations and  regional  stations  with  extended
programs  also  do air  and particulate   sampling  (procedures  are  not  yet
standard).

Parameters:   sample volume, $042-, ci", NH^,  Ca2+, Mg2+, Na2+, K+, N0a~,
             alkalinity or acidity, electrical  conductivity, pH.

Collection period:  1 month; some European  stations have adopted the 24-hour
sampling  period  of  the  Economic Commission  for Europe  (ECE)  Cooperative
Program for Monitoring  and  Evaluation of  the Long-Range Transmission of Air
Pollutants in Europe (EMEP).

Quality control;  U.S.  Environmental Protection Agency - sponsored reference
precipitation sample exchanges.

Contact:   Secretary  General,  World Meteorological  Organization,  Geneva,
Switzerland.   Directors, National  Meteorological  Services.

Data/Reports/References:   WMO  1974, WMO  Operations Manual  for Sampling and
Analysis Techniques for  Chemical  Constituents  in  Air  and Precipitation, WMO
No. 299, Geneva.

                       WMO/EPA/NOAA,  'Atmospheric  Turbidity and Precipitation
Chemistry Data  for  the  World1,  Environmental  Data Service,  NCC,  Asheville
(annually).
                                    8-13

-------
       TABLE 8-4.  CHARACTERISTICS OF THE ECONOMIC  COMMISSION  FOR  EUROPE
                 (ECE)  AIR POLLUTION  NETWORK  (WHELPDALE  1979)
Program  name:    COOPERATIVE  PROGRAM  FOR MONITORING  AND EVALUATION  OF THE
LONG-RANGE TRANSMISSION OF AIR  POLLUTANTS IN  EUROPE.

Organizatlon/Country/Agency:  Economic Commission  For  Europe.

Purpose:   to  provide  governments  with  information  on the  deposition and
concentration of air pollutants, as well  as  on  the  quantity  and  significance
of long-range transmission of pollutants and  fluxes  across boundaries.

Number  of stations;   operating  or  planned  by  1979  -  precipitation, 42;
aerosol, 52;  gas, 53 (- 1 station/105  km2).

Location:  Europe and Scandinavia

Period of program:   1977 to 1980 (first phase).

Collector  type:    for  precipitation:   open  polyethylene  gauges  and  some
automatic collectors;  for air:  pump  and bubbler going to  pump and  filter
pack; for particles: pump and bubbler  going  to pump  and filter pack.

Parameters:  precipitation: pH, $042-; optional  -  H+,  NOa", NH4+, Mg2+,
                            Na+, Cl",  Ca2+
             aerosol: S042-;  optional  - TSP,  H+, NH4+

             gas: S02;  optional - N02

Collection period:   24 hours

Quality control:  inter-laboratory sample exchange (NILU); laboratory  quality
assurance  programs;  statistical  analysis   of  data;  cation-anion balance,
acidity-pH checks.

Special  features:   (1) network  is  part of a  larger  program which includes
modeling, and comparison of field measurements and model  calculations;

                     (2)  some of these stations are  stations  in the EACN (see
Table 8-5) and  were stations in the  Long Range Transport of Air Pollutants
(LRTAP) network.

Contact:  H. Dovland, Norwegian Institute for Air Research (NILU),
          Box 130, 2001 Lillestri6m, Norway.
                                    8-14

-------
                             TABLE 8-4.   CONTINUED
Data/Reports/References:   ECE  1977,  Cooperative Program  for Monitoring  and
Evaluation  oftfieRing-Range  Transmission  of Air  Pollutants in  Europe  -
Recommendations of the ECE Task Force,  ECE/ENV/15,  Annexe  11,  10 pp.

     Data listings will  be published  regularly by NILU.
                                    8-15

-------
       TABLE 8-5.  CHARACTERISTICS OF THE  EUROPEAN  ATMOSPHERIC CHEMISTRY
                       NETWORK  (EACN)  (WHELPDALE  1979)
Program name:  EUROPEAN ATMOSPHERIC  CHEMISTRY  NETWORK  (EACN)

Organi zatl on/Country/Agency :    International Meteorological  Institute  (IMI),
Stockholm, Sweden.

Purpose:  initially, to study the transport from the atmosphere to the  ground
of  some  nutrients,  particularly  nitrogen.    It  now  has  a  more  general
atmospheric  chemistry  direction, including long-range  transport and  acidic
rain.

Number  of  stations:    a  maximum of  about 120  in  1959, currently  about 50
( ~ l station/iu5 km2).

Location:  Scandinavia  and western Europe.

Period of program:  started  in 1946  in Sweden,  expanded  to western Europe in
1955; continuing.

Collector type:   funnel  and bottle  thermostated to collect either  rain or
snow; automatic wet-only collectors  (Granat type, AAPS  type)  coming into  use.
Parameters;  precipitation amount,  pH,  conductance,  acidity,  SQtfi' , Cl",

             N03",  NH4+,  Na+,  K+, Ca2+, Mg2+,  HC03~-

Collection period:   1 month

Quality  control;     inter- laboratory  sample  exchanges;   laboratory  quality
assurance  programs;  cation-anion  balance, measured-calculated conductivity,
acidity-pH checks;  much analysis  of data.

Special features:   (1) supplementary measurement programs in Swedish part of
network examine network design aspects;

                    (2) several  sites are  equipped with air and particle  sam-
pling  systems,  primarily to  investigate  anthropogenic  acidity-related  phe-
nomena.

Contact;   L.  Granat, Department  of Meteorology,  University  of Stockholm,
Arrhenius Laboratory, S-106 91 Stockholm,  Sweden.

Data/Reports/References :    Granat,  L.,  1972, Deposition of  sulfate  and  acid
with  precipitation  over  northern  Europe,  Report  AC   20,  University of
Stockholm, Department of Meteorology/International Meteorological Institute,
Stockholm, 19 pp.
                                     8-16

-------
                             TABLE 8-5.   CONTINUED
                          Granat,  L.,  Soderlund,  R.  and  Back!in, L.,  1977,
The  IMI  Network  in Sweden.   Present  equipment  and  plans  for improvement,
Report AC40, University of Stockholm.

                          Granat,  L.,  1978,  Sulfate  in precipitation  as
observed by European  Atmospheric  Chemistry  Network,  Atmospheric Environment
12:413-424.

     Data  for  period  1955-59 published  in  Tellus by Eriksson.   Subsequent
data available from Granat.
                                   8-17

-------
In Chapter A-7, Section 7.3,  considerable attention  has  been  given  to methods
by which dry deposition fluxes  can  be  measured.   The techniques discussed are
those used  for  detailed  case studies  of deposition fluxes,  intended to pro-
vide  information  on  the processes  that contribute  to  the  net  transfer of
pollutants to the  surface, and usually designed to help formulate the depo-
sition process.  The  emphasis in Section 7.3 is on  trace gases and submicron
particles, which appear to be  of major  interest  in  the  context of  acidic and
acidifying deposition.   Few of  the  methods discussed are capable of long-term
routine operation. The  material that follows addresses similar questions, but
the present emphasis  will be on methods  suitable for long-term monitoring of
air pollution deposition fluxes either by direct measurement or by applica-
tion of the deposition  parameterizations resulting  from  the studies described
in  Chapter  A-7.   Many of  the comments  made earlier are equally  applicable
here.  Repetition will  be avoided as much as possible.

8.3.2  Methods for Monitoring Dry Deposition

Essentially two schools of thought  on monitoring dry deposition  exist.  The
first advocates the  use of  collecting  surfaces and  the  subsequent careful
chemical analysis of material  deposited  on  them.  For particles sufficiently
large that deposition is controlled by gravity,  surrogate surface and collec-
tion  vessels  have  obvious  applicability.  Furthermore,  they  provide samples
in  a  manner suitable  for chemical   analysis using   fairly conventional  tech-
niques.   Collecting  vessels  have  been  used  for generations in  studies of
dustfall;  standards  governing  the methods  used have  been  in place  for  a
considerable time (ASTM D 1739-70), and  intercomparisons between measurement
protocols  have  been  conducted (Foster  et  al.  1974a).    Collection vessels
gained considerable  popularity  following their  successful  use  in   studies of
radioactive  fallout  during  the  1950's  and  1960's.  For  some  gaseous  pol-
lutants,  species-specific  surrogate  surface techniques have  been  used to
evaluate  air  concentrations  rather than deposition fluxes.  Standards exist
concerning  sulfation plates  used  to  monitor  sulfur dioxide concentrations
(ASTM D  2010-65),  and  once  again  technique intercomparisons have been con-
ducted (Foster et al. 1974b).

The second school  of thought prefers  to infer deposition rates from routine
measurements  of air  concentration  of  the pollutants of concern and of rele-
vant  atmospheric  and surface quantities.   These inferential methods  assume
the  eventual   availability  of  accurate  deposition  velocities  suitable for
interpreting concentration measurements,  and they  assume  that accurate con-
centration measurements can  be made.   They  are  applicable in cases in which
deposition  is not controlled  by   gravity,  i.e., for  trace  gases  or small
particles.   They do  not provide samples as  convenient  for chemical analysis
as  do the  various surrogate  surface methods,  but they  do not  impose any
artificial  modification to  the detailed  nature  of the  surface  on  which
deposition is normally occurring.

Clearly,  a comprehensive  monitoring   program  would use  both concentration
monitoring  and  surrogate  surface   methods,  since  contributions  of neither
trace  gases nor  large particles   can be  rejected  on   the basis  of present
knowledge.
                                    8-18

-------
8.3.2.1   Direct Collection Procedures—There  is  no question that  the  depo-
sition  of large  particles is  adequately  monitored by  collection  devices
exposed  carefully over  the  surface of  interest.   Deposit gauges and  dust
buckets have been in  use  in  geochemistry for a considerable time,  and  their
use  is  well  accepted  for measuring the  rate of deposition  of  soil  and  other
airborne particles sufficiently large that their deposition  is   controlled by
gravity.   In  the era  of concern about radioactive  fallout, dustfall  buckets
were  used  to obtain  estimates of  radioactive  deposition,  especially of  so-
called  local  fallout  immediately  downwind of  explosions.   There was  much
concern about  how well deposited  particles  were  retained  within  collecting
vessels.   Some workers  used  water in the  bottom of collectors to minimize
resuspension of deposited material, and others  used various  sticky  substances
for  the same purpose.  It was recognized that the collection  vessels failed to
reproduce  the  microscale  roughness features of natural  surfaces.   However,
this  was  not viewed  as  a major  problem because  the  need  was  to  determine
upper limits on deposition so possible hazards  could be  assessed.

Much  farther  downwind,  so-called global   fallout  was shown  to  be  associated
with  submicron  particles similar to  those of interest in  the context  of acid
deposition.  However,  most of the distant radioactive fallout was transported
in the upper troposphere and lower stratosphere, and deposition was mainly by
rainfall.  The  acknowledged inadequacies of collection buckets  for  dry  depo-
sition collection of global fallout were of relatively  little concern  because
dry  fallout was a small fraction of the total  surface flux.

Special wet and dry collecting vessels were developed and  deployed  worldwide.
In their most highly-developed form, these devices employed  covers  that  moved
automatically  to  expose  a wet collection  bucket  when  precipitation  was  de-
tected and to cover it and expose a dry collection bucket  at all other times.
The  convenience and relative simplicity  of these  devices  have  contributed to
their  continued acceptance to this  day.   A major  factor that led to  their
general acceptance was the finding that dry and wet collection  buckets of the
same geometry provided answers that satisfied the  global budget of  strontium-
90 (Volchok et al. 1970).  However, as mentioned above,  worldwide radioactive
fallout was primarily delivered to  the surface  via  precipitation (as  much as
95 percent in some locations).   Consequently, an error of a factor  of two or
three  in  the  measurement  of the residual dry  deposition  component  might not
have been too obvious.

Concern regarding the meaning  of  collection-vessel  data is not only  recent.
Hewson (1951) comments that the limitations of  deposit gauges  are  like  those
of rain gauges.  Deposit gauges are  funnel-like collection  devices  that have
been used for generations.  They are familiar to most meteorologists,  and the
drawbacks involved are well known (Owens  1918,  Ashworth  1941).

Bucket dry deposition  data collected by  the NADP  have been  examined for evi-
dence  of  bird droppings  and locally  suspended soil particles  (Hicks 1982).
The  results  of chemical  analyses of  twice-monthly  dryfall  collections  were
examined for phosphate and calcium  concentrations.   High  levels of  phosphate
were considered  to  be evidence of  contamination  by guano, and calcium  was
used as an  indicator  of  soil-derived particles.  The data  indicate frequent
                                    8-19

-------
contamination of samples by bird droppings and by  soil  particles,  presumably
of local  origin.   It  is  obvious,  however,  that  relatively simple  remedial
steps can be taken.  Prongs arranged around collecting  vessels  can  be used  to
minimize  the  effects  of  perching  birds  and the  collectors  can  be  placed
sufficiently far  above the surface  that wind-blown soil  particles will  be
collected only under extreme conditions.

A recent  comparison  of collection devices  (Dolske and Gatz 1982)  indicates
that buckets of the kind  normally  used  in wet/dry collectors yield sulfate
dry deposition rates averaging about three times the values  provided by  flat
surrogate surfaces.  Hardy and Harley (1958)  report large differences between
radioactive fallout  dry deposition  rates to  buckets   and  other  artificial
collection devices and to natural  vegetation.

On all of the grounds  mentioned above, there is  reason to be concerned about
the use of  bucket  collection  devices for  studies  of acidic dry  deposition.
Surrogate surfaces  such  as flat, horizontal  plates,  share many of  the  con-
ceptual problems normally  associated with collection vessels,  yet appear  to
have  considerable  utility  in  some  special  circumstances (see Chapter  A-7,
Section 7.3).  For example, Lindberg and Harriss (1981) and Lindberg  et al.
(1982) show that the deposition of trace  metals  to surrogate surfaces mounted
within  a  forest canopy  is quite  similar to  the  deposition  to  individual
leaves, when expressed on a unit area basis.   Later work (Lindberg  and Lovett
1982) has extended these studies to  particle-associated sulfate,  nitrate, and
ammonium.   In general,  it seems that the  rates of deposition to  surrogate
surfaces are within  a  factor  of about two of the  rates measured to foliage
elements.  It is not yet clear how data concerning individual canopy elements
can be combined to evaluate the net removal  by a  canopy as a whole.

8.3.2.2  Alternative Methods—The acknowledged limitations  of  surrogate-sur-
face and  col lection-vessel  methods  for evaluating  dry  deposition  have caused
an  active search  for alternative  monitoring methods.    In general, these
alternative methods  have  been  applied  in studies  of specific  pollutants for
which  specially  accurate and/or rapid response  sensors are available.  The
aim  of these experiments  has  not been  to measure the  long-term  deposition
flux,  but instead  to develop  formulations  suitable   for  deriving  average
deposition  rates  from other,  more  easily obtained  information  such  as air
concentrations, wind speed, and vegetation characteristics.

Chapter  A-7  discusses  the processes  involved  and summarizes  a  number  of
recent  experimental  case  studies.   The  results  obtained in  these  detailed
experiments are  conveniently  expressed  in terms  of the  familiar  deposition
velocity, which  enables  deposition  fluxes to be deduced  directly  from meas-
urements  of  air concentration.   The  special  case  studies  are providing  a
rapidly expanding  body of  information concerning  the  factors  that determine
deposition  velocities.   Once  the  important  deposition  processes are form-
ulated  and  quantified, it  will  no  longer be necessary  to  measure  dry depo-
sition fluxes directly because measurements of atmospheric concentration  made
in  an  appropriate manner  could be  used  to infer  them.   This  philosophy has
formed the basis for monitoring networks in  Scandinavia  (Granat  et al. 1977)
and  in Canada  (Barrie et  al.  1980).    It  should be  noted  that using the
concentration-monitoring procedure  does  not remove completely  the  necessity


                                    8-20

-------
for conventional dustfall  monitoring because the purpose of the concentration
measurements is to  permit  evaluation of dry  deposition  rates only of  those
materials that do not fall  under the control of gravity.

Several  initiatives  are  underway to develop micrometeorological methods  for
monitoring the surface fluxes of particular pollutants.  Hicks  et  al.  (1980)
have  summarized a  range  of  potential  micrometerological  methods and  have
evaluated their potential  as routine monitors of dry deposition fluxes.   They
conclude that "at present, the most promising methods for monitoring  are eddy
accumulation, modified Bowen  ratio,  and variance."   The first of these  has
been of  special interest, because it offers the possibility of using  slowly-
responding chemical  monitors to deduce  deposition fluxes, bypassing the  usual
eddy-correlation requirement for a fast-response chemical sensor.   The method
compares air in updrafts with air in downdrafts (the former  having  slightly
lower concentrations of depositing  pollutants)  by  measuring each  in  separate
sampling systems.  Estimates of deposition  velocity are  readily obtained from
such  concentration  differences,  provided   the  samples   are  collected in  an
appropriate  manner.   The  method has  been  demonstrated for  meteorological
variables (e.g., sensible heat;  Desjardins  1977)  for which updraft/downdraft
differences  are large but  has yet to be  successfully  demonstrated  for  a
slowly depositing quantity.

The techniques loosely classified as "modified  Bowen  ratio"  all sidestep  the
need for direct measurement of  the pollutant  flux  itself by  relating  some
feature  of  pollutant concentration,  such   as  the  vertical  gradient or  the
concentration  variance  in  a  selected   frequency  band,   to  the same  charac-
teristic  of some better  understood quantity  for which  the flux is  known.
Easy  interpretation of  this   sort  of  information  requires  assumptions  of
similarity and of pollutant source and  sink distributions that are often hard
to verify, such as when researchers  are working over forests.  The method has
been  used  in  tests  involving carbon dioxide  (Allen  et al. 1974) and  ozone
(Leuning et al. 1979), but has yet to be used to monitor pollutant fluxes.

Methods  for  deducing fluxes of  atmospheric quantities  from measurements  of
the variance of their concentration  have been developed  and applied primarily
in studies of  the transfer  of sensible heat, moisture,   and momentum.   Tech-
niques of this kind might be especially attractive  for  some pollutants,  but
once again a successful  system has not  been demonstrated.  These three micro-
meteorological  methods  are  identified  by  Hicks et  al.  (1980) as "possibly
worthy for development for use in monitoring."  However, each imposes  special
sensor  requirements that  appear difficult to  satisfy.    Methods based  on
measurement of concentration variance require rapidly responding sensors with
low noise levels and linear response,  and  the  eddy accumulation and  modified
Bowen ratio  methods  involve the acccurate  measurement of  concentration  dif-
ferences on the order of 1 percent.

Attempts  to improve  sampling  by surrogate-surface  methods are  continuing.
Recent  comparisons  between  different   kinds  of  surfaces  and/or  collection
vessels  have  been  reported  by  Dolske and Gatz  (1982),  Dasch (1982),  and
Sickles  et  al. (1982).   Models  of  deposition  processes  are also being  im-
proved,  and  considerable  emphasis  is being given to the  role  of microscale
                                    8-21

-------
surface roughness  features  (e.g.,  in the model studies reported  by  Davidson
et al. 1982).  It must be expected that  the lessons  learned  in  such  modeling
exercises will be  used  to improve the similarity between  artificial  collec-
tion devices and natural  surfaces.

In some circumstances, deposition fluxes can be measured directly using  some
special  technique  unique to  the occasion.   Efforts  must be encouraged  to
compare fluxes determined by  any micrometeorological,  surrogate-surface,  or
collection vessel  technique to the  answers obtained in such special  situa-
tions,  which include  suitably calibrated  watersheds  (Eaton et al.  1978,
Dillon  et al. 1982),  snowpacks  and icefields (Dovland and Eliassen  1976;
Barrie and Walmsley 1978; Butler  et  al.  1980;  Section 8.5),  some  lakes,  and
mineral surfaces.

8.3.3  Evaluations of Dry Deposition  Rates

The paucity of accurate information  on dry deposition  rates to  natural  land-
scapes is a continuing problem to ecologists,  geochemists,  and meteorologists
alike.   Although  relatively  few data exist  on  which  to  base  estimates  of
deposition rates using the techniques outlined  above  (and explained in  detail
in Chapter A-7), it is appropriate to  consider in  some detail a  selected set
of information to  illustrate  the techniques  involved as  well  as to  derive
some  initial  estimates  of deposition  fluxes.   The  data  set  reported  by
Johnson et al. (1981)  has been selected for this purpose.   These data  were
obtained by using a limited  network  of particle samplers, modified to provide
aerosol samples  suitable  for  subsequent  analysis  by infrared spectroscopy.
The sites used were confined  to  the  northeast  quadrant of  the United States:
State College, PA;  Charlottesville, VA; Rockport,  IN; Upton,  Long  Island,  NY;
and Raquette Lake, NY.  Between  two  and  three  years  of data were  obtained at
each site,  starting  during  1977, except for  the Raquette Lake  site,  where
observations started late in  1978.   Size-resolved measurements  were made  of
sulfate,  nitrate,  ammonium,  and total  acidity  of  the aerosol.   For  the
present, main attention will be given to  the  three  chemical species.

A unique  feature of  the  Johnson et  al.  data set is  the fine time resolution
of the  data, designed specifically  to  enable detailed  analysis   of  rapidly
time-varying atmospheric phenomena.   Figures 7-12,  7-13, and 7-14  demonstrate
the inherent time dependence of the  factors that control dry deposition,  and
the resulting strong diurnal cycle of the depositional  flux.   The  data  set of
Johnson et  al.   permits  the  effects of  this  variability  to be   taken  into
account.

Figure 8-1 presents average diurnal  cycles of  sulfate,  nitrate, and  ammonium
in aerosol measured in the surface boundary layer (at  about  2 m   elevation),
     is  appreciated  that  these data  might be  influenced  by  sampling  dif-
 ficulties,  especially  for  ammonium  and  nitrate (see  Chapter A-5).    The
 intent here is to demonstrate the method by which deposition fluxes   can be
 evaluated from suitably detailed concentration data.  The  purpose  is  not to
 attempt to quantify the various fluxes in an  unequivocal  manner.


                                    8-22

-------
                 SULFATE
                     AMMONIUM
                                      NITRATE
    A
                  L J	L
0.9

0.8

0.7
1.4

1.2
            1.0 I  I   I  I
              3
              1
              3
                   I  II
                 PU
                   I  I   I
                   I  I   I
                            0.5
                 0.4
                                     	i
                0.5
                1.6

                1.2

                0.8
                1.2

                0.8

                0.4
                0.9

                0.7

                0.5
                                 J1
                                      iT
                                   I  I   I
                       I   I  i
                        il	I
0.2

0.1

  0
0.1
                                                   i  I  I
  0
0.3

0.2

0.1
0.2

0.1
                                                   i  I   I
                                        I  I  I
                                             0.2
                                             0.1
                                                    i  i  i
                 0   12  24
                     0    12   24
                     TIME OF DAY
                                     0    12    24
Figure 8-1.
Average diurnal cycles of near-surface concentrations of
sulfate, ammonium, and nitrate aerosol, as reported by
Johnson et al.  (1981) for rural sites located  at Raquette
Lake  (NY; A), Upton, Long Island (NY; B), Rockport (IN; C),
Charlottesville (VA; D), and State  College (PA; E).
Concentrations  are all in yg m~3
                    8-23

-------
as given  by Johnson et  al.  (1981).   Figure 8-2  shows  the average  diurnal
cycle of the  aerodynamic  resistance to transport  between  2  m elevation  and
the surface,  deduced from data  presented  by Hicks (1981)  for arid  grassland
(actually the Wangara meteorological experiment; see Clarke et al.  1971)  and
by Hicks  and Wesely (1980)  for transfer  to a pine  plantation.   These  two
examples are  selected  to  demonstrate  the  large  differences that  occur  in
atmospheric  transport  above  surfaces of  different  aerodynamic roughness.
Averages are  constructed  over  the  same time intervals  as were  used in  the
aerosol sampling program.

For the  aerosols under present  consideration,  surface  and/or canopy resis-
tances are  not  accurately known.   However, scrutiny of  Table  7-6   (Chapter
A-7)  and consideration  of the related discussion leads to  the  conclusion  that
a value of  about 1.5  s cm"1  is  likely to be appropriate for the pine plan-
tation case and  about  5  s  cnrl for  grassland.   It should  be  emphasized,
however, that considerable disagreement about these values  remains,  with  many
workers  preferring  to  continue  with the  approximation  0.1 cm  s~l for  the
deposition  velocity, regardless  of  the  nature of  the surface or the atmos-
sphere.  The  various arguments that are involved will not  be  discussed here.
Instead, we will apply the results  of the  experimental programs  and  overlook
the fact that many of the detailed deposition models  fail  to agree.

To estimate  deposition  velocities  suitable for  interpreting   the  data  of
Figure 8-1, we  must add  these  estimates  of  surface  resistance  to the time-
varying aerodynamic  resistances  of Figure 8-2, yielding  (as  the inverse  of
the resulting sums)  deposition  velocities  that have  a  small  diurnal varia-
tion,  averaging about 0.59  cm   s-1  for   the  pine  forest  and  about 0.17  cm
s"1 for the grassland.   It should  be noted,  in passing,  that the lack  of  a
strong diurnal  cycle of the deposition velocity  is  a direct consequence  of
the assumption  that the  surface resistance is relatively  large  but  constant
with time,  which  is known to be  erroneous for the  case  of trace  gas transfer
but  is presently assumed  for  particles  in  the  lack of  sufficient under-
standing  to  permit a  better  assumption,  notwithstanding the  evidence  of
Figure 7-15 (Chapter A-7).   Once  again,  it  is clear that surfaces of  dif-
ferent kinds will receive substantially different  dry deposition  fluxes.

Table  8-6   summarizes  the  deposition fluxes  evaluated  using the deposition
velocities  determined  above  and   the  diurnally-varying  concentrations  of
Figure  8-1.    It must  be  emphasized  that the  values   quoted  are  indeed
estimates;  several   potentially   important  factors  are  disregarded.    For
example, the  special  circumstances of snow cover  have  not been  considered.
The  evaluations  given  in   Table   8-6  are  intended  to  provide  realistic
estimates of  dry deposition  rates  to specific ecosystems  rather  than precise
determinations appropriate for detailed analysis.

Sheih  et al. (1979)  have combined deposition data  from many  experimental
sources  with  land-use  and meteorological  information to  produce deposition
velocity  "maps"  for sulfate aerosol.   Figure  8-3  (from  Masse  and  Voldner
1982)  is   a  recent extension   of  this   approach.   If  time-averaged  con-
centrations of   sulfate  in   air  near  the surface  are  known,  then  average
deposition  rates can be  estimated by using the  mean deposition velocities
illustrated in the diagram.


                                    8-24

-------
    c

    u

   oo
   UJ
   o
   oo
   1/1
   UJ
   o
   o
   o
   C£.
   UJ
                                                      o
                                                      I
                                      12


                                 TIME OF DAY
18
Figure 8-2.   Average diurnal variability of atmospheric resistance to
             pollutant transfer to the surface from convenient measuring
             heights above the surface, for the cases of a pine plantation
             (open circles), and grassland (solid circles).  Standard error
             bars are drawn wherever they are large enough to be visible.
                                   8-25

-------
       TABLE 8-6.   ESTIMATES OF AVERAGE DRY DEPOSITION LOADINGS TO
      AREAS OF  FOREST  AND GRASSLAND  IN THE NORTHEAST UNITED STATES,
      BASED ON SULFATE,  NITRATE, AND  AMMONIUM PARTICLE CONCENTRATION
                DATA REPORTED  BY JOHNSON ET AL. (1981).a
Location
Raquette Lake (NY)
Upton, Long Island (NY)
Rockport (IN)
Charlottesville (VA)
State College (PA)
Sulfur
(S04 - S)
0.7
(0.5)
0.2
(0.8)
0.4
(1.3)
0.3
(0.9)
0.2
(0.8)
Nitrogen
(N03 - N)
0.01
(0.03)
0.01
(0.03)
0.02
(0.07)
0.01
(0.03)
0.02
(0.05)
Nitrogen
(NH4 - N)
0.2
(0.6)
0.3
(1.0)
0.6
(2.0)
0.3
(1.2)
0.3
(1.0)
aThe particle size  range  measured was  0.3  to 1.0  ym diameter.   Fluxes to
  forests  are  given  in brackets.    Units  are  kg  ha-1  yr1  of elemental
  sulfur and nitrogen delivered by  each chemical  species.   Note  that these
  flux  estimates  are  based  on  preliminary  data,  including  rather crude
  evaluations of appropriate  deposition  velocities.   Errors on the order of a
  factor of two must be  expected.
                                   8-26

-------
As mentioned above, biological factors play an  important  role  in  determining
deposition velocities appropriate for the deposition of trace  gases.   Stoma-
tal  resistance  to  sulfur  dioxide transfer can vary by more than  an order  of
magnitude  between  day and  night (see Chamberlain  1980,  for  example).    In
consequence, exceedingly strong diurnal  cycles of deposition must  be expected
and  interpretation of  trace  gas  concentration  data  obtained   over  long
averaging  times might be quite difficult.  At this  time,  we lack  rural  trace
gas  concentration data that can  be  used  to  illustrate this point.  However,
the  difficulties involved can  be illustrated  by the conceptual example  of a
situation  in which  the  atmosphere aloft  supplies some  trace  gas  to  surface
air  at  a constant rate, with  concentrations  building at night when  surface
deposition is  prohibited  by biological  factors.   In  daytime, the vegetated
surface  will  act as an efficient sink  and  airborne concentrations near the
surface will  be  reduced.   In  this situation,  measurements of  nighttime  con-
centrations are essentially irrelevant to  depositional  flux calculations, yet
they contribute  most  of the  impact on average  air quality that may be  of
considerable importance  for other reasons.

Figure 8-4  (also from Masse and Voldner  1982)  shows isopleths of estimated
sulfur dioxide deposition  velocity for eastern North America.   The diagram  is
derived by combining land-use  descriptions with meteorological  and biological
factors, as  in  the case  of Figure 8-3  for  sulfate aerosol.   The analysis
follows initial work reported  by Sheih et  al.  (1979).   Both of  the deposition
velocity maps  reproduced  here  provide  estimates  typical of  conditions  in
April.  At other times,  different distributions of deposition velocity apply.

At this  time, no monitoring program in the United States  reports  air  concen-
trations of pollutants in  a manner such that dry deposition fluxes of acidic
and acidifying pollutants  can  be readily evaluated,  although several networks
offering suitable  information  have operated  for limited periods (see  Hidy
1982, and  see  Figure  8-5).   Such networks  are in operation elsewhere,  par-
ticularly in Scandinavia (Granat et al.  1977) and  in  Canada  (Barrie  et al.
1980).   A  wet-chemical  device is  used  in the Scandinavian network,  whereas
filter-packs are used in the  Canadian.   No  measurement method permits  accu-
rate measurement of all  of  the trace  gases and  small  particles of importance
in  the  context  of acid  precipitation.   Sampling  artifacts  are discussed
elsewhere  in  this  document,  as  are problems  associated with   isokinetic
sampling of particles.  Furthermore,  it  is  obvious that  the  quality  of dry
deposition data evaluation from  any such  concentration information is at the
mercy of the deposition velocity  assumptions made as the  intermediate steps.
If the  need  exists  for  accurate evaluations  of average dry deposition  rates
of gases and small particles, then  it seems  necessary to place almost  equal
emphasis on the requirements for accurate  concentration data and for reliable
and appropriate deposition velocity evaluations.  At  the same time,  it  must
be remembered that none  of the various methods for interpreting concentration
data is  intended for use in the  case  of large particles  that  fall under the
influence  of  gravity.   In  this  particular case,  use of collection  devices
remains an obvious preference.
                                    8-28

-------
        DRY DEPOSITION VELOCITY OF S02 FOR APRIL (cm s'1)
                                                    0.1 -  0.3

                                                    0.4 -  0.5

                                                    0.6 -0.7

                                                    0.8 -1.0
Figure 8-4.   Calculated deposition velocities  appropriate  for  sulfur
             dioxide over eastern North  America.   Adapted  from Masse
             and Voldner (1982).

                                  8-29

-------
                                                        1-HOUR
                                                     S02  (ppb)
                                                      24-HOUR

                                                    2-  (W) T3
Figure 8-5.   Examples  of pollution  concentration  isopleth  information
             of the kind suitable  for  applying  deposition  velocity
             maps such as in  Figures 8-3  and  8-4.   Shown are  the
             arithmetic (for  sulfur dioxide)  and  geometric (for sulfate)
             means of  data obtained during  5  months  between August  1977
             and July  1978.   Adapted from Hilst et  al.  (1981).

                                  8-30

-------
8.4  WET DEPOSITION NETWORK DATA WITH APPLICATIONS TO SELECTED  PROBLEMS
     (6. J. Stensland)

8.4.1  Spatial Patterns

There  is  a  vast amount of  precipitation chemistry  data  available.    This
section will  discuss  the general  spatial  patterns for the  United  States and
Canada.   The  first  set of  contour  maps will  be based  on  data from  the
National  Atmospheric  Deposition Program  (NADP).  Although data  from  other
recent networks  could have been included,  this would not  have altered  the
general patterns and could have added some additional uncertainties since,
for example,  sampling  intervals  other than weekly were us.ed.   At this  time
the NADP is the only network with sites throughout the United States  and thus
the NADP data will allow for comparisons between the  West  and the East,  where
the acidic precipitation problem is generally  perceived to  occur.  New  sites
are currently being added, as part of  the National Trends Network, that will
increase site density in the West.

Concentration and deposition maps will  be presented,  with the  contours  drawn
by hand instead  of  by computer.   Different  objective analysis  and  computer
plotting packages do not produce identical contour maps.   Likewise  hand-drawn
contour maps  are somewhat subjective  and  thus, will  not  be  identical   when
drawn by different analysts.  Since data  values will  be shown  on the contour
maps in this section, the reader can determine  if  he  agrees  with the contour
shapes.   Sites  with  only a  few  samples  can  produce  "bulls-eye"  contour
patterns;  this  effect has been  minimized by  using  the  hand-drawn  contours
instead  of  computer-produced   contours.    Because  there  are  year-to-year
variations in the average site  concentrations  of the  ions  it would  be best in
determining the  general  spatial  patterns  to include only sites  with several
years of data.   However, at this time we do  not  have enough  data  to  adopt
this rule.  Therefore for  the hand-produced contours  in this section, we did
not  try  to  precisely contour  the  site  data  values  but  instead did   some
subjective smoothing.

For  some  ions both the  weighted-mean  concentrations  and  the  median concen-
trations will be included to allow  for a comparison of these two measures  of
central tendency.  For sites with a relatively small  total  sample  number the
median probably gives a better  estimate of central  tendency  than  the  weighted
means because in the latter, one or two samples with  unusually large volumes
can produce  unreasonably  large  weighted means.  No corrections  for  sea-salt
influences have been made for the NADP data shown in  this  chapter.

For the combined picture of  the United States  and Canada,  data  maps adapted
from the  U.S./Canada  Memorandum of  Intent (MOD  report  (U.S./Canada  1982)
were used.   In  the MOI  report  only  1980 data were  used, and therefore  the
reader has yet another type of  contour map for purposes of comparison.

For  many  studies related  to effects  annual   deposition  values  are needed.
Other  chapters  in this  document may  have selected  deposition  values   from
monitoring networks  which  provided greater  space densities in  the  area  of
concern as well  as  longer time records.  These data can  be compared to  the
1980 deposition maps included in this  chapter.   Some maps have been  included


                                   8-31

-------
in this chapter for specific use in effects  studies,  an example being the wet
deposition nitrogen map  which  includes both nitrate  and  ammonium inputs of
nitrogen.

The National  Atmospheric  Deposition Program (NADP) began  in  July 1978.   By
October 1978, 20 sites were operating, mostly  in  the Northeast.   Figure 8-6
shows the  number  of weekly samples as of approximately the  end  of 1980 for
weeks when  at  least 0.02  inches  of liquid  equivalent  precipitation  was
collected  (NADP  1978, 1979,  1980).   The  data were  screened  at  the NADP
Central  Analytical  Laboratory to remove data for samples that were obviously
contaminated or collected  by  nonstandard procedures.  The quantity  of data
varies  from  6 weekly  samples  for  a  California site  to  128  for  the West
Virginia site.

Figure 8-7 shows the  median concentration contour  pattern for sulfate.  The
low site  density  in  some  areas and  the short data  record for  some  sites
suggest that the  depicted  patterns will  be subject to change as more data
become  available.   The  medians  displayed  on  the contour  map   are   better
indicators of central  tendency  for small  data sets  than are other  statistical
parameters.   The  site data  values are  shown  on  the  maps to indicate  the
degree of  subjective  smoothing  involved  in drawing  the  contour  lines.  For
example  the  2.0   mg  £-1   contour  line in   Figure  8-7,  cutting  through
northern Wisconsin, could have  been placed  further north  to accommodate the
2.2 mg  £-1  value  at the  Isle  Royale National  Park  site.   However, from
Figure  8-6  one  notes that the  2.2  mg  £-!   value  is the  median of only
eight values and  thus can  not  be considered  very  reliable.   The 2.0  mg
Cl contour  line  passes  through  the  north-central Wisconsin site  having  a
median  value of 1.3  mg  Jr1  illustrating  that a subjective decision  was
made to show rather smooth contour  lines instead of lines  bent to match each
site value.  On most of the maps  in this  section,  contour  lines  to the left
of an  imaginary  line from  northwestern  North  Dakota  to  southeastern  Texas
have been dashed to indicate that in these areas the site density and  length
of data record are  such that the contour  lines  probably do  not well represent
the true patterns.

Sulfate in precipitation has  a strong  seasonal  pattern  for  sites in  the
Northeast  (Bowersox and dePena  1980,  Pack and  Pack 1980,  Pack 1982).   Thus,
several  years of data will  be  required before  a very  stable  annual  average
pattern can  be expected.   Figure  6-15  in Chapter  A-6  shows the seasonal
pattern for  sulfate and  also  indicates  the  great variability  among  event
samples for sulfate and nitrate  at the Pennsyvania  State MAP3S site.

Consistent with the known  emission  pattern for sulfur dioxide,  the   higher
sulfate concentrations in  Figure  8-7  are  in  the  Northeast.    The  contour
values decrease eastward across New York and  New England.  The limited data
for Arizona  show  a  sulfate maximum  in  the Southwest.    Because  a  similar
maximum is present in  the calcium  map  (see Figure 8-11), soil dust is thought
to be the major source for  this maximum.   Possible sample evaporation  after
collection  or enhanced  raindrop  evaporation   must  also  be considered  as
partial  explanations  for the  high concentrations of  all  the  ions in  the
precipitation of the  Southwest.   The  arid  site at Bishop, CA,  also  has an
extremely  large  sulfate  value,  but only six  samples are available.    The


                                    8-32

-------
                               %--/,;.  *  "l-fi^'l

                                " "<.    X.26.    /-^'X'128/
Figure 8-6.  Map of National Atmospheric Deposition  Program  site
             locations and number of wet deposition  samples  for
             each site through approximately December  1980  (using
             data from NADP 1978, 1979, and 1980).
                                  8-33

-------
Figure 8-7.   Map of median  sulfate  concentrations  (mg  r1 as SO^-)
             for NADP wet deposition  samples through approximately
             December 1980  (using data  from NADP 1978, 1979, and 1980)
                                 8-34

-------
sample-volume-weighted-average sulfate values  shown in Figure  8-8 are gen-
erally similar to those for the median  values  (but not  true  for  Bishop, CA).

Pack (1980) found  the  MAP3S and  EPRI-SURE precipitation chemistry data from
August  1978  to  June 1979  to be  comparable.   The precipitation-weighted-
average  sulfate  values  in  an  area  from  central   Illinois  to  western
Massachusetts  were  2.9  mg  £-1  or  greater.   The maximum  sulfate  values
were 3.3,  3.4, and  3.7  mg £-1  for  three  sites  in Ohio  and  Pennsylvania.
The  five  NADP  sites in Ohio  and  Pennsylvania have sample-volume-weighted-
average concentrations of  3.3,  3.5,  3.6, 3.7,  and 4.0 mg  -1  for the data
record  indicated  in Figure  8-8.   These  values  are  very  similar to those
reported by Pack.

Figure 8-9 shows the nitrate pattern, which has general similarities to that
for  sulfate.   Again the  higher  values in  the  northeastern quadrant  of  the
United  States  are  consistent with  the  known  anthropogenic  NOx emission
pattern.  One difference is that  in Figure 8-9 the values  in South  Dakota  and
Nebraska are about the  same as those  in Ohio but this is not true for sulfate
in Figure 8-7.   The rather high  nitrate values at  the  upper  plains sites  do
not  seem  to  be consistent with known  anthropogenic combustion NOX sources.
The nitrate maximum in  east central California is  questionable because of  the
small number of samples (see Figure 8-6).  Recent  research has indicated that
most of  the  available  air quality data for nitrate in the Northeast are  of
limited value  because  of  sampling  problems  (Spicer  and  Schumacher  1977);
therefore,  the  precipitation  nitrate  data  patterns become  increasingly
important.

Figure 8-10  shows the contour pattern for  the ammonium  ion.   The general
pattern has  some similarities to  that for  nitrate in Figure  8-9.   As  for
nitrate,  the  values for the  northwest Indiana site  are  elevated, probably
indicating the effect  of the  upwind industrial areas.  There  is  a definite
maximum in the upper plains, probably due  to ammonia emissions from livestock
production.   In  particular, there are several  large cattle  feedlots in  the
vicinity  of  the  Nebraska  site.    The  site  just  east  of  Lake Ontario  had
elevated  values  for both  ammonium  and  nitrate  but  only  17  samples were
available (see Figure 8-6).  The  ammonium  values are lowest  in the  Northwest;
the median values of 0.02 are analytical  detection limit values.

Figure 8-11 shows the calcium concentration  pattern, the values  for which  are
very high  in  the Southwest and relatively  high in the upper Plains.   Dust
from  soils  and unpaved  roads  probably accounts  for  the  generally elevated
calcium levels in  the  central  United States.   Urban and industrial sources
may  account  for  the relatively  high  values at  the site  in Indiana.    The
central Illinois  site  with  a median value  of 0.28 mg  &'1  is an   example  of
a site surrounded by an area of Intensive  cultivation,  with  corn and soybeans
being the major crops in the area.  The median calcium  concentration there is
surprisingly  low,  considering the  surroundings,  and  indicates  that  the
sampler is quite  successful  in preventing  dust leakage into the  collection
vessel.

Figure 8-12  shows the  chloride concentration  pattern.   Sites closer to  the
major chloride source,  the sea, have  higher  levels.


                                   8-35

-------
                                                        6 /<
                                            ^--7-^3.7  «jn
                                            '4.5 «3.5 • ,'.  _-r
                                                          1.7
Figure 8-8.   Map of volume-weighted-average-sulfate concentrations

             (mg rl as SO^-)  for NADP wet deposition samples

             through approximately December 1980 (using data from

             NADP 1978, 1979, and 1980).
                                 8-36

-------
Figure 8-9.   Map of median  nitrate  concentrations  (mg  rl as N03~)
             for NADP wet deposition  samples  through approximately
             December 1980  (using data  from NADP 1978,  1979, and 1980)
                                8-37

-------
Figure 8-10.
Map of median ammonium ion concentrations  (mg  r^  as
NH4+) for NADP wet deposition samples  through  approxi-
mately December 1980 (using data from  NADP 1978,  1979,
and 1980).
                                 8-38

-------
Figure 8-11.
Map of median calcium concentrations (mg rl)  for NADP
wet deposition samples through approximately December
1980 (using data from NADP 1978, 1979, and  1980).
                                 8-39

-------
Figure 8-12.
Map of median chloride concentrations (mg r*)  for NADP
wet deposition samples through approximately December
1980 (using data from NADP 1978, 1979,  and 1980).
                                8-40

-------
 In  addition to the  ions displayed  in  Figures 8-7 through 8-12,  magnesium,
 potassium,  and sodium are measured in NADP and most other networks.  The data
 in  Table 8-7  demonstrate  the relative importance of  all  the ions  at  three
 NADP  sites.   The  concentrations in  Table  8-7 are expressed  in  microequiva-
 lents  per  liter in order to allow a direct evaluation  of the  contribution of
.each  ion to the anion or cation sum.  If all  ions  were  being  measured and if
 there  were no  analytical  uncertainty, then  the  anion  sum would equal  the
 cation  sum.  In Table  8-7,  the values for  hydrogen  ion concentration,  H+,
 were  calculated  from  the  measured  median  pH  value,  and  the  values  for
 bicarbonate,  HC03~,   were  calculated  by   assuming that  the   sample was  in
 equilibrium with  atmospheric  carbon  dioxide.    Although  the  sulfate  and
 nitrate  levels shown are  similar at  the  MN  and  NY   sites,  the  pH  differs
 greatly  due to the much higher levels of  the ammonium,  calcium,  magnesium,
 sodium,  and potassium ions at  the  Minnesota  site.   These ions are  frequently
 associated  with basic  compounds.    The data in Table  8-7  suggest  that the
 concentrations of all the major ions must be considered if the time and  space
 patterns  of pH are  to   be  fully  understood.   Currently sites in Ohio, New
 York,  Pennsylvania, and  West Virginia have the feature shown for the  New York
 site   in  Table  8-7  where  H+,  SC^2',  and   NCh~  are  the  dominant  ions.
 For  the New York site,  the acidity (H+)   could  be 98 percent accounted for
 if  all  the SCty2- had   been  sulfuric  acid  while nitrate,  as nitric  acid,
 could  have accounted  for about  55 percent  of  the  acidity.   By  applying
 multiple   linear   regression   analysis,   Bowersox   and  dePena  (1980)   have
 concluded  for  a central  Pennsylvania site that on the  average the principal
 contributor to (H+)  is   sulfuric acid, but  the acidity in snow is  determined
 principally by nitric acid.

 Figure  8-13 shows the  median  pH  from the  NADP  data.    Except  in  Minnesota,
 western  Wisconsin, and  southern Florida,  the  region east of  the Mississippi
 River  has median pH values less than 5.0,  while the Northeast has values less
 than  4.7.   The pH data  are frequently reported as the pH calculated from the
 sample-volume-weighted  hydrogen ion  concentration, which  will  be  referred to
 as  the  weighted  pH   values  in this  chapter.   When  weighted pH  values are
 considered, the  Northeast  still  has average  pH  values  less  than  4.7.
 However,  the weighted  pH values  at the Nebraska  and  southwestern Minnesota
 sites  are 4.95 and 5.14, respectively,  compared  to median values of  5.95 and
 6.19.   Therefore, the averaging procedure  needs  to be specified in  detailed
 analyses and comparisons of pH patterns.

 Figures 8-14 through 8-23  show data consolidated  for the  single year 1980
 from  NADP,  MAP3S,  and CANSAP, as  well  as  the APN  and Ontario Ministry of the
 Environment (OME) networks  (Barrie and  Sirois  1982,  Barrie et  al.  1982).
 Site  data  were  included in  the analysis if the site had been in operation for
 at  least two-thirds  of  the year.   For the CANSAP and  MAP3S  sites,  precipi-
 tation-weighted-average concentrations  were  calculated  and  used  in  the
 figures.   For NADP  sites,  sample  volume-weighted-average concentrations were
 used.   Deposition values were calculated by multiplying the concentrations by
 the  1980  precipitation  amounts.    Contour lines  of  ion concentrations and
 depositions were  drawn  by  hand.   The structure in the concentration contours
 indicates  that all site values  were assumed to be equally valid or represen-
 tative.   The authors elected to not draw contour lines in the western United
 States due to the small number of sites.   The contour  lines for  deposition


                                     8-41

-------
        TABLE 8-7.   MEDIAN  ION  CONCENTRATIONS FOR 1979 FOR THREE
                           NADP SITES  (yeq JT1)
No. Samples
so42-
NOs"
cr
HC03~ (calculated)
Anions
NH4+
Ca2+
Mg2+
K+
Na+
H+
Cations
Median pH
42
38.9
.6
8.2
0.3
59.0
5.5
5.0
2.4
0.7
17.6
17.8
49.3
4.75
37
45.8
24.2
4.2
10.3
84.5
37.7
28.9
6.1
2.0
13.7
0.5
88.9
6.31
NYC
49
44.8
25.0
4.2
0.1
74.1
8.3
6.5
1.9
0.4
4.9
45.7
67.7
4.34
 The Georgia Station site in  west central Georgia.
bThe Lamberton site in southwest Minnesota.
cThe Huntington Wildlife site in northeastern  New York
                                   8-42

-------
Figure 8-13.   Map of median pH for NADP wet deposition  samples  through
              approximately December 1980  (using  data from  NADP 1978,
              1979,  and 1980).
                                8-43

-------
have more  structure than appears  justified.   This  resulted  from using the
concentration  field to  calculate  deposition values  at  the  250  Class  I
Canadian weather service sites and  on  a 100 km x  100  km grid in the United
States.  Thus,  the greater density  of weather sites  that measure  precipi-
tation amounts resulted in more structure in the deposition contours than  if
the precipitation amounts at the  smaller  number  of chemistry  sites had been
used.   The maps by Barrie et al.  (1982) were presented  with  the units  of
millimoles per  liter  and millimoles  per  square meter.    For  this chapter,
sites values were converted to the units  shown in  Figures  8-14  through 8-23;
the published contour lines are  used,  but  they  have been  redrawn.

Figure 8-14 shows data for  sulfate.   The Canadian sulfate data were corrected
for sea salt but the U.S. data were  not.   Corrections  for  sulfate  are gener-
ally negligible (< 5 percent) except at locations  within 5 km of open ocean
areas (Barrie et al. 1982).   The  general pattern  for sulfate in  the Northeast
is  similar  in  Figures 8-8  and  8-14.   However,  by  carefully  comparing the
location of  the 1.9  and 2.9 mg  £-! contours in  Figure 8-14  with  the 2.0
and 3.0  mg a'*-  contours in  Figure 8-8,   we  note  that  spatial differences
of more than 200  kilometers are  sometimes evident.  In  central  Illinois and
western New  York  the  NADP   and MAP3S  sulfate  values differ by  more  than  25
percent.   For  western New  York,  the NADP  and MAP3S sampling locations are
about 25 miles  apart.   In  the MAP3S program,  very small precipitation sam-
ples, which  generally  have high ion concentrations, are not analyzed.  The
actual  reasons for  the rather large  differences  in 1980  sulfate ion concen-
trations at  these  two locations are not  known  and would require a  detailed
study.

Figure 8-15 shows  the 1980  nitrate  concentration  pattern.   The  high nitrate
values in the western plains of  Canada  are attributed to  wind-blown dust.   In
the east the highest values are  in southern  Ontario.  The notch  in  the 1.9  mg
r* contour  in Pennsylvania  and  New York  might be rather  important  if  it
is  real.   Such features should be  useful in relating emission patterns  to
acid precipitation  patterns.   However, at  this  time,  the fine  structure  in
the  sulfate  and nitrate  patterns is  unreliable.    The  uncertainty  in the
location of the contour lines for different areas,  averaging times, averaging
procedures, site densities, and networks  has  not been  determined.  The cor-
relative evidence  for  a  general  link between known  emission  sources and the
composition of precipitation is, however,  convincing.  When quality  data are
available for a sufficiently long period of time  and the  uncertainties  in the
placement of the contour lines  are established,  it  may  then  be possible  to
use  such  patterns  to answer more  specific  questions such as transport dis-
tances and scavenging mechanisms.

Figure 8-16 displays the 1980 ammonium pattern.   The very high concentrations
observed in Figure 8-10 are not found in Canada.

Figures 8-17 and 8-18 show the weighted  pH and  hydrogen ion  concentrations.
The  lowest  pH values  are  found  in   Ohio,  Pennsylvania,  New  York,  West
Virginia,  and  southern Ontario.   The 5.0  contour line  through the. central
United  States  is  peculiar  to  the weighted-averaging  procedure as was dis-
cussed in relation to Figure 8-13.  The area in the United  States enclosed  by
                                    8-44

-------
                                                          ..*
CANADA
 • CANSAP
 • APN
 AOME
UNITED STATES
 • NADP
 • MAP3S
                                                                             1,0
1980   CS02-
     Figure 8-14.   Weighted-average-sulfate  ion  concentrations for 1980,
                   for wet deposition  samples  (mg  r1).  Adapted from
                   Barrie et al.  (1982).
                                      8-45

-------
CANADA
 • CANSAP
 • APN
 A ONE
UNITED STATES
 • NADP
 • MAP3S
1980  CNO-
     Figure 8-15.  Weighted-average-nitrate ion concentrations  for  1980,
                   for wet deposition samples (mg  rl).  Adapted  from
                   Barrie et al.  (1982).
                                      8-46

-------
           V   '	   '
            \r^    "-1	«P
             »       I
CANADA
 • CANSAP
 • APN
 A ONE
UNITED STATES
 • NADP
 • MAP3S
     Figure 8-16.   Weighted-average-ammonium ion concentrations for 1980,
                   for wet deposition samples (mg r1).   Adapted from
                   Barrie et al.  (1982).
                                      8-47

-------
CANADA      UNITED STATES
  • CANSAP    • NADP
  • APN        • MAP3S
  ACME
    Figure 8-17.   pH from weighted-average-hydrogen  ion  concentration for
                  1980,  for wet deposition  samples.  Adapted  from Barrie
                  et al.  (1982).
                                    8-48

-------
CANADA
  • CANSAP
  • APN
  A ONE
UNITED STATES
  • NADP
  • MAP3S
    Figure 8-18.  Weighted-average-hydrogen ion concentrations for 1980,
                  for wet deposition samples (peq r1).   Adapted from
                  Barrie et al. (1982).
                                     8-49

-------
the 4.2  contour  line is substantially larger  in  Figure  8-17 as compared to
Figure 8-13. The larger area  of intense acidity in Figure 8-17 is due to the
pH values of 4.17 and 4.20  in Illinois.   The  pH values  in Ohio in Figure 8-17
are lower than those  in  Figure 8-13.   The data in  Table  8-8 provide a com-
parison between 1979 and 1980 and between median and weighted pH values.  The
weighted pH values for sites  in Table  8-8 are on the average  about 0.07 units
lower than  the median  values.  On the  average,  the 1980 median pH values are
0.07 unit lower  than  the 1979  values; the  1980 weighted  pH  values  are 0.10
unit lower.  So both the year-to-year  variation and  the choice of weighted pH
instead of median pH contribute to the apparent larger and more intense area
of acidity  in  the United States in 1980 compared to 1979.   It is useful to
remember that a decrease of 0.10 pH unit  corresponds to a 26  percent  increase
in acidity (free hydrogen ion concentration).

Figures  8-19  to  8-23  depict wet  deposition for  1980.   The wet deposition
patterns  are  probably  more  variable  from  year-to-year  than concentration
patterns because of the added variability of  annual  precipitation patterns.

The  variability  in  concentration between  weekly precipitation  chemistry
samples  for  eight sites distributed  across  the  United  States is  shown in
Table 8-9.  The  negative values in the  table  represent analytical detection
limit concentrations.  The 90 percentile values for each site divided by the
corresponding  10  percentile  values  have  the  following  mean  and   standard
deviation values:

            Na+:   27.8 +_  8.5                CT:   7.3 +_  3.0

           NH4+:   24.8 +_ 14.3                NOs':   7.1^1.9

             K+:   18.2^12.3              S042~:   6.4^2.0

           Ca2+:   11.9 _+  4.6         conductance:   4.6 +_  1.0

           Mg2+:   11.5+_  3.4                  pH:  1.28 +_  .09

For  H+,  the inverse  ratio values (i.e., 10 percentile  divided by  90  per-
centile)  are  24.6  +_  16.1.    The  detection  limit  values were  used in the
calculations so  the  resulting  ratio  values  are lower  limits  (applies mainly
to  NH4+).    In   summary,  then,  the  cations   are  more  variable  than  the
anions or conductance.

8.4.2  Remote Site pH Data

Galloway  et al.   (1982)  have reported precipitation chemistry  data for the
five  remote sites listed  in Table 8-10.   The average  pH  values listed  in
Table 8-10  vary  from 4.78  to 4.96, far below the often used  reference value
of 5.6.   The  samples  were  collected within 24  hours after a  storm had  ended.
At sites where bulk deposition was sampled,  the collectors were  installed for
a maximum of 24  hours  before an event began  in order  to minimize  dry deposi-
tion amounts.  Galloway et al. noted that previous research at the San  Carlos
location had indicated that  the  precipitation  was acidic  (Clark et  al. 1980,
Herrera  1979,  Jordan  et al.  1980).   However,  since samples  analyzed for


                                    8-50

-------
TABLE 8-8.  NUMBER OF WEEKLY SAMPLES (N)  AND
     AVERAGE pH VALUES FOR 1979 AND 1980
1979


Bondville, IL
Salem, IL
Delaware, OH
Cal dwell, OH
Wooster, OH
N

32
-
49
44
45
Median
pH
4.34
-
4.34
4.22
4.29
Weighted
PH
4.35
-
4.25
4.15
4.25
N

38
23
45
44
44
1980
Median
PH
4.29
4.33
4.15
4.15
4.21
Weighted
pH
4.17
4.20
4.11
4.08
4.17
                     8-51

-------
CANADA      UNITED STATES
  • CANSAP    • NADP
  • APN        • MAP3S
  AOME
    Figure 8-19.  Sulfate ion deposition for 1980 for wet deposition
                  samples (kg ha-1).   Adapted from Barrie et  al.  (1982).
                                     8-52

-------
                                    £9-8
'(2861)  ' LB l
   uoj.q.Lsodap
Q86T
batu)
UOL
                                                             '02-8 9-tn6j.j
+HQ 0861
                       3WO*
               '        NdV"
          ddVN •     dVSNVO •
     S31V1S Q31INH      VQVNV3

                     QN3931

-------
                                                        „*•
CANADA
  • CANSAP
  • APN
  AOME
UNITED STATES
  • NADP
  • MAP3S
1980 DNO-
    Figure 8-21.  Nitrate ion deposition for 1980  for wet deposition
                 samples (kg ha-1).  Adapted from Barrie et al. (1982).
                                   8-54

-------
LEGEND
CANADA
  • CANSAP
  • APN
  A ONE
            UNITED STATES
              • NADP
              • MAP3S
    Figure  8-22.  Ammonium ion deposition for 1980  for wet deposition
                 samples (kg ha-1).  Adapted from  Barrie et al. (1982).
                                    8-55

-------
CANADA
  • CANSAP
  • APN
    ONE
UNITED STATES
  • NADP
  • MAP3S
    Figure 8-23.  Total nitrogen deposition (calculated from nitrate and
                  ammonium deposition) for wet deposition samples (kg ha~l).
                  Adapted from Barrie et al.  (1982).
                                     8-56

-------
                        TABLE  8-9.   TEN, FIFTY,  AND  NINETY  PERCENTILE  ION CONCENTRATIONS (mg   r1),
                                             pH,  AND CONDUCTANCE FOR EIGHT  NADP  SITES3
00
I
en
Sites
ME-MEC


ME-NY


WV


GA


Central JL


N-MN


NE-CO


NW-OR


Percentlles (Ca2+) (ng2+)
10
50
90
10
50
90
10
50
90
10
50
90
10
50
90
10
50
90
10
50
90
10
50
90
.04
.12
.36
.04
.13
.45
.08
.25
.78
.04
.10
.42
.06
.28
.98
.09
.29
1.04
.10
.43
2.08
.05
,17
.31
.006
.020
.071
.009
.022
.090
.010
.030
.080
.013
.030
.134
.011
.035
.143
.016
.043
.183
.013
.052
.245
.012
.036
.106
(K+)
-.002
.015
.049
.005
.018
.050
.014
.035
.084
.005
.027
.124
.007
.027
.094
.017
.044
.154
.009
.076
.391
.010
.033
.144
(Ma+)
.018
.088
.707
.017
.081
.623
.025
.100
.650
.065
.278
1.291
.015
.065
.195
.032
.139
1.014
.043
.189
1.222
.077
.288
2.150
(NH4+)
-.02
.08
.38
-.02
.21
.64
-.02
.21
.65
-.02
.11
.55
.16
.42
1.18
-.02
.30
1.01
.11
.68
2.51
-.02
.04
.14
(N03~)
.31
1.08
2.52
.58
1.88
4.49
.82
2.00
4.64
.30
.88
2.10
.92
1.96
4.26
.50
1.42
3.41
.90
1.48
5.42
.10
.33
1.10
(C1-)
.09
.16
.42
.06
.16
.35
.08
.18
.37
.14
.30
1.35
-.03
.20
.40
.07
.17
.35
.08
.19
.57
.20
.41
1.68
(S042-)
.64
1.98
3.50
.70
2.31
5.90
1.54
3.47
7.00
.91
2.00
5.91
1.91
3.27
5.60
.51
1.50
3.50
.67
1.88
5.44
.21
.73
1.77
PH
4.20
4.46
5.70
3.99
4.30
4.83
3.92
4.25
4.59
4.11
4.62
5.65
3.98
4.31
4.65
4.52
5.17
6.15
5.30
6.03
6.86
4.95
5.52
6.50
Ab
7.0
19.5
29.2
9.5
27.0
53.4
15.1
31.4
61.7
8.3
16.6
40.3
16.0
27.5
51.2
6.9
12.0
24.3
6.2
12.7
33.4
3.7
6.8
21.9
No. of
Sampl es
31


100


128


91


72


94


42


99


                                aA11  measurements were made at the central laboratory and all samples were weekly
                                 collections when the equivalent collected rainfall was >_ 0.05 cm (using data from NADP
                                 1978,1979, and 1980).

                                Conductance 1n ^Siemens cm'1.

                                cNE-Me Indicates a site by Identifying a region within a state and then the state.  In this
                                 example, the site Is the NADP site In the northeastern part of Maine (cf. Figure 8-6).

-------
                    TABLE 8-10.  pH AND CONTRIBUTIONS TO FREE ACIDITY (%)  FOR FIVE REMOTE SITES
                                        (ADAPTED FROM GALLOWAY ET AL. 1982)
oo
in
00

Collector Type
No. Samples'5
Average pHC
pH Ranged
H2S04
HN03
HXf
St. Georges,
Bermuda
W/Da and Bulk
67
4.79
3.8-6.2
< me
< 35
> o
Poker Flat,
Alaska
W/D
16
4.96
4.7-5.2
< 65
< 17
> 18
Amsterdam
Island
Bulk (Funnel
and Bottle)
26
4.92
4.3-5.4
< 73
< 14
> 13
Katherine,
Australia
W/D
40
4.78
4.2-5.4
< 33
< 26
> 41
San Carlos,
Venezuela
Bulk
14
4.81
4.4-5.3
< 18
< 17
> 65
          aW/D refers to an automatic sampler which collects a wet-only sample in  one  container and a
           dryfall sample in the second container.
          bThese samples were treated with chloroform at the field sites.   Samples with  volumes less
           than about 500 ml were not treated with  chloroform at the  field  sites.
          cAverage pH here refers to the pH corresponding to the weighted-average  hydrogen ion
           concentration.
          dThis range is for pH measurements made at the Virginia laboratory,  on the samples treated
           with chloroform.
          ^Values greater than 100% simply indicate that the equivalence of sulfate exceeded the
           equivalence of free acidity.
          fThe authors indicate that HX could be HC1, organic acids,  or H3P04  but  they believe it
           was organic acid.

-------
constituents   other   than   H+   were   collected  monthly  in  these  studies,
Galloway  et al. felt  dry  deposition effects  would have  been  too large  to
allow  for a valid comparison with their own samples.

In  the study  by Galloway et  al.  (1982)  samples  with adequate  volume  were
split  in the field into two 250 ml  aliquots.  One of the aliquots was  treated
with  chloroform to  prevent biological   activity.   They  found  that  the  un-
treated aliquots were subject to pH changes during storage and shipment,  with
the acidity decreasing.  This  evidence,  combined with  results from ion chro-
matograph measurements (Keene et al. 1983), indicated that the sample  changes
were associated  with  degradation of  organic acids  in  the  samples.  Estimates
of the importance of  organic  acids compared to  sulfuric and  nitric acids  at
the five  remote sites are shown in the  last line  of Table 8-10.   The impor-
tance  of organic acids is clearly site dependent and varied from _>_ 65  percent
at the Venezuelan site to a negligible contribution at the Bermuda site.   Al-
though the percentages can be  rather  large,  in  absolute units the values  are
less  than  about 16  veq  £-1  (the free  acidity  for  pH   equals  4.8).   The
presence of organic  acids  again illustrates that  a simple comparison of  pH
data   is  insufficient to  address  time  trends  of acidity  associated  with
anthropogenic emissions.

Measurements in June 1980 of the pH and the major inorganic ions for over  300
samples collected in  Hilo,  Hawaii  showed that  the  acidity was  due mainly  to
sulfuric acid  instead of nitric or  hyrochloric acid  (Miller et  al.  1984).
Since  about one to four weeks elapsed between collection and pH  measurements,
it is  possible that any significant organic acid contribution would have  been
missed due  to  sample  changes as reported by Galloway et al.  (1982).   In  the
same study, about 75  additional samples  collected  at different  elevations  on
the island  of  Hawaii  were  measured for pH  within 24 hours and  again  about 5
months later.   The  hydrogen  ion concentrations  were  observed  to  typically
decrease  by 10 to  20  yeq  Jr1.     For  some  of  the  samples,  pH  changes
related to the  slow dissolution of dust particles could be definitely ruled
out.   Thus  it  seems likely that organic acids  are  making  a significant con-
tribution to some rain samples collected in Hawaii.

It has often been stated that  the  pH of natural precipitation  is controlled
by  the  equilibrium  with   atmospheric  C02,  producing  pH  values of 5.6.
Charlson and Rodhe  (1982)  have examined various aspects  of  the  atmospheric
sulfur and  nitrogen cycles for areas unaffected by anthropogenic  perturba-
tions.  They conclude  that,  in maritime areas where basic  constituents such
as ammonia  gas  and  CaC03  have low concentrations,  substantial  variations  in
precipitation  pH should be expected,  perhaps in  the range  of pH 4.5  to 5.6,
due to the  variability of  the  sulfur cycle alone.   Charlson and Rodhe and
several other  authors have  thus pointed  out that it is  not  appropriate to use
pH =   5.6  as  a  reference   value  against  which human  influences should  be
judged.  Charlson and Rodhe emphasize that generally pH will  be a  poor indi-
cator of manmade acidification, and instead the natural  elemental  cycles must
be studied in order  that manmade influences on  these cycles can  be recognized
and quantified.
                                    8-59

-------
8.4.3  Precipitation Chemistry Variations Over Time

8.4.3.1   Nitrate Variation Since  the  1950's--Likens  (1976)  reported  signifi-
cant  increases  in the  annual  volume-weighted concentrations  of nitrate  in
data  from New York and  the Hubbard  Brook Experimental  Forest,  New Hampshire.
Additionally, various other  authors  conclude  that NOx emissions  from  fossil
fuel  combustion are  the  most important  sources of  precipitation  nitrate
increases in the eastern United States, but that the  role of increased  ferti-
lizer use has not been rigorously assessed.

The  Hubbard  Brook precipitation  chemistry  data  record is  important  because
the  record   is  relatively long,  weekly bulk  collections  having been  made
continuously since 1964.  A recent examination of the nitrate data at Hubbard
Brook suggests  an  erratic trend  of  increasing  nitrate  from  1964  to  about
1971, followed by a leveling off  or  a slight  decrease  from  1971  to 1981  (NAS
1983).   The  annual  average values  range from about 1.40 mg £~1 in  1965-66
to  1.74  mg   rl  in the  early 1970's.   The  NAS report (1983)  indicates  that
the NOX  emissions in the  Northeast  increased  by  26 percent  between 1960  and
1970  and then decreased 4 percent by 1978.  Thus, the NAS  report notes  that
wet  nitrate  concentrations  at Hubbard  Brook  appeared to reflect emissions
trends in the Northeast.

Comparing the 1955-56 Junge data  (Figure 8-24) with  the  current  NADP  data in
Figures  8-9  and  8-25,  reveals  a broad  spatial  picture of  the increased
nitrate  levels.    The   average  nitrate  concentrations  in  Figure 8-24  were
obtained  by  weighting  the  quarterly  values  of  nitrate reported  by Junge
(1958)  with  the  quarterly  precipitation  for  the  sites  (Stensland  1979).
Attention should be focused on the eastern  United States,  where the NADP  data
record is most  complete.   The nitrate concentrations are clearly greater in
the recent NADP data than they are  in the  1955-56 Junge  data.  The  approxi-
mate  magnitude  of  the  increase is  consistent with the reported  increase in
combustion-related NOX  emissions over the  same  time  period  (cf.   Chapter
A-2, Table 2-1 and Figure 2-7).   However,  it  would be  inappropriate  to infer
a  quantitative  relationship  between  NOx  emissions  and  increases  in  pre-
cipitation nitrate  concentrations because  error bars for  the emission  and
precipitation data are  not yet available and  the  transport,  transformation,
and wet and dry removal  processes  are not well  understood.

Brezonik  et  al.  (1980)  indicated that  nitrate  had increased by  a factor of
4.5 in Florida rainfall  since the mid-1950's.   They  found,  for a  Gainesville
site, that  the average  bulk  nitrate  value was  24  percent  larger  than  the
corresponding wet-only  value,  and thus  concluded that  differences  in  col-
lector type  explained   only  a small  fraction of  the  overall  large  nitrate
increase.

The volume-weighted-nitrate concentrations  in  Figure  8-25  are generally lower
than  the median values  shown  in  Figure 8-9.   The difference appears to  be
very substantial when the 2.0 contour  is compared in the two figures.   How-
ever, the extension  of  the 2.0 contour  in  Figure 8-9  into  South Dakota  and
Nebraska  results from data  at only  three  sites, and illustrates why  it  is
important to show the data values  at the sites instead of  only  contour  lines.
The volume-weighted concentrations for  the  75 sites in  Figure 8-25  are,  on


                                  8-60

-------
Figure 8-24.   Map of precipitation-weighted-average-nitrate concen-
              trations (mg r1 as N03~)  for the 1955-56 Junge data,
              Adapted from Stens!and (1979).
                                8-61

-------
Figure 8-25.
Map of volume-weighted-average-nitrate concentrations
(mg  r1) for NADP wet deposition samples through
approximately December 1980 (using data from NADP 1978,
1979, and 1980).
                               8-62

-------
the average, 14  percent  lower  than the median values in Figure 8-9.  By way
of comparison, the volume-weighted-sulfate values in  Figure  8-8  were only 5
percent lower than the median sulfate  values  in Figure 8-7.

8.4.3.2   Temporal  pH Variation Since  the  1950's—Cogbill  and Likens (1974)
and Likens  and  Butler (1981) have  published eastern U.S. maps  of precipi-
tation  pH for the mid-1950's,  1960's, and 1970's.    Likens  and  Butler have
concluded from this mixture  of  calculated  and measured  pH values that there
has been  a  large spread and probable intensification  of acid precipitation
(pH < 5.6) in eastern North America during  the past 25 years.  These specific
conclusions were  based  on  trends shown on the pH maps,  but trends in emis-
sions and precipitation concentrations of sulfur and nitrogen compounds were
qualitatively considered.

Stensland (1979)  also calculated the pH distribution  for 1955-56 from Junge's
data.  He found  it necessary to  apply a correction  factor to the calculated
pH values to bring  the  values  into  agreement with  measured pH  values,  the
largest correction being required for  calculated  pH  > 6.0.  The resulting pH
map for 1955-56  by Stensland is  very similar  to the Likens and Butler map for
1955-56.  Stensland (1979)  also presents a series of pH maps to demonstrate
that  the  calculated  pH  pattern  is very sensitive to  the concentrations of
calcium and magnesium.  Tables 8-11 and 8-12  demonstrate the significance of
these  sensitivity  tests  (Stensland and Semonin 1982).   The 1977-78 data in
Table 8-11  are  for 1 year of  sampling  at two MAP3S sites  with automatic,
wet-only deposition collectors.   The 1955-56  Junge data for a nearby site, at
Williamsport, PA, were from a bulk  collector.  However, because the operators
at the Junge sites were instructed  to  place the bulk  collectors out only when
precipitation was  imminent,  the procedure  can be described  as  manual,  wet-
only  collection.   The magnesium concentration at Williamsport was estimated
(Stensland 1979)  because Junge did  not  measure this  parameter.   The data in
the column  labeled  'change'  in Table  8-11 indicates that the difference in
the calculated pH for the two time  periods, 4.67  versus  4.18, is  due more to
the change in the cations instead of the change in the anions.

A similar analysis for Illinois is  shown in Table 8-12.  The 1953-54 data in
Table 8-12 are a  summary of the results of Larson and Hettick (1956).   The
Larson  and  Hettick  samples were wet-only  deposition samples for  which  the
collection funnel was rinsed, just prior to sample collection, to reduce the
possibility of contamination by dust  between  rain events.   The 1977-78 data
in Table 8-12 are also from an automatic, wet-only collector  at the same site
used  for  the Larson  and Hettick study.   The decrease in  calcium  plus  mag-
nesium is the major reason  for the  increased  acidity  of the 1977-78 Illinois
samples.    Comparison  with  the 1980  data  for the NADP  site  located  10
kilometers from  the Larson  and Hettick site results in the same conclusion.

Both  the  1953-54  Larson  and Hettick  samples  and the 1955-56  Junge samples
were  collected  during  the  severe  drought  of  the   1950's.   Stensland  and
Semonin (1982) have  hypothesized  that this drought  produced unusually  high
dust  levels  in  the  atmosphere.    In  turn,   the  high dust  levels  produced
unusually high pH values for the  available precipitation  chemistry  data  for
the 1950's.   When the  calcium  plus magnesium levels measured by  Junge  are
reduced to levels currently being measured, the calculated pH for the entire


                                    8-63

-------
         TABLE 8-11.  WEIGHTED AVERAGE CONCENTRATIONS* (yeq £-1)
   FOR MAP3S AND JUNGE DATA.  ADAPTED FROM STENSLAND AND SEMONIN (1982)
                 Cornell      Penn.
                 Univ. NY  State Univ.   Mean of
                 9/21/77-   9/24/77-       the
                 9/29/78    9/15/78     two sites
                                  Williamsport,
                                      PA         Ch
                                    7/1/55-     (yeq
                                    6/30/56
Na+
K+
NH4+
Sum
 5.4

 1.5
 1.5
  .6
13.4
22.4
 4.5

 1.1
 1.5
  .7
12.9
20.7
 5.0

 1.3
 1.5
  .6
13.2
21.6
38.4

15.6
20.9
 3.6
 5.0
83.5
               -47.7

               -19.4
                -3.0
                +8.2
N03-
ci-
Sum
55.4
27.4
 4.4
87.2
55.5
27.6
 4.5
87.6
55.4
27.5
 4.4
87.3
 72.5
 21.1
 11.3
104.9
              -17.1
               +6.4
               -6.9
Calculated        4.19
 PH
Measured,         4.15
 Weighted pH
Number of          55
 Samples
             4.17
             4.16
              80
           4.18
              4.67
 MAP3S data are sample volume-weighted averages and Junge data are
 precipitation-amount-weighted averages.
                                    8-64

-------
      TABLE 8-12.   MEDIAN  PRECIPITATION CONCENTRATIONS (ueq T1)  AT
     CHAMPAIGN,  ILLINOIS.  ADAPTED FROM STENSLAND AND SEMONIN (1982)

Cations
Ca2+
Mg2+
Na+
K+
NH4+
Sum
Anions
S042-
N03~
cr
Sum
Calculated
pH
Measured,
Weighted pH
Number of
Samples
5/21/77- 10/26/53
1/16/78 8/12/54
10.5 84. 5a
[+=12.9
2.4
1.9 7.1
0.5 2.2
17.7 18.6
33.0 112.4
78.9 64.5
29.8 20.2
4.8 7.3
113.5 92.0
4.09 6.52
4.02
63 30
Change
(yeq £-1)
-71.6
-5.2
-1.7
-0.9
+14.4
+ 9.6
- 2.5



a
 Measured hardness.
                                   8-65

-------
Northeast  is  less than  4.6.    Stensland  and Semonin  suggest (1)  that the
drought-corrected pH pattern for the  1950's  should  be  compared with current
data and  (2)  that the  error  bars  associated with  the  calculations make it
difficult to discern a  pH time  trend over  the last 25 years.

Hansen and Hidy  (1982)   have discussed  other  features  of the historical data
record that make  establishing  the magnitude  of  the pH time trend difficult,
and Barrie et al. (1982) have  reviewed information relative  to  acidity  trends
in North America and state:

     As a  consequence of  this  continuing  debate,  one  can conclude  that
     it is presently unsafe  to  utilize existing  network data to draw any
     reliable conclusions with  regard  to acidity trends in eastern North
     America.

The NAS report  (1983)  and the  recent  article by Hidy et al.  (1984) discuss
trends  in acid  precursor emissions  and  their precipitation  products, in-
cluding the  effect on   precipitation  acidity.   The precipitation  chemistry
discussion is focused on  the  Northeast, and  especially  on the Hubbard Brook
and USGS bulk precipitation chemistry  data sets.  The  NAS report points out
that a  linear regression analysis  of the Hubbard  Brook sulfate data shows
that  the  concentrations  at  that site declined  by about  33  percent from
1965-66 to  1979-80.   The SOe  emissions for  the U.S.  EPA-designated regions
of the eastern  United  States  were examined.   It was concluded that the de-
cline in wet  sulfate at Hubbard Brook  fairly closely reflected the  reduction
in  S02  emissions  within the  Northeast  itself,  and  did  not  reflect  the
trends  in S0£  emissions  in  the Midwest  and other distant source regions.
In the 1964-77 time  period, there was no  statistically  significant trend in
precipitation pH  or  hydrogen  ion deposition  at  the Hubbard Brook site.  The
USGS bulk  data  for the three  sites in northeastern and  north-  central  New
York showed  significant declines  in sulfate in the 1965-78 time period for
two sites  and no trend at the other  site (Hidy  et al.  1984).   In general
terms, these USGS data  thus  qualitatively  support  the Hubbard Brook  data.

Hidy et al.  (1984)  also state that  changes in  precipitation sulfate between
1965 and 1980 at rural  sites  in central New York State and New  Hampshire were
more  influenced  by S02  emissions changes in nearby  source regions than by
those from more distant source regions.   However, Hidy  et al.  found that no
proportional  relationship between  nearby source  region emissions  and wet
sulfate existed for other USGS bulk  deposition  sites in western New York and
north-central  Pennsylvania.

The concluding statement  in the NAS report concerning the influence of local
vs distant sources is the following:

     On the  basis of currently available  empirical data,  we cannot in
     general determine the relative  importance for the net deposition of
     acids  in  specific  locations  of  long-range transport from  distant
     sources or more direct influences of  local  sources.
                                    8-66

-------
Thus,  although  the  NAS  report  (1983)  mentions  that sulfate  and  nitrate
deposition data in the Northeast  appeared to reflect emissions trends  in  the
Northeast, a strong concluding  statement concerning the importance  of  nearby
sources versus distant sources was not made.

8.4.3.3   Calcium  Variation  Since the 1950's—Table 8-13 shows precipitation
calcium concentrations  for  various networks, sites, and  time periods.   The
calcium levels for the  MAP3S and NADP networks  are  small  relative to  those
for  the  other networks.   Bulk samples were collected in  the USGS  network,
probably accounting for the higher calcium levels for that  network.   However,
urban  areas  such as  the  Albany,  NY,  USGS site, can also  produce  relatively
high  atmospheric dust  levels,  and  thus,  high  calcium levels  in  air  and
precipitation  samples.   The NCAR  and  WHO  networks used automatic,  wet-only
collectors, but,  because of sampler design, the  covers probably did  not make
firm  contact  with the sampling bucket  (Stensland  and  Semonin 1984).   Thus,
dust  probably  leaked in during  nonprecipitation  periods, producing  the  rela-
tively  high  calcium  concentrations  and  shifting  the precipitation  pH  to
higher values.

If  dust  leaks into  the  sampling containers of wet-only collectors  or  is
included  in the  precipitation  sample  via  bulk  sampling, the  measured  pH  may
be significantly different than that for rain and  snow that falls  into  clean
containers.   This sampling deficiency will  often  be   strongly influenced  by
the local environment, and will  likely  be  quite variable  on  both  short-  and
long-term  time scales.   For a given collector,  the  problem will  be most
severe in arid regions.  The data in Table 8-13  suggest this problem can also
occur  in  the  eastern  United States.   The  magnitude  of  this  dust  leakage
effect should  be continuously evaluated at  all  sampling  sites through col-
lection,  analysis, and  reporting  of appropriate blank  samples.  These  steps
have  been taken  in very few networks in the past,  and they  are only  rarely
taken now.

For the Junge  1955-56 data,  it  has been shown that the precipitation calcium
concentrations for the Plains States and eastward were  on the  average about a
factor of 4  higher than current levels; for the western states, the  calcium
was about a factor of 6.0 higher  (Stensland and  Semonin 1984).  As  discussed
in the previous  section,  the higher values in the east may have been  due  to
drought;   for  the west,   the  higher levels  were  probably due  to  sampling
problems,  specifically   bias   due  to  dry  deposition   contamination   and
evaporation.    But,  regardless  of the reason for  the  higher levels  of  the
crustal  elements,  their  effect  on pH trends   is  significant and  must  be
considered.

8.4.4  Seasonal Variations

Herman and  Gorham (1957)  reported that  snow  sampled  in  the  early  1950's
contained lower  sulfur  and  nitrogen  concentrations   than did  rain  sampled
during the same  period.   They speculated that this  difference  might have
resulted from  snow's having  a  lower collection  efficiency than rain or from
arctic air  bearing  snows  being  cleaner  than   tropical  air.    In  the late
1960's, Fisher et al.  (1968)  observed  lower precipitation sulfate in  the cold
season.  Bowersox and  dePena  (1980),  Pack and Pack (1980),  and  Pack  (1982)


                                  8-67

-------
      TABLE 8-13.  CALCIUM CONCENTRATIONS  (mg  £-1)  FOR VARIOUS NETWORKS,
         SITES, AND TIME PERIODS.   ADAPTED  FROM  HANSEN ET AL. (1981)
                Sites  Jungea     NCARb     WMOC      US6Sd    MAP3S6     NADPf
                       1955-56    1960-66   1974-76   1966-78  1978-79    1979
Rocky Mountain

Alamosa, CO                                2.65
Grand Junction, CO     3.41       7.25
Pawnee, CO                                                           0.53

Midwest

Grand Island, NE       3.12       0.96
Huron, SD              2.40                 2.74
Lamberton, MN                                                        0.58
Mead, NE                                                             0.53
St. Cloud, MN          1.02       1.12

Northeast

Albany, NY                       1.97                2.83
Caribou, ME            0.63       0.39       0.36
Hinkley, NY                                         0.70
Huntington, NY                                                       0.13
Mays Point, NY                                      1.48
Ithaca, NY                                                 0.14
Williamsport, PA       0.77

Southeast

Charlottesville, VA                                        0.05
Georgia Station, GA                                                  0.10
Greenville, SC         0.31       0.30
Raleigh, NC            '                    0.20
Roanoke, VA            0.32
Sterling, VA                     0.67


aWeighted averages, manual  wet-only sampling,  July 1955-June  1956.
^Weighted averages, wet-only sampler,  NCAR/Public Health  Service.
^Medians, wet-only sampler.
"Medians, bulk sampler.
Medians, wet-only sampler,  July 1978-June 1979.
^Medians, wet-only sampler.
                                    8-68

-------
 reported strong  seasonal  variations  in sulfate  in precipitation  at MAP3S
 sites  in New York,  Pennsylvania, and Virginia.

 Bowersox and Stensland (1981) analyzed NADP  data  for  seasonal  variations in
 sulfate and  nitrate.   Because the data  base was  small, two to  seven sites
 were  grouped in five regions  in the  eastern  United States and the  data  for
 each  region  were  averaged for the cold  season (November to March)  and  the
 warm  season  (May to September).  The resulting warm-to-cold-period ratios  for
 sulfate varied  from  about  2.0  in  the New  England region  to  1.25  in  the
 Illinois region.   The  investigators  noted that aerosol  sulfate  has a similar
 seasonal  variation but  that  SOX  emissions   for the  Northeast  have  a rela-
 tively  small  seasonal variation.

 For  nitrate, Bowersox  and Stensland  (1981)   found  a maximum  warm-to-cold-
 period   ratio  of 1.5  for the region  in the  Southeast,  but  three  of  the
 remaining  regions  had  little or no seasonal  variation.   Determining whether
 different  patterns  of seasonality for  nitrate and sulfate are predicted by
 numerical  simulations  would  be valuable.   The acidity of  the  precipitation
 was greater  in  the  warm  period for all  the regions and reflected  the mixture
 of the  patterns for sulfate and nitrate.

 Bowersox and dePena (1980)  found  only  slightly  higher  nitrate  in  precipi-
 tation  in the  winter  than they did  in other  seasons at the MAP3S  site in
 Pennsylvania.   Hydrogen  had  a strong  maximum  in the^warm months  and sulfate
 was  the  principal  anion  affecting  acidity.   Nitrate,  at concentrations
 similar  to  those of  sulfate, did not  correlate  well with  hydrogen  ion  in
 liquid  precipitation but did correlate with  hydrogen ion  in  snow and frozen
 precipitation.

 The seasonal  pattern of  precipitation  sulfate concentration is different  for
 western  Europe  than  it  is  for  the eastern  United States.   Granat  (1978)
 averaged  the data  for many  European  sites  and reported  a maximum  sulfate
 concentration in the  spring,  being 1.6 times  greater than  the minimum value
 observed in  the fall.  The sulfur  emissions in the region are at  maximum in
 the winter (Ottar 1978).

8.4.5  Very Short Time Scale  Variations

The  concentrations of  the  major  ions  in precipitation  vary  considerably
 during a rainshower (Robertson et  al. 1980).   Samples collected sequentially
 during rainshowers  in Arizona  had  calcium variations up  to  1000 percent over
a sampling period of less  than 15  minutes (Dawson  1978).  Dawson  found that
 the correlation  between  ions having a  common source were not significantly
different from those between  components not having  a common  immediate source.
Therefore,  Dawson   concluded  that  the  observed concentration  changes were
primarily determined by precipitation  processes.

8.4.6   Air Parcel  Trajectory  Analysis

Attempts have been made  to link  the  precipitation  chemistry patterns  to  the
emission source regions  through  the use  of air parcel  trajectory analysis.
There  are many different  approaches to  calculate trajectories  of air  parcels.


                                 8-69

-------
Forland  (1973)  used  surface geostrophic  analysis to  determine air  parcel
trajectories.  This analysis involved using surface air pressure gradients  to
calculate the wind speed and direction, which dictate  the movement of the air
parcels.  Recently, many investigators have  calculated trajectories  with the
National  Oceanic  and  Atmospheric  Administration Air  Resources  Laboratory
(ARL) model, which uses as input data surface layer wind observations (Miller
et al.  1978,   Wilson et al. 1980, Miller et al.  1981).   With  the ARL  model,
an average wind for  the  surface  layer,  such  as the layer 300  to  1500  meters
above  the  ground, is  used to calculate  the trajectories.  Many scientists
argue that air parcel trajectory techniques  need  to be  further developed and
verified with field  experiments.  Especially questionable are  the trajectory
calculations in areas of very low and  variable wind speed and  in  areas  near
the separation of different air masses, i.e., near weather fronts.

Some  conclusions  from  recent  trajectory studies  are as  follows.    Forland
(1973)  found  that,  for a  site  at the southwestern tip  of Norway, the  pre-
cipitation  pH  values were 4 to  5   for  air parcels  originating in  central
Europe  or England  and  5.1  to 6.6 for  parcels  originating in the North  Sea.
He concluded that acidic precipitation in southern  Norway is mainly  a  result
of S02  emissions  from  northern  Europe.   Ottar (1978) reported  that  aerosol
sulfate  at  European  sites  examined  by sector  (air parcel)  analysis  showed
that  sectors associated  with high concentrations  are  directed  towards  areas
of major sulfur emissions.   Similar analysis for  precipitation  illustrated
that, to a  large  extent,  acidity is strongly  influenced  by  the availability
of ammonia, with air masses  passing  over  the sea  showing the least degree  of
neutralization.

Jickells et al.  (1982)  used the ARL  trajectory  analysis to stratify the pH  of
precipitation samples collected at Bermuda.   They  found that  pH was generally
less than 5.0 for  trajectories  originating  in  the eastern United States and
frequently  greater than  5.0 for trajectories originating  in areas southwest
to southeast  of Bermuda.   Thus, they concluded  that rainwater  originating
over  continental  North  America  was markedly more acidic than  rainfall  from
the other sectors.

Wolff et al. (1979) used trajectory  analysis to characterize  precipitation  pH
for samples from eight sites in  the  New  York City area.  They found higher  pH
values  for air parcels from the ocean or from  the north  and  lower pH  for air
parcels  from the south through northwest  sectors.   The lowest  average  pH was
for air parcels from the southwest sector.  They also  classified the  precipi-
tation  events according  to synoptic meteorological conditions  and found air
mass  thunderstorms  and  precipitation  associated  with  cold  fronts  in the
absence  of  closed lows  to  be  the  most  acidic.   Whether the low pH  cases
identified in this study  were  more  strongly related  to source  direction  or
to characteristics  of  the  scavenging  processes   taking  place  in these  con-
vective  types of  precipitation  events seems open to  question  since  showers
and thunderstorms are usually associated with southwesterly flow.

Raynor  and  Hayes  (1981)  also classified  pH  data  by  synoptic  type and found
the lowest  pH with  cold fronts  and  squall  lines, or  with thunderstorms and
rainshowers.  Although  these are predominately warm  season rainfall  types,
Raynor and Hayes found that the  low  pH was not  a function of  season alone.


                                    8-70

-------
The question  of  the importance of atmospheric transformation and  scavenging
processes in  explaining  the observed  association  between southwest trajec-
tories and low pH is discussed by Wilson et al.  (1980), who maintain that:

     Normally, trajectory  analysis of  individual  events will  lead to
     some basic  source-receptor  relationships.   Vital  information is
     still  missing on  the  overall  transport/transformation processes
     that take place  in the  atmosphere relevant to  the formation and
     deposition of  "acid  rain"....In  summary,  the known  source  regions
     for  precursor  gases to  "acid rain"  cannot yet  be unequivocally
     linked to receptor  with the meteorological, physical and chemical
     information  available today.

Wilson et al. (1982)  emphasize  the  importance  of  recognizing  the relation
between precipitation  amount and ion  concentration.  When they normalized the
MAP3S data for 1977-79 for  precipitation  amount  they  found  that the sulfate
deposition per centimeter  of precipitation was  about  the same  at the MAP3S
Illinois  site and  the  Pennsylvania  site.   Stated another  way,  much more
sulfate was deposited annually at  the Pennsylvania site  than at  the Illinois
site, mainly because of the greater annual  precipitation  amount.

Wilson et al. (1982)  used trajectory  analysis  to examine  the directional
variability  of wet deposition  at  the  Whiteface  Mountain  MAP3S site   in
northeastern  New York and  at the MAP3S site in  east-central  Illinois.  For
the New York site, the southwest sector was found to contribute 56  percent  of
the  water  deposited,  and  62   to  65   percent  of   the  H+,   S042",  and
N03~.   The same  sector  contributed  71  percent  of the  water  deposited and
64  to 76  percent of  the  H+,  S042',  and  N03"  at   the  Illinois  site.
The authors state that:

     The  fact that  few major pollutant  sources lie within the southwest
     sector traversed  by trajectories  arriving  at  Illinois indicates
     that a knowledge of air mass  histories  further back  in  time may be
     necessary to adequately identify  all  important source regions.


8.5  GLACIOCHEMICAL INVESTIGATIONS AS  A TOOL IN THE HISTORICAL DELINEATION
     OF THE ACIDIC PRECIPITATION PROBLEM (W. B. Lyons and P.  A. Mayewski)

Precipitation in  the Northern Hemisphere has been recently recognized to have
hydrogen  ion  concentrations 10 to  100 times  higher than  expected for natural
precipitation (Likens and  Bormann 1974, Cogbill  and Likens  1974,  Lewis and
Grant 1980).   However, controversy  has arisen  regarding the  nature  of the
acidity of  the precipitation  sampled and  whether,  indeed,  the  pH of North
American  precipitation  has  increased  over  time (Miller and  Everett 1979,
Lerman 1979,  Stensland  1980,  Sequeria 1981, Charlson  and Rodhe 1982).    In
most locations pH  records  have  been  constructed rather  imperfectly  due  to
differences in sampling, handling, and  analytical  procedures used  (Galloway
and Likens  1976,  1978;  Galloway  et al.  1979).   The lower  pH's  measured  in
Northern   Hemisphere precipitation are  thought  to be  due to  the  input   of
sulfur and nitrogen oxides from fossil  fuel-burning (Likens and Bormann 1974)
and in  some  cases  hydrogen chloride  (Gorham  1958a).   Few  baseline  data,


                                   8-71

-------
however, are  available  on the pH  of  precipitation in areas of the  Northern
Hemisphere remote from  North  American  and European sources of  anthropogenic
sulfur emissions.  In addition, monitoring records of  pH  and  acidic  chemical
species  are  of  rather  short  time duration  (  ~  15  to 20 years  at  most),
limited  geographic coverage,  and  provide little useful information  prior  to
the  early  1960's  (Hornbeck  1981).    Baseline studies  of  pH  and   related
chemical species as  well  as historical  time series data are warranted if  we
are to understand man's  effect on  the  environment.

The National  Academy  of  Sciences  (1978)  recommends that historical  studies  of
glacier  snow and ice should be conducted.  Such studies are  needed to better
understand the atmospheric transport of anthropogenically-introduced  chemical
species to remote areas.  In addition,  a more  recent  NAS report (1980)  states
that a major scientific  goal of the 1980's should  be  to "identify  the signif-
icant natural  and anthropogenic factors contributing  to acid  rain."  Detailed
glaciochemical studies should provide  this type  of needed  information.

Snow  and ice  cores  collected  from  appropriately chosen  glaciers   provide
samples  of  entrapped chemical species  that,  unlike  those derived  from any
other medium,  are nearly to entirely unaltered since  their deposition.  This
technique has barely been applied  to the  study  of acid precipitation despite
the fact that it provides  a very  sensitive record  of  precipitation chemistry.

8.5.1  Glaciochemical Data

Past glaciochemical  studies (early studies are reviewed in  Langway 1970) have
provided  information concerning  1) the  documentation of individual  storm
events (Warburton and Linkletter 1978, Mayewski et al. 1983a), 2)  the  dating
and  seasonal  accumulation  of snow  and  ice  (Langway  et al. 1975,  Herron and
Langway 1979,  Butler  et  al.  1980,  Mayewski et  al.  1983b),  as well  as  3)  long-
term climatic change  (Delmas et al. 1980b, Thompson and Mosley-Thompson 1981,
Johnson and Chamberlain 1981).  Our discussion  will  deal  primarily  with the
use of glaciochemical studies  in delineating  the  acid  precipitation  phenome-
non.  The text that follows is divided into  a  section  on primary measurements
including sulfate, nitrate, pH, and total acidity, and a  section  concerning
analog measurements  or  trace  metals.   For both  primary  and analog  measure-
ments the discussion is subdivided  into  results from  polar glaciers  and from
alpine glaciers.

The  glacier division adopted in this  text  is used primarily  as  a  means  to
separate the  results of glaciochemical  studies for  review purposes.   Polar
glaciers, including  the Antarctic and Greenland  ice  sheets,  are  character-
istically lower in temperature and  accumulation rate  and  larger  in size than
alpine  glaciers.   Hence,  polar glaciers  classically are used  to  retrieve
longer  glaciochemical   time-series, often with  less  subannual  detail than
time-series from  alpine glaciers.   Although  there  are  many fewer  glacio-
chemical  studies  available  from  alpine glaciers,  they   are  included here
because  these  glaciers  are less  remote  from  industrialized  sites  than are
polar  glaciers  and,  therefore,  have  considerable potential  as proxy  indi-
cators of man's effect on  the environment.
                                  8-72

-------
8.5.1.1   Sulfate  -  Polar  G1aciers--The early  work  by Koide  and Goldberg
(1971), Weiss et al.  (1975), and Cragin et al.  (1975)  and more recent work by
Busenberg and Langway (1979) has suggested that  the concentration of sulfate
in recent Greenland  snow and ice  (past  20  yr) has increased by  at least a
factor  of  two.    This  increase has  been attributed to fossil-fuel burning.
However, other  investigations  have suggested  that  these  enrichments  may be
also  linked  to  natural  processes  and/or local contamination (Boutron 1980,
Boutron and Delmas 1980).

Herron  (1982)  most  recently  indicates  that  S042'  has  been enriched  by a
factor of 1.6 to 3.7  in Greenland snow and ice in the  past  200 years and  that
this enrichment is due to the burning of fossil fuel.   No anthropogenic input
of  $042-  has   been  observed  in   Antarctic  ice  cores  (Delmas  and Boutron
1978, 1980; Herron 1982).  Recent work by Rahn (Kerr 1981)  indicates that the
northern  polar  regions  receive pollutant  $042-  on  a seas9nal  basis,  and
mass budget considerations indicate that approximately 2.5 times the natural
atmospheric emission  leaves eastern North America  every  year (Galloway and
Whelpdale 1980).  Shaw's (1982a) work confirms that of Rahn, indicating  that
the Arctic haze observed in Alaska has  its source  in Eurasia, with smelting
operations in Siberia being a possible major  contributor.

Natural  processes may  also have  a profound  effect  on  S042~  profiles in
glacier ice.  For  example, Bonsang  et  al. (1980)  have  shown  that aerosols of
marine origin have much  higher  S04/Na ratios  than seawater, indicating  that
$042-  enrichments in  precipitation need  not be  all  due  to anthropogenic
emissions.  Recent work by Hammer et al. (1980) indicates  that Greenland ice
concentrations  of  $042-  are  greatly  affected  by  world-wide  volcanism.
The active volcano Mt.  Erebus may be a major sulfate source  to the Antarctic
continent  (Radke  1982).   Volcanically  produced  $042- has  been  observed in
Antarctic and Greenland  ice  cores  (Kyle  et al. 1982,  Herron 1982).   As one
proceeds  away   from  the ocean  in   both  Antarctica  and Greenland,  sea  salt
becomes less  of  a contributor  to  the S042~  concentration  in the  ice  and
snow  (Boutron   and  Delmas  1980),   and  in Antarctica  gas-derived  S042- as
well  as N03~ and Cl"  becomes very important  (Delmas  et  al.  1982).

In addition  to  the  possible  volcanic  input  of S02  into  the  atmosphere,
biogenic emission, particularly in lower latitude  regions, may also  be an
important  contributor  of  S02  (Lawson  and  Winchester  1979,  Stallard  and
Edmond 1983, Haines 1983).   Due to  the very  long  residence  time of sulfate in
Antarctic aerosols (Shaw 1982b), the  oxidation of marine-derived gases  such
as  dimethyl-sulfide  may  be  a  major  contributor  of  sulfate  to  Antarctic
precipitation (Delmas  1982).    Herron  (1982)  has also  suggested  a biogenic
source for a portion of the sulfate observed  in Greenland ice.  Gas adsorp-
tion  onto  particles  may also  be  an  important  source  of  S042"  in   some
locations (Mamane et al. 1980).  It is also  thought that the sulfate present
in Arctic  aerosols  is  formed from  the  conversion of continentally-produced
pollutant S0£ during  transport (Rahn and  McCaffrey 1980).

8.5.1.2   Nitrate - Polar Glaciers—The  work  of Parker et al.  (1977,  1982)
shows  downholevariationsTntRe  NOa"  concentration  of  snow  and   ice.
Parker et  al.  (1977)  have suggested  that this historic  variation  is due to
changes in  sunspot,  auroral, and/or  cosmic  ray  activities  and not  due to


                                 8-73

-------
variations  in anthropogenic  inputs.   These  workers  have recently  observed
seasonal,  11-  and 22-yr  periodicities  as  well  as  long-term  changes  in
Antarctic ice (Parker et al.  1982).   The  highest values  were associated  with
winter  darkness  and  heightened  solar activity.   They  observed  no  anthro-
pogenic  N03-.   Kyle  et al.  (1982)  have  observed  volcanically-introduced
N03-  in  Antarctic ice.   However, Aristarain  (1980)  has  observed  on James
Ross  Island, Antarctica,  no variation  in  N03~,  on  at  least the  seasonal
level.  Risbo et al. (1981)  and Herron (1982), on the  other hand,  observed  no
relationship  of  NOa-  with  solar  activity in  Greenland.   Herron (1982) did
note  a  seasonal  variation  of N0s~  in  Greenland ice; however,  the  highest
values were  associated  with the  summer season.  He also  observed an  anthro-
pogenic  doubling  of  N03"  in  surface  samples,  indicating  for  the first
time  the introduction  of  N03- into  this region,  probably  through  fossil-
fuel burning.

8.5.1.3  pH  and  Acidity -  Polar  Glaciers—Hammer (1977,  1980; Hammer et al.
1980)  has  measured the acidity  of  Greenland ice cores  and found a "back-
ground" value of pH - 5.4 although much  lower values  appear during times  of
high volcanic input (e.g.,  Laki Eruption in  1783, pH of ice = 4.4).   However,
in  most cases  Hammer  has  not measured  pH  directly but  rather  has   used
conductivity techniques.

Berner  et  al. (1978)  first measured the acidity of  Antarctic ice by using
strong  acid  titrations.   They  observed values  ranging  from 6.0  to   7.5.
Delmas et  al. (1980a)  found  an average  pH  in Antarctic  ice of 5.3.  These
investigators,  like  Berner et  al.  (1978),  used the strong acid  titration
technique rather than direct  measurements of pH.  More  recent work  (Legrand
et  al.  1982)  has substantiated  the  fact  that Antarctic precipitation  is
acidic with maximum reported values of 7  yeq £-1.

Much of  the earlier pH  work on glacier snow and ice is unusable  due  to  pos-
sible sampling and handling  artifacts (e.g.,  filtration  and hence  degassing
prior to analysis,  and  sample storage  in glass  rather than plastic; Gorham
1958b; Elgmork et al.  1973).

The polar data on acid  anion  concentrations  suggest  there has been a negli-
gible contribution of fossil  fuel by-products transported to  Antarctica,  as
expected due  to  its great  distance  from  Northern  Hemispheric  sources.   The
most recent data, those of  Herron  (1982),  indicate  however that Greenland has
been  affected  by  fossil-fuel  burning   with S042"  and N03"   enrichments
in  surface  snows of  ~  2 above preindustrial  times.  However, it should  be
noted that  these enrichments are  based  on very  few  data points,  and   more
detailed study may be  warranted.

8.5.1.4  Sulfate - Alpine Glaciers—To our knowledge,  no  published data exist
for S042~ concentrations in glacier ice from alpine  areas.

8.5.1.5  Nitrate - Alpine Glaciers—Butler et al.  (1980)  have observed values
of  from  <  0.03  to 2.80 yM  in a  short core from Athabasca Glacier, Alberta.
They observed higher values during the warmer months  of  the  year.  In addi-
tion,  their  mean  N03~   value was  approximately 15  times  lower  than   that
observed in central Alberta snows  close  to  populated  areas.    High  elevation


                                    8-74

-------
surface  samples  from Kashmir,  India,  demonstrate values as  high  as 1.3 yM
in snow from pristine air masses (Mayewski  et al.  1983a)).   Nitrate  values of
between  <  0.1  and 4.4 yM  have been obtained  from  a  ~ 17  m  core on Sentik
Glacier in Kashmir,  India,  close to the surface  sampling  site discussed in
Mayewski  et al.  (1983a).   The  source of  the  N03-  is  unknown,   although
variations in air mass source and/or accumulation  rate  may be  important.

8.5.1.6   pH  and Acidity -  Alpine Glaciers—Although  identifying  the pH of
snow  and  ice may be more  complex than simply measuring strong mineral  acid
contributions, Delmas and  Aristarain (1979)  have observed  in the Mt. Blanc
area  of the French Alps  strong mineral  acid values that  increase  from ~ 0
peq  £-1   for   1963   to  above   10  peg   £-1  in   1976.     It  should  be
pointed  out,  however,  that  this  increase  from 1963 to  1976  is only repre-
sented  by 4  data points.   It does however  provide insight  into the  possible
usefulness of  high-altitude  alpine glaciers as historic  tools.   Delmas and
Aristarain  (1979)  have  argued that  this  strong  acid  increase  is  due to
increased fossil-fuel burning.

Clement and Vandour (1967) have reported pH  values of  snow  from the southern
French Alps in the range 4.2 to 7.0, noting changes in pH with time,  type of
snow, and elevation.  These authors have suggested that,  in  general,  low  pH's
correspond  to  winter snow  accumulation,   freshly  fallen snows,  and higher
elevation  snow.   Lyons  et al.  (1982)  and  Mayewski  et al.  (1983a)  have  also
observed an elevation vs pH  relationship for Himalayan surface snows.  These
authors  have  suggested  that  the  majority  of  the pH  vs  elevation trend
observed  is   a  function   of  increased   CO?  saturation  with  decreasing
temperature.   A number  of workers  (Scholander et  al. 1961,  Berner et al.
1978, Stauffer and Berner, 1978, Oeschger et al. 1982) have shown that polar
ice  and snow  are  easily  "contaminated" with C02.   If these  data and the
interpretations are  correct,  detailed  ionic balance studies  must be under-
taken  to  understand  completely  the  nature of  the  acidity and/or  pH of
ultrapure snow and ice.

More  recently  Koerner and  Fisher  (1982)  have discussed the adsorption of
C02  as it  related to  snow pH measurements and  snow density.   They  have
argued  that  the  pH contribution  due to  C02  "contamination"  should  increase
with  depth  in glacial ice.   If this  is true,  the pH of  snow and  ice, es-
pecially  downhole,  may  have  little  relevance  to  the  acid precipitation
phenomenon.   The  measurement of  acidity via titration eliminates this  con-
tribution of  C02  to pH from  the  ice  as well  as any  contribution   from the
ambient atmosphere  upon  melting.   The  newly developed acid  titration tech-
nique of Legrand et al.  (1982)  appears  to be  the best suited for snow and ice
pH work.

8.5.2  Trace Metals - General Statement

In studies  aimed  at determining  the effects of  fossil-fuel  burning on the
environment, various investigators  have used trace  metal  concentrations in
precipitation as well as lacustrine sediments and soils as  analogs  of acidic
compounds (Andren and Lindberg  1977, Galloway  and Likens 1979, Wiener 1979,
Anderssen et al. 1980, Jeffries and Snyder 1981).  Mass budget calculations
indicate that by burning fossil fuel man has contributed both metals  as  well


                                    8-75

-------
as acid  into  the  atmosphere  (Bertine  and  Goldberg  1971,  Lantzy  and  Mackenzie
1979).   However,  some  controversy  exists as  to  whether this  anthropogenic
metal  introduction  via  burning  is regional  or global in  scale  (e.g.,  Nriagu
1979, 1980; Landy et al. 1980; Boutron 1980; Boutron and  Delmas 1980).   This
is coupled  with the fact  that  contamination  problems  and analytical  uncer-
tainties severely limit the interpretation of  much  of the data and complicate
the  use  of trace metal concentrations as  acid  surrogates  (Murozumi et  al.
1969, Boutron and Delmas 1980, Ng and  Patterson 1981).

8.5.2.1  Trace Metals - Polar Glaciers—The original glaciochemical  analyses
of Pb in Greenland and Antarctic ice by Murozumi  et al.  (1969) indicated:   1)
a rise from  1 ng  kg-1  in  Greenland prior to  800  BC to  values  greater  than
200  ng  kg-1  in 1968  with the  sharpest rise   since  1940, and 2)  a rise  in
Antarctica from less  than  1  ng  kg-1  to 20  ng kg-1  in  1968.   These authors
suggest that  the sharp  rise  in Greenland  concentrations  post-1940 was  due to
the increased consumption  of leaded  gasoline.   The  lower  values  in Antarctica
were  because  most  of  the  fossil-fuel   burning   occurs  in   the   Northern
Hemisphere and  little  if  any troposphere mixing  occurs  across  the  equator.
The work  of Murozumi  et al.  (1969)  also  demonstrated  much more terrestrial
material  in Greenland ice compared to Antarctic ice  ( ~ 15 to 20 times more)
while  the Antarctic  ice  contained   about  twice  as  much  sea  salt  as  the
Greenland  precipitation.    Unpublished  work  by  Boutron  and Patterson  now
indicates little  if any increase (possibly a  factor of  2;  from 1.5 ng  kg-1
to 3  to  4  ng kg-1) in Pb in the  surface  snows  of Antarctica compared  to
older ice samples, and that all  previous data  were  erroneously high.

The work  of Weiss et al.  (1975)  showed that  in  Greenland ice  (Camp Century
and Dye  3),  Hg, Cd, and  Cu  were enriched  in  the  surface  layers,  and  they
suggested  that  this  enrichment  was  due  to  increased  fossil-fuel   burning.
Similar  surface  enrichments  were  measured  for   Ag  in  Antarctic   ice   and
attributed to weather modification programs such as cloud seeding (Warburton
et al. 1973).

The work of Herron et al.  (1977) suggested for the  first  time that  "natural"
enrichments of several  orders of magnitude for several  trace metals  occur in
the atmosphere.  This work was  corroborated by additional investigations  on
Alaskan  snow  (Weiss  et  al. 1978).    The process causing  this   "natural"
enrichment for metals such  as Zn, Pb, Cd,  Cu, As, Se,  Hg,  and even Na  was
suggested to  be volcanism.   Although  volcanism may have  a pronounced  effect
on atmospheric aerosol  chemistry  great  distances  from  its source (Meiner  et
al. 1981), volcanic emission studies  are  in conflict as  to whether  volcanism
is a  major  source of volatile  trace  metals to the  atmosphere  (Unni et  al.
1978, Lepel et al. 1978).

Due to its remoteness from North American emissions, it  is  now  apparent  that
any  enrichments of  trace metals,  with  the   possible  exception  of  Pb  in
Antarctic ice may not be due  to  pollution but  possibly to volcanism  (Boutron
and Lorius 1977, 1979; Boutron 1979a,  Boutron  1983).  Although metal enrich-
ment factors  show temporal changes,  these changes  do not  vary systematically
on a short-term or long-term basis  (Boutron and  Lorius  1979, Landy and  Peel
1981).   In  addition,  the  present day  metal  fluxes  of Cd, Cu, Zn, and Ag  are
similar to those 100 years ago,  again suggesting  little to no  anthropogenic


                                    8-76

-------
 input  (Boutron 1979a).  However, manmade radionuclides are measurable in Ross
 Ice Shelf  samples  in Antarctica  as  well  as in Greenland  (Koide et  al.  1977,
 1979).    The detectable  concentrations of  these  weapon  test  products  in
 Antarctic  ice do indicate  that  some high altitude  interhemispheric  transport
 of  manmade products does  occur  (Koide  et  al.  1979).  Obviously the  mode  of
 transport,  the  altitude   of  transport, and  the  size of  the  transporting
 particles  all affect pollutant dispersion and distribution.

 In  Greenland,  the  recent findings of Ng and Patterson  (1981)  have  confirmed
 the earlier  work of Murozumi et al. (1969).   Their data indicate that  the
 concentration of  "naturally"  occuring Pb in ice during pre-industrial  times
 was less  than 1  ng kg-1  and  that  surface  snows  show a  ~ 200-to-300  fold
 increase above this background level.  These data,  along with those  collected
 by  Patterson and his colleagues  in  the SEAREX group, confirm  the hypothesis
 that  Pb  introduced  by  human   activities  is  ubiquitous  in  the   Northern
 Hemisphere.   Furthermore,  these data  allow  for a  better  understanding  of
 pollutant  dispersion from  Northern  Hemispheric  sources  and  provide  an inven-
 tory of  current  background levels  of  Pb  in continental  as  well  as  oceanic
 areas  (Shirahata et al. 1979, Schaule and Patterson 1981,  Settle  et  al.  1982,
 Flegal  and  Patterson   1982).    Whether  the  record  of   anthropogenically-
 introduced trace metals other than  Pb can  be  discerned  in Greenland snow and
 ice is still controversial  (Herron  et al.  1977; Boutron 1979a,b; Boutron  and
 Delmas  1980; Nrigau  1980;  Boutron 1980).   Much  more  data  gathering  and
 detailed sampling should be accomplished in Arctic  areas before this question
 can be adequately answered.

8.5.2.2   Trace Metals  -  Alpine Glaciers—Few data  are  available  on  time-
 series profiles of trace metals  in alpine glacier  ice and  snow. Jaworoski  et
 al. (1975)  reported Cd  and Pb  values  from  Storbreen Glacier, Norway.   The
 1954-72 profiles of Pb  show no  trend with  depth  but a slight  increase  in  Cd
 since  1965 appears.  These authors  have  recently published  metal data from a
 number of  alpine glaciers  including samples  from Norway,  the  Austrian  Alps,
 the  Nepalese  Himalayas,   the  Peruvian  Andes,  and  the   Ugandan   Ruwenzori
 (Jaworoski et al.  1981).   However,  their Pb values  from  Antarctic  snow and
 ice are  orders of  magnitude higher  than  accepted  values  (Murozumi  et al.
1969,  Boutron and  Lorius  1979,  Ng  and  Patterson 1981);  hence, their entire
 data set must be considered suspect.

Briat (1978) has measured  various trace metals in a  profile  (1948-74) on Mt.
Blanc   at  4280  m.   Much temporal variation  occurs  in  the  data,  but  Briat
 argues that there has been a two-fold increase  of Pb, Cd,  and V since  1950  in
the precipitation deposited at the Mt.  Blanc  site.

Based  on the review of  the literature,  with the  possible exception of  Pb, Zn,
and possibly V, one would  be hard put to argue  that the previous glaciochem-
ical work  has shown  that  fossil-fuel  burning has affected the precipitation
of glaciated areas.  One  of the problems with this  interpretation, however,
is the lack of data, especially  from alpine  glaciers in both areas  close  to
and remote from man's  activities.   In addition, the  previous  alpine  glacio-
chemical  studies  have produced time-series  of only a  few years.
                                    8-77

-------
In conclusion, the  alpine  glacier  data available could be considered  sparse
at best, unreliable at worst,  and the  limited  number of  glaciers sampled does
not  provide  an adequate  picture as  to the  regional  effect of  fossil-fuel
burning.

8.5.3  Discussion  and Future Work

With  the  exception   of   Pb,   S042-,  and   NOs"   in   the  northern  polar
regions, little conclusive  evidence  is available from  glacier  ice and snow
samples to interpret  with  any  certainty the effect of  fossil-fuel emissions
through  time.   The  large  majority  of  stratigraphic  information regarding
trace metals and anionic  acid  species concentrations  is from Antarctica and
Greenland.   Few if  any data come from  glacier ice and  snow in lower latitude
areas.  Because a  very large percentage of  fossil-fuel burning takes place in
the Northern Hemisphere,  the  Antarctic data provide little  historic  insight
into past and present anthropogenic emissions.  It is apparent, however, that
Antarctic data do  provide information  concerning background concentrations of
various chemical  constituents in frozen  precipitation.   Until  recently, the
glacier  data  can  be  termed  controversial  in  that different  workers have
interpreted the results  in  different  ways  (Herron et  al.  1977,  Murozumi  et
al. 1969,  Boutron  1980,  Nriagu 1980,  Landy et al.  1980, Boutron and  Delmas
1980).   The  most  recent work of  Ng  and Patterson  (1981)  and  Herron  (1982)
indicates more than a  two-order-of-magnitude  increase in Pb  in the Greenland
area and a factor  of two  increase in  sulfate and nitrate.

Even less information is  available  from alpine glaciers.  Although there is  a
suggestion that trace metal emissions  have  increased  in  alpine  ice  (Briat
1978) and that anthropogenic nitrate inputs occur in Canadian Rocky glaciers
(Butler et al. 1980),  it must be emphasized  that little definitive informa-
tion  is  available  at  this  time to  eludicate long-term historic trends  in
regions  where they  should  be  easily  detected(i. e.,  midlatitude   alpine
regions both close to and remote from  emission sites).

Owing  to  the  potential   post-depositional  modifications  inherent  in many
temperate  ice sampling  areas,  the  majority of  time-series  relationships
sought through ice  and snow  analyses  have  been conducted on polar glaciers.
Information concerning climatic events  and  hence  records  potentially per-
tinent  to  resolution  of chemical  time-series  in  polar  regions have been
retrieved  for  periods on  the  order  of 100  to 104 years  (i.e., Cragin et
al. 1975, Hammer et al. 1980).   Polar glaciers,  however, owing  to their low
accumulation rates  (mm to  cm  yr~l)  and unique  geographic  location  provide
only a portion of the  potential  snow  and ice  core record.    Full  realization
of the potential  climatic and,  therefore, chemical  sequences recoverable from
snow and ice studies  is currently in progress with the  addition of temperate
glacier  snow  and  ice cores (i.e., Thompson 1980;  Mayewski  et al. 1983a,b).
These  glaciers,  by  virtue  of  their  higher   accumulation  rates  (cm  to   m
yr-1),  provide short-term  time series  (10°  to  102  yr)  with  considerable
sub-annual  detail.   Proper selection of temperate  glacier core  sites, most
particularly with respect to elevation  and latitude is  necessary  if pristine
snow and ice samples, unaffected by post-depositional effects such as  melting
and  diffusion are  to  be  recovered   (Murphy  1970, Oeschger  et  al. 1977,
Thompson 1980, Davies et al. 1982, Mayewski  et al. 1983b).   As Hastenrath


                                    8-78

-------
 (1978)  has demonstrated through direct  measurement  of net short-  and  long-
 wave radiation and albedo on Quelccaya ice cap, Peru,  a  condition  of zero to
 negligible glacier surface melt  can  be maintained  if the sampling  site  is at
 a high enough altitude, in this case 5400 m,  even at 13°  56'  latitude.

 Although the recent  work  of  Herron (1982) has contributed greatly  to under-
 standing  the  effect  of  fossil-fuel  burning  on   precipitation  in  remote
 northern polar regions, more  detailed  ice sampling  and analyses of  the  past
 100  to  150 years record would provide a  better  comparison with  records  such
 as fossil-fuel burning through time in the Northern Hemisphere.

 Sampling on glaciers  requires  great  care  in  sample  collection, handling,  and
 analysis (Murozumi et al.  1969,  Vosters et  al.  1970,  Boutron 1979c, Boutron
 and Martin 1979,  Boutron  and  Delmas  1980).   With the  advent  of  "ultraclean"
 laboratories  and  procedures  as  well  as  more  sophisticated coring and/or
 sampling devices  (e.g., teflon coated augers  and PICO's new all kevlar coring
 unit) this, we believe, can be accomplished  for  at least the  anionic species
 of interest.   If care  in  sample acquisition and  handling  is taken, modern
 analytical techniques such  as isotope dilution mass spectrometry,  flameless
 atomic absorption, auto-analyzer visible spectrophotometry,   and ion chroma-
 tography can be used  to determine the various chemical  species of interest at
 extremely low levels.

 To ascertain what is controlling the pH of  the  snow and ice  sampled,  ionic
 balances must also be undertaken (Granat 1972).   This should  at least involve
 determining  N03~, and  S042'  as well as  Cl"  and  NH4+.  If possible  Na+,  K+,
 CaZ+,  Mg2*,  and  P043-  should  also  be  determined  in  each   sample.    With
 this  information  the  strong  mineral   acid   contribution  to the  total   H
 concentration can be  determined  independently of pH or acid  titration meas-
 urements.  In  addition to the glaciochemical studies,  more   information  is
 needed on  possible aerosol-snow  fractionation  and  aerosol source  location.
 Perhaps the most  serious concern raised  regarding the  use  of  glaciochemistry
 as an  historic  time-series  tool  is  the  possibility that atmospheric compo-
 sitions  are  not   fully  represented  in  resultant  surface snow compositions.
 Although the correlation between  the compositions at the  South Pole were  good
 (Zoller et al. 1974), similar studies in the Arctic yielded  no correlation
 (Rahn and McCaffrey 1979).

 Superimposed on these problems are the  effects of seasonality  of  transport in
 the northern polar  region  {Rahn and McCaffrey  1980, Rahn et al.  1980),  as
 well  as the time  lapsed between  precipitation events (i.e., dry  vs wet depo-
 sition)  and snow-air  fractionation (Rahn and McCaffrey 1979,  Davidson et  al.
 1981).   Rahn and  McCaffrey (1980) have suggested that  winter  Arctic  aerosols
 originate  from  polluted European  sources and  hence contribute fossil-fuel
 emission products to  northern  polar  ice  and  snow.   In addition, in  the case
of sulfate, the record in ice  cores may  be dampened  with respect to what  is
 observed in  the   atmosphere  (Scott 1981).   This demonstrates  the   need  for
complimentary  air and snow/ice studies to evaluate  properly  the results  of
 the  latter.    Little doubt  exists  that  the  aerosol-snow   link   requires
extensive  study  and  that  aerosol  studies  are  needed in  conjunction  with
 surface snow and  ice  sampling to enhance the  resolution capabilities of such
snow/ice  studies  (Davidson et  al.  1981).
                                  8-79

-------
In  addition,  aerosol  source and  possible cyclicity  in source(s)  must be
investigated  in  more detail.   Source  discrimination for  certain  chemical
species  has  been undertaken  in some glaciochemical  studies (Gorham 1958a,
Cragin et  al. 1975, Busenberg  and Langway 1979,  Herron 1982).   An effort
should be  made  to better qualify  the  source  of acids to  the snow  and  ice.
Samples  could be analyzed  for  F~  using ion  chromatography (Herron 1982).
Samples  with  high  F~  concentrations may  have  had  a significant  input of
volcanic acid  (Lazrus  et al. 1979, Stoiber et al. 1980).   Table 8-14  sum-
marizes the potential sources of chemical  species in  the  atmosphere and hence
glacier snow  and  ice,  with  estimations  of spatial  and  temporal  controls on
the input  of  these  species  to glacier sampling sites.  As an example of  the
type of data  needed  to quantify  the approach  taken in Table 8-14, decreases
in chemical concentration as a function  of  distance  in Antarctica  (Boutron et
al. 1972, Johnson and Chamberlain  1981)  have been investigated.  This type of
information is needed if  a  more quantitative  assessment  of  anthropogenic vs
natural sources is  to be  made.   Determining metal  or acid  sources may  also
clarify the nature and cause of the high aerosol enrichment  factors  observed
for most  volatile elements, even  in  remote  areas  (Dams and  DeJonge  1976,
Davidson et al. 1981).  Knowledge of the acid  source  in  frozen precipitation
is necessary if the problem of acid precipitation is to be completely under-
stood.

8.6  CONCLUSIONS

The following conclusions may be drawn from the preceding discussion  of depo-
sition monitoring.

 o   Although precipitation sampling networks  have  been  operated  many times
     at  many  locations,   assessments  of national  or regional  patterns   and
     trends must be cautiously used because of variability in the methods of
     collection and analytical  techniques.  Usually  the networks have been of
     limited spatial  or temporal  extent  (Section  8.1).

 °   Bulk  sampling,  used  in many networks, does not  generally provide  data
     useful in determining  quality  of precipitation,  although this  approach
     has some potential to estimate  total deposition  (Section 8.2.3).

 °   Automatic devices designed  to exclude dry  deposition   can  produce  wet
     deposition samples contaminated by  dry deposition if the protective lid
     does  not seal  the  collection bucket  tightly.   Wet  deposition  networks
     should be designed to estimate  dry  deposition contamination,  by  site and
     by chemical  element  (Section  8.2.3).

 o   Most  precipitation  chemistry networks have only measured  the  soluble
     fraction of the major inorganic ions.  This procedure is reasonable  for
     acidic wet deposition studies because  these  complements  generally can be
     used  to  predict a  pH that is close to the measured pH, especially  for
     samples with pH less  than 5.0  (Section 8.2.3).

 o   Understanding reasons for pH changes sometimes observed during  handling
     and storage  requires consideration of other  chemical  constituents and
     measurement of both  the soluble and insoluble fractions  (Section 8.2.3).


                                   8-80

-------
       TABLE 8-14.  POTENTIAL SOURCES FOR CHEMICAL SPECIES FOUND
                        IN SAMPLES OF GLACIER ICE
 Chemical
 Species
*1,4,5

Jipgenic
Emission
1,2,4,5,6

Crustal
Weathering
1,2

Lightning
Discharge
                                           1,2,4,5
Seasalt
         2,4,5
Volcanism
1,2,3,4,5!
Anthropo^
genie    j
Emission
Volatile
trace
metals
(Pb. Hg)
* Source Characteristics
                                  ? - species production from
                                      this source uncertain.
Temporal Distribution
     1 - cyclic (seasonal)
     2 - non-cyclic (inter-annual  &/or intra-annual)
     3 - significant only as of post-AD 1850
Spatial Distribution and magnitude of species
     4 - distance &/or elevation source to site
     5 - atmospheric circulation pattern source to site
     6 - aerial distribution of local  ice-free terrain
(increasing importance of factors  such as 5 (i.e., monsoonal  flow)  and  6
increase likelihood of 1 compared  to 2)
                                  8-81

-------
0   Sampling networks should be operated for periods  of many years to deter-
    mine  variability  in  the  general  patterns  of  precipitation  quality.
    Deposition patterns over time  are  highly variable because they include
    the variability  of both the  ion  concentration  and  the precipitation
    amount patterns (Sections 8.2.3 and 8.2.4).

0   Regional and national wet  deposition networks with automatic collectors
    have been operated  continuously in the  United States  and Canada since
    the late 1970's (Section 8.2.4).

0   These networks provide reasonable resolution  of major ion  concentrations
    for eastern  precipitation   but,  to  date,  only  an  indication  of  what
    western patterns might  generally be.   The  difference  in sampling site
    density accounts for  the difference in  our  knowledge  of precipitation
    chemistry in the two areas.  Inadequate site  density in the west will be
    corrected in the near  future through the National  Trends Network (Sec-
    tion 8.4.1).

0   Maximum sulfate, nitrate,  and  hydrogen  ion  concentrations  in precipi-
    tation  are  observed in  the northeast  quadrant  of the  United  States.
    Levels  decrease  to  the west,  south, and northeast  toward New England.
    Elevated  levels  extend into  southeastern   Ontario,  Canada  (Section
    8.4.1).

0   Highest  calcium concentrations  occur   in  the central  regions  of  the
    United States (Section 8.4.1).

°   Highest chloride concentrations occur along the coasts, consistent with
    a marine source (Section 8.4.1).

0   Patterns  for  each  of  these  ions are  fairly  consistent  with the known
    source regions (Section 8.4.1).

°   On the  broad  scale,  nitrate in U.S. precipitation has likely increased
    since  the 1950's,  in  conjunction with  NOX emissions increases (Section
    8.4.3.1).

o   Calcium measured  in U.S.  precipitation  has  decreased,  perhaps  due to
    lack of extreme drought recently as compared to the  1950's, but  more
    certainly due to improved sampling  procedures (Section 8.4.3.3).

o   A combination  of  drought effects,   the  mixing  of urban  data with more
    regionally representative  data,  and the mixing  of  bulk  data and lower
    quality wet-only  data with  higher quality  wet-only  data,   has  led to
    statements  concerning  increasing  acidity  of  precipitation  which  are
    quantitatively difficult to  support.   In general,  it appears difficult
    to use historical U.S. network data to  discern  the precipitation pH time
    trend as related to the acid precursor  emissions  (Section  8.4.3.2).

o   The  most reliable  long-term  trends  for  precipitation  chemistry  are
    available for  the  Hubbard  Brook Forest site in  New Hampshire (record
    continuous since  1964).    The  nitrate  data  record  suggests  an erratic


                                   8-82

-------
trend  of  increasing  nitrate  from  1964  to  about  1971,  followed by  a
leveling off or  slight  decrease  from  1971  to 1981.  Wet  sulfate  at  the
site declined by about 33 percent from 1965-66 to 1979-80.  Emissions of
NOX  and  SOX are  generally  consistent with  these observations for  wet
sulfate and nitrate.  Although the emissions for the Northeast track  the
wet deposition record especially  well,  it is not yet  possible  to reli-
ably and quantitatively separate out the contribution from long-range vs
short-range transport.  From 1964-77 there was no statistically signifi-
cant  trend in  precipitation  pH  at  the  Hubbard  Brook  site  (Sections
8.4.3.1 and 8.4.3.2).

Sulfate and hydrogen  ion  concentrations are much  higher  in  warm  season
precipitation in  the  eastern  United States  than in cold season precipi-
tation.  The trend  follows  the aerosol  sulfate trend  but  not  the trend
of SOX emissions (Section 8.4.4).

Although  precipitation  pH  in the northeastern  United States  has been
reported to have decreased  in the past 20  to 30 years,  several  recent
revaluations have  suggested  that the data  do not support the  idea of a
sharply decreasing pH trend (Section 8.4.3.2).

Remote  site  pH  data indicate that  the  common  reference  to  the  C02
atmospheric equilibrium value of pH  5.6  is  of limited  value.   Recent
measurements in  Hawaii  and other locations  not strongly  influenced  by
alkaline dust, have  indicated  that the  average  precipitation pH is less
than 5.0.  Samples at some remote sites  have been found to be chemically
unstable,  with  pH  rising with time,  due to  organic acid loss.   These
relatively acid  samples at  remote  sites need to be  explained  to  better
understand  the  acidic  samples  in  areas  with  strong  anthropogenic
influences (Section 8.4.2).

Air trajectory analysis,  frequently applied  to  precipitation  chemistry
in attempts to identify important source  regions  for receptor  sites,  is
qualitative at best.   Degree of success probably  varies  with  location.
Applying this fairly  simple approach  to such  a  complex problem leads to
doubts about the utility of the approach (Section 8.4.6).

Wet and dry deposition  processes are  roughly  of  equal  importance  in  the
average deposition of specific chemical  species (Section 8.3.1)

Direct methods of monitoring  dry deposition  consist of collecting ves-
sels,  surrogate  surfaces,   and  concentration  monitoring  from   which
deposition rates  are inferred.  The  latter applies to trace  gases  and
small particles  (<  1  to 5 ym  diameter),  i.e.,  where deposition  is  not
controlled by gravity.  Surrogate surface methods  apply to particles  of
a  size controlled  by  gravity  and   gases  for  which  species-specific
surfaces are used to evaluate air concentrations (Section 8.3.2.1)

Micrometeorological methods have been developed  as alternative monitor-
ing  techniques  for  surface  fluxes.   These  include  eddy-accumulation,
modified Bowen ratio, and variance (Section 8.3.2.2)
                               8-83

-------
Limited data are available on which to base estimates  of  dry  deposition
rates using concentration  techniques.   A  study  conducted for  sulfate,
nitrate, and ammonium in aerosol measured  in the  surface  boundary  layer
had a resolution of four-hour intervals and gave  average  diurnal cycles
of near-surface concentrations (Section 8.3.3)

Snow and ice cores collected  from appropriately chosen  glaciers  provide
samples of entrapped chemical  species.   This  technique has barely been
applied  to  the  study  of acid  precipitation  despite  the  fact that  it
provides a  very sensitive record  of  precipitation  chemistry.   Little
definitive information is available at this time  to  elucidate long-term
historic trends  in regions where  they  should  be  easily detected (i.e.,
midlatitude alpine regions both close  to  and remote  from emission sites)
(Section 8.5.3).
                             8-84

-------
8.7  REFERENCES

Allen, L. H., Jr., R.  J.  Hanks,  J.  K.  Aase, and H. R. Gardner.  1974.  Carbon
dioxide uptake by wide-row grain sorghum computed by the profile Bowen-ratio.
Agronomy J. 66:35-41.

Andren,  A.  W.  and S.  E. Lindberg.   1977.    Atmospheric input  and  origin
of selected  elements   in  Walker  Branch  Watershed,  Oak  Ridge,  Tennessee.
Water, Air, Soil Pollut.  8:199-215.

Anderssen, A. M., A. H. Johnson, and T. G. Siccama.  1980.  Levels of Pb, Cu
and  Zn  in  the  forest floor  of the  northeastern  U.S.    J.   Environ.  Qua!.
9:293-296.

Aristarain, A. J.  1980.  Otude glaciologique de la callote polaire de L'ile
James  Ross  (Peninsule Antarctique)  CNRS  Lab.  de Glaciol.  et Geophys.  de
L'Environ. 322. 130 p.

Ashworth, J. R.  1941.   Atmospheric pollution and the deposit gauge.  Weather
3:137-140.

Barrie,  L.   A.  and  A.  Sirois.   1982.    An   analysis   and  assessment  of
precipitation chemistry measurements made  by CANSAP (The Canadian Network for
Sampling  Precipitation):  1977-1980.    Report   AQRB-82-003-T.    Atmospheric
Environment Service, Downsview,  Canada.  163 pp.

Barrie, L. A., J. L. Walmsley.  1978.  A  study of sulphur dioxide deposition
velocities to snow in  northern Canada.  Atmos. Environ. 12: 2321-2332.

Barrie,  L.  A.,  J.  M.  Hales,  K.  G.  Anlauf,  J.  Wilson,  A.  Wiebe,  D.  M.
Whelpdale,  G.  J.  Stensland,  and  P.  W.  Summers.    1982.   Preliminary  data
interpretation.  U.S.  Canadian MOI, Monitoring  and Interpretation Sub-Group,
Report No. 2F-I.

Barrie, L.  A.,  H.  A.  Wiebe,  K.  Anlauf, and P. Fellin.  1980.   The Canadian
air  and  precipitation  monitoring  network  APN,  pp. 355-365.   _Ini Atmospheric
Pollution 1980.   M. M.  Benarie, ed.   Studies in Environmental Science, Vol.
8.  Elsevier Scientific Publishing  Company, Amsterdam.

Berner,  W.,  B.  Stauffer,   and  H.   Oeschger.    1978.    Past  atmospheric
composition and  climate,  gas  parameters  measured  on  ice  cores.    Nature
276:53-55.

Bertine, K. K.  and  E.  D. Goldberg.   1971.   Fossil  Fuel  combustion  and the
major sedimentary cycle.   Science  173:233-235.

Bonsang,  B.,  B.  C. Nguyen,  A. Gaudry,  and  G.  Lambert.   1980.   Sulfate
enrichment in marine aerosols owing to biogenic gaseous sulfur compounds. 0.
Geophys. Res. 85:7410-7416.
                                  8-85

-------
Boutron, C.  1979a.   Past and present day troposheric fallout fluxes of Pb,
Cd,  Cu, Zn  and  Ag  in  Antarctica  and  Greenland.    Geophys.  Res.  Lett.
6:159-162.

Boutron, C.  19795.   Trace element of  snows  of Greenland along an east-west
transect.  Geochim.  Cosmochim.  43:1252-1258.

Boutron, C.   1979c.    Reduction  of  contamination problems  in  sampling  of
Antarctic  snows  for   trace  element  analysis.    Analytica  Chimica  Acta
106:127-130.

Boutron, C.   1980.    Trace  metals in  remote Antarctica  snows:   Natural  or
anthropogenic?  Nature 284:575-576.

Boutron, C.   1983.    Respective influence of  global  pollution and volcanic
eruptions on the  past variations  of  the  trace  metals  content of Antarctica
snows since 1880's.   J. Geophys. Res.   In  press.

Boutron, C.  and R.  Delmas.   1980.   Historical  record of global atmospheric
pollution revealed in polar ice sheets.  Ambio  9:210-215.

Boutron, C.  and C.  Lori us.   1977.  Trace element  content  in East Antarctic
snow samples.   I.A.H.S. Publ.  118:164-171.

Boutron, C.  and C.  Lori us.   1979.   Trace metals  in  Antarctica snows since
1914.   Nature 277:551-554.

Boutron, C.  and S.  Martin.   1979.   Preconcentration  of  dilute solutions at
the  10"12  g/g  level  by  nonboiling  evaporation with variable  variance
calibration curves.   Analytical Chem.  51:140-145.

Boutron, C.,  M. Echevin,  and  C.  Lori us.   1972.   Chemistry  of polar snows.
Estimation of  rates  of deposition  in Antartica.  Geochim.  Cosmochim.  Acta
36:1029-1041.

Bowersox, V. C. and R. G.  dePena.   1980.   Analysis  of  precipitation chemistry
at a central Pennsylvania site.  J.  Geophys.  Res. 85:5614-5620.

Bowersox, V. C. and G. J.  Stensland.   1981.   Seasonal  patterns  of  sulfate and
nitrate in  precipitation  in  the United States.  Jji  Proceedings of the 74th
Annual  Meeting, Air  Pollution  Control  Association,  Philadelphia,  PA.  June
21-26.   Paper No. 81-6.1.

Brezonik,  P.   L.,  E.  S.  Edgerton,  and   C.  D.  Hendry.     1980.  Acidic
precipitation and sulfate deposition  in Florida.  Science 208:1027-1029.

Briat,  M.   1978.  Evaluation  of level  of  Pb, V, Cd,  Zn and Cu in  the  snow of
Mt. Blanc during the  last 25 years.   Atmos.  Pollut. 1:225-227.

Busenberg,  E.  and C.  C.  Langway, Jr.   1979.   Levels of ammonium,  sulfate,
chloride,  calcium and  sodium in snow  and  ice  from  southern Greenland. J.
Geophys. Res. 84:1705-1708.


                                  8-86

-------
Butler, D., B. Lyons, J. Hassinger, and P. A. Mayewski.   1980.  Shallow core
snow chemistry  of  Athabaska Glacier, Alberta.   Canadian J.  Earth  Sciences
17:278-281.

Chamberlain, A.  C.   1980.   Dry  deposition  of  sulfur dioxide, pp. 185-197.  In
Atmospheric Sulfur  Deposition.   D.  S.  Shriner,  C. R.  Richmond,  and S.  FT
Lindberg, eds.  Ann Arbor  Science,  Ann Arbor, MI.  568 pp.

Charlson,  R.  J.  and H. Rodhe.  1982.   Factors controlling  the  acidity  of
natural rainwater.   Nature 295:683-685.

Clark, H. L., K. E. Clark, and  B.  L.  Haines.   1980.   Acid rain in Venezuelan
Amazon.   In  Tropical Ecology and  Development,  p.  683, International  Society
of TropicaT Ecology, Kuala Lumper,  Malaysia.

Clarke, R. H., A. J. Dyer, R.  R. Brook, D. G.  Reid, and  A. J.  Troup.   1971.
The Wangara  experiment:    Boundary-layer  data.   Tech. Paper  No.  19,  CSIRO,
Division of Meteorological  Physics, Aspendale,  Victoria,  Australia.  362  pp.

Clement, P. and J.  Vandour.  1967.  Observations on the pH of melting snow in
the Southern French Alps.   Arctic Alpine Environ. 205-213.

Cogbill, C. V.  and G.  E.  Likens.   1974.   Precipitation  in the  northeastern
United States.  Water Resour. Res.  10:1133-1137.

Cowling,  E.  B.  1982.    Acid  precipitation  in  historical  perspective.
Environm. Sci. Techno!. 16(2):110A-123A.

Cragin, J. H., M. M. Herron, and C.  C. Langway, Jr.  1975.  The chemistry of
700 years  of  precipitation  at  Dye  3, Greenland.   Cold  Regions  Research and
Engineering Laboratory, Research Report 341.

Dams,  R. and J. DeJonge.   1976.  Chemical composition of Swiss aerosols from
the Jungfraujoch.  Atmos.  Environ.  10:1079-1084.

Dana,   M.  T.   1980.    Distribution  of  contaminants.    Research  Report  To
American  Electric   Power   Corporation  by Battelle  Columbus  Laboratories,
Columbus, OH.  79 pp.

Dana,  M.  T.,  J.  M.  Hales,  and M.  A. Wolf.   1975.   Rain scavenging  of SQz
and sulfate from power  plant plumes.  J. Geophys. Res. 80:4119-4129.

Dasch, J.  M.   1982.   A comparison of  surrogate  surfaces for dry deposition
monitoring,  Proceedings   4th  International   Conference   on  Precipitation
Scavenging, Dry  Deposition,  and  Resuspension, Santa  Monica,  California,  23
November to 3 December  1982.

Davidson, C. I., J.  M. Miller, and M.  A. Pleskow.  1982.   The influence  of
surface  structure  on  predicted  particle  dry deposition  to  natural  grass
canopies.  Water, Air,  and Soil Pollut. 18:25-43.
                                  8-87

-------
Davidson, C. I., C. Liyang, T. C. Grimm, M. Nasta, and M. P. Qumooss.  1981.
Wet  and dry  deposition  of  trace  elements  into the  Greenland  ice  sheet.
Atmos. Environ. 15:1-9.

Davies,  T.  D., C.  E. Vincent,  and  P.  Brimblecombe.   1982.   Preferential
elution of strong acids  from a Norwegian  ice cap.  Nature 300:161-163.

Dawson,  G.  A.   1978.   Ionic  composition of rain during  sixteen convective
showers.  Atmos. Environ.  12:1991-1999.

Delmas, R. J.  1982.  Antarctic sulphate  budget.  Nature 299:677-678.

Delmas,  R.  and  A.  Aristarain.   1979.  Recent Evolution of Strong Acidity of
Snow  at Mt.  Blanc.   Proc.  13th   Inter.  Coll.,  Atm.  Pollut.,  Volume  1,
Elsevier, Amsterdam 233-237.

Delmas,  R.  and  C. Boutron.   1978.   Sulfate  in  Antarctic snow:   Spatio-
temporal distribution.  Atmos. Environ. 12:723-728.

Delmas,  R.   and  C.   Boutron.    1980.    Are   the  past  variations  of  the
stratospheric  sulfate burden  recorded  in  central   Antarctic  snow  and  ice
layers?  J.  Geophys. Res.  85:5645-5649.

Delmas,  R.,  A.  Aristarain, and  M.  Legrand.    1980a.   The  acidity  of polar
precipitation:   a  natural  reference level  for acid rains.   In Ecological
Impact  of  Acid Precipitation. 0.  Drablos  and  A. To!Ian,  eds. Proc.  of  an
International Conference,  Sandefjord, Norway.  SNSF Project,  Oslo.

Delmas,  R., J. M.  Ascencio, and  M. Legrand.  1980b.   Polar  ice evidence that
atmospheric C02 20,000 year B.  P.  was 50% of  present.   Nature 284:155-157.

Delmas,  R., M. Briat, and  M. Legrand.  1982.   Chemistry of south polar snow.
J. Geophys. Res. 87:4314-4318.

dePena,  R.   G.,  J.  A.  Pena,  and  V.  C.  Bowersox.    1980.   Precipitation
collectors  intercomparison  study.    Dept.  of  Meteorology,  The Pennsylvania
State University, State College.  57  pp.

Desjardins,  R.  L.   1977.    Energy budget by  an eddy  correlation method.   J.
Appl. Meteorol. 16:248-250.

Dillon,  P.  J., D.  S. Jeffries, and W.  A.  Scheider.   1982.    The  use  of
calibrated  lakes and  watersheds  for  estimating  atmospheric  deposition near a
large point source, Water, Air, and Soil  Pollut. 18:241-250.

Dolske, D. A. and F. D.  Gatz.  1982.  A  field  intercomparison of sulfate dry
deposition  monitoring  and  measurements  methods:    Preliminary  results,
presented at the  American  Chemical  Society Acid Rain Symposium, Las Vegas,
Nevada.  30 March, 1982.

Dovland,  H., and  A.  Eliassen.  1976.    Dry  deposition on  a  snow surface,
Atmos. Environ. 10:783-785.


                                 8-88

-------
Eaton, J.  $.,  G.  E.  Likens, and F. H. Bormann.  1978.  The input of gaseous
and particulate sulfur to a forest ecosystem.   Tell us  30:546- 551.

Elgmork, K., A. Hagen,  and A. Langfland.  1973.   Polluted  snow in southern
Norway during the winters, 1968-1971.   Environ.  Pollut. 4:41-52.

Eriksson,  E.   1952.    Composition  of  atmospheric  precipitation.   Tell us
4:280-303.

Fisher,  D.,  A. Gambell, G.  Likens,  and  F.  Bormann.   1968.   Atmospheric
contributions to water  quality  of streams  in the Hubbard Brook Experimental
Forest, New Hampshire.  Water Resour.  Res.  4:1115-1126.

Flegal,  A.  R.  and  C.  C.  Patterson.    1982.    Lead  in the  central  Pacific
(Abstract).  Amer. Geophys. Union:L041A-15.

Forland,  E.  J.   1973.    A study  of the acidity  in the  precipitation  in
southwestern Norway.   Tell us 25:291-298.

Foster, J. F., G. H.  Beatty,  and J. E.  Howes,  Jr.   1974a.   Final  report on
inter!aboratory  cooperative  study  of  the  precision and  accuracy of  the
measurement of dustfall  using ASTM method D1739, ASTM  Data Series Publication
DS 55-S4.  45 pp.

Foster, J. F., G.  H.  Beatty, and J.  E.  Howes,  Jr.   1974b.   Final  report on
interlaboratory cooperative study  of  the  precision  and  accuracy  of  the
measurement of sulfation in the  atmosphere  using ASTM  method D2010, ASTM Data
Series Publication DS 55-S2.  45 pp.

Galloway,  J.  N.  and  G.  E. Likens.    1976.    Calibration  of  collection
procedures for the  determination of precipitation chemistry.  Water,  Air and
Soil Pollut.  6:241-258.

Galloway, J.  N. and G. E. Likens.  1978.  The collection of precipitation for
chemical analysis.  Tellus 30:71-82.

Galloway, J. N.  and G.   E.  Likens.   1979.   Atmospheric enhancement of metal
deposition in Adirondack lake sediments.  Limnol.  Oceanogr. 24:427-433.

Galloway, J.  N. and D. M. Whelpdale.   1980.   An  atmospheric sulfur budget for
eastern North America.  Atmos. Environ.  14:409-417.

Galloway, J. N.,  B. J.  Cosby, and G. E. Likens.  1979.  Acid precipitation:
Measurement of  pH and acidity.  Limnol.  Oceanogr.  24:1161-1165.

Galloway, J. N., G.  E.  Likens,  W. C. Keene,  and  J.  M. Miller.  1982.   The
composition of precipitation in  remote areas of the world.  J. Geophys. Res.
87:8771-8786.

Gatz, D. F.  1980.   Associations  and mesoscale spatial  relationships  among
rainwater constituents.   J. Geophys.  Res. 85:5588-5598.
                                  8-89

-------
GCA Corporation.   1980.   Acid  Rain Information Book -  Draft Final Report.
DOE Contract AC02-79EV10273-1, GCA Corp.,  Bedford, MA.

Gorham, E.   1958a.   Atmospheric pollution by hydrochloric  acid.   Quart. J.
Royal  Meteorol.  Soc. 84:274-276.

Gorham, E.  1958b.  The  salt content of  some  ice samples  from Nordaustlandet
(North East Land)  Svalbard. J.  Glaciol.  3:181-186.

Granat, L.  1972.   On the relation between pH  and the chemical composition in
atmospheric precipitation.  Tell us 24:550-560.

Granat, L.   1978.   Sulfate  in  precipitation  as  observed by  the European
atmospheric chemistry network.   Atmos.  Environ.  12:413-424.

Granat, L., R.  Soderlund, and  L. Back!in.   1977.  The IMI  network  in Sweden -
present  equipment,  methods  and   plans   for   improvement.     International
meteorological  Institute in Stockholm  Report AC-40  (November 1977).

Haines, B.   1983.   Forest ecosystem  S04-S  input-output discrepancies and
acid rain:  Are they related.  Oikos 41:139-143.

Hammer,  C.  U.    1977.    Past   volcanism  revealed by  Greenland  ice   sheet
impurities.  Nature 270:482-486.

Hammer, C.  U.   1980.   Acidity  of  polar  ice cores in  relation to absolute
dating, past volcanism and radio-echoes.   J. Glaciology 25:359-372.

Hammer, C.  U.,  H.  B.  Clausen,  and W.  Dansgaard.  1980.   Greenland  ice  sheet
evidence  of  past-glacial  volcanism  and  its  climatic   impact.    Nature
288:230-235.

Hansen, D.  A.  and  G.  M.  Hidy.   1982.   Review of questions regarding  rain
acidity data.  Atmos. Environ.  16:2107-2126

Hansen, D.  A.,  G.  M.  Hidy, and G.  J.  Stensland.   1981.   Examination of the
basis  for  trend interpretation  of  historical rain  chemistry  in  the Eastern
United  States.    ERT  P-A097,  Environmental  Research  and Technology,   Inc.,
Westlake Village, CA.

Hardy, E. P., Jr.  and 0.  H. Harley, eds.   1958. Environmental  contamination
from weapons tests.  U.S. AEC  Health and  Safety  Laboratory Report  HASL-42A.

Hastenrath,  S.   1978.    Heat  budget measurements on the  Quelccaya ice  cap,
Peruvian Andes.  J. Glaciology 20:85-97.

Herman, F.  and F. Gorham.  1957.  Total mineral  material,  acidity, sulfur and
nitrogen in rain and snow at Kentville, Nova Scotia. Tell us 9:180-183.

Herrara,  R. A.    1979.    Nutrient distribution  and cycling  in  an  Amazonian
caatinga  forest on spodosols  in Southern Venezuela.  Ph.D. Thesis, Dept. of
Soil Science, Univ. of Reading,  England.


                                  8-90

-------
Herron,  M.  M.    1982.    Impurities  of  F- ,  Cl",  NOr-,  and  S0d2-  in
Greenland and Antarctic precipitation.  J. Geophys. Res. 87(C4)-.3052-3060.

Herron, M. M. and C. C. Langway, Jr.  1979.   Dating  of  Ross Ice Shelf cores
by chemical analysis.  J.  Glaciology 24:345-357.

Herron,  M.  M.,  C.  C. Langway,  H.  V.  Weiss,  and  J.  H.  Cragin.    1977.
Atmospheric trace metals  and  sulfate in  the  Greenland  ice sheet.   Geochim.
Cosmochim. Acta 41:915-920.

Hewson, E. W.  1951.   Atmospheric pollution, pp.  1139-1157.  ^n Compendium of
Meteorology.  American Meteorological  Society, Boston, MA.,  1334 pp.

Hicks, B. B.  1981.   An analysis  of  Wangara micrometeorology: surface stress,
sensible  heat,  evaporation,  and dewfall.    NOAA  Technical Memorandum  ERL
ARL-104.  June 1981.   34 pp.

Hicks, B. B.  1982.   Measurement techniques:   dry deposition.   Presented at
the National Research Council  (Canada) Symposium  on Monitoring and Assessment
of  Airborne Pollutants with  Special  Emphasis on  Long Range  Transport  and
Deposition  of Acidic  Materials.    August  30-  September   1,  Ottawa,  Canada
(NOAA/ARATDL Cont.  No. 82/7).

Hicks, B.  B.  and M.  L. Wesely.  1980.  Turbulent  transfer processes  in the
canopy  and  at  vegetation  surfaces,  pp.  199-207.    In  Atmospheric  Sulfur
Deposition.  D. S. Shriner, C. R. Richmond, and  S.  E.Ti'ndberg,  eds.).   Ann
Arbor Science, Ann  Arbor,  MI.   568  pp.

Hicks, B.  B., M. L.  Wesely, and J.  L. Durham.  1980.  Critique of methods to
measure dry deposition (workshop  summary).  U.S.  EPA  Report EPA-600/9-80-050,
69 pp.  NTIS PB81-126443.

Hidy, G.  M.  1982.   Bridging  the  gap between air  quality  and precipitation
chemistry.  Water,  Air, Soil  Pollut.  18:191-198.

Hidy, G. M., D. A. Hansen, R.  G. Hendry, K. Ganesan,  and  J. Collins.   1984.
Trends  in  historical  acid  precursor  emissions   and  their  airborne  and
precipitation products. J.  Air Pollut. Contr. Assoc.  31(4)-.333-353.

Hilst, G.  R.,  P.  K. Mueller,  G.  M. Kidy,  T.  F.  Lavery,   and J.  G.  Watson.
1981.  EPRI Sulfate  Regional Experient:   Results and  Implications.  Electric
Power Research Institute,  Palo Alto, California.  Report No.  EA-2165-SY-LD.

Hornbeck,  J.  W.    1981.   Acid  rain:   Facts and  fallacies.   J.  Forestry
79:438-443

Jaworoski,  Z., M.  Bysiek, and L. Kownacka.    1981.   Flow  of metals into the
global atmosphere.   Geochim.  Cosmochim. Acta 45:2185-2199.

Jaworowski, Z., J.  Bilkiewicz, E. Dobosz, and L. Wodkiewicz.   1975.   Stable
and  radioactive  pollutants  in  a   Scandinavian  glacier.    Environ.  Pollut.
9:305-315.


                                  8-91

-------
Jeffries, D.  S.  and W. R.  Snyder.   1981.   Atmospheric  deposition  of heavy
metals in central  Ontario.   Water,  Air,  Soil  Pollut. 15:127-152.

Jickells, T., A.  Knap,  T.  Church,  J. Galloway, and J. Miller.   1982.   Acid
rain on Bermuda.   Nature 297:55-57

Johnson, B.  B.  and J. M.  Chamberlain.   1981.   Sodium, magnesium, potassium
and  calcium  concentrations  in  ice  cores  from the  Law  Dome,  Antarctica.
Geochim. Cosmochim. 45:771-776.

Johnson, S. A.,  R. Kumar,  P. T. Cunningham,  and T. A. Lang.  1981.  The MAP3S
aerosol  sulfate   acidity  network:     a   progress  report  and   data  summary.
Argonne National  Laboratory Report  ANL-81-63.   139 pp.

Jordan, C.,  F. Golley,  J.  Hall, and J.  Hall.  1980.  Nutrient scavenging of
rainfall by the  canopy of an Amazonian rain  forest.  Biotropica 12:61.

Junge, C. E.  1958.   The distribution of ammonia and  nitrate  in rain water
over the United  States.  Trans. Amer. Geophys.  Union 39:241-248.

Junge,  C.  E.  1963.   Air  Chemistry  and Radioactivity.   Academic Press, New
York.  382 pp.
Keene, W. C., J. N. Galloway,  and J. D. Hoi den.  1983.  Measurements of weak
organic acid-
Res. 88:5122-
organic acidity in  precipitation  from remote areas of the world.  J. Geophys.
     	1-5130.
Kerr, R. A.   1981.   Pollution of the Arctic  atmosphere  confirmed.   Science
212:1013-1014.

Koerner, R. M. and D.  Fisher.  1982.  Acid  snow in  the Canadian high arctic.
Nature 295:137-140.

Koide, M.  and E. D.  Goldberg.   1971.   Atmospheric sulfur  and  fossil  fuel
combustion.  J. Geophys. Res. 76:6589-6595.

Koide,  M.,  E.  D.   Goldberg,  M.  M.  Herron,  and  C.  C.  Langway.    1977.
Transuranic  deposit!onal  history in  South  greenland  firn  layers.   Nature
269:137-139.

Koide, M. K., R. Michel, E. D. Goldberg,  M.  M. Herron, and C. C. Langway, Jr.
1979.   Deposit!onal   History of  artificial   radionuclides  in  the  Ross  Ice
Shelf, Antarctica.  Earth Planetary  Sci.  Lett. 44:205-223.

Kyle, P., J.  Palais, and R.  Delmas.   1982.  The volcanic  record of Antarctic
ice  cores:    Preliminary  results and  potential for  future investigations.
Annals of Glaciology 3:172-177.

Landy, M.  P.  and D.  A. Peel.  1981.  Short term fluctuations in heavy metal
concentrations in Antarctic snow. Nature 291:144-146.
                                  8-92

-------
Landy, M.  P.,  D.  A. Peel,  and  E.  W. Wolff.   1980.   Trace metals in remote
Arctic snows:  natural  or anthropogenic?   Nature 284:574-574.

Langway, C.  C., Jr.   1970.   Stratagraphic analysis of a  deep  ice core from
Greenland.  The Geological  Society  of America.  Special Paper 125.

Langway,  C.  C.,  J.  H.  Cragin,  G.  A.  Klouda,   and  M.   M.  Herron.   1975.
Seasonal variations  of chemical continuents  in  annual  layers  of Greenland
deep ice deposits.  CRREL Rpt.   347.  5pp.

Lantzy, R. J. and F. T.  Mackenzie.   1979.   Atmospheric trace metals:  global
cycles and assessment of man's  impact. Geochim. Cosmochim. Acta 43:511-525.

Larson, T.  E.  and  I.  Hettick.   1956.   Mineral   composition  of rainwater.
Tellus 8:191-201.

Lawson, D. R. and J. W.  Winchester.   1979.   Sulfur, Potassium and Phosphorus
Associations in  Aerosols  from  South  American Tropical   Rain  Forests.   J.
Geophys. Res. 84:3723-3727.

Lazrus, A. L.,  R.  D. Cadle, B. W.  Gandrud,  J. P.  Greenberg,  B. J.  Huebert,
and W. I.  Rose.  1979.   Sulfur and  halogen chemistry of the stratosphere and
of volcanic eruption plumes. J. Geophys.  Res.  84:7869-7875.

Legrand, M. R., A. J. Aristarain,  and R. J.  Delmas.  1982.  Acid titration of
polar snow.  Anal. Chem.  54:1336-1339.

Lepel, E.  A., K.  M.  Stefansson  and W. H. Zoller.   1978.   The enrichment of
volatile elements in the atmosphere by volcanic activity:   Augustine Volcano
1976.   J.  Geophys. Res. 83:6213-6220.

Lerman, A.   1979.   Geochemical  Processes:   Water  and Sediment Environment.
John Wiley and Sons, New York.   481  pp.

Leuning, R., M.  H.  Unsworth,  H.  N.  Newman, and  K.  M.  King.    1979.   Ozone
fluxes  to  tobacco and  soil under  field  conditions.    Atmos.  Environ.  13:
1155-1163.

Lewis, W.  M., Jr. and M. C. Grant.   1980.   Acid precipitation in the Western
United States.   Science 207:176-177.

Li, Ta-Yung and H. E. Landsberg.   1975.   Rainwater pH  close to a major power
plant.  Atmos.  Environ. 9:81-88.

Likens, G. E.  1976.   Acid precipitation.  Chem. Eng. News 54:29-44.

Likens, G.  E.  and F.  H. Bormann.   1974.   Acid  rain:    a  serious  regional
environmental problem.   Science  184:1176-1179.

Likens, G. E. and  T.  J. Butler.   1981. Recent  acidification of precipitation
in North America.   Atmos. Environ.  15:1103-1109.
                                 8-93

-------
Lindberg, S.  E.,  and G.  M.  Lovett, 1982;   Dry deposition  of particles to
inert  and   foliar   surfaces  in   a   foresting  canopy,   Proceedings   4th
International  Conference  on  Precipitation Scavening,  Dry  Deposition,  and
Resuspension, Santa  Monica,  California, 28  November  to 3 December 1982.

Lindberg, S. E. and  R.  C.  Harriss.   1981.   The  role  of atmospheric deposition
in an eastern U.S. deciduous forest. Water, Air, Soil Pollut.  16:13-31.

Lindberg,  S.  E.,  R. C.  Harriss,  and  R. R.  Turner.   1982.   Atmospheric
deposition of metals to  forest vegetation.  Science  215:1609-1611.

Lyons, W.  B., P. A. Mayewski,  and N.  Ahmad.   1982.    Acidity of Recent
Himalayan Snow.  38th Eastern Snow Conf.

Mamane, Y.,  E. Ganor, and A.  E. Donagi.  1980.  Aerosol composition  of urban
and desert origin in the  eastern Mediterranean  I.   Water,  Air, Soil Pollut.
14:29-43.

Masse, C., and E.  Voldner.  1982.   Estimation of dry deposition velocities of
sulfur  over  Canada  and  the United  States  east  of  the   Rocky  Mountains,
Proceedings  4th  International Conference  on   Precipitation  Scavenging,  Dry
Deposition,  and  Resuspension, Santa Monica,  California,   28  November to  3
December 1982.

Mayewski, P. A., W.  B. Lyons,  and  N. Ahmad.  1983a.  Chemical  composition of
a high altitude fresh snowfall in the Ladakh Mountains.   Geophys. Res. Lett.
10:105-108.

Mayewski,  P.  A.,  W.  B.  Lyons,  and N.  Ahmad.    1983b.    Reconnaissance
Glacio-chemical studies in the Indian Himalayas, 38th  Eastern Snow  Conf. In
press.

Meiner, F.  X., H. W. Georgii, G. Ockelmann,  H. Jager, and R.  Reiter.  1981.
The arrival of the Mt.  St. Helens eruption cloud over Europe.  Geophys. Res.
Lett. 8:163-166.

Miller, J.  M.  1981.   Trends in  precipitation composition and deposition.
Work Group 2, Report 2-14, United States-Canada.  Memorandum  of Intent, July.
Chapter II.

Miller,  M.  L.  and  G.  E.   Everett.   1979.    A  detailed  analysis  of  the
scientific evidence  concerning acidic precipitation.  Am. Chem. Soc. Meeting.
Washington, D.C.  155-157.

Miller, J. M., J.  N. Galloway, and G. E.  Likens.  1978.  Origin of air masses
producing  acid  precipitation at  Ithaca,  New  York.   Geophys.  Res. Letters
5:757-760.

Miller, J.  M., G. J. Stensland,  and R.  G. Semonin.   1984.    The chemistry of
precipitation on  the island of Hawaii.   NOAA Technical Report, Air  Resources
Laboratory, Rockville,  MA. In press.
                                  8-94

-------
Murphy,  E.  J.   1970.   The generation  of electromotive  forces  during  the
freezing of water.  J. Colloid  and Interface Sci.  32(1):1-11.

Murozumi, M., T. J. Chow, and C. Patteson.   1969.   Chemical  concentrations of
pollutant lead  aerosols, terrestrial  dusts  and sea  salts  in Greenland  and
Antarctic snow strata.  Geochim. Cosmochim. Acta 33:1247-1294.

National Atmospheric Deposition Program.   1978.   National  Atmospheric  Program
Data  Reports.   Vol.   I  (1-2).   Available from  NADP .Coordinator's  Office,
Natural Resource Ecology Laboratory, Colorado State University,  Fort Collins,
CO.

National Atmospheric Deposition Program.   1979.   National  Atmospheric  Program
Data  Reports.   Vol.   II  (1-4).  Available from  NADP Coordinator's  Office,
Natural Resource Ecology Laboratory, Colorado State University,  Fort Collins,
CO.

National Atmospheric Deposition Program.   1980.   National  Atmospheric  Program
Data  Reports.  Vol.  Ill (1-4).   Available from  NADP Coordinator's  Office,
Natural Resource Ecology Laboratory, Colorado State University,  Fort Collins,
CO.

National  Academy  of  Sciences.    1978.    The  Tropospheric  Transport  of
Pollutants  and  other  Substances  to  the  Oceans.    National  Academy  Press,
Washington, D.C.  243  pp.

National  Academy  of  Sciences.   1980.   The Atmospheric  Sciences:  National
Objectives for the 1980's.  National Academy Press, Washington,  D.C.  130 pp.

National Academy  ofSciences.   1983.   Acid  Deposition:  Atmospheric  Processes
in  Eastern North America.  National Academy Press, Washington, D.C.  375 pp.

Ng, A.  and C. C. Patterson.  1981.  Natural concentrations of lead in  ancient
Arctic  and Antarctic  ice.  Geochim. Cosmochim. Acta 45:2109-2121.

Niemann,  B.   L.,  J.  Root, N.  Van Zwalenburg,  and  A. L.  Mahan.   1979.   An
integrated  monitoring network for  acid  deposition:   A  proposed  strategy.
Interim report R-023-EPA-79.   236 pp.

Nriagu, J.  1979.  Global inventory of natural and anthropogenic emissions of
trace  metals  to the atmosphere.  Nature 279:409-411.

Nriagu, J.  0.    1980.    Trace metals in  remote Arctic  snows:   natural  or
anthropogenic?  Replies.  Nature 284:575-577.

Oeschger, H., U.  Schotterer,  B. Stauffer,  W.  Haeberli, and H. Rothlisberger.
1977.   First results  from alpine core drilling projects.  Ziet.  fur. Gletsch.
und Glazial 13:193-208.

Oeschger, H., B.  Stauffer, A.  Neftel, J.  Schwander,  and  R.  Zumbrunn. 1982.
Atmospheric  C02  content  in the past  deduced  from ice-core analysis.   Annals
of  Glaciology 3:227-232.


                                  8-95

-------
Ottar, B.  1978.  An assessment of the OECD  study on long range transport of
air pollutants (LFAP).   Atmos.  Environ. 12:445-454.

Owens, J.  S.   1918.   The  measurement of atmospheric pollution.   Quart.  0.
Roy. Meteorol. Soc.  44:149-170.

Pack, D. H.  1980.   Precipitation  chemistry  patterns: A two-network data set.
Science 108:1143-1145.

Pack, D. H.  1982.   Precipitation chemistry probability—the shape of things
to come.  Atmos. Environ.  16:1145-1157.

Pack, D. H. and D.  W.  Pack.  1980.  Seasonal  and  annual behavior of different
ions  in acidic  precipitation.    World  Meteorological  Organization  Special
Environmental  Report No. 14,  303-313.

Parker,  B.  C.,  E.  J.   Zellar, L. E.  Heiskell,  and  W.  J.  Thompson.   1977.
Nitrogen in South Polar ice  and  snow:  Tool to  measure  past solar,  auroral
and cosmic ray activities.   Antarctic  J.  133-134.

Parker,  B. C.,  E. J.  Zellar, and A. J.  Gow.  1982.  Nitrate fluctuations in
Antarctic  snow  and  firn:  Potential  sources  and  mechanisms  of   formation.
Annals of Glaciology 3:243-248.

Peden,  M.  E.   and L.  M. Skowron.   1978.   Ionic stability of  precipitation
samples.  Atmos. Environ.  12:2343-2349.

Radke,  L.  F.    1982.    Sulphur  and sulphate  from  Mt.   Erebus.    Nature
299:710-712.

Rahn,  K. A.  and R. J.  McCaffrey.  1979.   Compositional  differences between
Arctic aerosol and snow.  Nature  280:479-480.

Rahn,  K. A. and  R.  J.  McCaffrey.   1980.   On the  origin  and transport of  the
winter Arctic  aerosol.  J_n Aerosols:  Anthropogenic  and  Natural,  Sources  and
Transport.  Annals N.Y. Acad. Sci. 338:486-503.

Rahn,  K. A.,  E. Joranger,.A.  Semb,  and T. J. Conway.   1980.   High winter
concentration  of S02  in  the  Norwegian  Arctic and  transport  from Eurasia.
Nature 287:824-826.

Raynaud,  D.,   R. Delmas,  J.  M.  Ascencio,  and   M.  Legrand.    1982.     Gas
extraction from  polar  ice  cores:  a  critical  issue  for  studying  the evolution
of  atmospheric  C0£  and ice  sheet surface  elevation.   Annals  of  Glaciology
3:265-272.

Raynor,  G.  S.  and  J. V.   Hayes.    1981.   Acidity  and  conductivity  of
precipitation on central Long Island,  New York, in relation  to meteorological
variables.  Water, Air, Soil Pollut. 15:229-245.

Risbo, T., H.  B. Clausen,  and K.   L. Rasmussen.  1981.  Supernovae  and nitrate
in  the Greenland ice sheet.  Nature 294:637-639.


                                  8-96

-------
Robertson, J.  K., T.  W.  Dobzine,  and  R.  C. Graham.   1980.   Chemistry of
precipitation   from   sequentially   sampled   storms.      U.S.   EPA   Report
600/4-80-004.  p.  116.

Schaule,  B.  K. and  C.  C.  Patterson.   1981.   Lead concentrations  in  the
northeast Pacific:   Evidence  for  global  anthropogenic  perturbations.   Earth
Planet.  Sci. Lettr.  54:97-116.

Scholander, P.  F., E.  A.  Hemmingsen, L. K. Coachman,  and D. C. Nutt.  1961.
Composition of Gas Bubbles in  Greenland  Ice Bergs. J. Glaciology 3:813-822.

Scott,  B.  C.  1981.   Sulfate  washout ratios in  winter  storms.    S.  Appl.
Meteorol. 20:619-625.

Settle, D. M., C. C. Patterson,  K.  K.  Turekian,  and J.  K.  Cochran.  1
Lead  precipitation fluxes at tropical  oceanic  sites determined  from P
measurements.  J.  Geophys. Res.  87(C2):1239-1245.

Sequeria, R.   1981.   Acid Rain:   some  preliminary  results from global  data
analysis.  Geophys. Res.  Lett. 8:147-150.

Shannon,  J.   D.   1981.    A  model  of  regional   long-term   average   sulfur
atmospheric  pollution,  surface  removal,  and net  horizontal  flux.    Atmos.
Environ. 13:1155-1163.

Shaw, G.  E.   1982a.   Evidence for  a  central  Eurasian  source area of  Arctic
haze  in Alaska.  Nature 299:815-818.

Shaw, G. E.  1982b.  On the residence time  of the Antarctic ice sheet sulfate
aerosol.  J.  Geophys. Res. 87(C6):4309-4313.

Sheih, C. M., M. L. Wesely,  and B. B. Hicks.   1979.  Estimated  dry deposition
velocities of  sulfur over the eastern United  States and  surrounding regions,
Atmos. Environ.  13:1361-1368.

Shirahata,  H.,  R.  W.  Elias,   C.  C.   Patterson,  and   M.   Koide.    1979.
Chronological  variations  in  concentrations  and  isotopic  compositions of
anthropogenic  lead in  sediments  of  a remote subalpine  pond.    Contribution
#3186:  Division of Geological and Planetary Sciences.  Cal.  Inst. Tech.

Sickles,  J.  E. Ill,  W.  D.  Bach,  and L.  L.  Spiller.   1982.   Comparison of
several  techniques  for  determining  dry   deposition  flux,  Proceedings  4th
International  Conference  on   Precipitation  Scavenging,  Dry  Deposition   and
Resuspenion,  Santa Monica, California, 28  November to 3 December  1982.

Smith, R. A.   1872.  Air and  Rain - The  Beginnings of a Chemical  Climatology.
Longmans, Green, and  Co., London,  England.

Spicer,  C.   W.  and  P.  M.  Schumacher.     1977.   Interferences  in  sampling
atmospheric particulate nitrate.   Atmos. Environ. 11:873-876.
                                  8-97

-------
Stallard, R.  F.  and  0.  M. Edmond.   1983.   Geochemistry  of the Amazon  I:
Precipitation chemistry and the marine contribution  to  the  dissolved  load  at
the time of peak  discharge.  J.  Geophys. Res.  88:9671-9688.

Stauffer, B.  and W.  Berner.    1978.   C02  in Natural   Ice.   J.  Glaciology
21:291-300.

Stensland, G. J.   1979.  Calculation of precipitation pH, with application  to
the Junge data,  pp.  79-108.    In  Study  of Atmospheric  Pollution Scavenging.
DOE Contract EY-76-S-02-1199.  TTlinois State  Water Survey,  Urbana, IL.

Stensland, G. J.   1980.   Precipitation  chemistry  trends in the  Northeastern
United States, pp. 87-108.  In Polluted  Rain. Y.  Taft,  M.  W. Miller, and  P.
E. Morrow, eds.  Plenum Press~T~New York.

Stensland, G. J.  and R. G. Semonin.   1982.  Another  interpretaion of the  pH
trend in the United States.  Bull. Amer. Meteorol. Soc. 63:1277-1284.

Stensland, G.  J.  and R.  G. Semonin.  1984.   Response  to comment on  another
interpretation of the pH trend  in  the United  States.   Bull.  Amer. Meteorol.
Soc. 6:in press.

Stoiber,  R.  E.,  S.  N. Williams,  and L.  L.  Malinconico.    1980. Mount  St.
Helens,  Washington,  1980  volcanic eruption:   Magmatic  gas  component during
first 16 days.  Science 208:1258-1259.

Thompson,  L.  G.    1980.    Glaciological  investigations  of   the  tropical
Quelccaya ice cap, Peru.   J. Glaciology 25:69-84.

Thompson, L.  G.  and  E. Mosley-Thompson.   1981.  Microparticle concentration
variation  linked with climatic:   Evidence from  polar ice  cores.   Science
212:812-815.

U.S./Canada  Memorandum of  Intent on  Transboundary Air  Pollution.    1982.
Atmospheric  Sciences  and  Aanalysis,  Work Group 2.   Final  Report, November,
1982.

Unni,  C.,  W. Fitzgerald, D. Settle,  G.  Gill, B.  Ray,  C.  Patterson,  and  R.
Duce.    1978.   The impact of  volcanic  emissions on  the global atmospheric
cycles of S, Hg and Pb.  Trans. Am. Geophys. Un. 59:1223.

Volchok, H. L. and R. T. Graveson.  1975.   Wet/dry fallout  collection.  Proc.
of the Second Federal Conf. on the Great Lakes.   Interagency Commission  on
Marine  Science and  Engineering,  Argonne National  Laboratory,  Argonne,  IL.
pp. 259-264.

Volchok,  H.  L.,  M.  Feiner, H.  J. Simpson, W.  S.  Broecker,  V. E. Noshkin,  V.
T. Bowen, and  E. Willis.  1970.   Ocean fallout -  the Crater  Lake experiment.
J. Geophys. Res. 75:1084-1091.
                                  8-98

-------
Vosters, M., F. Hanappe, and P.  Buat-Menard.   1970.  Determination of Cl, Na,
Mg,   K,   and   Ca,   in   firn  sample   66-A-2   *rom   New   Byrd   Station,
Antarctica-Comparison with  work of  Murozumi,  Chow,  and Patterson.   Geochim.
Cosmochim. Acta 34:399-401.

Wallen,  C. C.  1981.  Monitoring  potential  agents of climatic change.  Ambio
9:222-228.

Warburton, J.  A.  and G. 0.  Linkletter.   1978.   Atmospheric process and the
chemistry  of  snow   on   the  Ross  Ice  Shelf,  Antarctica.    0.  Glaciology
20:149-162.

Warburton, J.  A.,  G. 0.  Linkletter,  M.  S.  Owens,  and L.  G. Young.   1973.
Measurements of  Silver  content of  Antarctic  snow  and  firn.   International
Symposium on the chemistry of Sea/Air  Particulate Exchange Processes, Oct.

Weiss, H.  V.,  K.  Bertine,  M. Koide,  and E. Goldberg.   1975.   The chemical
composition of a Greenland glacier.  Geochim.  Cosmochim. Acta 39:1-10.

Weiss, H. V., M. M.  Herron,  and C. C.  Langway.  1978.   Natural enrichment of
elements in snow.  Nature 274:352-353.

Whelpdale,  D.  M.    1979.    Tabulations  of features  of major  networks  and
programs.  In Ecological Effects of  Acid Precipitation, Workshop Proceedings,
EPRI EA-79-^LD, Electric Power Research Institute, Palo Alto, CA.

Wiener,  J.  G.   1979.   Aerial inputs  of  cadmium, copper  lead and  manganese
into a freshwater pond  in the vicinity  of a  coal-fired power  plant.   Water,
Air, Soil Pollut. 12:343-353.

Wilson,  J.  W., V.  A.  Mohnen, and  J.  A.  Kadlecek.   1982.   Wet deposition
variability as observed  by MAP3S.  Atmos.  Environ. 16(7) :1667-1676.

Wisniewski, J. and J. D. Kinsman.  1982.   An overview of acid rain monitoring
activities in North  America.  Bull.  Am.  Meteorol. Soc. 63:598-618.

Wilson,  J.,  V.  Mohnen, and  J.   Kadlecek.    1980.    Wet  deposition  in  the
northeastern United  States.    State University of New  York,  Albany,  NY, ASRC
Pub. 796, December 1980.  pp. 139.

Wolff,  G.  T.,  P.   L.   Lioy,  H.   Golub,  and  J. S.  Hawkins.    1979.   Acid
precipitation  in the   New   York  metropolitan  area:  its  relationship  to
meteorological  factors.   Environ.  Sci. and Tech. 13:209-212.

Zoller,  W.  H.,  E. S.  Gladney,  and  R.  A.  Duce.    1974.    Atmospheric
concentrations and sources of trace metals at the South  Pole.   Science 183:
198-200.
                                  8-99

-------
               THE ACIDIC DEPOSITION PHENOMENON AND ITS EFFECTS

            A-9.  LONG-RANGE TRANSPORT AND ACIDIC  DEPOSITION  MODELS

                       (C. M. Bhumralkar and  R. E.  Ruff)

9.1  INTRODUCTION

The previous  chapters have  described  our state-of-knowl edge  of the  funda-
mental   physical  and  chemical  processes  that  affect effluents  as they  are
transported between sources  and  receptors.   When   transport  covers  distances
of 500 kilometers and above, models  that  numerically simulate  these physico-
chemical  processes are called Long-Range  Transport (LRT) models.   Currently,
justifiable concern about the  adequacy of these models  leads  researchers  to
test LRT  model  performance  quantitatively by  comparing model  calculations
with  field  measurements.  However,  such  comparisons  have  been  severely
hindered by data bases that  are limited in spatial  and  temporal  coverage  and
in the  types  of parameters  that have  been measured.  As a  result,  how well
model  results  compare with  the  real  world  is not known.    Current research
attempts to improve this  situation.

Dozens of different LRT models have been used to establish  quantitative rela-
tionships between acidic  deposition  and emission levels.  Most of  these have
dealt strictly  with sulfur dioxide  and  sulfate.   There is  large variation  of
the  inherent  detail  from  simple  to  complex  models.   The complex  models
attempt  to incorporate  the  most detailed (but not  necessarily established)
treatments that the state-of-knowledge will  permit.  However, in practice,  no
conclusive evidence indicates that detailed models can outperform the simpler
models.  Both types have  given unverified answers,  but  the  simpler ones have
done  so at  a  much  more attractive  cost.    Fortunately,   researchers  have
recognized the need  to  continue work  on simple  and  complex models  while
awaiting improved data bases that will  help  resolve existing questions about
performance and applicability.

Several  of the models discussed  in this  chapter  have  been studied by  the
modeling group (U.S./Canada 1982)  established under  auspices of  the  Memo-
randum of  Intent  (MOD  on Transboundary  Air  Pollution  signed by  the  United
States and Canada  on  5 August 1980.   However,  some of  the  models  studied  by
this group,  hereafter referred  to  as  the  MOI group,  are  not specifically
mentioned  by   name.    Rather,  this  chapter  focuses  on  generic model  types
representative  of the various approaches employed  to date.

9.1.1  General  Principles for Formulating Pollution Transport and
       Diffusion Models

The problem of transport can be reduced  to  solving an equation representing
the  conservation of  mass.   Written  in   terms  of  the concentration  of  a


                                     9-1

-------
particular chemical  species,  say Cj,  this equation  is
     _1 + ?-vCj  = S1  - RJ  + k1V2Ci                                      [9-1]
      at

where:

         ^  = velocity vector,
         Sj = sources  of species i,
         R-j = sinks of species  i, and
         kj = molecular diffusivity  of species i.

The  process  of  physical   transport  is complicated  because  the  atmospheric
velocity field is not constant  in either  time or  space.   To incorporate  the
effect  of the  fluctuation  in   velocity  field on  transport,  an  averaging
assumption is  introduced by which all  the variables are  redefined as mean
values:

     Ci = Cj  + Cj'.                                                     [9-2]

where  Cj  is  the  average concentration and Cj1  is  the instantaneous  devia-
tion from the average.

Equation 9-1  is then averaged using  mean values to give:
      aC.         _____        	>.
        1     •   Cj = Sj - RJ  + k1v2Ci -  v •  cV                       [9-3]
       3t

where  the  last term  is  called the  turbulent correlation term.   Generally,
the turbulent correlation term is  interpreted  as  a flux of species  i  across
some surface due to the turbulent velocity,  V, i.e.,

     V - CjVj'' = - V  KjVCj                                              [9-4]

which  formally  defines  Kj,  the eddy diffusivity  of the i species.   Because
the eddy diffusivity  Kj  is much  larger than  the molecular diffusivity  kj,
the latter term can be neglected  in Equation 9-3.   Thus the equation

     3 C         -    -
     	L + V • V Cj = S,-  - R,- + V • Ki V Cj                                [9-5 ]
      3t         ill        11

can be used as a representation of the  conservation of mass.

Significance has  been attached  to the  difference between  the  second  term on
the left  side and  the  last  term  on  the right  side  of  Equation  9-5.   The
former represents advection or bulk movement  of the average concentration by
the  average  velocity; the  latter represents  the diffusion  of material  by
theturbulent velocity  field.   Most  considerations in  atmospheric  transport


                                     9-2

-------
 and  diffusion modeling  are based  on  a  simplification  and  idealization  of
 either or both of these processes.

 9.1.2  Model Characteristics

 Air  quality models  have  a  variety of characteristics that can be  defined  in
 terms of:

     0 Frame of reference
     o Average temporal and spatial scales
     ° Treatment of turbulence
     o Transport
     0 Reaction mechanisms
     ° Removal mechanisms.

 These models  may be steady  state  or  time  dependent;  may  incorporate  the
 effect of  complex terrain on  wind  flow and deposition; and  may  treat emis-
 sions from point sources  or  area  sources or  both,  perhaps distinguishing
 between  elevated  and ground  emissions.   Table 9-1  shows  some of  the signif-
 icant  characteristics  of the  three  model   types  classified  by  frame  of
 reference.

 Most LRT models  are  related  to  a coordinate system or  reference  frame that
 may  be fixed  either at the Earth's surface,  at  the source of the  pollutant
 (for either  fixed or moving  sources),  or on  a puff  of  pollutant  as it moves
 downwind from  the source.  Models  whose reference  frames  are  fixed  at  the
 surface, or on the source, are called Eulerian models;  those whose frames are
 fixed on the puff of  pollutant are called Lagrangian.   Lagrangian  models are
 usually  more practical than Eulerian models in accounting  for emissions from
 individual  source locations  and  describing diffusion  as the pollutants  are
 carried  by the wind.   Eulerian  models  are more  capable  of accounting  for
 topography, atmospheric  thermal  structure,  and  nonlinear  processes such  as
 those governing reactive pollutants.

 9.1.2.1    Spatial  and Temporal  Scales--Atmospheric motions  span  a  range  of
 spatial   scalesfrom  the  microscale  (centimeters)  to  large  synoptic  scales
 (1000 km).  LRT models employ input data representative of the synoptic scale
 because  of  the  large transport distances  (500  to 2500  km).   This  includes
 incorporation  of  data  from  the  rather  sparse  upper  air  network in  North
 America  (approximately 50  stations  for  the eastern  United States  and  south-
 eastern   part of Canada;  these stations  measure winds and  temperatures aloft
 twice a  day).  When  source-to-receptor  distances  of less than 500  km  become
 important, a model  capable of treating  sub-synoptic scale motions  should  be
 employed.  In general, LRT models do not have this capability.

 For  temporal  scales, the assumption  has  been  that the physical and  biological
 effects   are  dominated by  long term  (e.g.,  annual) dosages  of acidic  pre-
 cursors.    However,   it  appears that  insufficient  interaction  has  occurred
 among the modelers and effects researchers on  this subject.

9.1.2.2    Treatment  of Turbulence—Atmospheric  turbulence dilutes  and  mixes
gaseous  and particulate pollutants as they are transported by the  mean wind.


                                     9-3

-------
                     TABLE 9-1.   CHARACTERISTICS OF POLLUTION  TRANSPORT  MODELS BY  FRAME OF  REFERENCE9
10
i
Model class
( f rane of
reference)
Eulerlan






Lagrangian






Hybrid
(mixed
Eulerian-
Langrangian



Types
of models
Rollback
Statistical
Gaussian plume
and puff
Box and multi-
box
Grid
Gaussian plume
and puff
Trajectory
Box and
mul tlbox
Statistical
trajectory
Trajectory
Partlcle-
In-cell
Puff-on-cell
Physical


Space
Site-
specific/
local
Regional



Site-
specific/
local
Regional



Local
and
regional




Time
Dally
(Episodic)





Dally or
long-term
(monthly
seasonal
annual)


Dally or
long-term
(monthly
seasonal
annual)


Treatment
of
turbulence
Implicit
Eddy
diffusivltles
Complex formu-
lation (higher
moment theory)

Well-mixed
volume
Eddy
diffuslvltles



Implicit
Eddy
diffusivitfes
Complex formu-
lation (not
applicable to
physical models)
Reaction
mechanism
Nonreactlve
Nonllnearly
reactive




Nonreactlve
Heavily
parameterized.
linearly
reactive


Nonreactlve
Nonllnearly
reactive




Removal
mechanism
Implicit
Dry and
wet




Dry and
wet





Dry and
wet





Ability
to quantify
source-receptor
relationship
Possible but
difficult to
Implement




Yes






Yes






           aAdapted from Drake et al. (1979) and Hosker (1980).

-------
Turbulence, one of the  most  important atmospheric phenomena, is produced by
the wind, temperature, and,  to a lesser extent,  humidity gradients  that occur
in the atmosphere.

In a given model,  atmospheric turbulence may be  represented by a  well-mixed
volume, semi-empirical diffusion coefficients,  eddy  diffusivities,  Lagrangian
statistics, or  more   complex  (higher-moment)  turbulence models.   The well-
mixed  volume  approach  basically  ignores  turbulence  except  in   a   loosely
implicit manner.   The most  common parameters in  current pollution transport
models are semi-empirical diffusion  coefficients  determined  from   field dif-
fusion studies over flat terrain, usually under neutral  stability conditions.
Most working-grid  and multibox  models use the eddy diffusivity formulation,
which is based on  theoretical,  physical,  and numerical studies of  the plane-
tary boundary layer (PBL).

To account for some of  the  physical  inconsistencies in the  eddy diffusivity
formulation,  several   investigators  have developed more complex formulations
of turbulence.  These require specifying more parameters and thus introduce
additional  uncertainties and  increase computational  costs.

Some models have  incorporated turbulence  effects  by applying Lagrangian sta-
statisties generated  from field  data.   This presents  a problem because most
field data are obtained in  an Eulerian framework.

9.1.2.3  Reaction Mechanisms--LRT models describe  the  fates of airborne gases
and  particles.   As these pollutants  are transported,  physical  and  chemical
changes may occur.  These may be nonreactive mechanisms,  reactive (photochem-
ical  and  nonphotochemical)  mechanisms,  gas-to-particle conversions,  gas/
particle processes,  and particle/particle  processes.   However, not all  of
these processes are explicitly treated in LRT models.

Both  the  SOg/sulfate  and   photochemical  models  have  gas-to-particle  com-
ponents  to account for the  production  of particles directly from gases  via
gaseous  reactions  or  condensation.   In  LRT  models  this treatment most fre-
quently  is limited to  the  conversion of sulfur  dioxide  to  sulfate.  Other
acidic  precursors  (e.g.,   N02)  usually  are  ignored.    The  gas/particle
components in the models take into account particle  growth by condensation or
gas  absorption.   Particle/particle  processes in aerosol  models treat coagu-
lation, breakup, condensational  growth, and  diffusion  of particles.

9.1.2.4   Removal  Mechanisms—Removal  is the reduction of  mass of  airborne
pollutants by either wet or dry deposition.   Wet deposition  is the  removal of
pollutants  by  precipitation  elements,   by  both  below-cloud  and  in-cloud
scavenging processes.   Dry deposition  is  the removal of pollutants by trans-
fer  from the air to exposed surfaces.

Removal mechanisms used in pollution  transport models  can vary  widely.  Some
models listed in  Table 9-1  (such as rollback or  statistical  models)  are  not
well  suited  to  deposition  modeling  because  they  do not treat  physical pro-
cesses  explicitly.   Others  (such  as  Gaussian  or  Langrangian   trajectory
models)  treat these  processes  in  a  rather  straightforward manner.  Grid
models are especially well  suited to use complex precipitation  scavenging  and


                                     9-5

-------
cloud dynamics In  treating  wet  deposition,  although this capability has not
been exercised very often.

9.1.3  Selecting Models for Application

9.1.3.1  General—LRT  modeling  specialists  have  made progress in developing
new techniques to  meet the  challenges of simulating pollution transport and
deposition.  A number  of excellent comprehensive reviews of transport models
have  been  prepared,   for   example,  Fisher  (1978),  Drake et  al.  (1979),
MacCracken  (1979), Smith and  Hunt  (1979), Bass   (1980),  Eliassen (1980),
Hosker (1980), Niemann et  al.  (1980), and  Johnson  (1981).   These and other
review papers have indicated that most of the existing models have been used
to:

     0  Estimate contributions  of given sources to receptors of
        interest.

     0  Estimate consequences of projected emissions changes.

     0  Fill gaps between observations.

     °  Assist in field study planning, determining  such  factors  as
        which variables to  measure and where to site stations.

     0  Assist in interpreting  data,  e.g., by inferring
        transformation or deposition  rates.

Most of these tasks can be accomplished only by using models in  concert with
field measurements where available.

9.1.3.2  Spatial  Range of Application—Model  calculations have  been  performed
over spatial scales classified  as short range (< 100 km), intermediate range
(100  to  500 km),  and  long  range (>  500 km).   Table 9-2  lists  some of the
model  types  that are  commonly  used  for each  of these  ranges.   Terminology
specific to  spatial  scales has  changed  over the years.   Lately, the terms
regional  and  long-range transport have  both been  used to  describe models
capable of treating distances of 100  km and  greater.

Generally,  the  Gaussian plume  model  has  been  the choice  for   short-range
calculations.  For intermediate  ranges,  a  Gaussian  plume model  is  sound  if
uncertainty about  dispersion  coefficients  at these distances  is taken into
account.   Applying intermediate range  Gaussian models in this  range  presents
problems  if wind  and  precipitation  distributions  vary significantly.    In
complex terrain,  shorelines, or forested terrain,  a  trajectory  model, with
plume or puff dispersion,  is more appropriate  for  intermediate  ranges.  For
long-range transport,  trajectory ensemble models, box models,  or  grid models
can be used.

9.1.3.3   Temporal  Range  of Application—Table  9-3 lists general  types   of
models on the basis of the  averaging  time used in their applications.  A host
of Lagrangian trajectory-type LRT models have been  used  for long-term appli-
cations.   Some modelers (e.g.,  Bhumralkar et al. 1981) have also  developed a

                                                                       -*•
                                   9-6

-------
TABLE 9-2.  MODEL TYPES USED WITH DIFFERENT SPATIAL RANGES
     Spatial Range
   Model  Type
    Short
      (< 100 km)
    Intermediate
      (100-500 km)
Gaussian plume
Trajectory
Particle-in-cell
Puff-on-cell

Gaussian plume
Trajectory
Grid
Particle-in-cell
Puff-on-cell
    Long
       (> 500 km)
Trajectory
Grid
Box
                            9-7

-------
     TABLE 9-3.  SHORT-TERM AND LONG-TERM MODELS
     Temporal Range                Model  Type
Short term                    Gaussian puff
  (hourly, daily)              Lagrangian trajectory
                              Particle-in-cell
                              Puff-on-cell
                              Grid

Long term                     Gaussian plume and puff
  (monthly, seasonal,         Lagrangian trajectory
  and annual)                  Statistical  trajectory
                         9-8

-------
short-term  model,  modifying  the  long-term  model  by incorporating  a more
detailed treatment of boundary layer and  diffusion processes.  A few Eulerian
models have been developed for long-range and short-term applications  (e.g.,
Durran et al.  1979).

9.2  TYPES OF LRT MODELS

Table 9-4  lists  some of  the  LRT  models  that have been  developed to  date.
Their properties are discussed below.

9.2.1  Eulerian Grid Models

The Eulerian grid model  divides the geographical  area  of  the  volume of  inter-
est into  a  two-  or three-dimensional array of grid  cells.   Advection, dif-
fusion,  transformation,  and  removal  (deposition)  of  pollutant emissions  in
each grid cell are simulated by  a  set of mathematical expressions, generally
involving  the K-theory  assumption (that the  flux of a  scalar quantity  is
proportional  to  its  gradient).   Some finite-difference  technique  is usually
employed in the numerical solution of these equations.

The major advantages of the Eulerian grid approach are:

     0  Eulerian grid models are capable  of sophisticated
        three-dimensional physical treatments.

     °  The approach can handle nonlinear chemistry.

     o  Data  input is simplified on the Eulerian  grid.

The disadvantages of the Eulerian grid approach are:

     o  Such  models usually require large amounts of computer time,
        computer storage, and input data.

     0  These models, when they incorporate nonlinear modules,
        are cumbersome to use to determine contributions  from
         individual sources.

     0  Artificial (computational) dispersion can be significant.

9.2.2  Lagrangian Models

9.2.2.1    Lagrangian  Trajectory  Models--A characteristic  feature of  these
models  is  that calculations  of  pollutant  diffusion,  transformation,  and
removal  are performed in a moving  frame of reference tied to each of a number
of air  "parcels"  that  are  transported  around   the  geographical  region  of
interest  in accordance with an observed or calculated wind field.

As indicated in Table  9-4,  Lagrangian  trajectory models can be  either  re-
ceptor  oriented,  in  which trajectories are calculated backward in time from
the  arrival  of an air parcel  at  a receptor of interest,  or source oriented,
                                     9-9

-------
      TABLE  9-4.   LONG  AND  INTERMEDIATE  RANGE  TRANSPORT  MODELS
                Model  Type
                and  Method
               Investigator
 Eulerian

   Finite Differencing
   Pseudospectral  method



 Lagrangian

   Statistical  trajectory
   Receptor oriented3



   Source oriented




 Hybrid:  Mixed

   Lagrangian-Eulerian

     Particle-in-cell (PIC)
     Atmospheric diffusion
     Particle-in-cell (ADPIC)
     Puff-on-Cell
Lavery et al. (1980); Durran et
al. (1979); Carmichael and Peters
(1979); Egan et al. (1976);
Nordj6 (1974, 1976); Pedersen
and Prahm (1974)

Berkowicz and Prahm (1978); Prahm
and Christensen (1977),
Christensen and Prahm (1976);
Fox and Orsag (1973)
Fay and Rosenzweig (1980);
Venkatram et al. (1980); Shannon
(1979); Fisher  (1975, 1978);
Mills and Hirata (1978); Sheih
(1977); McMahon et al.  (1976);
Bolin and Persson (1975); Scriven
and Fisher  (1975); Rodhe (1972,
1974)

Samson (1980);  Henmi  (1980);
Olson et al.  (1979);  Ottar
(1978); Szepesi (1978);  Eliassen
and Saltbones  (1975)
Bhumralkar  et al. (1981);
Bhumralkar  et al. (1980); Heffter
(1980); Powell  et al. (1979);
Johnson et  al.  (1978);  Maul
(1977); Wendell et al.  (1976)
Sklarew et al.  (1971)

Lange  (1978)
Sheih  (1978)
aReceptor oriented models usually have options to compute forward
 (source oriented) and backward trajectories.
                                 9-10

-------
in which  trajectories  are calculated forward  in  time from the release of  a
pollutant-containing air parcel  from an  emission source.

Most  source-oriented  Lagrangian trajectory models  simulate continuous pol-
lutant emissions by discrete  increments or "puffs" of emission occurring at
set time intervals, usually between 1 and 24 hr,  depending  upon the  designed
averaging  time  of  the model  outputs.    Such  models  simulate movement and
behavior of a pollutant plume from a continuous  source, as shown by one of
the four approaches illustrated in Figure 9-1  (Bass 1980).

Some of the advantages of Lagrangian trajectory models are:

     0 The models may be used to estimate contributions from
       individual sources.

     o The models are relatively inexpensive to run on a computer.

     ° Pollutant mass balances are easily calculated.

     o Individual sources or receptors can be treated  separately.

The disadvantages of these models are:

     ° The extension to three dimensions is not straightforward.

     0 Nonlinear physical and chemical formulations are difficult  to
       incorporate.

     0 Horizontal and vertical diffusion are highly parameterized.

The  two  most  important features  of the Lagrangian trajectory model  are  its
capability  for  calculating  detailed  source-receptor  contributions  and  its
computational  efficiency.   To achieve  the  latter, most models of this type
are  more  highly parameterized  and  thus  are potentially  less  physically
realistic than some Eulerian grid approaches.

9.2.2.2    Statistical   Trajectory  Models--As  shown  in Table 9-4,   several
Lagrangian  modelsarecharacterizedas   statistical  trajectory   models.
Although  many  different  kinds  of statistical  trajectory models  exist, each
has one or more of the following characteristic features that distinguish  the
type:

     0  Large numbers of air trajectories are calculated either
        forward in time from source areas or backward  in time from
        receptor areas, and the results are statistically
        analyzed to give average pollutant contributions.

     o  Meteorological variables are frequently averaged over long
        time periods before such parameters as concentrations and
        depositions are calculated.
                                     9-11

-------
    CONTINUOUS PLUME  MODEL
SEGMENTED PLUME MODEL
                                                                  m
                                                                   %
                                                                  m*
     PUFF SUPERPOSITION MODEL
"SQUARE PUFF" MODEL
Figure 9-1.   Trajectory modeling  approaches.  Adapted  from Bass  (1980).
                                  9-12

-------
Statistical  trajectory model s  have the  following  advantages:

     0  Computational  requirements are  modest.

     °  The models are cost efficient for repeated  runs  using
        alternative emissions  scenarios.

     o  The models do not suffer from computational  dispersion.

     o  The models may be used to estimate contributions from
        individual sources.

     0  Pollutant mass balances can be  estimated.

Disadvantages of statistical  trajectory model s  are:

     0  Most types are not adaptable to short averaging  times  (i.e.,
        episodes).

     o  Dispersion and other processes are usually  highly
        parameterized.

     0  Some types ignore dependence between meteorological  variables
        (e.g., wind and precipitation).

In  summary,  the low computational  cost  of statistical  trajectory models  is
often obtained at the expense of physical realism.

9.2.3  Hybrid Models

In  the hybrid (mixed Lagrangian/Eulerian) approach, pollutants,  whose  distri-
bution  is represented  by  Lagrangian  particles or puffs,  are  transported
through a fixed Eulerian grid that divides physical  space into several  cells.
The particles or puffs are moving horizontally  in a derived  velocity  field  in
the model domain.  The hybrid approach offers advantages of  both Eulerian and
Lagrangian models.  For  example,  hybrid  models can  provide  treatment  of non-
linear  reactions  between the compounds  of interest (in the  Eulerian  frame-
work)  and the  source-receptor  relationship (in the Lagrangian  framework).
One  of  the  main  disadvantages  of  the  hybrid  approach  (especially the
particl e-in-cell method)  is  that to simulate spatial distribution of  pollu-
tion  satisfactorily,  a  large number  of  particles  must be used.   This has
been  obviated considerably by  the  POC  (puff-on-cell)   method  developed  by
Sheih (1978).

9.3  MODULES ASSOCIATED WITH CHEMICAL  (TRANSFORMATION)  PROCESSES

9.3.1  Overview

Primary air pollutants undergo reactions in the  atmosphere,  forming secondary
pollutants  such as ozone  from N0x-hydrocarbon  reactions  and  sulfates  from
S02  oxidation  reactions.  The compounds that  appear in rainwater are mainly
                                     9-13

-------
sulfate and nitrate anions and hydrogen and ammonium cations; they typically
account for more than 90 percent  of the  ions in  rainwater.

Theoretical, laboratory,  and  field experiments  seem  to  indicate  that both
homogeneous and heterogeneous processes are important. However, the range of
transformation  rates,  the conditions  by  which they  vary,  and  the   actual
mechanisms still largely remain beyond  simulation capabilities.

9.3.2  Chemical  Transformation Modeling

As  source  emissions  are  changed  from  gases  to  aerosols,  or  (through  a
reaction  with  other materials in the  atmosphere)  to  different compounds,
their  wet and  dry removal  rates will  change, affecting  their subsequent
concentrations.   Furthermore, the chemical  transformations at any given time
will depend on prior transformation,  dilution,  and  removal.

Considerable research has been performed to understand  the combined processes
of atmospheric transport, diffusion, wet/dry removal, and chemical transfor-
mation.    The  LRT model normally incorporates  a separate module that  treats
each of these processes.  As  is the case with most  modules, chemical routines
are  most  often  gross  simplifications  of more  detailed  kinetic models that
were developed independently  of the overall modeling  effort.

There are two approaches to modeling chemical  transformations:

     o By approximation with  simplified first-order reactions.   As
       described in Chapter A-4,  the conversion of  S02  to sulfate
       is usually treated this way.

     o With more complex sets of  reactions  describing transformations
       among many compounds.   However,  only a  few developmental  models
       (e.g., Carmichael and  Peters 1979) employ nonlinear mechanisms.

The  simplified  first-order approximations  can be used  with all  approaches  to
the  modeling of  pollution  transport:    Eulerian,  statistical   or  Lagrangian
trajectory,  and  hybrid  models.   The multireaction  schemes are  most  suitable
for  implementation  in Eulerian  or  hybrid  models.   Lagrangian  models, under
some  special circumstances, can  use multireaction  schemes.   In  general,  this
is possible  only when the emissions from one source can be treated separately
from  those  of  other sources.    Thus,  such models  can  treat  the  chemical
transformations  taking  place in   a  plume from  an  isolated source  within  the
vicinity  of that  source, extending out  to  the  point  where  it  begins  to
overlap significantly with plumes from other major  sources.

9.3.2.1    Simplified  Modules—Currently,  many  models treat  transformations
either by assuming that they take place at a constant rate or by using  simple
first-order  reactions.   This type  of  treatment  usually ignores  secondary
pollutants  (e.g., ozone,  HC)  and  their dependence on time of day, season,  and
latitude  (Altshuller 1979).   This simplified  treatment  usually  ignores  any
heterogeneous reactions that may  take place.  Please refer to Chapter A-4  for
a detailed discussion on  linearity vs nonlinearity in transformation models.
                                     9-14

-------
The currently used simple modules of chemical transformation are chosen such
that the model results  are  consistent with observations rather  than  on the
basis  of their  consistency with  theory.   Because most  models  have been
trajectory models and, therefore, superposition of plumes is assumed,  linear
chemistry is  required  to treat transformation.   It  is  common  for models to
assume that  about 1  percent of the  S02  is converted  to  SO^"  each hour.
Many models have  yet  to consider dependence on temperature, relative  humid-
ity,  photochemical  activity,  time  of  day/year,  particulate   loading,  or
concentrations of other  pollutants.    To illustrate dependencies  of model
calculations to such parameters, a recent set of model  calculations has made
the transformation rate a function of zenith angle and of source type.  This
resulted  in  a variation of 5 to  10  percent  in predicted  S02  and SO/r
concentrations in comparison with  results from  the same model  using  a fixed
transformation rate.

9.3.2.2  Multireaction Modules—A!though more realistic treatment is possible
with  multireaction  simulations  (particularly  with  Eulerian  models), their
implementation is often  difficult.   For example,  the model  reaction  schemes
frequently emphasize photochemical  processes because  those processes are more
easily defined.   The  reactions between  the  pollutants may  be well  known and
characterized.    The  chemical  models  may  simulate  laboratory   smog-chamber
experiments,  with their  well-defined  conditions   and  concentrations, quite
reasonably.  Nevertheless, the  application of these mul tireaction sets to the
real  world  is often difficult because  of the  wide  variety  of   ambient con-
ditions  and  pollutant  concentrations that  occur.  The detailed knowledge
required  for  simulating  many  of  the  reactions  calls   for  air  quality  or
meteorological data not  available on  a  sufficiently dense  spatial   scale,
horizontally  and  vertically.    Data  assumptions that must  then be  made to
exercise  the  detailed  chemical   modules  are  often  not  very  different,
philosophically,  from the cruder reaction  assumptions in  simpler  models.

Another  major  weakness of  most  chemical  transformation modules is   the way
heterogeneous reactions  are handled.    Under conditions  of  high humidity or
weak  sunlight,  these  reactions  are  important.    In  the context  of  acidic
deposition,  many  of  the  more  important  heterogeneous  reactions   involve
conversion from sulfur dioxide  to sulfate.  Among  the catalysts  and reactants
are:

     °  Oxygen
     o   Iron
     o  Manganese
     °  Carbon (soot)
     °  Ozone
     o  Hydrogen  peroxide.

Freiberg  and  Schwartz (1981)   have  pointed  out some of  the difficulties in
handling  heterogeneous  reactions involving  sulfur  compounds.  They note that
heterogeneous  formation of sulfate  can  take  place over  a number  of dif-
ferent  paths,  including  uncatalyzed  oxidation,  reactions  with oxidizing
agents (e.g., ozone or  hydrogen  peroxide),  oxidation catalyzed by  transition
metal  ions,  or surface-catalyzed reactions.  Furthermore, all  the  processes
are  complicated   by  finite mass  transfer  rates  between phases.   Although


                                     9-15

-------
 heterogeneous  transformations  are undeniably  important,  their inclusion  in
 chemical transformation modules has heretofore been  cursory  at  best.

 Chapter  A-4  describes a  variety  of the  chemical  transformation mechanisms
 that  have been  proposed.   However, incorporating  such mechanisms  into a
 long-range transport  model  with  spatial  resolutions of  tens  of kilometers
 (typically 80  km)  is  not always  consistent with the  sub-grid scale of  the
 actual  physical  process.   In general, the  spatial  scale is more consistent
 with  urban modeling  (typically  less  than  5 km).  For this  reason, some com-
 promise must be  struck between  a  comprehensive chemical  scheme and practical
 application  in  LRT modeling.   A number  of  factors must  be  considered  in
 striking this  compromise;  these factors will  relate to the intended appli-
 cations  of  the model.   For example, if  only source/receptor relationships
 entailing  total  amounts  of sulfur  are required,  chemical transformations
 involving sulfur compounds are  important only to the degree that they affect
 removal processes.  When  pH  is  important,  the number of  important reactants
 and reactions increases dramatically to include a broad  range  of sulfur-  and
 nitrogen-containing compounds,  oxidants, potential  catalysts,  and precursors
 to all of these.

 9.3.3  Modules for N0y Transformation
Until quite recently, treatment of nitrogen  pollutants  in LRT models had been
set aside in favor of work on  sulfur  pollutants.   This is partially because
of  the  emphasis on  sulfur  pollutants in the  past few years  and partially
because nitrogen chemistry has been considered too complex for incorporation
into a  simple model.   One problem has been  how to incorporate NOX chemistry
into present models  that  require  a linear parameterization;  another problem
is  the  difference   in  the  time  scales on  which  NOX  and  SOX  chemistry
occurs.  For example, in LRT models, because of  the  relatively  slow rate of
conversion of  S02  to  $042-,  it  is  possible to  use  coarse  emission  grids
and  a  3-hr integration  time  step, which enables these  models  to be used
economically.    However,  with  the  more  rapid  NOX  chemistry,   such  coarse
spatial  and temporal  resolution cannot  be  justified,  thereby making  model
application impractical.

The  problem  of modeling  NOX  conversion  in  the  atmosphere can  also  be
attributed to  two other considerations.   First,  the  primary  end  products of
NOX  conversion  in  the  atmosphere (mainly,  HN03 and PAN)   do  not  appear
until after most of  the NO  has been converted to N02,  which  takes approxi-
mately 2  to 3  hours.  This  reaction  delay  for  fresh  emissions  into  an air
column must be  preserved  in  a  transport model.  The second point  is that most
of  the  end products  in  both the  simulation  and measurements in  urban air
masses  are gaseous.    These  account  for at least  90  percent of converted
nitrogen in the atmosphere.   Aerosol  nitrates  constitute  only about 5  to 10
percent of the  end  product (Spicer  et  al.  1981).

Despite the difficulties  discussed above,  researchers  have  started to in-
corporate NOX chemistry  into LRT  models.   However, these NOX modules have
not yet been evaluated by comparison  of results  with reliable measurements.
Most  of  the  researchers have assumed  that the NOX  conversion  could  be
handled by simple  first-order  rate  equations analogous  to  those  for S02.


                                    9-16

-------
Recently, an intermediate product, PAN, was introduced into the calculations
in a  short-term  version  of  the ENAMAP model (Bhumralkar et al.  1982).  The
research  suggests  application  of the  simplified  set of  reactions  and con-
stants given in  Table 9-5.   In this approach  first-order rate equations are
used  to determine the concentrations of the reaction products.  For example,
the rate equation for N0£ is:

               = -a(kn[N02])+b(kp[PAN]).                                [9-6]


The   other  reaction  products  (PAN,   HN03,   and  N03~)   are  governed  by
similar equations.   In  this example,  the  partition  constants,  a  and b, are
unity.  For the  other products, these constants are different and are  chosen
to give the partition percentages given in  Table 9-6.  Table 9-6 shows  that a
large proportion of  PAN  is   formed during  the  day but is  removed  at  night.
This  removal is  caused by thermal decomposition and is accompanied by  a con-
version of PAN to N02-

The  above formulation  neglects  the  explicit  incorporation  of hydrocarbons
(HC) , primarily  the  influence  of the HC/NOX ratio.  As described in Chapter
A-4,  this  ratio  appears to  have a  strong  influence on the  N02  conversion
rate  and on the ratio of PAN to HN03.

9.4  MODULES ASSOCIATED  WITH WET AND  DRY DEPOSITION

9.4.1  Overview

Existing pollution transport models  represent pollution deposition removal in
several different ways.   The simplest  approach involves incorporating  a non-
specific decay form intended to treat both  wet  and dry processes.  As pointed
out by a number of reviewers, such as MacCracken (1979), Eliassen (1980), and
Hosker (1980), the values of deposition coefficients used  in  various  pollu-
tion  transport models vary  widely,  sometimes by  more than a  factor  of ten.
This  is   partly  caused   by  the  different  model   formulations,  but  it also
reflects,  in a major  way, a  basic lack of knowledge in the area.  The problem
of incorporating removal  by  deposition  in  LRT  models is made more difficult
because the measurements  of  deposition coefficients for many chemical  species
of interest are  either nonexistent  or exhibit  a  major degree of variability
even  when  stratified,  indicating that the values  of coefficients  are  in-
fluenced by a number of  factors.  Some of  the  factors known to have signifi-
cant effects on wet and dry  depositions are:

Wet deposition:

     °  Atmospheric properties

        -  Precipitation  rate and type
        -  Cloud type  and  size
        -  Storm intensity
        -  Temperature and humidity.
                                    9-17

-------
    TABLE 9-5.  AN EXAMPLE OF CHEMICAL REACTIONS AND RATES
       (HR-1) FOR AN NOX MODULE (BHUMRALKAR ET AL. 1982)
                                            Reaction Rate
           Reaction                        Day          Night
  N02 -»•  PAN + HNOa + N°3"                 0-1           °-02
  PAN  + PAN + HN03 + N03"                 0.1            0



      kP
  PAN  + N02                                0            0.02
aThe ratio N02/NO is assumed to be at equilibrium with
 a value of 2 during the day and 50 at night.
                              9-18

-------
 TABLE 9-6.   PARTITION  OF  CONVERSION PRODUCTS OF
 EXAMPLE NOX REACTIONS  (BHUMRALKAR ET AL. 1982)
                               Day           Night
  Product                      (%)             {%)
HN03 (gas)                      40             85


PAN (gas)                       50              0


N03" (aerosol)                  10             15
                       9-19

-------
     0  Pollutant properties

        - Form (and size distribution  if  particulate)
        - Solubility and reactivity
        - Concentration vertical  profile
        - Location relative to  clouds.

Dry deposition:

     o  Atmospheric properties

        - Solar radiation
        - Wind speed
        - Atmospheric stability
        - Surface aerodynamic roughness
        - Humidity.

     o  Pollutant properties

        - Form (and size distribution  if  particulate)
        - Concentration vertical  profile
        - Solubility and reactivity.

     0  Vegetation properties
        - Type, size, leaf area,  density
        - Stomatal condition
        - Growth stage
        - Stress condition
        - Wetness.

     o  Other surface (non-vegetation) properties

The current models account for wet and  dry deposition  with  highly parame-
terized treatments that do not explicitly include many of the factors in the
above lists.   Some of the effects of these variables can be considered to be
"averaged out"  over the long  travel  distances and  large  spatial  averaging
areas involved  in  interregional-scale modeling.   Comparing model-calculated
depositions to available measured values produces information useful to help
select  suitable  values  for  such  "integrated" values  of  deposition  coef-
ficients.  In  general,  however,  much  additional  fundamental knowledge about
the deposition processes is needed  to  facilitate further  progress in develop-
ing models for studying acidic  deposition problems.

The discussion in  this chapter is  strictly confined to modules for treatment
of wet and dry deposition  in current  pollution transport models.   The basic
theory and principles pertaining  to these have  been  described in Chapters A-6
and A-7.

9.4.2  Modules for Wet Deposition

9.4.2.1  Formulation  and Mechanism—Various  parameterization techniques are
used  for modeling  washout interms of rainfall  rate  and  characteristic
                                     9-20

-------
scavenging efficiency.  These offer at  least  the capability to describe wet
deposition formally.   Precipitation rates  can be highly  variable,  and spa-
tially limited, especially during  active  convective situations.   Therefore,
it is difficult, if  not  impossible, to  categorize rainfall rate  on  a scale
adequate to describe the  fate  of  a  plume, especially in its early stages.

In existing models,  removal by wet  deposition  has  been parameterized in terms
either of  the  scavenging  coefficient,  A,  or  washout  ratio, W,  (Dana 1979;
refer to  Chapter  A-6 for a more comprehensive  discussion of the scavenging
coefficient).  The former results from the assumption  that wet deposition is
an exponential  decay process obeying the equation:

     Ct = C0 exp (-At)                                                C9-7]

where:

      Ct = atmospheric concentration at  time t,
      CQ = atmospheric concentration at  initial  time, and
      A  = scavenging coefficient (in  units of time-I).

The concept of a washout ratio is used  frequently in steady-state models.  It
is defined  as  the concentration of contaminant  in  precipitation divided by
its concentration in air (usually at the surface level); i.e.,

     W = I                                                             [9-8]
         C
where:

     X = concentration of contaminant  in precipitation,
     C = concentration of contaminant  in unscavenged air,  and
     W = washout ratio (dimensionless).

The  spatial  and temporal  distribution  of the concentrations determine how A
and  W are related.    For  example,  for  the simple case  of pollutant  washout
from  a column  of air having a uniform  concentration  over height,  h, one
obtains:

     A = WR                                                            [9-9]


where:

      R = the precipitation intensity.

The  values  of  washout   coefficients,  at  least for  S02  and  $042-,  vary
widely  among various modelers,  with  disagreement even on  which  pollutant is
scavenged most  efficiently.

9.4.2.2   Modules Used in  Existing Models--Wet  deposition is  usually  calcu-
lated by  using  Equation 9-7 and  allowing  A to vary with  position to  account
for  precipitation  changes over  the region of interest.    However, the basic
problem  in  applying  Equation  9-7 is  the  actual  evaluation  of A, which


                                     9-21

-------
depends on  the  characteristics  of the rainfall  and  the scavenged effluent.
Also, because the scavenging  rate  approach  inherently assumes an irreversible
collection  process,  it  is  suitable  for  gases  only  if they  are extremely
reactive.  For gases that form simple solutions in water, it is essential to
account  for possible  desorption  of gas  from droplets as  they  fall   from
regions of high concentrations toward the ground  (Hosker  1980).

The wet deposition of soluble gases in Gaussian plume models has been calcu-
lated  under simplifying  assumptions of  steady   state,  negligible chemical
reactions, and vertical fall   of raindrops.   However,  many  gases of interest
become acids when  in  solution,  and  their solubility then becomes  a function
of  pH.    Inability  to  calculate actual  pH  forces an empirical  approach to
estimating washout ratios, W,  for gases,  similar to those for  particulates.
However, some empirical approaches (e.g., Barrie  1981) have suggested ways of
estimating improved S02 washout ratios.

Some models  represent wet deposition  in  terms  of  wet  deposition velocity,
Vw, given by

     v  =  	   wet flux                .                       [9-10]
      w    concentration in air at the surface


This has been estimated from  empirically  determined  washout ratios  W given by
Equation  9-8  (SIinn 1978).   Because wet flux to the  surface  is  simply X-R
(where R is the precipitation rate),  Vw has been  estimated by using

     Vw  =  WR  .                                                      [9-11]

The wet deposition velocity has been used in models for  the wet removal  pro-
cess.   In some cases,  the washout  ratio has  been  used directly  to give an
exponential decay term for a plume if the thickness  of the wet  layer of  plume
is  known  (Heffter et al. 1975, Draxler 1976).

In  Lagrangian  puff and trajectory models (e.g.,   Bhumralkar et  al. 1981) wet
deposition  is generally treated via  an exponential decay term  (Equation 9-7)
where  the parameter depends  on the  characteristics  of  the effluent and the
precipitation.   This  technique is applicable  to irreversible   scavenging of
particles and highly reactive gases.

In  Eulerian grid models,  wet deposition is  generally handled  by  an exponen-
tial  decay term,  exp(-  A t),  although  some models simply  assume that all
the effluent  is  scavenged   immediately  when precipitation  is  encountered
(e.g., Peterson and Crawford  1970, Sheih 1977).   An  interesting variation is
contributed  by  Bolin  and Persson  (1975), who  calculate  the wet removal  rate
from


       3 /Xdz.                                                         [9-12]
        0
 The  coefficient  B is an  "expected"  overall  scavenging rate that  takes  into
 account the probability of rainfall, its likely duration  and  intensity,  and


                                     9-22

-------
the  actual  scavenging  rate  3  expected  for  such precipitation  (Rodhe  and
Grandell 1972).   Evidently  3  can vary with  locale  and season;  the method
seems  best  suited   to  long-term-average  investigations.     Wet  deposition
velocities, washout ratios, or  both,  do  not seem to have been  used  in grid
models to any extent.  However,  work  on such  formulations is in progress.

Complex  numerical   models dealing   with   wet deposition,  including  cloud
dynamics, have been described by  Molenkamp (1974),  Hane (1978),  and others.
These models deal with the equations of motion for cloud formation, precipi-
tation  formation, and  the various scavenging phenomena that  may  apply.  For
example, an  interactive  cloud-chemistry  model   has  been  used  to calculate
effects of cloud droplet growth and S02 oxidation  within  the  droplet on pH.
With this approach,  nucleation  scavenging can  be  examined for different types
of clouds (e.g., wave cloud and stratus cloud).   This  type  of work is still
in a research phase.  It  requires parameterizations of only partially under-
stood processes and  (like  most  deposition models)  is  still  unvalidated.  Such
research, while  potentially  useful,  is  presently unsuitable  for practical
appl ication.

In hybrid  (Lagrangian plus  Eulerian)  transport models  (e.g.,  particle-in-
cell),  treatment of  wet  deposition  is  more complicated.    Whereas  it  is
relatively easy to deal  with  aerosol s/particulates, problems occur in dealing
with gases.   However, the wet  deposition  velocity concept can be  used  for
gases in these types of  models.

9.4.2.3  Wet Deposition Modules for  Snow--It  is  sometimes  necessary  to dif-
ferentiate between  wet  deposition by snow and rain.   This is based  on  the
following considerations:

     «  The scavenging coefficients vary with  season and depend on the
        precipitation intensity.

     0   The scavenging coefficient is  a function  of raindrop and
        snowflake size distribution and effective  scavenging area.

     °  The scavenging coefficient is  strongly dependent on the type
        of snow (e.g., plane dendrites are  much more effective as
        scavengers than  grouped);  no  such differentiation is applicable
        to rain.

To date,  very few  LRT models  have  incorporated  the  above  considerations
explicitly in the modeling of wet  deposition.

9.4.2.4  Wet  Deposition Modules for NOX—Very little  information  is avail-
able in  the  literature  concerning treatment  of  wet deposition  of nitrogen
compounds in  transport models.   As a general  rule,  the information  that has
been  given  is expressed  as  a  fraction  of the  rates  estimated  for  sulfur
compounds.   The approach  is obviously crude,  and this is certainly  an area
where extensive use could be made of data  bases  that have been collected  in
recent years.
                                    9-23

-------
McNaughton  (1981)  has made  some  progress in developing  relationships  among
sulfate, nitrate, and precipitation pH for use in modeling.   He  has  used  wet
deposition  observations  available from  a number of research  and  monitoring
networks, including MAP3S  (Multistate Atmospheric Power  Production  Pollution
Study),  EPRI  (Electric Power Research  Institute), NADP  (National  Atmospheric
Deposition  Program),  CANSAP  (Canadian Network  for  Sampling  Precipitation),
and  Ontario  Hydro,  in model  evaluation  studies  (e.g., McNaughton  1980).   It
may  be  noted  that,  whereas deposition  networks  are  not as dense  as  modelers
of   pollution   transport  and  deposition  would prefer,  considerable  wet
deposition data exist for model verification.

9.4.3   Modules for Dry Deposition

9.4.3.1    General   Considerations—The   dry  deposition  rate  of  gases  and
particles  has usually been  parameterized  using a  deposition velocity  Vj,
defined by the equation

      Vd = F/C                                                        [9-13]

where

     F = the flux of material ,
     C = the ambient concentration at a particular height, and
     Vd (which is a function of height)  refers  to the  same level  as  the
     concentration measurement.

This simplified  treatment  of a deposition velocity conveniently  ignores  the
complexities of the governing  processes  as described in  Chapter A-7.  How-
ever, such  simplifications are consistent with other treatments  imbedded  in
LRT models.  Sehmel (1980)  has  summarized many of the parameters  that affect
dry deposition rates;  these concepts  are  examined in  Chapter  A-7.

A common  approach  used  in  many  models   has been  to assume  a constant  dry
deposition  velocity for each  pollutant  over  the  entire model  domain.   Of
course,  this,  is unrealistic because pollutants  are  absorbed differently  by
different surfaces  (e.g.,  vegetation,  soil, or  water),  and  because atmos-
pheric  stability can also be a factor,  particularly during nighttime.

Recently, models  have  used dry deposition velocities that are functions  of
land-use types and atmospheric  stability.  Sheih et al.  (1979) have  prepared
maps  of  surface  deposition  velocities  for  sulfur  dioxide   and   sulfate
particles  over eastern  North  America  that take  into  account  land  use,
atmospheric stability, and seasonal  differences.   Variations in deposition
rates  for  nitrogen  compounds  can  also  be mapped  in  a  similar   fashion,
although the  necessary  field  studies  for characterizing different  surfaces
and  stabilities are only  beginning to be  conducted.

Among the reasons  for characterizing deposition  rate  according to season  is
that the  character of the  Earth's  surface  changes  from season  to  season--
deciduous vegetation changes  with the  growth and loss of leaves; in grass-
lands,   the  grass dies and  is  replaced  by  a  snow cover.    The  reason  for
including atmospheric  stability  as  part  of the  categorization  scheme   is


                                     9-24

-------
that dry  deposition  depends on the  concentration  of material in the  lowest
layers,  just  above  the  surface.    These  low-level  concentrations  in  turn
depend  on  the rate  at  which material is  transported from higher layers  to
replace that  which  is lost  to  the  surface; these  transfer  rates depend  on
atmospheric stability.   The latter  effect  can be  simulated more  directly  if
the  atmosphere   is  subdivided  into  layers  for  purposes of  modeling.    A
compromise  can  be  struck  between  detailed  simulation  of the   vertical
structure  of  the atmosphere and stability-based  parameterization,  using  a
surface  layer  formulation,  which   controls  deposition  based  on  observed
vertical distribution of the material of concern.

Verifying  dry deposition  simulations is  currently difficult  because we  lack
monitoring  instrumentation.     A   number  of  carefully controlled   field
measurements of dry deposition fluxes have  been made, principally  by  the eddy
correlation method.    The   results  can  be   used  in examining  the  scientific
validity of the parameterization used in the models.

9.4.3.2    Modules  Used  in  Existing Models-- In  Lagrangian  puff /trajectory
models,  generally  Ffie  vertically  integrated  concentration of  puffs  is
depleted by an exponential  factor
                                                                       [9-14]

where:

     ^  _          dry deposition flux _
          _
          vertically integrated concentration

Most of these models compute the dimensionless value for kd from
 where  h  is the height of the surface layer.  For simpler models there is only
 one  uniformly mixed layer, so h is simply the mixing height.  Some Lagrangian
 models  (e.g.,  Shannon 1981)  incorporate several layers  in  the  vertical, and
 dry  deposition processes are allowed to remove material from only the surface
 layer.    Eddy  diffusivity controls  the  redistribution between  the  vertical
 layers.   These  models sometimes also  include  treatments that allow  the dry
 deposition  velocities to vary  with  season,  time of day,  type  of underlying
 surface,  and atmospheric stability.

 In Eulerian grid  type models,  dry  deposition is treated in a  way similar to
 that  discussed  above.   These  models  are  especially  well  suited to  use the
 relation  between  mass flux,  dry deposition velocity,  and concentration at or
 just  above the  surface.   Constant  values  for  V^  are often  used,  probably
 for  simplicity,  although some  grid  models  (Durran et  al. 1979)  include an
 algorithm that  allows V
-------
9.4.3.3   Dry Deposition Modules  for  N0x--As stated previously, most  models
treat  the  sulfur  oxide-sul fate  cycle  exclusively.   The  nitrogen  oxides-
nitrate cycle  is being  treated  in only  a  few models (e.g.,  Bhumralkar et al.
1982).  For these models, the mathematics of dry deposition  treatment remains
the  same  is it was for the sulfur version.   However, the values for the dry
deposition velocity are different.  Chapter A-7 gives a  comparison  of experi-
mentally determined dry deposition velocities.

9.4.4  Dry Versus Wet Deposition

The  relative  significance of dry and wet deposition in  LRT models  has not
been examined in a systematic way, but is now being studied  via  field experi-
ments.     In  early field experiments,  the emphasis was  on the wet removal
process; consequently,  few  data on dry deposition were collected  and  hence
large uncertainties exist on dry deposition velocities.

A  reasonable  comparison between dry  and  wet removal  rates can be made  when
the deposition modules  incorporate the roles  of  pollutant release  height and
precipitation  frequency.   For example, whereas  dry deposition  will   play  an
important  role in  removing  pollutants near ground level, wet deposition can
be  expected to  be spotty  and   intermittent  because of  naturally-occurring
spatial and temporal  variation in precipitation events.

9.5  STATUS OF LRT MODELS AS OPERATIONAL TOOLS

9.5.1  Overview

The  ability to  simulate complex physical  and  chemical   processes  of  the
natural environment is  essential  for  making  regulatory and  policy  decisions.
There is no economical  way  to gather  enough observations to determine,  from
the  data  alone,  all  the  possible combinations  that can occur  in  the  real
world.  In addition, the effect of altering the  existing situation cannot  be
assessed  by collecting  observations  before  such alterations  take  place.
Thus, modeling supported by  monitoring  data is  the only  practical means  by
which the efficacy and advisability of control  strategies can be assessed.

The  past  decade  has  seen  increasing  concern  about  production  and long-
distance travel of  pollutants such as sul fates  and nitrates and  deposition
of  these  precursors  of  acid  on sensitive  areas  at  long  distances  from
sources.   Such concern  has given  impetus  to developing  and applying  several
LRT models, not  only  for studying acidic  deposition processes but  also for
policy-making  and regulatory purposes.

The understanding of the complex  processes  that act to  transform  and  trans-
port  pollutants  is  incomplete,  and   the capacities  of  even   the   largest
computers do not  permit easy simulation  of the almost infinite combination  of
physics, chemistry,  and hydrodynamics  of the  real  world.   It  is  therefore
necessary  to  simplify  and parameterize  the  mathematical  simulations.   The
effects of these  simplifications  are  not fully  understood  and understanding
will   not  be achieved  until  the  models  undergo  rigorous   evaluation.   The
evaluation is  not  limited  to the model itself,  but must  extend  to  the  data
base  that  drives  the  model  and  the  data  base  that  is  used   to  assess
                                     9-26

-------
performance.   In  the  remainder  of  this  section,  model  applicability  and
performance are discussed  along  with  their  attendant data requirements.

9.5.2  Model Application

9.5.2.1   Limitations  in  Applicability—Ideally,  the choice  of  a particular
model as an operational  tool  is  based on  the  specifications of the particular
application at hand; how  well  the model  has  performed  in  comparable appli-
cations; and the availability of  suitable data to drive the model.   In turn,
the  specifications  of  the application should  be determined by  certain  air
quality  regulations  (when   applicable)  and  the  ecological  effects  being
addressed.  Such criteria determine  the  spatial  and temporal  scales and the
chemical compounds that the  selected  model  must treat.

The spatial ranges of concern might require treatment of long-range  transport
(> 500  km), intermediate-range transport (100 to 500 km),  short-range trans-
port  (< 100 km),  or combinations of all  three.   The  discussion  here  has
focused on the long-range problem with the assumption that the resolution is
sufficient for smaller (spatial)  scale problems.  When the  receptor  locations
of interest are  influenced by large  sources  within  distances of 500 km, the
resolution in these LRT models may be inadequate  (unless  they include smaller
scale treatments). Obviously, for some applications, this is a serious limi-
tation  in almost all existing LRT models.

In most LRT models, temporal scales  germane  to  acidic  deposition  have been
assumed to be long-term (e.g., monthly,  seasonal, and annual averages).  The
underlying  assumption  is  that the effects of  acidic  deposition result from
long-term  build-ups,  not short-term episodes.   Only  a limited  number  of
models  have been developed  to address the  short-term (e.g., 3-hr averages).
Most  of these  applications  have  focused  on ground level concentrations, not
depositions, of certain  acid  precursors (primarily $02).   Until  recently,
treatment of wet deposition  was  ignored  in most short-term  models.  Now,  a
host  of  short-term models  treat  both wet  and  dry  depositions  of  acid
precursors.  However, much  less effort has been put  into  the evaluation of
these  long-range,  short-term models  in  comparison with  those  designed for
long-term  calculations.   As a  result  almost  no   knowledge  exists  on  the
performance  of  short-term   models  in  calculating  depositions of acidic
compounds.

A major problem  is that there are certain  types of  applications  for which no
single  model may be appropriate.  The majority  of LRT  models have been de-
signed  to  calculate  long-term   concentrations  and  depositions  of sulfur
dioxide  and sulfate.   Some  of these models  also  treat nitrogen oxides and
nitrates, but much less is known  about model  performance for nitrogen oxides
or  any  other  reactive compounds (other  than  sulfur).    For  more  complete
chemical  systems, LRT models are still in the research phase and, in general,
are  not ready as operational  tools.

9.5.2.2    Regional  Concentration  and Deposition  Patterns—A  better under-
standing  of LRT model  design  and application can  be  obtained  by  examining
                                     9-27

-------
one particular Lagrangian modeling approach—the  EURMAP/ENAMAP—on the  basis
that it can be considered as a  typical  example  of  such  models.   There  are two
versions  of  EURMAP  (European   Regional  Model  of Air  Pollution):   EURMAP-1
(Johnson et al. 1978) is a long-term  model  that  calculates  monthly, seasonal,
and annual  values;  EURMAP-2  (Bhumralkar  et al.  1981) is a  short- term  model
that calculates 24-hr values.   ENAMAP-1, Eastern  North American Model of Air
Pollution  (Bhumralkar et  al. 1980)  is  a closely  related version  of  EURMAP-1
that has been adapted for application to the geographical  region covering the
eastern United States and southeastern  Canada,  as  illustrated in Figure  9-2.

The  EURMAP  and   ENAMAP  models  are  designed  to have  minimal  computation
requirements  for  making  long-term   calculations  while simulating  the  most
important  processes  involved   in the  transboundary  air-pollution  problem.
These  models  can  be used to calculate  daily,  monthly, seasonal,  and annual
S02   and   S042-    air    concentrations;   S02   and   $042-   dry   and   wet
deposition  patterns;  and  interregional  exchanges  resulting from the  S02  and
$04^-  emissions   over  a  specified  domain.   The  models  use long  sequences
of  historical meteorological   data  as  input,  retaining  all  the  original
temporal and  spatial detail  inherent  in the data.

The short-term models,  EURMAP-2 and  ENAMAP-2,  use the  same general  design as
the long-term models but  have   a  number  of important differences, which  are
necessary  to  incorporate  more   details  into the emissions  and  meteorological
simulations to  be consistent with the  much shorter (24-hr) averaging  time.
In  particular,  atmospheric  boundary-layer  processes  have  been treated  in  a
more detailed manner than in long-term  versions.

The results from  both EURMAP and  ENAMAP  models  are  obtained in  the  following
forms:

     °  Graphical  displays of the distribution of  $03
        and $04^- concentrations

     o  Graphical  displays of the distributions of S02
        and 504^- wet and dry depositions

     °  Tabulated results showing the interregional  exchanges
        of  sulfur pollution between  individual source and
        receptor  regions.

Examples are  presented  in Figures 9-3 and  9-4 and Table 9-7, respectively, of
each of the above types of  products  resulting from the ENAMAP application.

9.5.2.3  Use  of  Matrix Methods to  Quantify  Source-Receptor Relationships--
For long-range transport,  environmental  assessment must  consider  potential
impacts of emissions on areas   far removed  from the  source. Transport across
the boundaries  of air quality  planning  regions, states,  and even nations can
be  important.  At the present   state-of-the-art of  modeling,  the  models that
have  been  used to  quantify source-receptor  relationships are based on the
principle  of  tracking the trajectories  of  emitted  pollutants.   These models
are used  to  compute "transport matrices" (e.g.,  Table  9-7) that permit
                                     9-28

-------
                                            SOUTH QUEBEC
          viii-     |-—--\

          SOUTH     /     VH ';
          -~L	|	,
                (a)  EPA Regions  used  in  this  study
33
3C
2C
10
1























































VIII-NORTH
























VII
SOU!















































>


































































































































VII


























































































VI-EAST









1 ]
















































V-NORTh





























































































































































1







































































































ON





V-SOUTH













































J



































































SOUTH
TAR 1C

















































i i
(SOUTH




i













IV-NORTI






V-SOUT



























,

H






















































ii














































ii














































OUE



























































































M


1



























1
:c



i























































































0 20 30 40 43

                    (b) Emission Grid and Model  Domain



Figure 9-2.   Eastern North American domain and EPA regions  used in  the

             ENAMAP modeling study.  Adapted from Bh •    " -    +-

             (1980).

                                  9-29

-------
   Local maximum values shown apply at points marked by plus signs.
Figure 9-3.   Calculated SO? and S04   concentrations (yg m~^) for August
             1977.   Adapted from Bhumralkar et al .  (1980).
                                    9-30

-------
                                                 DRY  DEPOSITION
                                   16
                                                WET  DEPOSITION
  Local  maximum values shown apply at points marked by plus signs

Figure 9-4.   Calculated annual  dry and wet depositions of SQqp- (10 mg m~3)
             for 1977.  Adapted from Bhumralkar et al. (1980).
                                  9-31

-------
                   TABLE  9-7.   ANNUAL INTERREGIONAL  EXCHANGES OF SULFUR DEPOSITION FOR 1977
                          AS  CALCULATED BY THE ENAMAP -  1  MODEL (BHUMRALKAR ET AL. 1980)
UD
I
co
ro
TOTAL CONTRIIIUTION TO S DEPOSITIONS WITHIN RECEPTOR REGIONS
EMITTER
REGION
1 VIII - NORTH
2V- NORTH
3 S. ONTARIO
4 VII
5 VIII - SOUTH
6 VI - EAST
7 V - SOUTH
8 IV - SOUTH
9 IV - NORTH
10 III
11 II
12 I
13 S. QUEBEC
TOTAL (K TON S )

EMITTER
REGION
1 VIII - HORTH
2 V - NORTH
3 S. ONTARIO
4 VII
5 VIII - SOUTH
6 VI - EAST
7V- SOUTH
8 IV - SOUTH
9 IV - NORTH
10 III
11 II
12 I
13 S. QUEBEC

1
10.
3.
0.
1.
0.
1.
2.
0.
0.
0.
0.
0.
0.
18.


1
55.
19.
3.
3.
0.
7.
9.
1.
0.
2.
1.
0.
0.

2
1.
655.
66.
43.
0.
4.
186.
8.
19.
11.
1.
0.
2.
997.


2
0.
66.
7.
4.
0.
0.
19.
1.
2.
1.
0.
0.
0.

3
0.
290.
820.
10.
0.
1.
145.
7.
24.
57.
53.
1.
105.
1514.


3
0.
19.
54.
1.
0.
0.
10.
0.
2.
4.
4.
0.
7.

4
2.
46.
2.
367.
0.
40.
135.
16.
11.
3.
0.
0.
0.
621.
PERCENT

4
n.
7.
0.
59.
0.
6.
22.
3.
2.
1.
0.
0.
0.

5
0.
0.
0.
0.
0.
1.
0.
0.
0.
0.
0.
0.
0.
1.

6
0.
3.
1.
26.
0.
401.
14.
44.
13.
1.
n.
0.
0.
503.
CONTRIBUTIONS TO

5
6.
0.
0.
0.
0.
92.
1.
0.
0.
0.
0.
0.
0.

6
0.
1.
0.
5.
0.
80.
3.
9.
3.
0.
0.
0.
0.

7
0.
229.
49.
137.
0.
7.
1566.
31.
221.
178.
1.
1.
1.
2422.

8
0.
6.
2.
22.
0.
35.
59.
949.
108.
14.
1.
0.
0.
1197.
S DEPOSITIONS WITHIN

7
0.
9.
2.
6.
0.
0.
65.
1.
9.
7.
0.
0.
0.

8
0.
0.
0.
2.
0.
3.
5.
79.
9.
1.
0.
0.
0.

9
0.
24.
7.
41.
0.
6.
425.
279.
929.
141.
4.
2.
0.
1856.
RECEPTOR

9
0.
1.
0.
2.
0.
0.
23.
15.
50.
8.
0.
0.
0.
(kllotons)

10
0.
7R.
74.
12.
0.
1.
520.
25.
159.
1363.
37.
9.
2.
2280.
REGIONS

10
0.
3.
3.
1.
0.
0.
23.
1.
7.
60.
2.
0.
0.

11
0.
50.
87.
3.
0.
0.
92.
2.
15.
179.
204.
91.
8.
732.


11
0.
7.
12.
0.
0.
0.
13.
0.
2.
24.
28.
12.
1.

12
0.
18.
4f).
2.
0.
0.
30.
1.
7.
56.
65.
207.
41.
467.


12
0.
4.
9.
0.
0.
0.
6.
0.
1.
12.
14.
44.
9.

13
0.
23.
87.
2.
0.
0.
26.
2.
6.
?1.
14.
22.
204.
407.


13
0.
6.
21.
0.
0.
0.
6.
1.
1.
5.
3.
5.
50.

-------
assessment of  air  pollution  impacts  for  multiple  scenarios  of emissions.
(Please refer  to Chapter  A-2  for  discussion  on  natural  and anthropogenic
emissions  that  contribute  to  acidic deposition.)    The  transport matrix
concept is based on the assumption that the average  concentration deposition
of  a  pollutant  in  one geographic  region  (the  "receptor")  is  the  sum of
contributions received from emissions in every other region (the "sources").
The  matrix  method  has been  used  in  several   assessment  studies  and  for
analyses of policy issues  (Ball  1981).

Table  9-8  (from  Ball  1981)  exemplifies  some  of the  features of results
presented  in  the matrix  format.   The Brookhaven  National  Laboratory (BNL)
AIRSOX model  (Meyers  et  al.  1979)  was used  to generate the  results which
quantify the transport of sul fates from one Federal  (EPA) region to  another.
Terms  along  the diagonal  of  the  matrix  are  the  intraregional   (locally
produced)  contributions.  Summation of the off-diagonal contributions of the
receptor  regions gives  the imported  fraction  of  sul fate concentrations.
Table 9-8  shows  that  the  imported  fraction  varies from 6 percent (Region 9)
to  92  percent (Region 1).   Examining  the  individual contributions  to the
Region 1 totals  in  the  first  column, it is seen  that slightly over  one-half
the  total  impact of  5.461  yg  m~3  is  calculated to originate  from Region
5, which has an incremental  contribution of  2.817 yg  m~3.

While  the  matrix method  is a reasonable  way  to present the  source-receptor
relationship results of the transport models in  a  convenient form,  important
questions remain about their validity in general  and  also about  the  accuracy
of  matrices  derived with  current models.   Chemical  and  physical   processes
that transform and  remove  air  pollutants  such  as  sulfur oxides  from the air
often are not linear in terms of the  amount of pollutant present.    However,
most large-scale,  long-range  transport models  in current  use are  based on
linear  approximations.    This  is  due to  the  difficulties  in simulating
nonlinear processes and  lack of  knowledge  about the processes.

Finally, all  the model results must be regarded  as preliminary.  The results
presented  previously  (Figures  9-2,  9-3;  Table 9-7)   primarily  indicate the
type of  information and  the format that can be provided  for use by others.
The  results  (Tables  9-7  and 9-8)  also  give   some  useful   indications,  or
trends, regarding  the  relative  importance of  various  source  regions on the
sensitive  receptor  areas  presently  of  interest.    But  at  this   time  the
absolute values  of  the  numbers  in  the matrices  should not be given  too much
importance, and certainly  the  results of any one model  should not be  taken in
preference to the others.

9.5.3  Data Requirements

9.5.3.1   General—Figure  9-5  shows schematically how  the components  of a
general transport model  are interconnected  and how they interact with basic
data sources.  The  diagram represents a model   that is meteorologically diag-
nostic in  that it  does not attempt  to generate meteorological  information
from dynamic principles but instead  makes maximum use of available meteoro-
logical observations.   Two other categories  of  input  information  are  required
in  addition  to meteorological  data:  geographical  information (e.g., surface
characteristics and topography)  and detailed emissions data from both point


                                     9-33

-------
                             TABLE 9-8.   INTERREGIONAL  CONTRIBUTIONS TO SULFATE  CONCENTRATIONS
                                                 AMONG FEDERAL  REGIONS (BALL  1981)
00
          Emltter
                                                                  Receptor
                                                                                                                      10
1
2
3
4
5
6
7
8
9
10
Local
Import
Total
0.453
0.540
1.232
0.646
2.817
0.035
0.174
0.008
0.008
0.000
0.453 (81)
0.461 (921)
5.914
0.059
1.199
2.212
0.934
4.120
0.058
0.295
0.014
0.019
0.000
1.199 (131)
7.712 (87t)
8.911
0.009
0.328
4.728
2.559
5.640
0.098
0.322
0.007
0.014
0.000
4.728 (34%)
8.976 (661)
13.704
0.002
0.037
0.518
3.832
1.730
0.228
0.283
0.006
0.011
0.000
3.832 (58%)
2.815 (421)
6.647
0.000
0.009
0.171
1.042
4.420
0.293
0.966
0.114
0.041
0.007
4.420 (63%)
2.642 (37%)
7.062
0.000
0.000
0.012
0.256
0.121
1.032
0.169
0.059
0.484
0.004
1.032 (48%)
1.105 (52%)
2.137
0.000
0.000
0.001
0.209
0.617
0.755
1.113
0.243
0.287
0.012
1.113 (34%)
2.124 (66%)
3.237
0.000
0.000
0.000
0.007
0.026
0.278
0.050
0.530
0.791
0.080
0.530 (30%)
1.232 (70%)
1.762
0.000
0.000
0.000
0.000
0.000
0.068
0.000
0.026
1.848
0.026
1.848 (94%)
0.121 (6%)
1.969
0.000
0.000
0.000
0.000
0.000
0.003
0.000
0.061
0.250
0.316
0.316 (50%)
0.314 (50%)
0.630
         Note:  Values are from BNL AIRSOX model for average of January and July 1974 meteorology; units are mlcrograms per cubic meter.

-------
                     COLUMN 1
    COLUMN 2
                                                                                    COLUMN 3
CO
cn
                    PRIMARY  DATA
TIME-VARYING FIELDS
             METEORLOGICAL DATA
                • SURFACE (HOURLY,
                    3 HOURLY)
                • UPPER AIR (6-,
                    12-HOURLYJ
                • SYNTHESIZED  FROM
                    NUMERICAL  WEATHER
             GEOGRAPHICAL INFORMATION
                • TOPOGRAPHY
                • SURFACE CHARACTER-
                   ISTICS (LAND USE)
             EMISSIONS
                t MAJOR POINT SOURCES
                  -  SPECIES
                  -  TIME VARIATION
                  -  LOCATION  (3-d)
                  -  OTHER
                     CHARACTERISTICS
                • DISTRIBUTED SOURCES
                  -  SPECIES
                  -  TIME VARIATION
                  -  LOCATION (2'-d)
                                              PRECIPITATION
                                                -RATE
                                                -TYPE
  3-d HUMIDITY
13-3.
HATION
H
    3-d WIND
   - HORIZONTAL
   - VERTICLE
  3-d TURBULENT
    DIFFUSION
 CHARACTERISTICS
   2-d SURFACE
      UPTAKE
 CHARACTERISTICS
          n
  3-d  SOURCE  FLUX
   DISTRIBUTIONS
    BY SPECIES
                             MAJOR COMPONENTS
                                   OF A
                        POLLUTION TRANSPORT MODEL
                         	A	
                                       CHEMICAL
                                    TRANSFORMATION
                        TRANSPORT
                           AND
                        DILUTION
                                                WET
                                              REMOVAL
                                               REDISTRIBUTED
                                               CONCENTRATIONS
                                                                                     VISIBILITY
                                                       DRY
                                                     REMOVAL
                                      T       t
   Figure 9-5.   Interaction among the data sources  and components of a pollution transport model.

-------
and distributed  sources.   Input data  requirements  are shown in column  1  of
the figure.

All LRT models are to a  large  extent  driven  by a set  of  time-varying  scalar
and vector  fields  like  those shown in column  2  of  Figure 9-5.   Some  of the
input  data  required  in  transport  model  simulations,  such  as  rainfall  rate
(used in  calculating  wet deposition) and  humidity  (used in  chemical  trans-
formations), can be generated from data processing components external  to the
LRT model.  The boxes in column  3  represent  the  major components of a  model.
Although some processes must be simulated in all  types of models (Lagrangian/
Eulerian), the choice of formulation  influences  the  character of the model's
other components.

9.5.3.2   Specific  Characteristics of Data  Used  in Model Simulations--It  is
evident that  to  obtain accurate,  meaningful,  and  useful   information  from
models, the input data must be of a  quality  and  quantity  consistent  with the
structure and assumptions of the model in question.   The following  discussion
examines these aspects in some detail.

9.5.3.2.1   Emissions.   Characterization  of emissions  directly  affects model
results.   Comprehensive sulfur  emission  inventories  have been  prepared for
western  Europe  (Semb  1978)  and  North  America  (Mann 1980,  Mueller  et  al.
1979).   The  SURE,  Sulfate  Regional  Experiment, emissions  (Mueller  et  al.
1979),  and MAP3S  (MacCracken  1979)  emission  inventories were  specifically
prepared to meet the needs of LRT models.

Two major  sources  of error  in emission  inventories can  be  identified.   The
first of these relates  to  the  surrogates for emissions that  are used  (e.g.,
fuel  consumption rates, population  densities,  employment  figures,  traffic,
and industrial production  rates).   The second  potential  source  of  error lies
in the  factors or  algorithms  used  to convert these  surrogates into estimates
of emissions  at  a particular  time  and place.   These uncertainties must  be
quantified  because they will  directly affect  any  model's  performance.  For
example, a  major  uncertainty  is the importance of primary  sulfates  (e.g.,
SOX emitted  from  the  stacks  already in  the form  of  sulfate).    This  has
become  a  controversial  issue   during  the  last year because  of  possible
implications  involving  comparisons  of local  sources  and  distant sources and
their relative contributions to sulfur concentrations and depositions.

The inventories are normalized to annual  average emission rates  with seasonal
and diurnal  adjustment  factors  (multipliers)  incorporated.  However, these
factors are average values and are  subject to  large  errors  at any particular
simulation  time.   Spatial   resolution  is  typically  80  km because  the  inven-
tories  are gridded to  that  size.   Emissions from  large point  sources are
usually inventoried separately  such  that the modeler  has the option to treat
these sources  separately or to  combine their  emissions  into the 80 x 80 km
grid  cells.

Klemm and  Brennan  (1981)  have  estimated  the  uncertainties in annual emission
rates in the  SURE  inventory.   Their estimates  were  separated by broad source
categories.   For sulfur dioxide emissions,  the  error ranged from 12  percent
for electric  utility sources to  32  percent  for commercial sources  and had an


                                      9-36

-------
overall  error value of  17  percent.   In other words, the estimated emissions
were said to be  within  17  percent of the actual  emissions  from the sources
inventoried.   (Their  analysis  was restricted  to  anthropogenic  sources.)
Their error estimate for NO emissions was also 17 percent but was thought to
be low because of  the  high uncertainty for transportation source emissions.
Errors in  sulfate,  nitrogen dioxide,  and  hydrocarbon emission  values were
estimated to be several  times higher than  those  for  sulfur dioxide.

9.5.3.2.2  Meteorological Data.  Existing LRT models operate in  the diagnos-
tic mode using  available meteorological  measurements, which are  quite sparse.
To date  the wind  fields for the LRT models  are interpolated  directly from
these  measurements and  have  not  been coupled  with  the  calculations  of
boundary layer models  (BLMs).   The  BLMs use the meteorological  measurements
as initial  conditions  to solve the  hydro-dynamic  equations  that govern the
wind flow.  The marriage of BLM and  LRT models  is a  current research topic.

Most  of  the meteorological  data  for   North  America  are obtained  from the
National   Climatic  Center  (United   States)  and   the  Atmospheric Environment
Service  (Canada).   Some special data  (e.g.,  meteorological  tower  data) are
also available.  Most LRT models require  upper  air  winds (e.g., 500 m) that
must be  derived  from an estimated  50  upper air stations (for  eastern  North
America)  taking measurements every 12 hr.  These measurements must be  inter-
polated  in  time  (e.g.,  3-hr time steps)  and  space (e.g., 50-km resolution)
prior to  being operated on  by the  LRT model.    It is  recognized  that the
existing  density  of stations  (less than  one  every 100,000 km2)  is  insuf-
ficient  to  compute realistic  trajectories on  a short-term basis.    It is
assumed (with some supporting evidence)  that,  for long-term calculations, the
distribution of calculated trajectories is a reasonable  approximation of the
distribution of actual  trajectories.  However,  insufficient  field data  exist
to quantify the accuracy of this assumption.

Detailed cloud and precipitation data are  needed  by the  model  for the  esti-
mation of wet  removal.   These precipitation data are obtained  from standard
reporting  surface  stations.   Hourly data  are  available within the  United
States,  but only daily  values  are  reported  in Canada.   Cloud  data are not
currently  used by  any  of  the  models  and,  hence,  treatment  of in-cloud
processes  is completely ignored.   This  is  a  major limitation  in  the data
bases and models.  Cloud data are not available  in  a readily useful form and
as  a result,  it  appears that most modelers  have  chosen not  to pursue the
rather massive effort to incorporate such data.

Other important data for model  simulations  pertain  to atmospheric  stability,
mixing  height,  and surface  characteristics.   These are critical  in  calcu-
lating diffusion  coefficients.   Information  about surface characteristics
(land  use  type)   is  used  in   estimating  dry  deposition  velocities.   For
estimating  wet removal  parameters,  considerably detailed cloud  and precipi-
tation data are required.

9.5.4  Model Performance and Uncertainties

9.5.4.1   General--The  evaluation  of   model  performance must  consider ac-
curacies inherent in:
                                     9-37

-------
    0   the model  itself—i .e.,  the package of algorithms
        containing the mathematics designed to represent the  physical
        processees germane to acid deposition;

    0   the raw information  (unprocessed  input data)  that must be
        transformed into a format compatible with  the model;

    o   the preprocessors—i .e.,  the procedures that  operate  on  the raw
        information generating the model  compatible input;  and

    0   the test data base containing the measurements  that are  compare
        with the model  calculations.

A major limitation in most assessments of model  performance is that the cause
of disagreements  between calculations and measurements cannot be  isolated
among  the  four items mentioned  above.    Normally, the four items  are con-
sidered as  a  package with the assumption that,  if agreement is  "good,"  the
model is a "valid" representation of the  real  world.

The primary objectives of model  evaluation are to ensure that modeled physi-
cal  and chemical  processes  are as  representative  as possible   of real-world
conditions and  to quantify  the uncertainties  inherent in  the  model.   Some
progress has been made toward developing  an accepted  protocol for  performance
evaluation (Fox 1981).   A widely  accepted protocol  proposed  by Bowne (1980)
lists three steps in the evaluation process:

    0   Technical  evaluation:  "Does the  model  perform  as intended and
        is it scientifically  sound?"

    o   Operational evaluation:   "Does the model compute the  correct
        values?"

    0   Dynamic evaluation:   "Can  the model  be extended or  adapted to
        other regions?"

To answer  the  questions posed in  Bowne's protocol,  four  kinds of  analysis
should be performed:

    0   Accuracy analysis—use of accepted performance  measures  to
        quantify the model's  performance  relative  to  observed
        conditions

    0   Diagnostic analysis—identification of conditions associated
        with accuracies and  inaccuracies  in the model's performance

    o   Uncertainty analysis--quantification of the modeling
        uncertainties, both  inherent in  the model  and  in  the response
        of the model to uncertainties in  the input data

    °   Scientific Evaluation—a  comprehensive technical evaluation of
        the model's conformity with the appropriate physical  and
        computational principles.


                                     9-38

-------
With the  exception  of the last item  in  the  above list,  an appropriate data
base is essential  for the required  analysis.

9.5.4.2   Data Bases  Available  for Evaluating  Models--Extensive  data bases
that can  be used to  evaluate transport models  are  scarce;  however, enough
data exist  to calculate  performance  measures over  fairly  broad confidence
intervals.    Niemann's  (1981)   examination   of  the   available  data  bases
indicated that, while they may be  adequate  for initial evaluation of sulfur
pollution  transport  models  and  perhaps wet  sulfur  deposition,  they  are
inadequate for substantially refining  the current generation of models.

The data from the years 1978 and 1980 are most frequently used for LRT model
evaluation. The former  corresponds  to the second year of SURE, during which
the most  comprehensive  air quality data  base was collected.   However,  the
coverage  and  quality  of  precipitation  chemistry data  were  not up  to  the
standard  that existed  in  the  year 1980, when several  Canadian  and United
States  networks  were operational   (see  Chapter A-8).   Of the networks,  the
NADP offers the most coverage, having  approximately 100 sites  with the great-
est density in the eastern United  States. However, regional air  quality data
coverage was not comprehensive in  1980,  and  it appears that only  the  Canadian
APN network collected daily (regional) sul fate concentration data.   (The MOI
group has assembled this data base for 1980.)   Evaluation data bases  are also
available  from  other  parts  of the  world,  especially  from  western  Europe,
which has provided data bases that have been  used  to  evaluate performance  of
several LRT models  (e.g., Eliassen and  Saltbones  1975; Johnson et al . 1978;
Bhumralkar et al. 1980, 1981).

9.5.4.3   Performance  Measures—Various  groups have  been developing proce-
dures   for  evaluating  model s  (e.g.,  Martinez  et   al .  1980,   Ruff  1980,
U.S./Canada 1982).

Many of the  widely  used  performance  measures  require data bases from rela-
tively  dense  networks  of ground  stations.    Data bases for  evaluating per-
formance  of  pollution  transport models  often emphasize  airborne   sensors.
Many of the  performance  measures  are suitable  for  application  to  airborne
observations, but  some are not.   This  is a  weakness in current evaluation
methodologies.    There  seems  to  be  a  need  for performance  measures  and
evaluation  methodology that  can  take full  advantage of  all  the  available
airborne data.

Model   evaluation  statistics  and  displays  generally  try   to   answer  the
following questions:

     °  How closely does  a model  calculation match the corresponding
        observed val ue?

     0  How well  do the fluctuations in the  predictions follow the
        fluctuations  of the measured parameter in time and space?

For  the  most part,  paired  values   of  observations, C0,  and  predictions,
Cn,  are  used to  calculate  quantitative  measures  that address  the above
questions.  A difference, d, is defined such that:


                                     9-39

-------
     d = C0(x,t)  - Cp(x,t)  .                                            [9-15]


When answering the  first question in  the  above list,  we  often define  this
difference in terms  of  measurements  and predictions from the  same  place,  x,
and time, t.

If the difference, d, is always zero,  the model would  be considered perfect.
Most often, the average and standard  deviation of  d are computed because  they
are measures  of  the model  bias  and  precision,  respectively.   Correlation
coefficients are  also  used  as  performance  measures and accompanied  by  scat-
terplots  with  regression  coefficients.    These  statistics  and  graphical
displays of  scatterplots  (and  sometimes frequency  distribution comparisons)
have been  used  by  modelers since  the time  of the  early  model  evaluation
studies.  One of the reasons they  remain useful  is that they are more or  less
the universally accepted language  on  the subject.

9.5.4.4   Representativeness of Measurements—The evaluation  of model  per-
formance has been discussed in terms of how  well  the  results from the model,
or from one of its components,  agree with some  observed value.   This assumes
that the observed values are accurate and  representative.  To legitimize  this
assumption, extensive quality assurance measures  should govern  the  acquisi-
tion and verification of the data base.  Most data bases  have been subjected
to considerable  screening  to  ensure that  data  are consistent  and  reliable,
but it is not  clear  that  the measurements  (especially  precipitation)  are
representative of  conditions on  the scale  represented by  the model.   This
must be taken into account when comparisons are made.

9.5.4.5  Uncertainties—Modeling uncertainty  consists  of two components. One
part of  the   uncertainty can be  thought  of  as "reducible" by means  of im-
provements to  the  model  and its  prescribed input  data;   a  second  part  is
considered  "irreducible"  and  is   generally   attributed to the  uncertainty
inherent  in  the   small-scale and  short-term  fluctuations  in  atmospheric be-
havior, which never  can  be  completely  characterized by the finite  amount of
data used for input to existing LRT models.  To date little  progress has  been
made on this subject.

Some estimates of the reducible  uncertainty  could be  made by  conducting a
sensitivity analysis.   In  such  an analysis the model's sensitivity to input
errors  (or data  parameterization  errors)  can be qualified  and distinguished
from errors  in  the  basic  formulation.   Methods to estimate the irreducible
uncertainty are  currently  being  developed  by  the  research community.   For
instance,  a  recently  proposed  model  evaluation  framework   (Venkatram  1982)
incorporates statistics that attempt to quantify these uncertainties.

9.5.4.6   Selected Results—Numerous  examples of  LRT  model   evaluation  exer-
cises  exist in the  open  literature.   However, most of these are presented in
a qualitative manner or with very minimal  statistical  evidence.  Despite the
desirability  to  know  the  statistical   significance  of results,  researchers
have usually neglected  to compute confidence  intervals associated with their
comparisons  of  model calculations with field  observations.     However, re-
search  programs  underway will  greatly enhance existing  information  on the


                                     9-40

-------
 subject.   The MOI,  EPRI,  EPA,  and National Park Service  are  all  sponsoring
 such  studies,  and  results  are appearing in the literature,  e.g.,  Stewart et
 al. (1983).

 In  this presentation, example model  evaluation  studies are presented  to be
 more  or  less  illustrative  of  the  state  of  knowledge.    The  first  study
 (Voldner  et  al.  1981)   examined   seasonal  averages  of  concentration  and
 depositions calculated by  a modified Long-Range Transport of  Air  Pollutants
 (LRTAP)  program  and  compared them  with  atmospheric  sulfate  concentrations
 from  the SURE network  and  precipitation  sulfate  concentrations  from  the
 CANSAP  network.   For the  month  of October 1977, the  examination  found that
 the monthly  average computed sulfate concentrations and  depositions  agreed
 with  the  measurements within 60 percent.   This agreement held  for  the four
 combinations of  wet and  dry removal  parameteric values  that were  presented.
 The correlation  coefficient between measurements  and predictions varied from
 0.55  to 0.59 for atmospheric sulfate concentrations and from 0.86 to 0.91 for
 precipitation  sulfate concentrations.

 In  another  study (Mayerhofer et al.  1981), monthly averaged  sulfur  dioxide
 and sulfate atmospheric  concentrations  calculated  by the ENAMAP model,  were
 compared  with  measurements  from  the SURE  network  for  January and  August,
 1977.   Scatterplots of  the  sulfate  comparison  are  presented  in Figure 9-6.
 The  correlation  coefficients  are  0.51 and  0.23  for January  and  August,
 respectively, but substantial over-prediction occurred  at most  stations.  The
 sulfur  dioxide  concentrations  (Figure  9-7)  compared more favorably  with
 correlation coefficients of 0.71  (January)  and  0.48  (August).

 The preliminary Phase III  results of  the MOI group  addressed the comparisons
 of  observations  and  model  calculations  of sulfate  concentrations and  wet
 depositions.   The eight models listed in  Table 9-9 were exercised  to  calcu-
 late  annual and  monthly  averages  for the year 1978.   The  model  calculations
 were  compared  with measurements from the SURE, MAP3S,  and CANSAP  programs
 using performance measures described earlier in this section.  A very  limited
 partial listing  of the MOI  results is given  in  Table  9-10.    This  listing
 allows  one  to  visually compare the average model  calculation (C), the bias
 (d~),  and  the root-mean-square error (s
-------
     to
      fD
                         CALCULATED S04Z"  AIR  CONCENTRATION (yg nfJ)
                         CALCULATED S042"  AIR CONCENTRATION  (pg  m"3)
                                 00
                                       ro
                                                     ro
                                                     o
                                                        ro
ro
00
co
ro
co
cr>
ro
   3  CO
   OO
   3  O)
c* — • 0>
   «< -S
i— '
in < Q.
^j JU _i.
-vj _. Q)
•  C.    a
Q. O
Q> -h O
"0   -h
r+ CO
(D O O
Q-P> CT
    rot/>
-t>  i n>
-s   -s
o o <
3 O 0)
   rs Q.
2 O
O> fD 3
<< 3 O
ro c+ 3
-S -S r+
3- CU 3-
O r+ — •
-»,-••«<
fD O
-S 3 <
   I/) CD
fD   — •
r+ -h C
   O fD
QJ -S to
      fU  (SI


    (—•CO
    m &>  &*
   CV
     O)
     fD
     Q-
                   O
                   CO
                   CO
                   m
                   73
                    ro
               §  £
               co
                   o
                       ro
                       -Pa.
                       ro
                   •c
                   (Q
                 i   ro
                 co
                    co

-------
co
 i  i
  3.




 O
 o
 o
  CM

 O
 GO
 O
 _1


 O
CO
 I
  e

  O)
  o

  o
  o

  a:
  l-H
  •a:

   CM
  o
  co
               10    20    30    40    50    60    70

                 OBSERVED S02  AIR CONCENTRATION (ug

                                  (a)  JANUARY
                                          80

                                         -3x
           }.xT.
           11« * i« t •  « * i  »   t
90   100
Figure 9-7.
  6     12    18    24    30    36   42    48    54    60


     OBSERVED S02 AIR CONCENTRATION (yg m"3)

                    (b)  AUGUST


Scatter diagram of observed monthly values  vs  calculated

monthly values of SO? concentrations  for  January  and

August 1977.  Adapted from Mayerhofer et  al.  (1981).
                                   9-43

-------
           TABLE 9-9.  LONG-RANGE TRANSPORT MODELS ASSEMBLED
                 BY THE MOI REGIONAL MODELING SUBGROUP
             Model  Name
Acronym
   Reference3
 Atmospheric Environment Service
 Long-Range Transport Model

 Advanced Statistical  Trajectory
 Regional Air Pollution Model

 Center for Air Pollution Impact and
 Trends Analysis - Monte Carlo Model
 Planning, Ltd.

 Eastern North American Model  of
 Air Pollution

 Transport of Regional Anthropogenic
 Nitrogen and Sulfur (TRANS)  Model  of
 Meteorological and Environmental
 Planning, Ltd.

 Ontario Ministry of Environment
 Long-Range Transport Model

 University of Illinois Regional
 Climatological Dispersion Model

 University of Michigan Atmospheric
 Contributions to Interregional
 Deposition Model
  AES
  MEP
  MOE
Olson et al. 1979
 ASTRAP     Shannon 1981
 CAPITA     Patterson et al.
            1981
ENAMAP-1    Bhumralkar et al.
            1980
Weisman 1980
Venketram et al.
1980
 RCDM-3     Fay and Rosenzweig
            1980

 UMACID     Samson 1980
aSome of the model  characteristics may have been revised since these
 references were printed.
                                    9-44

-------
       TABLE 9-10.   MO I  PRELIMINARY  COMPARISON  BETWEEN MODELED
         AND MEASURED  SULFATE  CONCENTRATIONS  AND  DEPOSITIONS
Model

AES
MOEa
MEP
ENAMAP
UMACID
CAPITA
RCDM
ASTRAP

AESb
MOEb
MEPb
ENAMAP
UMACID
CAPITA
RCDM
ASTRAP

C

7.5
.
4.0
5.4
6.2
7.5
3.1
6.0
(b)
0.8
0.8
0.8
0.1
0.5
0.4
0.7
January
d
(a) Sul
-0.7
—
2.6
1.3
-0.4
-0.7
3.7
0.8

Sd
fate
2.4
_
1.7
1.8
0.4
1.8
1.6
3.0

~
C
July
d
Concentrations (yg
8.5
—
11.9
8.1
10.8
11.9
9.0
8.9
Sulfate Depositions
-0.1
0.0
0.0
0.3
0.2
0.3
0.0
0.6
1.0
0.2
0.3
0.7
0.7
0.8
0.9
0.2
0.7
0.1
0.8
0.4
0.9
2.7
—
-0.4
3.5
0.5
-0.3
2.6
2.6
(kg ha-1
0.2
0.7
0.6
0.9
0.3
0.7
0.2
Annual
Sd
^.
C
d
Sd
m-3), 1978
2.3
_
2.1
3.6
2.4
3.2
27
3.2
period"1
0.4
0.4
0.4
0.3
0.2
0.5
0.3
10.0
7.0
8.3
-
-
10.3
7.2
7.2
), 1978
9.7
10.1
6.5
-
-
6.4
6.6
8.1
-0.9
2.1
0.8
-
-
-1.2
1.9
1.9

1.2
0.8
4.4
-
-
4.5
4.2
2.7
1.7
2.3
0.9
-
-
1.6
1.6
2.8

4.0
2.8
2.5
-
-
3.5
4.1
4.3
Background of 2 ug m~3 added to the  calculation.

bBackground of 2 kg ha"1 added to the annual  calculations  only.
                                   9-45

-------
9.6  CONCLUSIONS

A  host  of  Eulerian  and  trajectory  models  have been  developed  to  treat
long-range transport (LRT)  problems.   The majority of these models have been
of  the  trajectory type—statistical  or  Lagrangian--and  primarily have been
developed to  calculate  long-term (monthly  and  annual)   averages  for sulfur
dioxide and sulfate concentration and depositions  over transport distances of
500 km  and  above.  The  Eulerian grid  model is capable  of treating complex
physical  and  chemical   processes  in   a more  realistic  manner  than  the
trajectory model, but this  capability has not been employed frequently on the
LRT scale.    Hence,  treatments  in  the most  detailed  Lagrangian trajectory
models are similar in complexity to  those in Eulerian models.

Current  LRT  models treat  the  processes  of transport,  diffusion,   chemical
transformation,  and  (wet  and  dry)  deposition, but  even  the  most  detailed
treatments represent gross simplifications of existing knowledge about these
processes.  The  effect of  these  simplifications on model performance has yet
to  be  determined.    These  limitations lead  to   somewhat  more   specific
conclusions described below:

    0    At present, calculations  from  LRT models  alone  are not a sufficient
        basis for estimating levels  of  acidic deposition  because  the  validity
        of  the   modeled  source-to-receptor  relationships   has  not  been
        established (Sections 9.4.1  and 9.5.4).

    o    In  a  limited number  of  model  evaluation studies  comparing  sulfur
        dioxide  and  sulfate  concentrations,  LRT   model  calculations  are
        moderately  correlated  with  field  measurements.  A  more definitive
        statement  on  this  subject  should be possible within  the next year
        when  the results of current model  evaluation  studies are  reported.
        Unfortunately,  such  a  statement probably will   address  sulfur com-
        pounds  only,  ignoring other compounds  germane  to acidic deposition
        (e.g., nitrogen oxides) (Section 9.5.4).

    °   In general,  LRT models  are capable  of  treating  only large  synoptic
        scale  processes.    As a  result, many  important smaller  (sub-grid)
        scale  processes  are  ignored  (Section 9.5.3).  These include lack  of
        treatment of:

        - processes in individual clouds and precipitation events (cloud data
        are  not treated by  existing  models and  precipitation  data are  not
        sufficiently resolved),

        - effects of nearby sources (e.g., within  100 km  of a receptor) whose
        effluents  may  dominate  acidic  precursor concentrations in  certain
        situations, and

        -  gross differences in  the  transport  winds  that might  occur  within
        the small  scale.

    0   Previous and existing  measurement  programs  have  not provided  suf-
        ficient  data to evaluate models or model components  to the  extent


                                     9-46

-------
        needed.  Additionally, the raw (Input)  data operated on  by  the models
        need improvement in spatial  and temporal  detail.   The sparcity of  the
        existing upper air meteorological network is a prime  example of this
        problem (Sections 9.4.1 and 9.5.3).

Current research programs  are  addressing many  of the topics  mentioned above
and progress  is inevitable.    Some  of this  effort is devoted to quantifying
model  accuracy and uncertainty using  existing data bases.   Better  guidelines
on how and when to use LRT results ultimately will  emerge.
                                     9-47

-------
9.7  REFERENCES

Altshuller, A.  P.   1979.    Model  predictions  of  the  rates  of homogeneous
oxidation of sulfur dioxide  to  sulfate  in  the troposphere.   Atmos. Environ.
13:1653-1661.

Ball,  R.  H.   1981.   Matrix  methods to  analyze  long-range  transport of air
pollutants.  Department of  Energy.   DOE/EV-0127.

Barrie, L.A.   1981.   The  prediction  of rain acidity  and  S02 scavenging  in
eastern North American.  Atmos.  Environ. 15:31-41.

Bass,  A.   1980.   Modelling long-range transport and diffusion, pp. 193-215.
In Proceedings of the  2nd  Joint Conference on Applications of Air Pollution
"Meteorology,  New Orleans,  LA,  March  24-27,  1980.   American  Meteorological
Society, Boston, MA.

Berkowicz, R.  and L.  P.  Prahm.   1978.   Pseudospectral  simulation  of dry
deposition from a point source.   Atmos.  Environ. 12:379-387.

Bhumralkar,  C.  M.,   R. M.  Endlich,  K.  C. Nitz,  R.  Brodzinsky,   and  P.
Mayerhofer. 1982.   Lagrangian  long-range  air  pollution model  for   Eastern
North  America.    13th  International   Technical  Meeting  of  Air Pollution
Modeling and Its Application NATO/CCMS.   He des Embiez, France.

Bhumralkar, C.  M.,  R.  L.  Mancuso,  D.  E.  Wolf,  R.  H.  Thuillier,  and W.  B.
Johnson.   1980.   Adaptation  and Application  of  the  EURMAP-1  Model  to  Eastern
North  America.    Final  Report,   Project 7760,  EPA  Contract 68-02-2959, SRI
International, Menlo Park, CA.

Bhumralkar,  C.  M.,  R.  L. Mancuso,  D.   E.  Wolf,  and W. B.  Johnson.   1981.
Regional  air  pollution  model for calculating  short-term (daily) patterns and
transfrontier exchanges of airborne sulfur in Europe.  Tell us 33:142-161.

Bolin, B. and  C. Persson.   1975.   Regional  dispersion  and deposition  of
atmospheric pollutants  with  particular  application  to  sulfur  pollution  over
western Europe.  Tell us 27(3):281-310.

Bowne, N. E.    1980.    Validation and  performance criteria  - air  quality
models.   In  Proceedings of  the  2nd Joint  Conference on Applications  of Air
Pollution~Meteorology.  American Meteorological  Society, Boston, MA.

Carmichael,  G.  R.  and L.  K.   Peters.   1979.   Numerical  simulation  of the
regional  transport of  S0£  and  sulfate in  the eastern  United States,  pp.
 337-344.   In  Proceedings of the 4th  Symposium  on  Turbulence, Diffusion, and
 Air  Pollution,  Reno,   NV,   January  15-18,  1979.    Anerican  Meteorological
 Society,  Boston,  MA.

Christensen,  0.  and  L.  P. Prahm.     1976.    A  pseudospectral  model  for
 dispersion of atmospheric  pollutants. J. Appl. Meteor.  15:1284-1294.
                                     9-48

-------
Dana, M.  T.   1979.  Overview of wet  deposition  and  scavenging,  pp.  263-274.
Jji Atmospheric Sulfur Deposition.  D.  S.  Shriner,  C. R. Richmond, and  S.  E.
Lindberg, eds.  Ann Arbor Science,  Inc., Ann Arbor, MI.

Drake,  R.  L.,  D.  J. McNaughton, and C. Huang.   1979.   Available air quality
models.   Appendix  D of Mathematical  Models for  Atmospheric  Pollutants,  EPRI
EA-1131, Project 805.  Electric Power Research Institute, Palo Alto,  CA.

Draxler,  R.  R.   1976.   A Diffusion-Deposition Scheme for  Use Within the ARL
Trajectory  Model.    Technical  memo   ERL-ARL-63.    National   Oceanic   and
Atmospheric Administration/Laboratories, Silver Spring,  MD.

Durran, D.,  M.  J.  Meldgin,  M. K. Liu, T. Thoem, and D.  Henderson.   1979.   A
study of long range air pollution problems related to coal  development in the
northern Great Plains.  Atmos. Environ. 13:1021-1037.

Egan, B.  A.,  K.  S. Rao, and A. Bass.   1976.   A three-dimensional  advective-
diffusive  model  for  long-range sulfate  transport  and  transformation,  pp.
697-714.   In Proceedings of  the 7th International Technical  Meeting  on Air
Pollution  Modeling  and  Its  Application, Airlie, VA, September 1976.   Report
of the  Air Pollution Pilot  Study,  NATO Committee on  the Challenges to Modern
Society.

Eliassen,  A.   1980.   A review of long-range  transport modeling.    J.  Appl .
Meteor. 19:231-240.

Eliassen,  A. and  J.  Saltbones.   1975.   Decay  and  transformation  rates  of
S02   as  estimated  from   emission  data,  trajectories   and  measured   air
concentrations.  Atmos. Environ. 9:425-429.

Fay,  J. A.,  and J. J.  Rosenzweig.  1980.   An  analytical  diffusion model for
long-distance transport of air pollutants.  Atmos. Environ.  14:355-365.

Fisher, B. E. A.  1975.  The long  range transport  of sulfur  dioxide.  Atmos.
Environ. 9:1063-1070.

Fisher, B. E.  A.   1978.  The calculation of  long  term  sulphur deposition in
Europe.  Atmos. Environ. 12:489-501.

Fox,  D. G.   1981.   Judging  air  quality model  performance-review  of the Woods
Hole  workshop.  Bull. Am. Meteorol. Soc. 62:599-609.

Fox,  D.  G., and  S.  A.  Orszag.    1973.    Pseudospectral  approximation  to
two-dimensional turbulence.  J. Comput. Phys. 11:612-619.

Freiberg,  J.  E.,  and S. E.  Schwartz.  1981.    Oxidation  of S02  in aqueous
droplets:   Mass-transport limitation  in  laboratory  studies  and  the ambient
atmosphere.  Atmos. Environ. 15:1145-1154.

Hane, C. E.  1978.   Scavenging  of  urban pollutants by thunderstorm rainfall :
numerical  experimentation.  J. Appl.  Meteorol.  17(5):699-710.
                                     9-49

-------
Heffter, J. L.   1980.   Air Resources Laboratories Atmospheric Transport and
Dispersion Model  (ARL-ATAD).   Technical  Memo  ERL-ARL-81,  National Oceanic and
Atmospheric Administration, Air Resources Laboratories,  Silver  Spring, MD.

Heffter,  J.   L.,  A.  D. Taylor,  and  G. J.  Ferber.    1975.    A  Regional-
Continental Scale Transport,  Diffusion,  and Deposition Model.   Technical memo
ERL-ARL-50.   National  Oceanic and Atmospheric Administration, Air  Resources
Laboratories,  Silver Spring,  MD.

Henmi, T.   1980.    Long-range transport model of  S02 and  sulphate and its
application  to  the  eastern  United  States.    J.   of  Geophys.   Research.
85:4436-4442.

Hosker, R. P., Jr.  1980.  Practical application of  air  pollutant deposition
models—current   status,   data  requirements,  and   research   needs.    ^n
Proceedings of the  International  Conference  on  Air  Pollutants  and  Their
Effects  on the  Terrestrial   Ecosystem,  Banff,  Alberta,  Canada,  May 10-17,
1980.  S. V.  Krupa and A. H.  Legge,  eds.  John Wiley  and Sons,  New York.

Johnson, W. B.  1981.  Interregional  exchanges of air pollution:  model  types
and applications.  Proceedings of the llth International  Technical Meeting on
Air Pollution Modeling and Its Applications,  Plenum Press, New  York.

Johnson, W. B.,  D.  E. Wolf,  and  R.  L.  Mancuso.   1978.   Long-term  regional
patterns and  transfrentier exchanges of  airborne sulfur  pollution in  Europe.
Atmos. Environ.  12:511-527.

Klemm,  H.A.  and  R.J.  Brennan.   1981.   Emissions  inventory for  the  SURE
region.  EPRI  Report EA-1913.  Electric  Power  Research  Institute, Palo  Alto,
CA.

Lange, R.  1978.  ADPIC—A three-dimensional  particle in cell  model  for the
dispersal  of  atmospheric  pollutants  and its  comparison to regional tracer
studies.  J.  Appl. Meteor.  17:320-329.

Lavery, T. F., R. L.  Baskett,  J.  W.  Thrasher,  N. J.  Lordi, A. C. Lloyd, and
G.  M.  Hidy.    1980.    Development  and  validation  of  a  regional  model to
simulate  atmospheric  concentrations of  S02  and  sulfate,  pp.  236-247. In
Proceedings  of the  AMS/APCA  2nd Joint  Conference  on  Applications  of ATr
Pollution  Meteorology,  New  Orleans,   LA,   March  24-27,  1980.     American
Meteorological Society, Boston, MA.

MacCracken, M. C.  1979.  Simulation of regional  precipitation  chemistry.  Ir±
Proceedings:   Advisory  Workshop  to  Identify  Research Needs on the  Formation
of  Acid  Precipitation,  D.  H. Pack,  eds., 2-75 to  2-92,  EPRI  Report EA-1074,
Electric Power Research Institute, Palo Alto, CA.

Mann,  C.  0.    1980.    NEDS   Information.   EPA  Report  450/4-80-013.    U.S.
Environmental Protection Agency, Washington,  D.C.
                                     9-50

-------
Martinez,  J.  R.,  H.  S.  Javitz, W.  F.  Dabberdt,  and  R.  E.  Ruff.    1980.
Development and application of  methods  for evaluating highway air  pollution
models, In 2nd Joint Conference on Applications of Air Pollution  Meteorology.
Anericari~Meteorological  Society, Boston, MA.

Maul, P.  R.   1977.   The mathematical  model  of  the  mesoscale transport  of
gaseous pollutants.  Atmos.  Environ. 11:1191-1195.

Mayerhofer, P. M., R. M.  Endlich, B.  E.  Cantrell, R.  Brodzinsky,  and    C.  M.
Bhumralkar.   1981.    ENAMAP-IA  long-term S02   and  sulfate  air  pollution
model:    Refinement  of  transformation  and   deposition  mechanisms.    Final
Report, EPA Contract 68-02-3424, SRI International,  Menlo Park,  CA.

McMahon, T. A., P. J.  Denison,  and R.  Fleming.   1976.   A long-distance air
pollution  transportation model   incorporating  washout  and  dry deposition
components.  Atmos. Environ.  10:751-761.

McNaughton, D.  J.   1980.   Initial  comparisons  of SURE/MAP3S sulfur  oxide
observations  with  long-term  regional model  predictions.   Atmos.  Environ.
14:55-63.

McNaughton, D.  J.    1981.   Relationships between  sulfate  and  nitrate ion
concentrations  and  rainfall  pH  for  use  in  modeling  applications.    Atmos.
Environ. 15(6):1075-1079.

Meyers, R. E., T.  Y.  Li,  R.  T.  Cederwall, and  L.  I. Kleinman.  1979.   A long
range  transport model  for  calculating  the atmospheric  impacts of residual
sulfur  oxides.   BNL 26185-RI.   Brookhaven National  Laboratory,  Upton, New
York.

Mills, M. F. and A. A.  Hirata.   1978.   A multi-scale transport and dispersion
model for  local  and  regional  scale sulfur dioxide/sulfate concentrations--
formulation and initial  evaluation.  Jji Proceedings of the 9th  International
Technical  Meeting  on  Air Pollution  Modeling  and  Its Application,  Toronto,
Ontario, Canada,  August  28-31,  1978.   A report  of the  Air Pollution  Pilot
Study, NATO Committee  on  the  Challenges  to Modern  Society.

Molenkamp, C.  R.   1974.   Numerical modeling  of precipitation scavenging  by
convective  clouds,  pp.  769-793.    Jji   Precipitation   Scavenging   (1974),
Proceedings of Symposium, Champaign, IL, October  14-18.   R. G.  Semonin  and  R.
W. Beadle, coordinators,  U.S.  ERDA Technical  Information Center, Oak  Ridge,
TN, 1977.  Available  from NTIS,  Springfield, VA,  as  CONF-741003.

Mueller,  P.  K.,  G.  M.  Hidy,  E.  Y. Tong,  and  M.  C.  MacCraken.    1979.
Implementation and coordination  of the sulfate regional experiment (SURE) and
related research programs, EPRI  Report EA-1066.   EPRI,  Palo Alto, CA.

Niemann, B. L.  1981.   Initial  data  bases  for  the  intercomparison of regional
air quality/acid deposition simulation  models--description and applications.
Presented at the 74th Annual  Meeting of  the  Air Pollution  Control  Associa-
tion, Paper 81-46.4.
                                     9-51

-------
Niemann, B. L., A. A. Hirata, B. R. Hall, M. T. Mills, P. M. Mayerhofer, and
L. F. Smith.   1980.   Initial  evaluation  of regional transport and subregional
dispersion models for sulfur dioxide and fine particulates, pp. 216-224.  In
Proceedings of  the  2nd  Joint Conference on  Applications of  Air  Pollution
Meteorology,   New  Orleans,  LA,  March  24-27, 1980.   American Meteorological
Society, Boston, MA.

Nordib, J.  1974.  Quantitative  estimates of long-range transport of sulphur
pollutants in Europe.  Ann  Meteor.  9:71-77.

Nordrf, J.  1976.   Long-range transport of air pollutants in Europe and acid
precipitation in Norway.  Water, Air,  and Soil  Pollut.  6:199-217.

Organization   for  Economic  Cooperation  and  Development (OECD).   1977.   The
OECD programme  on long-range  transport  or air pollutants:  Measurements and
findings.  Director of Information  OECD, Paris, France.

Olson, M.  P., E.  C.  Voldner,  K. K.  Oikawa,  and  A. VI.  MacAfee.   1979.    A
concentration/deposition  model  applied  to  the Canadian long-range transport
of air pollutants project:   a technical description.   Report No. LRTAP79-5,
Environment Canada,  Downsview, Ontario,  Canada.

Ottar, B.  1978.  An  assessment of the  OECD study on long-range transport of
air pollutants  (LRTAP).  Atmos.  Environ. 12:445-454.

Patterson, D. E., R.  B.  Husar,  W.  E.  Wilson,  and  L.  F. Smith.   1981.   Monte
Carlo simulation of daily regional  sulfur distribution:  Comparison with SURE
sulfate  data  and visual  range  observations during  August 1977.    J.  Appl.
Meteorol. 20:404-420.

Pedersen,  L.  B. and  L. P.  Prahm.  1974.  A method for numerical solution of
the advection equation.  Tell us 26:594-602.

Peterson,  K.  R. and  T.  V.  Crawford.    1970.   Precipitation  scavenging in  a
large-cloud  diffusion  code,  pp.   425-431.    In  Precipitation Scavenging,
Proceedings of  Symposium, Richland, WA.  R. J. Tngelmann  and W. G. N.  Slinn,
coordinators,  U.S.   AEC  Division  of  Technical  Information,  Oak  Ridge,  TN.
Available  from  NTIS,  Springfield, VA as CONF-700601.

Powell,  D. C.,  D. J. McNaughton, L. L.  Wendell,  and R.  L. Drake.   1979.   A
variable trajectory  model   for  regional  assessments  of  air pollution  from
sulfur   compounds.     Report   No.   PNL-2734,   Battelle   Pacific   Northwest
Laboratory, Richland, WA.

Prahm,   L.  P.  and  0.  Christensen.     1977.    Long-range  transmission  of
pollutants simulated by a  two-dimensional  pseudospectral dispersion  model.
J. Appl. Meteor. 16:896-910.

Rodhe,  H.   1972.    A study  of the  sulfur budget for  the atmosphere  over
northern Europe.  Tell us 24:128-138.
                                     9-52

-------
Rodhe,  H.    1974.    Some aspects  of the  use of  air trajectories  for the
computation of  large-scale  dispersion  and  fallout  patterns.    Advances in
Geophysics 188:95-109.

Rodhe, H., and J. Grandell.  1972.   On  the  removal  time of aerosol particles
from the atmosphere by  precipitation  scavenging.  Tellus  24(5):442-454.

Ruff, R. E.  1980.   Evaluation of  the RAM using  the RAPS data base—part 2.
Results.   Final Report,  Contract EPA/68-02-2770,   SRI  International,  Menlo
Park, CA.

Samson,  P.   A.     1980.     Trajectory  Analysis   of   Summertime  Sulfate
Concentrations in the Northeastern United States.  J.  Appl. Meteor.
19:1382-1394.

Scriven, R.  A., and B.  E.  A.  Fisher.   1975.   The  long-range transport of
airborne  material   and  its  removal  by  deposition  and  washout. I.  General
observations.  II.    The  effect  of  turbulent diffusion.    Atmos.   Environ.
9:49-68.

Sehmel,  G.  A.   1980.   Deposition  and reinsertion  processes.  Chapter 12 in
Atmospheric  Sciences  and  Power  Production,  D.   Randerson,  ed.,  U.  S.
Department of Energy.

Semb, A.  1978.  Sulfur emissions in  Europe.  Atmos.  Environ. 12:455-460.

Shannon,  J. D.    1979.   The  advanced   statistical   trajectory regional air
pollution  model,   pp.  376-380.    In  Proceedings of  the  4th  Symposium on
Turbulence,  Diffusion,  and  Air  Pollution,  Reno,  NV, January 15-18,   1979.
American Meteorological  Society,  Boston, MA.

Shannon,  J.  D.    1981.    A  model  of  regional   long-term   average  sulfur
atmospheric  pollution,   surface  removal, and  net  horizontal  flux.    Atmos.
Environ. 15:689-701.

Sheih,  C.  M.    1977.   Application of a  statistical  trajectory model to the
simulation  of  sulfur pollution over  the northeastern United States.   Atmos.
Environ. 11:173-178.

Sheih, C. M.  1978.  A  puff-on-cell  method  for computing  pollutant  transport
and diffusion.  J.  Appl. Meteor.  17:140-147.

Sheih, C. M., M. L. Wesely, and B. B. Hicks.  1979.   Estimated dry deposition
velocities  of sulfur over the  eastern United  States  and  surrounding  regions.
Atmos. Environ. 13:1361-1368.

Sklarew,  R. C., A. J.  Fabrik, and J. E. Prager.   1971.  A  particle-in-cell
method  for numerical  solution of the  atmospheric  diffusion  equation, and
application to  air  pollution problems.   Report No.  3SR-844,  Systems, Science
and Software, La Jolla, CA.
                                     9-53

-------
SI inn, W. G.  N.   1978.   Parameter!zations  for resuspension and for wet and
dry deposition of particles and gases  for  use  in  radiation dose  calculations.
Nucl. Safety 19(2):205-219.

Smith, F. B. and R.  D.  Hunt.  1979.  The dispersion  of  sulfur  pollutants over
western Europe.  Phil.  Trans.  Roy.  Soc.  London A290:523-542.

Spicer,  C.  W.,  G.  M.  Sverdrop,  and  M.  R.  Kuhlman.   1981.   Smog chamber
studies  of   N(L chemistry  in  power  plant  plumes.     Battelle-Columbus
Laboratories, Columbus, Ohio.

Stewart, D.  A.,  R.   E.  Morris, A. B.  Hudischewsky, and M.  K.  Liu.   1983.
Evaluation of episodic regional transport models of  interest  to  the  National
Park  Service.    Final  Report,  SYSAPP  831020, Contract  No.  PX-001-2-0739,
Systems Applications, Inc., San Rafael,  CA.

Szepesi, D. J.   1978.  Transmission of  sulfur dioxide  on local, regional and
continental  scales.   Atmos. Environ. 12:529-535.

United States/Canada.   1982.    Phase  II Interim Report of  Working Group  2.
Final   Report   submitted   to    the  U.S.  Environmental  Protection   Agency,
Washington, D.C., and Environment Canada,  Ottawa,  Ontario.

Venkatram,  A.   1982.    A  framework  for  evaluating  air  quality models.
Boundary-layer Meterology 24:371-385.

Venkatram,  A.,  B.  E.  Ley, and S. Y.  Wong.   1980.   A statistical model  to
estimate  long-term  concentrations of pollutants  associated  with  long-range
transport.   Internal  Report,  Ontario  Ministry of the  Environment,, Toronto,
Ontario, Canada.

Voldner,  E.G.,  M.P.  Olson, K.  Oikawa,  and  M. Loiselle.   1981.   Comparison
between  measured and computed  concentrations  of  sulphur compounds  in eastern
North America.  J.  Geophys Res. 86(C6):5339-5346.

Weisman,  B. 1980.    Long  range transport model  for  sulphur.   73rd  Annual
Meeting  of  the Air Pollutin Control Association.  Montreal, Canada.

Wendell, L. L.,  C. D. Powell, and R.  L. Drake.  1976.  A regional scale model
for  computing deposition  and  ground-level  air  concentrations of  S02  ancl
sulfates  from elevated and ground sources,  pp.  318-324.   jji Proceedings  of
the  3rd  Symposium  on  Atmospheric  Turbulence,  Diffusion   and   Air  Quality,
October  19-22, 1976, Raleigh, NC.  Published by the AMS, Boston, MA.  Also  in
Proceedings  of  the  7th  International  Technical  Meeting   on  Air  Pollution
Modeling  and  Its  Application, September 7-10,  1976,  Airlie House, VA.   A
report of the  Air Pollution Pilot  Study,  NATO Committee on  the Challenges  to
Modern  Society.
                                    9-54

-------
                                   TECHNICAL REPORT DATA
                            (Please read Instructions on the reverse before completing)
1. REPORT NO.
  600/8-83-016AF
                                                            3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
                                                            5. REPORT DATE
  The Acidic Deposition Phenomenon and Its Effects:
  Critical  Assessment Review Papers
  Volume I  - Atmospheric  Sciences	
             6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
  Editors - Rick A. Linthurst and Aubrey P. Altshuller
                                                            8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
  NCSU Acid Precipitation  Program
  North Carolina State University
  1509 Varsity Drive
  Raleigh, North Carolina   27606
             10. PROGRAM ELEMENT NO.



             11. CONTRACT/GRANT N6.
12. SPONSORING AGENCY NAME AND ADDRESS
  US EPA/ORD
  401 M Street, S.W.
  Washington, D.C.   20460
                                                            13. TYPE OF REPORT AND PERIOD COVERED
             14. SPONSORING AGENCY CODE
               EPA/ORD
15. SUPPLEMENTARY NOTES
  This project is  part of a cooperative agreement between EPA  and  North Carolina
  State University	
16. ABSTRACT                                                        _—^——
         This  document is a review and  assessment of the  current scientific
    knowledge  of  the acidic deposition  phenonemon and its effects.  The areas
    discussed  include both atmospheric  (Volume I) and effects  (Volume II) sciences.
    Specific topics  covered are:  natural  and anthropogenic  emissions sources;
    transport  and transformation processes; atmospheric concentrations and
    distributions of chemical substances;  precipitation scavenging and dry deposition
    processes; deposition monitoring  and modeling; and effects of deposition on
    soils, vegetation, aquatic chemistry,  aquatic biology, materials and human
    health, indirectly through ingested food or water.  Each of the above topics  is
    reviewed in detail using the available literature, with  emphasis on U.S. data,
    and where  possible, conclusions are drawn based on the available data.
17.
                                KEY WORDS AND DOCUMENT ANALYSIS
                  DESCRIPTORS
                                               b.lDENTIFIERS/OPEN ENDED TERMS
                           c.  COSATI Field/Group
18. DISTRIBUTION STATEMENT

  Release to Public
19. SECURITY CLASS (Tliu Report)
                                                                          21. NO. Or rAubo
                               750
                                               20. SECURITY CLASS (This page)
                                                                          22. PRICE
EPA Form 2220-1 (R«v. 4-77)   PREVIOUS COITION ic OBSOLETE
                                                               *U.S. OOVBRMMENT PRINTING OFFICE : 1984 0-421-082/529

-------