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
------- |