United States Office of EPA-600/8-83-016A
Environmental Protection Research and Development May 1983
Agency Washington, DC 20460
Research and Development
vvEPA The Acidic Deposition
Phenomenon and
Its Effects
Critical Assessment
Review Papers
Volume I Atmospheric Sciences
Public Review Draft
-------
THE ACIDIC DEPOSITION PHENOMENON AND ITS EFFECTS:
CRITICAL ASSESSMENT REVIEW PAPERS
Aubrey P. Altshuller, Editor
Atmospheric Sciences
Co-editors
John S. Nader
Lawrence" E. Niemeyer
Rick A. Linthurst, Editor
Effects Sciences
Co-editors
William W. McFee
Dale W. Johnson
James
John
N.
J.
Joan P.
Galloway
Magnuson
Baker
Project Staff
Rick A. Linthurst-Director
Betsy A. Hood-Coordinator
Gary B. Blank-Manuscript Editor
Clara B. Edwards-Production Staff
C. Willis Williams-G-rapHes
Mike Conley-Graphics
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 U.S. EPA/NCSU Acid
Precipitation Program, a cooperative agreement between the U.S.
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 Program and was funded by
U.S. EPA.
NOTICE
This document is a public review draft. It has not been formally
released by EPA and should not at this stage be construed to represent
Agency policy. It is being circulated for comment on its technical
accuracy.
U.S. Erv: - - - '• /"
Region
200 •- -
o w , - • •
C'nicajc, i.< •
-------
Environs- ; T':n Agency
-------
Authors
Chapter A-l - Introduction
A. Paul Altshuller - Consultant
John S. Nader - Consultant
Lawrence E. Niemeyer - Consultant
Chapter A-2 - Natural and Anthropogenic Emissions Sources
Elmer Robinson - Washington State U.
Jim B. Homolya - TRW
Chapter A-3 - Transport Processes
Noor V. Gillani - Washington U.
Jack D. Shannon - Argonne National Lab
David. E. Patterson - Washington U.
Chapter A-4 - Transformation Processes
David F. Miller - U. of Nevada
Dean A. Hegg - U. of Washington
Peter V. Hobbs - U. of Washington
Noor V. Gillani - Washington U.
Michael R. Whitbeck - U. of Nevada
Chapter A-5 - Atmospheric Concentrations and Distributions of Chemical
Substances
A. Paul Altshuller - Consultant
Chapter A-6 - Precipitation Scavenging Processes
Jeremy M. Hales - Battelle, Pacific Northwest Lab
Chapter A-7 - Dry Deposition Processes
Bruce B. Hicks - National Oceanographic and Atmospheric
Administration
-------
Chapter A-8 - Deposition Monitoring
Gary J. Stensland - Illinois State Water Survey
Bruce B. Hicks - National Oceanographic and Atmospheric
Administration
William B. Lyons - U. of New Hampshire
Paul A. Mayewski - U. of New Hampshire
Chapter A-9 - Long-Range Transport and Acidic Deposition Models
Chandrakant M. Bhumralkar - National Oceanographic and Atmospheric
Administration
Ronald E. Ruff - SRI International
Chapter E-l - Introduction
Rick A. Linthurst - North Carolina State U.
Chapter E-2 - Effects on Soil Systems
William W. McFee - Purdue U.
Fred Adams - Auburn U.
Christopher S. Cronan - U. of Maine
Mary K. Firestone - U. of California, Berkeley
Charles D. Foy - U.S. Department of Agriculture
Robert D. Harter - U. of New Hampshire
Dale W. Johnson - Oak Ridge National Lab
Chapter E-3 - Effects on Vegetation
Dale W. Johnson - Oak Ridge National Lab
Boris I. Chevone - Virginia Polytechnic Institute
Patricia M. Irving - Argonne National Lab
Samuel B. McLaughlin - Oak Ridge National Lab
Dudley J. Raynal - Syracuse U.
David S. Shriner - Oak Ridge National Lab
Lorene L. Si gal - Oak Ridge National Lab
John M. Skelly - Pennsylvania State U.
William H. Smith - Yale U.
Jerome B. Weber - North Carolina State U.
ii
-------
Chapter E-4 - Effects on Aquatic Chemistry
James N.Galloway - U. of Virginia
Dennis S. Anderson - U. of Maine
M. Robbins Church - U.S. EPA
Christopher S. Cronan - U. of Maine
Ronald B. Davis - U. of Maine
Peter J. Dillon - Ontario Ministry of Environment
Charles T. Driscoll - Syracuse U.
Steve A. Norton - U. of Maine
Gary C. Schafran - Syracuse U.
Chapter E-5 - Effects on Aquatic Biology
John J. Magnuson - U. of Wisconsin
Joan P. Baker - North Carolina State U.
Peter G. Daye - Daye Atlantic Salmon Corp.
Charles T. Driscoll - Syracuse U.
Kathleen Fischer - Environment Canada
Charles A. Guthrie - N.Y. State Dept. of Environ. Conservation
John H. Peverly - NY State College Agric. & Life Sciences
Frank J. Rahel - U. of Wisconsin
Gary C. Schafran - Syracuse U.
Robert Singer - Colgate U.
Chapter E-6 - Indirect Effects on Health
Thomas W. Clarkson - U. of Rochester
Joan P. Baker - North Carolina State U.
William E. Sharpe - Pennsylvania State U,
Chapter E-7 - Effects on Materials
John Yocom - TRC Environ. Consultants, Inc.
Norbert S. Baer - New York U.
iii
-------
PREFACE
The Acidic Deposition Phenomenon and Its Effects: Critical
Assessment Review Papers publ 1c review draTtT Is a technical review
document In two volumes, prepared and released for a 90-day period of
public technical comment. The Environmental Protection Agency will
develop an interpretive summary, The Acidic Deposition Phenomenon and
Its Effects: Critical Assessment Document, based upon the contenfoT
Th~e Review Papers and the public comments.
The Acidic Deposition Phenomenon and Its Effects: Critical
Assessment Review Papers was requestedliy tfie Clean Air Scientific
Advisory Committee (CASAC) of EPA's Science Advisory Board and will be
reviewed by that committee. The CASAC is comprised of independent
scientists who are quite knowledgeable in matters pertaining to
atmospheric pollution and its effects. These scientists will evaluate
the scientific adequacy of the Critical Assessment Document. As part of
this evaluation, the CASAC considers the comments and criticisms of the
general public and scientific community as they pertain to sc+entific
issues and questions. (Although the science of an issue may obviously
have implications for policy decisions, matters of policy per se are not
in the province of the document.) This review process is essential to
developing a scientifically unimpeachable assessment.
The document's original charge was to prepare 'a comprehensive
document which lays out the state of our knowledge with regard to
precursor emissions, pollutant transformation to acidic compounds,
pollutant transport, pollutant deposition and the effects (both measured
and potential) of acidic deposition.1 It was the decision of the
editors to provide the following guidelines to the authors writing the
Critical Assessment Review Papers to meet this overall objective of the
document:
1. Contributions are 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.
iv
-------
6. Potential vs known processes/effects is to be clearly noted to
avoid misinterpretation.
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 avoided.
10. Scientific knowledge is to be included without regard to
policy implications.
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 knowledge relative to 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 con-
clusions 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 participation of over 54 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 Award in Pollution Control, and recently
retired director of EPA's Environmental Sciences Research Laboratory;
Dr. Linthurst is an ecologist and serves as Program Coordinator for the
Acid Precipitation Program at North Carolina State University, Dr.
Bennett is the Director of the Acid Deposition Assessment Starf in EPA's
Office of Research and Development and provides liaison between the
section editors/contributors and CASAC scientific reviewers.
The United States and Canada in 1980 signed a Memorandum of Intent
to seek agreement on transboundary air pollution issues. A number of
working groups are compiling technical information to support the
negotiations called for by the Memorandum. Although the Critical
Assessment Document and the U.S.-Canada working group reports come from
different origins, and are intended for different purposes, there is
likely to be some overlap in their areas of coverage.
-------
The written materials to follow are contributions from one to eight
authors per chapter, integrated by the editors. Approximately 75
scientists, with expertise in the fields being addressed, have
participated in reviewing earlier drafts of the chapters. In addition,
200 individuals participated in a public workshop held for the review
of these materials in Novanber of 1982. Mumerous changes resulted from
these reviews, and this document reflects those comments. This is the
final public review draft and comments are welcome. However, several
guidelines and forms should be used to submit formal comments. Please
consult the last section of the volume for details.
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, recommendations, and deadlines, leading to this fourth version
of the document. These dedicated persons 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 producing 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. Their concern for the truth, as we know it now,
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.
vi
-------
THE ACIDIC DEPOSITION PHENOMENON AND ITS EFFECTS:
CRITICAL ASSESSMENT REVIEW PAPERS
Table of Contents
Volume I
Atmospheric Sciences
Note: Comment forms and guidelines to be used by reviewers can be found at the ends of
Volumes I and II
Page
GLOSSARY (not available)
ACRONYM LIST xxiii
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 1-1
1.2.5 Atmospheric Concentrations and Distributions of Chemical 1-2
Substances 1-2
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 Biogenic Emissions of Sulfur Compounds 2-5
2.2.1.4 Geophysical Sources of Natural Sulfur Compounds 2-16
2.2.1.4.1 Volcanism 2-16
2.2.1.4.2 Marine sources of aerosol particles and
gases 2-20
2.2.1.5 Scavenging Processes and Sinks 2-22
2.2.1.6 Summary of Natural Sources of Sulfur Compounds 2-23
2.2.2 Nitrogen Compounds 2-24
2.2.2.1 Introduction 2-24
2.2.2.2 Estimates of Natural Global Sources and Sinks 2-25
2.2.2.3 Biogenic Sources of NOX Compounds 2-29
2.2.2.4 Tropospheric and Stratospheric Reactions 2-31
2.2.2.5 Formation of NOX by Lightning 2-32
2.2.2.6 Biogenic NOX Emissions Estimate for the United States ... 2-33
2.2.2.7 Biogenic Sources of Ammonia 2-34
2.2.2.8 Oceanic Source for Ammonia 2-38
2.2.2.9 Biogenic Ammonia Emissions Estimates for the United
States 2-39
2.2.2.10 Meteorological and Area Variations for NOX and Ammonia
Emi ssions 2-40
2.2.2.11 Scavenging Processes for NOX and Ammonia 2-40
2.2.2.12 Organic Nitrogen Compounds 2-40
2.2.2.13 Summary of Natural NOX and Ammonia Emissions 2-41
2.2.3 Chlorine Compounds 2-41
2.2.3.1 Introduction 2-41
2.2.3.2 Oceanic Sources 2-42
2.2.3.3 Volcanism 2-46
2.2.3.4 Combustion 2-46
vn
-------
Table of Contents (continued)
Page
2.2.3.5 Total Natural Chlorine Sources 2-47
2.2.3.6 Seasonal Distributions 2-47
2.2.3.7 Environmental Impacts of Natural Chlorides 2-47
2.2.4 Natural Sources of Aerosol Particles 2-49
2.2.5 Precipitation pH in Background Conditions 2-50
2.2.6 Summary 2-54
2.3 Anthropogenic Emissions 2-55
2.3.1 Origins of Anthropogenically Emitted Compounds and
Related Issues 2-55
2.3.2 Historical Trends and Current Emissions of Sulfur Compounds 2-58
2.3.2.1 Sulfur Oxides 2-58
2.3.2.2 Primary Sulfate Emissions 2-66
2.3.3 Historical Trends and Current Emissions of Nitrogen Oxides 2-72
2.3.4 Historical Trends and Current Emissions of Hydrochloric Acid {HCD 2-75
2.3.5 Historical Trends and Current Emissions of Heavy Metals Emitted
from Fuel Combustion 2-79
2.3.6 Historical Emissions Trends in Canada 2-87
2.3.7 Future Trends in Emissions 2-96
2.3.7.1 United States 2-96
2.3.7.2 Canada 2-96
2.3.8 Emissions Inventories 2-98
2.3.9 The Potential for Neutralization of Atmospheric
Acidity by Suspended Fly Ash 2-100
2.4 Conclusions 2-105
2.5 References 2-109
A-3 TRANSPORT PROCESSES
3.1 Introduction 3-1
3.1.1 The Concept of Atmospheric Residence Times 3-1
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 3-11
3.3.1 The Planetary Boundary Layer 3-11
3.3.2 Structure of the Transport Layer 3-13
3.3.3 Dynamics of the Transport Layer 3-15
3.3.4 Effects of Mesoscale Complex Systems on Transport Layer Structure
and Dynamics .,. 3-28
3.3.4.1 Effect of Mesoscale Convective Precipitation Systems
(MCPS) 3-28
3.3.4.2 Complex Terrain Effects 3-32
3.3.4.2.1 Shoreline environment effects 3-32
3.3.4.2.2 Urban effects 3-35
3.3.4.2.3 Hilly terrain effects 3-36
3.4 Mesoscale Plume Transport and Dilution 3-39
3.4.1 Elevated Point-Source Emissions 3-39
3.4.2 Broad Areal Emissions Near Ground 3-62
3.5 Continental and Hemispheric Transport 3-68
3.6 Conclusions 3-91
3.7 References 3-94
A-4 TRANSFORMATION PROCESSES
4.1 Introduction 4-1
4.2 Homogeneous Gas-Phase Reactions 4-3
4.2.1 Fundamental Reactions 4-3
-------
Table of Contents (continued)
Page
4.2.1.1 Reduced Sulfur Compounds 4-3
4.2.1.2 Sulfur Dioxide 4-4
4.2.1.3 Nitrogen Compounds 4-10
4.2.1.4 Halogens 4-16
4.2.1.5 Organic Acids 4-16
4.2.2 Laboratory Simulations of Sulfur Dioxide and Nitrogen Dioxide
0x1 dati on 4-18
4.2.3 Field Studies of Gas-Phase Reactions 4-21
4.2.3.1 Urban Plumes 4-21
4.2.3.2 Power Plant Plumes 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 1n Solution 4-38
4.3.5 Production of ^504 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 SOg^O to H2S04 ... 4-48
4.3.5.4 Heterogeneous Production of H2S04 in Solution 4-53
4.3.5.5 The Relative Importance of the Various H2S04
Production Mechanisms 4-54
4.3.6 Neutralization Reactions 4-62
4.3.6.1 NeutralIzati-on by NH3 4-62
4.3.6.2 Neutralization by Particle-Acid Reactions 4-63
4.3.7 Summary 4-64
4.4 Transformation Models 4-64
4.4.1 Introduction 4-64
4.4.2 Approaches to Transformation Modeling 4-67
4.4.2.1 The Fundamental Approach 4-67
4.4.2.2 The Empirical Approach 4-70
4.4.3 The Question of Linearity 4-70
4.4.4 Some Specific Models and Their Applications 4-75
4.4.4.1 Detailed Chemical Simulations 4-75
4.4.4.2 Parameterized Models 4-77
4.4.5 Summary 4-81
4.5 Conclusions _. 4-83
4.6 References .". 4-87
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
5.2.3 Sulfate 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 5-15
5.2.3.4 Nonurban Composition Measurements 5-19
5.2.3.5 Concentration and Composition Measurements at Remote
Locations 5-22
5.2.4 Particle Size Characteristics of Particulate Sulfur Compounds .... 5-23
-------
Table of Contents (continued)
Page
5.2.4.1 Urban Measurements 5-23
5.2.4.2 Nonurban Size Measurements ; 5-25
5.2.4.3 Measurements at Remote Locations 5-26
5.3 Nitrogen Compounds 5-27
5.3.1 Introduction 5-27
5.3.2 Nitrogen Oxides 5-27
5.3.2.1 Historical Distribution Patterns and Current
Concentrations of Nitrogen Oxides 5-27
5.3.2.2 Measurements Techniques-Nitrogen Oxides 5-28
5.3.2.3 Urban Concentration Measurements 5-28
5.3.2.4 Nonurban Concentration Measurements 5-29
5.3.2.5 Measurements of Concentrations at Remote Locations 5-33
5.3.3 Nitric Acid 5-35
5.3.3.1 Urban Concentration Measurements 5-35
5.3.3.2 Nonurban Concentration Measurements 5-38
5.3.3.3 Concentration Measurements at Remote Locations 5-43
5.3.4 Peroxyacetyl Nitrates 5-44
5.3.4.1 Urban Concentration Measurements 5.44
5.3.4.2 Nonurban Concentration Measurements 5-46
5.3.5 Ammonia 5.43
5.3.5.1 Urban Concentration Measurements 5-50
5.3.5.2 Nonurban Concentration Measurements 5-50
5.3.6 Particulate 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 5-55
5.3.7 Particle Size Characteristics of Particulate 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 Concentration Measurements 5-64
5.5.2 Nonurban Concentration Measurements 5-65
5.5.3 Concentration Measurements In Rainwater 5-65
5.6 Chlorine Compounds 5-66
5.6.1 Introduction 5-66
5.6.2 Hydrogen Chloride 5-66
5.6.3 Particulate Chloride 5-67
5.6.4 Particle Size Characteristics of Particulate Chlorine Compounds .. 5-67
5.7 Metallic Elements 5-68
5.7.1 Concentration Measurements and Particle Sizes in Urban Areas 5-69
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 5.74
5.8.1 Fine Particle Concentration and Light Scattering Coefficients .... 5-74
5.8.2 Light Extinction or Light Scattering Budgets at Urban Locations .. 5-75
5.8.3 Light Extinction or Light Scattering Budgets at Nonurban
Locations 5-77
5.8.4 Trends 1n Visibility as Related to Sulfate Concentrations 5-79
5.9 Conclusions • 5-79
5.10 References 5-85
A-6 PRECIPITATION SCAVENGING PROCESSES
6.1 Introduction 6-1
6.2 Steps in the Scavenging Sequence 6-3
-------
Table of Contents (continued)
Page
6.2.1 Introduction 6-3
6.2.2 Intermixing of Pollutant and Condensed Water (Step 1-2) 6-7
6.2.3 Attachment of Pollutant to Condensed Water Elements (Step 2-3) ... 6-8
6.2.4 Aqueous-Phase Reactions (Step 3-4) 6-15
6.2.5 Deposition of Pollutant with Precipitation (Step 4-5) 6-15
6.2.6 Combined Processes and the Problem of Scavenging Calculations .... 6-18
6.3 Storm Systems and Storm Climatology 6-18
6.3.1 Introduction 6-18
6.3.2 Frontal Storm Systems 6-19
6.3.2.1 Warm-Front Storms 6-20
6.3.2.2 Cold-Front Storms 6-25
6.3.2.3 Occluded-Front Storms 6-25
6.3.3 Convective Storm Systems 6-28
6.3.4 Additional Storm Types: Nonldeal Frontal Storms, Orographic
Storms and Lake-Effect Storms 6-28
6.3.5 Storm and Precipitation Climatology 6-30
6.3.5.1 Precipitation Climatology 6-32
6.3.5.2 Storm Tracks 6-32
6.3.5.3 Storm Duration Statistics 6-35
6.4 Summary of Precipitation-Scavenging Field Investigation 6-35
6.5 Predictive and Interpretive Models of Scavenging 6-51
6.5.1 Introduction 6-51
6.5.2 Elements of a Scavenging Model 6-54
6.5.2.1 Material Balances 6-54
6.5.2.2 Energy Balances 6-55
6.5.2.3 Momentum Balances 6-56
6.5.3 Definitions of Scavenging Parameters 6-56
6.5.4 Formulation of Scavenging Models: Simple Examples
of Microscopic and Macroscopic Approaches 6-62
6.5.5 Systematic Selection of Scavenging Models:
A Fl ow Chart Approach 6-65
6.6 Practical Aspects of Scavenging Models: Uncertainty Levels and Sources
of Error 6-68
6.7 Conclusions 6-72
6.8 References 6-75
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 7-12
7.2.5 Surface Adhesion 7-15
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 7-19
7.2.9 Fog and Dewfall 7-20
7.2.10 Resuspension and Surface Emission 7-21
7.2.11 The resistance Analog 7-22
7.3 Methods for Studying Dry Deposition 7-28
7.3.1 Direct Measurement 7-28
7.3.2 Wind Tunnel and Chamber Studies 7-31
7.3.3 Mlcrometeorological Measurement Methods 7-33
7.4 Field Investigations of Dry Deposition 7-39
7.4.1 Gaseous Pollutants , 7-39
7.4.2 Particul ate Pollutants 7-46
7.4.3 Routine Handling in Networks 7-51
xi
-------
Table of Contents (continued)
Page
7.5 Micrometeorologlcal Models of the Dry Deposition Process 7-52
7.5.1 Gases 7-52
7.5.2 Particles 7-55
7.6 Summary 7-56
7.7 Conclusions 7-60
7.8 References 7-63
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 Equfpment for Wet Deposition Networks .... 8-4
8.2.4 Wet Deposition Network Data Bases 8-7
8.3 Monitoring Capabilities for Dry Deposition 8-11
8.3.1 Introduction 8-11
8.3.2 Methods for Monitoring Dry Deposition 8-17
8.3.2.1 Direct Collection Procedures 8-18
8.3.2.2 Alternative Methods 8-20
8.3.3 Evaluations of Dry Deposition Rates 8-21
8.4 Wet Deposition Network Data With Applications to Selected Problems 8-28
8.4.1 Spatial Patterns 8-28
8.4.2 Remote Site pH Data 8-50
8.4.3 Precipitation Chemistry Variations Over Time 8-59
8.4.3.1 Nitrate Variation Since 1950's 8-59
8.4.3.2 pH Variation Since 1950's 8-61
8.4.3.3 Calcltm Variation Since the 1950's 8-65
8.4.4 Seasonal Variations 8-67
8.4.5 Very Short Time Scale Variations 8-68
8.4.6 Air Parcel Trajectory Analysis 8-68
8.5 Glaciochemical Investigations as a Tool in the Historical Delineation of
the Acid Precipitation Problems 8-70
8.5.1 Glaciochemical Data 8-70
8.5.1.1 Sulfate - Polar Glaciers 8-71
8.5.1.2 Nitrate - Polar Glaciers 8-72
8.5.1.3 pH and Acidity - Polar Glaciers 8-72
8.5.1.4 Sulfate - Alpine Glaciers 8-73
8.5.1.5 Nitrate - Alpine Glaciers 8-73
8.5.1.6 pH and Acidity - Alpine Glaciers 8-73
8.5.2 Trace Metals - General Statement 8-74
8.5.2.1 Trace Metals - Polar Glaciers 8-74
8.5.2.2 Trace Metals - Alpine Glaciers 8-76
8.5.3 Discussion and Future Work 8-76
8.6 Conclusions 8-79
8.7 References 8-83
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 9-5
9.1.2.3 Reaction Mechanisms 9-5
9.1.2.4 Removal Mechanisms 9-5
XI1
-------
Table of Contents (continued)
Page
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 9-8
9.2 Types of LRT Models 9-8
9.2.1 Etilerian Grid Models 9-8
9.2.2 Lagrangian Models 9-11
9.2.2.1 Lagrangian Trajectory Models 9-11
9.2.2.2 Statistical Trajectory Models 9-13
9.2.3 Hybrid Models 9-13
9.3 Modules Associated with Chemical (Transformation) Processes 9-14
9.3.1 Overview 9-14
9.3.2 Chemical Transformation Modeling 9-14
9.3.2.1 Simplified Modules 9-15
9.3.2.2 Multireaction Modules 9-15
9.3.3 Modules for NOX Transformation 9-16
9.4 Modules Associated with Wet and Dry Deposition 9-20
9.4.1 Overview 9-20
9.4.2 Modules for Wet Deposition 9-21
9.4.2.1 Formulation and Mechanist! 9-21
9.4.2.2 Modules Used in Existing Models 9-22
9.4.2.3 Wet Deposition Modules for Snow 9-24
9.4.2.4 Wet Deposition Modules for NOX 9-24
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 9-26
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 9-27
9.5.1 Overvi ew 9-27
9.5.2 Model Application 9-27
9.5.2.1 Selection Criteria 9-27
9.5.2.2 Regional Concentration and Deposition Patterns 9-28
9.5.2.3 Use of Matrix Methods to Quantify Source-Receptor
Relationships 9-29
9.5.3 Data Requirements „ 9-34
9.5.3.1 General 9-34
9.5.3.2 Specific Characteristics of Data Used in Model
Simul ations 9-37
9.5.3.2.1 Emissions 9-37
9.5.3.2.2 Meteorological Data 9-38
9.5.4 Model Performance and Uncertainties 9-38
9.5.4.1 General 9-38
9.5.4.2 Data Bases Available for Evaluating Models 9-40
9.5.4.3 Performance Measures 9-40
9.5.4.4 Representivity of Measurements 9-41
9.5.4.5 Uncertainties 9-41
9.5.4.6 Selected Results 9-42
9.6 Conclusions 9-47
9.7 References 9-49
xm
-------
THE ACIDIC DEPOSITION PHENOMENON AND ITS EFFECTS:
CRITICAL ASSESSMENT REVIEW PAPERS
Table of Contents
Volume 11
Effects Sciences
Note: Comment forms and guidelines to be used by reviewers can be found at the ends of
Volumes I and II
Page
E-l INTRODUCTION
1.1 Objectives 1-1
1.2 Approach 1-1
1.3 Chapter Organization and General Content 1-2
1.3.1 Effects on Soil Samples 1-3
1.3.2 Effects on Vegetation 1-3
1.3.3 Effects on Aquatic Chemistry 1-4
1.3.4 Effects on Aquatic Biology 1-4
1.3.5 Indirect Effects on Health 1-5
1.3.6 Effects on Materials 1-5
1.4 Acidic Deposition 1-5
1.5 Linkage to Atmospheric Sciences 1-6
1.6 Sensitivity 1-6
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 ... 2-1
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 2-19
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 2-25
2.3.3.1 Effects on Cation Nutrient Status 2-29
2.3.3.2 Effects on S and N Status 2-31
2.3.3.3 Acidification Effects on Plant Nutrition 2-34
XIV
-------
Table of Contents (continued)
Page
2.3.3.3.1 Nutrient deficiencies 2-34
2.3.3.3.2 total ion toxicities 2-34
2.3.3.3.2.1 Aluminum toxicfty 2-35
2.3.3.3.2.2 Manganese toxicity 2-36
2.3.4 Reversibility of Effects on Soil Chemistry 2-36
2.3.5 Predicting Which Soils will be Affected Most 2-37
2.3.5.1 Soils Under Cultivation 2-37
2.3.5.2 Uncultivated, Unamended Soils 2-37
2.3.5.2.1 Basic cation-pH changes In forested soils .... 2-40
2.3.5.2.2 Changes in aluminum or other metal concen-
tration in soil solution In forested soils ... 2-41
2.4 Effects of Acidic Deposition on Soil Biology 2-41
2.4.1 Soil Biology Components and Functional Significance 2-41
2.4.1.1 Soil Animals 2-41
2.4.1.2 Algae 2-42
2.4.1.3 Fungi 2-42
2.4.1.4 Bacteria 2-42
2.4.2 Direct Effects of Acidic Deposition on Soil Biology 2-43
2.4.2.1 Soil Animals 2-43
2.4.2.2 Terrestrial Algae 2-44
2.4.2.3 Fungi 2-44
2.4.2.4 Bacteria 2-45
2.4.2.5 General Biological Processes 2-45
2.4.3 Metals—Mobilization Effects on Soil Biology 2-47
2.4.4 Effects of Changes in Mlcrobial Activity on Aquatic Systems 2-48
2.4.5 Soil Biology Summary 2-48
2.5 Effects of Acidic Deposition on Organic Matter Decomposition 2-49
2.6 Effects of Soils on the Chemistry of Aquatic Ecosystems 2-50
2.7 Conclusions 2-56
2.8 References 2-59
E-3 EFFECTS ON VEGETATION
3.1 Introduction 3-1
3.1.1 Overview 3-1
3.1.2 Background 3-2
3.2 Plant Response to Acidic Deposition 3-5
3.2.1 Leaf Response to Acidic Deposition 3-5
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 3-10
3.2.3 Summary 3-17
3.3 Interactive Effects of Acidic Deposition with Other Environmental
Factors on PI ants 3-18
3.3.1 Interactions with Other Pollutants 3-18
3.3.2 Interactions with Phytophagus Insects 3-21
3.3.3 Interactions with Pathogens 3-21
3.3.4 Influence on Vegetative Hosts That Would Alter Relationships
with Insect or Mlcrobial Associate 3-24
3.3.5 Effects of Acidic Deposition on Pesticides 3-25
3.3.6 Simmary 3-26
3.4 Biomass Production 3-27
3.4.1 Forests 3-27
3.4.1.1 Possible Mechanisims of Response 3-28
3.4.1.2 Phenological Effects 3-30
3.4.1.2.1 Seed germination and seedling establishment .. 3-31
3.4.1.2.2 Mature and reproductive stages 3-33
3.4.1.3 Growth of Seedlings and Trees 1n Irrigation
Experiments 3-33
XV
-------
Table of Contents (continued)
Page
3.4.1.4 Studies of Long-Term Growth of Trees 3-34
3.4.1.5 Oieback and Decline in High Elevation Forests 3-37
3.4.1.6 Summary 3-41
3.4.2 Crops 3-42
3.4.2.1 Review and Analysis of Experimental Design 3-42
3.4.2.1.1 Dose-response determination 3-43
3.4.2.1.2 Sensitivity classification 3-44
3.4.2.1.3 Mechanisms 3-45
3.4.2.1.4 Characteristics of precipitation simulant
exposures 3-45
3.4.2.1.5 Yield criteria 3-46
3.4.2.2 Experimental Results 3-46
3.4.2.2.1 Field studies 3-47
3.4.2.2.2 Controlled environment studies 3-51
3.4.2.3 Discussion 3-59
3.4.2.4 Summary 3-62
3.5 Conclusions 3-62
3.6 References 3_65
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 4-1
4.2.1 Receiving Systems 4-1
4.2.2 pH, Conductivity, and Alkalinity 4-4
4.2.2.1 pH 4-4
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 4-6
4.3.1 Atmospheric Inputs 4-6
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 4-8
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 4-9
4.3.2 Characteristics of Receiving Systems Relative to Being Able to
Assimilate Acidic Deposition 4-10
4.3.2.1 Canopy 4-10
4.3.2.2 Soil 4-12
4.3.2.3 Bedrock 4-14
4.3.2.4 Hydrology 4-15
4.3.2.4.1 Flow paths 4-15
4.3.2.4.2 Residence times 4-17
4.3.2.5 Wetlands 4-17
4.3.2.6 Aquatic 4-18
4.3.2.6.1 Alkalinity 4-18
4.3.2.6.2 International production/consumption
of ANC 4-22
4.3.2.6.3 Aquatic sediments 4-24
4.3.3 Location of Sensitive Systems 4-25
4.3.4 Summary—Sensitivity 4-30
4.4 Magnitude of Chemical Effects of Acidic Deposition on
Aquatic Ecosystems 4-31
XVI
-------
Table of Contents (continued)
Page
4.4.1 Relative Importance of HNOa vs H2S04 4-31
4.4.2 Short-Term Acidification 4-37
4.4.3 Long-Term Acidification 4-38
4.4.3.1 Analysis of Trends based on Historic Measurements of
Surface Water Quality 4-44
4.4.3.1.1 Methological problems with the evaluation
of historical trends 4-44
4.4.3.1.1.1 pH 4-44
4.4.3.1.1.1.1 pH-early metho-
dology 4-44
4.4.3.1.1.1.2 pH-current metho-
dology 4-46
4.4.3.1.1.1.3 pH-comparability
of early and cur-
rent mesurement
methods 4-47
4.4.3.1.1.1.4 pH-general
problems 4-47
4.4.3.1.1.2 Conductivity 4-48
4.4.3.1.1.2.1 Conductivity
methodology 4-48
4.4.3.1.1.2.2 Comparability of
early and current
measurement
methods 4-48
4.4.3.1.1.2.3 General problems.. 4-48
4.4.3.1.1.3 Alkalinity 4-49
4.4.3.1.1.3.1 Early methodology. 4-49
4.4.3.1.1.3.2 Current
methodology 4-49
4.4.3.1.1.3.3 Comparability of
early and current
measurement
methods 4-50
4.4.3.1.1.4 Summary of measurement
techniques 4-51
4.4.3.1.2 Analysis of trends 4-51
4.4.3.1.2.1 Introduction 4-51
4.4.3.1.2.2 Canadian studies 4-53
4.4.3.1.2.3 United States studies 4-61
4.4.3.1.3 Summary—trends in historic data 4-74
4.4.3.2 Assessment of Trends Based on Paleol imnological
Technique 4-77
4.4.3.2.1 Calibration and accuracy of paleol imnological
reconstruction of pH history 4-78
4.4.3.2.2 Lake acidification determined by
paleolimnological reconstruction 4-78
4.4.3.3 Alternate Explanations for Acidification-Land Use
Changes - = 4-79
4.4.3.3.1 Variations in the groundwater table 4-79
4.4.3.3.2 Accelerated mechanical weathering or
land scarification 4-79
4.4.3.3.3 Decomposition of organic matter 4-80
4.4.3.3.4 Long-term changes in vegetation 4-80
4.4.3.3.5 Chemical amendments 4-80
4.4.3.3.6 Summary—Effects of land use changes
or acidification 4-80
4.4.4 Sunmary--Magnit.ude of Chemical Effects of Acidic Deposition 4-81
4.5 Predictive Modeling of the Effects of Acidic Deposition
on Surface Waters 4-82
XVI1
-------
Table of Contents (continued)
Page
4.5.1 Aimer/Dick son Relationship 4-83
4.5.2 Henriksen's Predictor Nomograph 4-88
4.5.3 Thompson Cation Denudation Rate Model (CDR) 4-91
4.5.4 Summary of Predictive Modeling 4-94
4.6 Indirect Chemical Changes Associated with Acidification
of Surface Maters 4-94
4.6.1 Metals 4-94
4.6.1.1 Increased Loading of Metals From Atmospheric
Deposition 4-95
4.6.1.2 Mobilization of Metals by Acidic Deposition 4-97
4.6.1.3 Secondary Effects of Metal Mobilization 4-98
4.6.1.4 Effects of Acidification on Aqueous Metal Speciation .... 4-98
4.6.1.5 Indirect Effects on Metals in Surface Waters 4-98
4.6.2 Aluminum Chemistry in Dilute Acidic Maters 4-99
4.6.2.1 Occurrence, Distribution, and Sources of Aluminum 4-99
4.6.2.2 Aluminum Speciation 4-102
4.6.2.3 Aluminum as a pH Buffer 4-102
4.6.2.4 Temporal and Spatial Variations in Aqueous
Aqueous Levels of Aluminum 4-104
4.6.2.5 The Role of Aluminum in Altering Element Cycling
Within Acidic Waters 4-106
4.6.3 Organic? 4-108
4.6.3.1 Atmospheric Loading of Strong Acids and Associated
Organic Micropollutants 4-108
4.6.3.2 Organic Buffering Systems 4-109
4.6.3.3 Organo-Metallc Interactions 4-109
4.6.3.4 Photochemistry 4-110
4.6.3.5 Carbon-Phosphorus-Aluminum Interactions 4-110
4.6.3.6 Effects of Acidification on Organic Decomposition
in Aquatic Systems 4-110
4.7 Mitigative Strategies for Improvement of Surface Water Quality 4-111
4.7.1 Base Additions 4-111
4.7.1.1 Types of Basic Materials 4-111
4.7.1.2 Direct Water Addition of Base 4-115
4.7.1.2.1 Computing base dose requirements 4-115
4.7.1.2.2 Methods of base application 4-119
4.7.1.3 Watershed Addition of Base 4-123
4.7.1.3.1 The concept of watershed
application of base 4-123
4.7.1.3.2 Experience in watershed 1 Iming 4-124
4.7.1.4 Water Quality Response to Base Treatment 4-126
4.7.1.5 Cost Analysis, Conclusions and Assessment of Base
Addi tion 4-128
4.7.1.5.1 Cost analysis 4-128
4.7.1.5.2 Summary—base additions 4-130
4.7.2 Surface Water Fertilization 4-130
4.7.2.1 The Fertilization Concept 4-130
4.7.2.2 Phosphorous Cycling in Acidified Water 4-132
4.7.2.3 Fertilization Experience and Water
Quality Response to Fertilization 4-133
4.7.2.4 Summary-Surface Water Fertilization 4-134
4.8 Conclusions 4-134
4.9 References 4-137
E-5 EFFECTS ON AQUATIC BIOLOGY
5.1 Introduction 5-1
5.2 Biota of Naturally Acidic Waters 5-3
xvm
-------
Table of Contents (continued)
Page
5.2.1 Types of Naturally Acidic Waters 5-3
5.2.2 Biota of Inorganic Acidotrophic Waters 5-4
5.2.3 Biota in Acidic Brownwater Habitats 5-6
5.2.4 Biota in Ifltra-Oligotrophic Waters 5-8
5.2.5 Summary 5-9
5.3 Benthic Organisms 5-15
5.3.1 Importance of the Benthic Comnunity 5-15
5.3.2 Effects of Acidification on
Components of the Benthos 5-16
5.3.2.1 Microbial Community 5-17
5.3.2.2 Periphyton 5-18
5.3.2.2.1 Field surveys 5-18
5.3.2.2.2 Temporal trends 5-19
5.3.2.2.3 Experimental studies 5-21
5.3.2.3 Microinvertebrates 5-22
5.3.2.4 Crustacea 5-23
5.3.2.5 Insecta 5-25
5.3.2.5.1 Sensitivity of different groups 5-25
5.3.2.5.2 Sensitivity of insects from different
microhabitats 5-30
5.3.2.5.3 Acid sensitivity of insects based on food
sources 5-31
5.3.2.5.4 Mechanisms of effects and trophic
Interactions 5-31
5.3.2.6 Mollusca 5-32
5.3.2.7 Annelida 5-33
5.3.2.8 Summary of Effects of Acidification on Benthos .......... 5-34
5.4 Macrophytes and Wetland Plants 5-39
5.4.1 Introduction ; 5-39
5.4.2 Effects on Acidification on Aquatic Macrophytes 5-43
5.4.3 Summary 5-45
5.5 Plankton 5-45
5.5.1 Introduction 5-45
5.5.2 Effects of Acidification on Phytoplankton 5-47
5.5.2.1 Changes in Species Composition 5-47
5.5.2.2 Changes in Phytoplankton Biomass and Productivity 5-54
5.5.3 Effects of Acidification on Zooplankton 5-57
5.5.4 Explanations and Significance 5-70
5.5.4.1 Changes in Species Composition 5-70
5.5.4.2 Changes in Productivity ; 5-72
5.5.5 Summary 5-75
5.6 Fishes 5-76
5.6.1 Introduction 5-76
5.6.2 Field Observations 5-77
5.6.2.1 Loss of Populations 5-78
5.6.2.1.1 United States 5-78
5.6.2.1.1.1 Adirondack Region of
New York State 5-78
5.6.2.1.1.2 Other regions of the eastern
United States 5-81
5.6.2.1.2 Canada 5-82
5.6.2.1.2.1 LaCloche Mountain Region of
Ontario 5-82
5.6.2.1.2.2 Other areas of Ontario 5-86
5.6.2.1.2.3 Nova Scotia 5-86
5.6.2.1.3 Scandinavia and Great Britain 5-91
5.6.2.1.3.1 Norway 5-91
5.6.2.1.3.2 Sueden 5-95
5.6.2.1.3.3 Scotland 5-95
XIX
-------
Table of Contents (continued)
Page
5.6.2.2 Population Structure 5-97
5.6.2.3 Growth 5-100
5.6.2.4 Episodic Fish Kills 5-103
5.6.2.5 Accumulation of fetals in Fish 5-105
5.6.3 Field Experiments 5-105
5.6.3.1 Experimental Acidification of Lake 223 Ontario 5-105
5.6.3.2 Experimental Acidification of Nor Ms
Brook, New Hanpshi re 5-108
5.6.3.3 Exposure of Fish to Acidic Surface Waters 5-108
5.6.4 Laboratory Experiments 5-112
5.6.4.1 Effects of Low pH 5-113
5.6.4.1.1 Survival 5-113
5.6.4.1.2 Reproduction 5-116
5.6.4.1.3 Growth 5-123
5.6.4.1.4 Behavior 5-124
5.6.4.1.5 Physiological responses 5-124
5.6.4.2 Effects of Aluminum and Other Metals in Acidic Waters ... 5-127
5.6.5 Sunmary 5-129
5.6.5.1 Extent of Impact 5-129
5.6.5.2 Mechanism of Effect 5-131
5.6.5.3 Relationship Between Water Acidity and Fish
Population Response 5-133
5.7 Other Related Biota 5-137
5.7.1 Amphibians 5-137
5.7.2 Birds 5-138
5.7.2.1 Food Chain Alterations 5-138
5.7.2.2 Heavy Metal Accumulation 5-139
5.7.3 Mammals 5-140
5.7.4 Summary ; 5-141
5.8 Observed and Anticipated Ecosystem Effects 5-144
5.8.1 Ecosystem Structure 5-144
5.8.2 Ecosystem Function 5-146
5.8.2.1 Nutrient Cycling 5-146
5.8.2.2 Energy Cycling 5-146
5.8.3 Summary 5-147
5.9 Mitigative Options Relative to Biological Populations at Risk 5-148
5.9.1 Biological Response to Neutralization 5-148
5.9.2 Improving F1sh Survival in Acidified Waters 5-150
5.9.2.1 Genetic Screening 5-150
5.9.2.2 Selective Breeding 5-151
5.9.2.3 Acclimation 5-152
5.9.2.4 Limitations of Techniquest to Improve Fish Survival 5-153
5.9.2.5 Summary 5-154
5.10 Conclusions 5-154
5.10.1 Effects of Acidification on Aquatic Organisms 5-155
5.10.2 Processes and Mechanisms by Which Acidification
Alters Aquatic Ecosystems 5-161
5.10.2.1 Direct Effects of Hydrogen Ions 5-161
5.10.2.2 Elevated Metal Concentrations 5-161
5.10.2.3 Altered Trophic-Level Interactions 5-162
5.10.2.4 Altered Water Clarity 5-162
5.10.2.5 Altered Decomposition of Organic Matter 5-162
5.10.2.6 Presence of Algal Mats 5-163
5.10.2.7 Altered Nutrient Availability 5-163
5.10.2.8 Interaction of Stresses 5-163
5.10.3 Biological Mitigation 5-164
5.10.4 Summary 5-164
5.11 References 5-165
XX
-------
Table of Contents (continued)
Page
E-6 INDIRECT EFFECTS OM HEALTH
6.1 Introduction 6-1
6.2 Food Chain Dynamics 6-1
6.2.1 Introduction 6-1
6.2.2 Availability and Bioaccumulation of Toxic Metals 6-2
6.2.2.1 Speciation (Mercury) 6-2
6.2.2.2 Concentrations and Speciations in Water (Mercury) 6-5
6.2.2.3 Flow of Mercury in the Environment 6-5
6.2.2.3.1 Global cycles 6-6
6.2.2.3.2 Biogeochemical cycles of Mercury 6-6
6.2.3 Accumulation in Fish 6-10
6.2.3.1 Factors Affecting Mercury Concentrations in Fish 6-11
6.2.3.2 Historical and Geographic Trends in Mercury Levels in
Freshwater Fish 6-22
6.2.4 Dynamics and Toxicity in Humans (Mercury) 6-24
6.2.4.1 Dynamics in Man (Mercury) 6-24
6.2.4.2 Toxicity in Man 6-25
6.2.4.3 Human Exposure from Fish and Potential for Health
Risks 6-31
6.3 Ground Surface and Cistern Waters as Affected by Acidic Deposition 6-34
6.3.1 Water Supplies 6-34
6.3.1.1 Direct Use of Precipitation (Cisterns) 6-35
6.3.1.2 Surface Water Supplies 6-36
6.3.1.3 Groundwater Supplies 6-40
6.3.2 Lead 6-43
6.3.2.1 Concentrations in Noncontaminated Waters 6-43
6.3.2.2 Factors Affecting Lead Concentrations
in Water, Including Effects of pH 6-43
6.3.2.3 Speciation of Lead in Natural Water 6-45
6.3.2.4 Dynamics and Toxicity of Lead in Humans 6-45
6.3.2.4.1 Dynamics of lead in humans 6-45
6.3.2.4.2 Toxic effects of lead on humans 6-46
6.3.2.4.3 Intake of lead in water and potential for
human health effects 6-53
6.3.3 Aluminum 6-57
6.3.3.1 Concentrations in Uncontaminated Waters 6-57
6.3.3.2 Factors Affecting Aluminum Concentrations in Water 6-58
6.3.3.3 Speciation of Aluminum in Water 6-58
6.3.3.4 Dynamics and Toxicity in humans 6-58
6.3.3.4.1 Dynamics of aluminum in humans 6-59
6.3.3.4.2 Toxic effects of aluminum in man 6-59
6.3.3.5 Human Health Risks from Aluminum in Water 6-59
6.4 Other Metals 6-60
6.5 Conclusions 6-60
6.6 References 6-63
E-7 EFFECTS ON MATERIALS
7.1 Introduction 7-1
7.1.1 Long Range and Local Effects 7-2
7.1.2 Inaccurate Claims of Acid Rain Damage to Materials 7-5
7.1.3 Complex Mechanisms of Exposure and Deposition 7-5
7.1.4 Laboratory vs Field Studies 7-7
7.1.5 Measurement of Materials Damage 7-7
7.1.5.1 Metals 7-7
7.1.5.2 Coatings 7-8
7.1.5.3 Masonry 7-8
7.1.5.4 Paper and Leather 7-8
7.1.5.5 Textiles and Textile Dyes 7-8
XXI
-------
Table of Contents (continued)
Page
7.2 Mechanisms of Damage to Materials 7-8
7.2.1 Metals 7-9
7.2.2 Stone 7-10
7.2.3 Glass 7-12
7.2.4 Concrete 7-12
7.2.5 Organic Materials 7-12
7.2.6 Deposition Velocities 7-13
7.3 Damage to Materials by Acidic Deposition 7-13
7.3.1 Metals 7-13
7.3.1.1 Ferrous Metals 7-15
7.3.1.1.1 Laboratory Studies 7-18
7.3.1.1.2 Field Studies 7-19
7.3.1.2 Nonferrous Metals 7-23
7.3.1.2.1 Aluminum 7-23
7.3.1.2.2 Copper 7-25
7.3.1.2.3 Zinc 7-25
7.3.2 Masonry 7-26
7.3.2.1 Stone 7-26
7.3.2.2 Ceramics and Glass 7-30
7.3.2.3 Concrete 7-30
7.3.3 Paint 7-31
7.3.4 Other Materials 7-35
7.3.4.1 Paper 7-35
7.3.4.2 Photographic Materials 7-35
7.3.4.3 Textiles and Textile Dyes 7-36
7.3.4.4 Leather 7-36
7.3.5 Cultural Property 7-37
7.3.5.1 Architectural Monuments 7-37
7.3.5.2 Museuns, Librarties and Archives 7-37
7.3.5.3 Medieval Stained Glass 7-38
7.3.5.4 Conservation and Mitigation Costs 7-38
7.4 Economic Implications 7-40
7.5 Mitigative Measures 7-42
7.6 Conclusions 7-43
7.7 References 7-44
xxn
-------
Acronym and Abbreviation List
ADI (acceptable daily intake) E-6
AL (Aeronomy Laboratory, NOAA)
6-ALA (s-aminolevulinic acid) E-6
ANC (acid neutralizing capacity) E-4
ARL (Air Resources Lab, NOAA)
ARS (Agricultural Research Service, DOA)
BCF (bioconcentration factor) E-6
BLM (Bureau of Land Management, DOI)
BLMS (boundary layer models) A-9
BM (Bureau of Mines, DOI)
BNC (base neutralizing capacity) E-4
BNC aq (aqueous base neutralizing capacity) E-4
BOD (biologic oxygen demand)
BS (base saturation) E-4
BSC (base saturation capacity) E-4
BUREC (Bureau of Reclamation, DOI)
BWCA (Boundary Water Canoe Area)
CANSAP (Canadian Sampling Network for Acid Precipitation)
Cp (base cation level) E-4
CDR (cation denudation rate) E-4
CEC (cation exchange capacity) E-2
CEQ (Council on Environmental Quality)
CSI (calcite saturation index) E-4
CSRS (Cooperative State Research Service, DOA)
xxiii
-------
DOA (Department of Agriculture)
DOC (dissolved organic carbon) E-4
DOD (Department of Defense)
DOE (Department of Energy)
DOT (Department of Interior)
DOS (Department of State)
ELA (experimental lakes area) E-4
ENAMAP (Eastern North America Model of Air Pollutants)
EPA (Environmental Protection Agency)
EPRI (Electric Power Research Institute)
ERDA (Energy Research and Development Agency (defunct)
ESRL (Environmental Sciences Research Laboratory, EPA)
FA (fulvic acid) E-4
FDA (flourescein diacetate) E-2
FEP (free erythrocyte protoporphyrin) E-6
FGD (Flue Gas Desulfurization)
FS (Forest Service, DOA)
FWS (Fish and Wildlife Service, DOI)
GTN (Global Trends Network)
HHS (Department of Health and Human Services)
ILWAS (Integrated Lake Watershed Acidification Study) E-4
LAI (leaf area index) A-7
LIMB (Limestone Injection/Multistage Burner)
LRTAP (Long-Range Transboundary Air Pollution)
LSI (Langelier Saturation Index) E-6
MAP3S (Multi-State Atmospheric Power Production
Pollution Study)
xxiv
-------
MCPS (Mesoscale convective precipitation systems) A-3
MOI (Memorandum of Intent, U.S.-Canada)
NADP (National Atmospheric Deposition Program)
NASA (National Aeronautics and Space Administration)
NATO (North Atlantic Treaty Organization)
NBS (National Bureau of Standards, DOC)
NCAR (National Center for Atmospheric Research)
NECRMP (Northeast Corridor Regional Modeling Program) A-2
NOAA (National Oceanic and Atmosperic Administration, DOC)
NPS (National Park Service, DOI)
NSF (National Science Foundation)
NSPS (New Source Performance Standards)
NTN (National Trends Network)
NWS (National Weather Service, NOAA)
OECD (Organization for Economic Cooperation and
Development)
OMB (Office of Management and Budget)
ORNL (Oak Ridge National Laboratory)
OSM (Office of Surface Mining, DOI)
PAN (peroxyacetyl nitrate) E-3, A-5
PBCF (practical bioconcentration factor) E-6
PBL (planetary boundary layer) A-4
PGF (pressure gradient force) A-3
PHS (Public Health Service)
RSN (Research Support Network)
SAC ($04 adsorption capacity) E-4
SEAREX
XXV
-------
SAES (State Agricultural Experiment Station, DOA)
SCS (Soil Conservation Service, DOA)
SURE (Sulfate Regional Experiment, EPRI)
TFE (total fixed endpoint alkalinity) E-4
TIP (total inflection point alkalinity) E-4
TVA (Tennessee Valley Authority)
US6S (United States Geological Survey, DOI)
VOC (Volatile Organic Compounds)
WMO (World Meteorological Organization)
xxvi
-------
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 and leading up to the presence of
acidic and acidifying 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 emitted from natural and anthropogenic sources. The chemical
species of principal concern are the hydrogen ion (H+), ammonium ion
(NH4+), sulfate ion (S042-), and the nitrate ion (NOs-). Chloride, in
the form of hydrogen 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 includes volcanic and sea spray contributions, the latter, soil
and vegetation contributions. The anthropogenic source categories
include electric utilities, 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 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
1-1
-------
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,
catalytic 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 orginally 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 transformations (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 geographic locations provide a
complex picture of concentrations of these substances 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 constituents 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 interactive physical and chemical phenomena that occur prior to and
during the precipitation process. Cloud droplets form and evaporate,
airborne pollutants are incorporated into and released by condensed
water, chemical reactions occur, ice crystals form and melt, energy is
exchanged, and hydrometeors are created and evaporate. These and a
multitude of additional processes create a continually changing
1-2
-------
environment for pollution elements within a storm system. The final
stage of these complex scavenging processes 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 atmosphere 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 associated 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 collection vessels and with surrogate surfaces specific to varous
receptor surfaces 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
processes for gases and for particles.
1.2.8 Deposition Monitoring
Deposition monitoring networks have been established to collect wet
deposition 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 essentially wet deposition monitoring networks. Dry
deposition monitoring networks 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
1-3
-------
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 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 compared 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 include 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 neutralization 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 of hydrogen ion
concentration. A pH of 6.0 is ten times more acid than a pH of 7.0.
Atmospheric water droplets are in equilibrium with the geophysical
concentrations 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 measureed
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 deposition. Acidic or basic substances can
collect on vegetation or soil surfaces and subsequently be washed into
the soil by rainfall. Once substances are within an ecosystem,
1-4
-------
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.
1-5
409-261 0-83-2
-------
THE ACIDIC DEPOSITION PHENOMENON AND ITS EFFECTS
A-2. NATURAL AND ANTROPOGENIC EMISSIONS SOURCES
2.1 INTRODUCTION
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 atmospheric substances, and for deposit to
surfaces. Chapter A-2 discusses where acidic and acidifying substances
originate, thus setting the stage for further 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
emitted anthropogenically affect acidic deposition, either by catalysis
or direct reaction with sulfur, nitrogen, and chlorine containing
compounds? In contrast to acidic or acidifying substances, what sources
exist for neutralizing substances—including ammonia, soil-related or
cement plant dusts, and alkaline 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
dioxide, are ubiquitous trace constituents of the Earth's atmosphere
even in very remote, natural areas. Thus, we must assume that these
common pollutants result from natural sources in the unperturbed
2-1
-------
environment. Concentrations in most backgound situations are low, and
sampling and analysis problems are major factors that limit the
determination of 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 concentration 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 potential 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 influencing 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 technique described below, Eriksson (I960) estimated
natural sulfur sources, as sulfur, to be 77 x 10^ mT (77 Tg S) from
land areas and 190 x 106 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
estimate 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
2-2
-------
accounted for by wet and dry deposition. 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 have been
estimated, but until recently (Adams et al. 1980, 1981a), the much
larger biological component had to be estimated from only fragmentary
data. 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-i (Eriksson
1963). To place the biogenic contribution in perspective, Granat et al.
(1976) estimated anthropogenic 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 contribution 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 sulfur compounds, tabulated tropospheric
concentration data for these compounds. In Table 2-1, background
concentration data are presented from the tabulations of Sze and Ko
(1980) that are considered to be most applicable to the northeastern
U.S. conditions without anthropogenic influences. 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.
2-3
-------
TABLE 2-1. BACKGROUND CONCENTRATIONS OF SULFUR COMPOUNDS
(ADAPTED FROM SZE AND KO 1980)
Compound
Concentration
yg m-3
Location
S02
S042-
COS
H2S
(CH3)2S
CS2
0.52 +_ 0.23
0.25 +_ 0.12
0.05
1.26 +_ 0.15
0.007 - 0.07
0.15
0.31
Western U.S. and
Canada above
boundary layer
Western U.S. and
Canada within
boundary layer
Remote ocean areas
67°N-57°S
Southern Florida
Wallops Island, VA
England
2-4
-------
2.2.1.3 Biogenic Emissions of Sulfur Compounds—The initial estimates
of biogenic emissions, such as those by Eriksson (1960), assigned the
total biogenic estimate to hydrogen sulfide (^S) 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 compounds were also part of the natural background. It should be
noted, however, that all of the authors dealing with the sulfur cycle
recognized the probable complexity of the natural emission cycle, and
the assumption that the total emission was H2$ was recognized as a
simplification of the probable 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 biogenic component of the
sulfur cycle, i.e., the emissions from natural sources, was one of
having suitable analytical methodology. It was not until the 1970's
that the measurement technology for ^S and the organic sulfur
compounds that might be expected to come from natural sources was
developed. The nature of potential biogenic sulfur emissions had
emphasized ^S as the probable compound (e.g., see Eriksson 1960)
..,
although earlier Conway (1942) had concluded that non-sea-salt sulfur in
precipitation away fron anthropogenic sources may be due to volatile
sulfur compounds such as H2S or possibly mercaptans. Lovelock et al .
(1972) showed that (CH3)2S (dimethyl sulfide) was present in sea
water and given off by enclosed soils, and they proposed ((^3)2$ as
an important 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 biogenic 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 H2S. Maroulis and Bandy (1977)
used gas chroma tographic techniques for atmospheric studies of
(CH3)2S. Delmas et al . (1980) carried out a number of measurements
of the rate of evolution of \\2S 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 biogenic 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 chroma tography to soil
emissions in the form of air samples collected from a small stirred
chamber placed over selected soil and water surfaces. This gas chroma-
tographic analytical technique was capable of detecting six potential
biogenic sulfur emissions compounds: H2S, (CHa^S, (^3)282, COS, C$2,
and CH3SH. In the sampling program used by Aneja et al . (1981) the
2-5
-------
detectable emission rate for H2$, ((^3)2$, and COS was 0.01 g S nr2 yr'1
and for C$2, CHaSH, and (CH3)2$2 it was 0.05 g S irr2 yr"1. 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 emissions 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 compounds to be above the
detection limit while "saline marsh mud flat" showed detectable
emissions only of h^S and COS.
Further improvements in sulfur gas analysis by gas chroma tography
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 deposition, the
results of this study by Adams et al . as reported in the several
available papers and reports will be used as a basis for the following
evaluation 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. I^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 sul fide (COS) and carbon di sul fide (CS2) were typically
higher than those of dimethyl sul fide [(CI^S]. Dimethyl di sul fide
C(CH3)2S2l was found in low concentrations in a large proportion
of the samples, and methyl mercaptan (C^SH) 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) .
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. 1981a)
ro
Average sulfur flux, g S m~2
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
-------
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
' Celeryville, OH
Elba, NY
E. Wareham, MA
Brunswick Co., NC
Belle Glade, FL
Lakeland, FL
Jeanerette, LA
Fair hope, AL
H2S
0.018
0.046
0.044
0.084
0.042
0.047
0.158
0.09
0.005
0.069
0.01
COS CH3SH
0.008
0.011
0.024
0.01
0.012
0.023
0.007
0.002
0.001
0.001
(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
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
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
Coastal Soils
Georgetown,
SC
Mollisols
Ames, LA
Linneus, MO
Yankeetown, IN
Stephenville, TX
0.008
0.147
0.104
0.073
0.008
0.017
0.009
0.023
0.002
0.002 0.005
0.003
0.003
0.002
0.001
0.016
0.005
0.021
0.004
0.0005
0.0015
0.023
0.18
0.12
0.12
0.008
-------
TABLE 2-2. CONTINUED
ro
Average sulfur flux, g S rrr2 yr-1
Soil types/locales
Alluvial Soils
Clarkedale, AR
Al f 1 sol s
WadesvMle, IN
Kearnysvllle, WV
R.T.P., NC (Wooded)
R.T.P., NC (Cultivated)
Jeanerette, LA
Shreveport, LA
Stephenvllle, TX
Inceptlsols
Phllo, OH
Belle Valley, OH
Spodosols
1 W. Wareham, MA
Ul ti sol s
Calhoun, GA
Falrhope, AL
Hastings, FL
Freshwater Pond
Belle Valley, 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 „. (CH3)2S2
0.003
0.002 0.002
0.022 0.0001
0.001
0.001
0.0004
0.005
0.003
0.001 0.0014
0.010 0.002
0.0002
0.011 0.0001
0.005 0.0001 0.0003
0.002 0.0003 0.0007
0.028 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
aUn1dent1f1ed sulfur gases.
-------
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 statistical 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~l, and it is
associated with a total SURE region biogenic emission of about 0.12 Tg S
yr~l.
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). Some surface 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 conclusion 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 emissions 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
2-10
-------
TABLE 2-3. SUMMARY OF ANNUAL SULFUR FLUX BY SOIL GROUPINGS
WITHIN THE STUDY AREA (ADAMS ET AL. 1981)
Soil grouping Sulfur flux Land area Emission density
g S yr-1 m2 g S m-2 yr~l
Coastal wetlands 48,822 x 1Q6 2.56 x IQH 0.191
Inland high organic 13,451 x 1Q6 6.85 x 10*1 0.020
Inland mineral 56,843 x 1()6 27.26 x IflU 0.021
Total 119,116 x 106 36.7 x Iflll 0.032
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
-------
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 biogenic 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
identification 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, Louisiana, and
Georgetown, South Carolina. This rapid increase of sulfur flux
southward is interpreted as being a result of an increase in
temperatures, an increase in wetland areas, 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 biogenic
emission area. For Florida, the area along the northern border near
30°N has an indicated annual flux density in units of metric tons (10-*
kg) of about 350 mT S per 6,400 km2, or about 0.05 g S nr2 yr'1;
while 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 statewide
anthropogenic emissions of S02 for Florida in 1978 were about 606,000
mT S02 yr-1 or 303,000 mT S yr-1 (Section 2.3.2.1). Thus, the estimated
biogenic emissions on a statewide basis in Florida are about 5% 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~l. 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 anthropogenic 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
2-13
-------
2,000
1,000 -
CM
'1
O
o
of.
>-
500 -
300 -
80 km INTERVALS
Figure 2-2.
Total 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 km2) (103 mT S
yr-1 equals 0.16 g S nr2 yr-1). Adapted from Adams
et al. (1981b).
2-14
-------
estimates. The individual 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 number 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 $042- 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 mT 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 relatively long reaction times required for sul fate
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 land areas is about 64 Tg S yr-1. This may be
compared with Granat'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 m"^
yr-1 over the 149 x 1012 m2 global land area. A similar figure
for Granat's estimate is about 0.22 g S nr2 yr-1. The model shown
in Figure 2-2 when extended to equitorial latitudes predicts an emission
value that is within the range of the measurements 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. Similarly, 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 rrr2 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~l. If this same emission density is
extended to the contiguous United States, an area of 7.824 x 10l2
m2, the resulting natural source is 0.23 Tg S yr~l. This latter
figure should be considered a maximum upper limit because it 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.
2-15
-------
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 km'2 yr"1. Open ocean areas are
estimated to have an emission of less than 10 kg knr2 yr'I, although
open ocean emissions were not measured by Adams et al. (1980). South of
SURE grid, soil emissions are expected to increase generally, as
indicated by the latitudinal distribution 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 estimated 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, more recent estimates by
Cadle (1980) may change the evaluation of the importance of volcanic
emissions.
2.2.1.4.1 Volcanism. 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 background 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 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).
2-16
-------
>30 kg
22.5 - 29 kg knT2 yr'1
HI 15 - 22.5 kg km-2 yr"1
<15 kg km~2 yr'1
OCEAN, V 10 kg km'2 yr"1
Figure 2-3. Annual biogenic sulfur emission pattern for the SURE grid
over the northeastern U.S. Adapated from Adams et al.
1980.
2-17
-------
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-1, or only a few percent of the total estimated
global sulfur cycle.
Since Granat1s 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-1 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 H?S, 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 volcanic emissions was that they
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-1.
The major sulfur compound from volcanic action, as noted by Cadle, is
S02- Cadle (1980) also considered the volcanic emissions of H2$,
COS, and C$2 and concluded that they were unimportant on a global
scale relative to S02-
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., that 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.
2-18
-------
TABLE 2-4 ESTIMATES OF VOLCANIC SULFUR GAS FLUX VALUES
(ADAPTED FROM DATA IN CADLE 1980)
Authors
Bartels
Kellog et al .
Friend
Stoiber and Jepsen
Naughton et al .
Granat et al.
Cadle
Date
1972
1972
1973
1973
1975
1976
1980
Estimated Flux
(Tg S yr-1)
17
0.8
2
5
24
3
2.1-8
2-19
-------
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
precipitation 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). 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
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 sul fate'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
2-20
-------
trace levels of SO? at coastal sites in Antarctica, and Lovelock et
al . (1972) measured dimethyl sul fide 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. Dept. of Commerce 1968). On an annual
basis this gives an onshore transport of marine air of about 1.2 x
m y across tne Gulf Coast (a
IQ m y i across tne Gulf Coast (about 160° 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, (^3)2$, and SO^2" 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 yg S m"3 in the marine
air masses that could be expected to participate in acidic deposition
processes. Considering the estimated total annual air mass volume
transported across the Gulf and Pacific Coasts given above (1.2 x 1018
rrr 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'1 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 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
contiguous 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.
2-21
-------
2.2.1.5 Scavenging Processes and Sinks — Ultimately, reactive materials
such as the sulfur compounds return to the 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
atmosphere, generally tending toward oxidation to SO^- 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 H£S 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 SO? atmospheric
lifetimes 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.
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 H2$ via the liquid droplet route is given by Granat
et al . (1976) as a day or more; and for (CH3)2S the reaction rate is
even slower. The reaction of (CH3)2S 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, C$2, COS, (^3)2$, (CH3)2S2,
and CH3SH, are given by Graedel (1978), Sze and Ko (1980), and others.
These reactions proceed to H2$04 and/or sul fates, but not always
through S02 as an intermediate compound. The common sulfate compound
in the atmosphere is ammonium sulfate [(NH^SO^ as a result of
the reaction, presumably in liquid droplets, between the two common
gases ammonia (H^) and
As mentioned above, pollutants are deposited on the Earth's surface
by either wet or dry processes and these topics are discussed in detail
in other chapters (Chapters A-6 and A-7) of this document. However,
2-22
-------
briefly with regard to acidic deposition, the precipitation scavenging
mechanisms are directly involved in the precipitation pH or acidic
deposition controversy, and it is useful to mention some aspects of
deposition in this discussion. Various authors have pointed out that
surface waters may be affected by deposited pollutants, whether they
arrive as part of the precipitation chemistry or are deposited on the
ground in a dry state and then are incorporated into the surface water.
Resuspension of sulfur compounds is probably minor because of their
general solubility and thus rapid incorporation into the soil. Desert
areas and agricultural regions with exposed soils may create situations
where strong winds may cause blowing dust. This would resuspend both
the deposited material and natural soil constituents.
Granat et al. (1976) have attempted to estimate the relative
importance of precipitation and dry deposition processes. They argue
that dry deposition increases in relative importance for situations
where the value of the dry deposition velocity, Vj, is large. This
would occur where gaseous compounds are a relatively large fraction of
the total atmospheric sulfur. Granat et al. (1976) also point out that
wet deposition increases in importance when the S02 to S04
particle formation rate and the boundary layer mixing depth increase.
This reduces the value of Vj and increases the probability of water
droplet interaction. For background sulfur emissions such as H2$,
V(j would probably be similar to that for SO^. The mixing depth
would be relatively great because atmospheric residence times of the
trace compounds would be relatively long (e.g., one or more days). For
this type of situation, the deposition process would be expected to be
dominated by aerosol formation and wet, rather than dry, deposition
processes. Thus, the natural sulfur emissions could be expected to
affect the precipitation chemistry of an area more than would the dry
deposition accumulation onto the Earth's surface.
2.2.1.6 Summary of Natural Sources of Sulfur Compounds—There are many
problems remaining in the natural sulfur cycle and many unknown factors;
however because of the importance of the acidic deposition problem it is
useful to summarize the natural sulfur cycle and relate it to
anthopogenic emissions. For land areas, probably the most important
natural sources of sulfur compounds are the emissions from biological
actions in the soil although on a global basis volcanos may also be
significant. In mid-latitudes, and specifically in the United States
east of the Mississippi River, an average biogenic sulfur emission rate
of about 0.03 g S nr2 yr'1 is indicated by extensive, although still
incomplete, field experiments. The emission rate from soil sources
increases with ambient temperature and in coastal wetlands. Thus, there
is a rapid increase in the emission rate of sulfur compounds from the
soil from north to south. Coastal wetlands, although they cover only
about 7 percent of the land area, have an average emissions rate of
about 10 times the rate of inland soils and account for about 40 percent
of the biogenic sulfur emissions in the area east of the Mississippi.
Figure 2-3 summarizes, on a grid basis, the results of a measurement
program on gaseous sulfur emissions from soils in the midwest and
eastern United States. The more arid and alkaline soils in the west
2-23
-------
would be expected to have lower biogern'c 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 101Z m2, results in
an estimated total biogenic emission of about 0.23 Tg 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)2S2, and CH3SH.
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 sea water 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 compounds. Estimates of oceanic
transported sulfur were made using a 3-km mixing depth, an 8-m-sec"1
average onshore wind, and background sulfur concentrations of 0.18 x
10"6 g S m"3 for gaseous compounds and 0.02 x 10~6 g S m"3 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-1 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 contributor 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 contribu-
tions 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 anthropogenic sulfur oxide emissions that are in
the range of 12 to 15 Tg S yr-*.
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 (NpO), as potentially acidic nitric oxide (NO) and nitrogen
dioxide (N02), and as potentially acid-neutralizing ammonia (NH3).
2-24
-------
The sources for these 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 NH^
based on relatively recent research, which are probably applicable to
non-anthropogenic-affected locations.
2.2.2.2 Estimates of Natural Global Sources and Sinks—A first
approximation of the global magnitude of natural sources 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 significantly 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, N02, and NH3
or NH4+. The nitrogen oxides, NO and NO?, were combined as NOX
for this estimate, and the MH3 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 emissions of NOX compounds totaling 19 Tg N
yr-l 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 emphasize 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 NH3 (N) while deposition
terms are shown in reference to NHd"1" (N). In a detailed paper
submitted for publications, Logan (1982) derived a nitrogen cycle with
several important differences relative to that given in Table 2-6.
Biogenic emission of NOX is estimated by Logan at 8 Tg N yr"1 with a
2-25
-------
TABLE 2-5. ATMOSPHERIC BACKGOUND CONCENTRATIONS OF
NITROGEN OXIDES AND AMMONIA
Constituent
Concentration
yg m"3
Reference
NOX (afternoon)
as N02
NO (afternoon)
(afternoon)
(land)
NH3 (ocean)
0.47
0.025 - 0.062
0.038 - 0.043
0.7 - 1.4
0.6
Kelly et al.
(1980)
Kelly et al.
(1980)
Kelly et al.
(1980)
Hoell et al.
(1980)
Ayers and Gras
(1980)
2-26
-------
TABLE 2-6. GLOBAL EMISSIONS OF NITROGEN COMPOUNDS3
Total
Tg N yr'1
Global emission
density
g N m"* yr-1
Terrestrial
NOX
NO
emissions0
x wet deposition
NOX dry deposition
NH3 emissions'1
Nfy wet deposition
NH4 dry deposition
Organic N wet deposition
Atmospheric Reactions (global)'
loss via OH
x formation
NOX lightning formation
NH3
NO,
N0xfrom N20
Oceanic^
UV
NOX wet deposition
NOX dry deposition
wet deposition
NH4 dry deposition
Organic N emissions
River flow to ocean
NOX
NH4
Organic N
21
13
19
109
30
61
89
30
53
232
60
126
10 - 100
3-8
3-8
?
0.3
5 - 16
6-17
8-25
11 - 25
10 - 20
5 -
< 1
8 -
11
13
0.14
0.09
0.13
0.73
0.20
0.41
- 0.59
- 0.20
- 0.36
- 1
56
0.40
0.85
0.07 - 0.67
0.006 - 0.016
0.006 - 0.016
0.0006
0.014
0.017
0.022
0.03
0.03
- 0.04
- 0.05
- 0.07
- 0.07
- 0.05
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 ielude anthropogenic contributions.
Original reference includes 4 to 12 Tg N yr'1 from coal combustion,
Deposition terms include anthropogenic contributions.
eGlobal area: 5.13 x 1014 m2.
fOcean area: 3.64 x 1014 m2.
2-27
-------
range of 4 to 16 Tg N yr'1 in a comparison to a value of 21 to 89 Tg N
yr-1 in Table 2-6. Logan (1982) also estimates lightning as a
potential source of 8 Tg N yr'1 (range 2 to 20 Tg N yr-1). Logan
estimated NOX from fossil fuel sources at 21 Tg N yr"1 plus an
additional 12 Tg N yr-1 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
An estimate of the wet deposition of organic nitrogen compounds,
e.g., ami no 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 indicated by the order of magnitude range for the
estimate of terrestrial deposition. 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 nr2 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.
Galbally (1975) has made separate estimates of NOX and HN3
sources and sinks, based on a boundary layer gradient method analogous
to a calculation of dry deposition. For the Northern Hemisphere, he
obtained an NOX emission 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 in making his estimates. His
estimates may be converted to average emission densities of 0.32 g N
m-2 yr-l for NOX and 0.55 g N nr2 yr-1 for NH4+. These values are
comparable to those derived from Soderlund and Svensson (1976)
and listed in Table 2-6. Gal bally 's estimating procedure would appear
to be relatively insensitive to local high concentrations of anthro-
pogenic emissions. In Table 2-6 natural NOX emission densities of
0.14 to 0.60 g N m"2 yr"1 are indicated. More recent estimates
(e.g., Logan 1982) arrive at lower values of natural emissions because
they relate to newer and lower ambient background NOX concentrations.
The nitrogen compounds N02 and HN4+ 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 generally 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 Galbally (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
2-28
-------
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.
2.2.2.3 Biogenic Sources of NOy Compounds—It seems to be generally
concluded that the major natural sources of NOX are found in the
terrestrial biosphere (Junge 1963, Galbally 19/5, 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 NOX 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 19/0) 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 N02 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 experiments 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 N2; 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, N2, N20, and N0£.
In experiments on five soils in the pH range of 4.8 to 6.0, held for 2
days at 25 C, the evolved NO? accounted for 55 percent of the applied
nitrite. At near neutral pH (6.6 to 7.0), 28 percent of the nitrite was
2-29
409-261 0-83-3
-------
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 conditions. However, they do indicate clearly the evolution of NO
and N02 from 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 N02
evolved from nitrite applied to the soils as NaN02. Prior experiments
by Makarov (1969) were related to applications of nitrate as NH4N03
and the results showed a decrease in the evolution of NO? from these
field soils when microbiological processes were reduced by the addition
of inhibiting substances to the test soil field plots. Thus it was
hypothesized that N02 soil emissions were related to microbiological
activity. Perhaps trie 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 N02 evolution during one experimental
period averaged 0.6 g ha~l hr"1 from May 31 to September 26 (119
days). This N02 production is 0.17 g nr2, 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 nr2
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 N02 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 NOX emissions were estimated to have
an emission density of 0.14 to 0.6 g N nr2 yr-1. 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 NOX 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 followed by an atmospheric reaction
2-30
-------
to form N02 was hypothesized earlier by Robinson and Robbins (1970b).
In the Australian field measurements by Gal bally and Roy, the observed
NO emission density, If integrated over a year, amounted to a value of
0.1 g N m~2 yr~l. If this rate is extended to a global land area
value, it produces a total nitrogen emission of 10 Tg N yr-1.
Bui la 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 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 that 1 Tg N
yr-1 and thus is insignificant in the global nitrogen oxide cycle.
2.2.2.4 Tropospheric and Stratospheric Reacjtions--A small transport of
N02 into the troposphere from the st'rartosp1he>e probably occurs.
Soderlund and Svensson (1976) estimate this flow at 0.3 Tg N yr-1,
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 nr3 as N02 in surface air to about 0.38
yg m~3 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-1
(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
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).
2-31
-------
Atmospheric reactions of NH3 in the troposphere involving
reactions with OH radicals have been proposed as another source of
NOX. Soderlund and Svensson (1976), usinq reaction systems
suggested by Crutzen (1974) and McConnel (1973), estimated a formation
rate of NOX from NHj 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 m-2 yr-l. Thus, this is also an
inconsequential source of NOX-
2.2.2.5 Formation of NOX by Lightning—The question of nitrogen
fixation oy lightning has been studied for more than 150 years, and no
definitive answer is yet at hand. Soderlund and Svensson (1976)
leave the possibility of lightning fixation as still a questionable
atmospheric source of NOX, as indicated in Table 2-6. They note one
reference on the question of lightning fixation of nitrogen, dated 1827
and authored by J. Kiebig.
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 N0£ 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 that observational evidence
linking atmospheric NOX to electrical discharge is for the most part
still lacking. However, modeling and theoretical analyses done since
the early I9601s 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 NO? production rate of 14.4 Tg yr-1 or 4.4 Tg N yr~l.
Dawson (1980), in an article published back-to-back with Hill et al.
(1980), concluded that liqhtning may produce about 3 Tg N yr~l.
Dawson also used Noxon1s (1976, 1978) data on solar spectral
measurements of enhanced N02 around thunderstorms to deduce a global
annual N02 production rate of 7 Tg N yr"1 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 (1982) 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-1.
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
2-32
-------
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 significant NOX formation, the fact that thunderstorms and their
accompanying lightning are frequent in the midwestern and eastern
regions of the United States could be important considerations with
regard to acidic deposition in the northeastern states and southeastern
Canada.
2.2.2.6 Biogenic NOX Emissions Estimate for the United States--
Quantitative measurements of NOX emissions for a wide variety of
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 procedure has already been described,
and, as noted in Table 2-6, the nonanthropogenic 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 emission 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 difusivity and concentration gradient.
With this calculation procedure and a surface layer average concentra-
tion 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 biogenic NOX emission of 2.4 Tg N
yr-1 and 0.69 TgN yr'1 for the contiguous 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 (1982) using NO and NO? emission measure-
ments from pasture plots of Gal bally and Roy (1978), has estimated the
global NOX biogenic source to be 8 Tg N yr-i. This is a value of
about 0.06 g N nr2 yr"1 or about 20 percent of the emission density
2-33
-------
calculated above from Gal bally (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~I, respectively.
Measurement techniques for NOX that are applicable to background
situations have been available only in recent years and it appears that
general NOX background concentrations may be significantly lower than
the values used by Galbally (1975) and Soderlund and Svensson
(1976). This may be especially 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 yg m~3 as
shown in Table 2-5. This is very much lower than the 6 yg nr* used
by Galbally (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 (1982) global emission estimate may be
high by about a factor of 3 or so.
Table 2-7 summarizes these several estimates of the biogenic 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 less
than 5 percent of the estimated anthropogenic emissions in both the
contiguous United States and 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 inferred. First, ammonium is found
in relatively high concentrations in rainwater, 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, increasing
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 surface
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,
2-34
-------
TABLE 2-7. SUMMARY OF BIOGENIC NOX ESTIMATES
FOR THE UNITES STATES
Contiguous U.S. east of
U.S. Mississippi River
Author Tg N yr"1 Tg N yr'1
Soderlund and Svensson 1.25 - 5.30 0.36 - 1.52
(1976)
Galbally (1975) 2.4 0.69
Logan (1982) 0.47 0.13
Boundary Layer Cone.
= 0.25 ppb (see text) 0.15 0.04
1978 Anthropogenic 10.7 8.9
(this chapter)
2-35
-------
although there have been a large number of investigations by soil
scientists and agronomists examining NH3 losses as a function 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 NOX by placing small
plastic hoods over areas of topsoil in pine-, oak-, and 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'1 as NH3 is taken for the forest soil emissions rate,
it would translate into an annual nitrogen flux of about 13 g N m'2
yr"1, a figure about an order of magnitude higher than that estimated
for ammonia emissions by Soderlund and Svensson (1976) and listed in
Table 2-6. Even if the NH3 emissions estimate 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 that needs to
be corrected for some significant amount of canopy or vegetation
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 NH3 from soil, based on "unsophisticated physical chem-
istry 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 evaporation, 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 NHi yr-1 (27 Tg
N yr'1) from the Northern Hemisphere and 14 Tg NH3 yr~* 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, accompanied by excessively high temperatures, which also
are not conducive to high NH3 emission. NH3 emissions are modeled
as having a maximum emissions rate in a temperature range from about 18
to 24 C. These model calculations agree well with the latitudinal
2-36
-------
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. The total amounts of NH3 on a global basis 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
day"1 (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 Demead et al. (1974) is larger than
Kim's (1973) estimated rate from ungrazed grass sod of 8 g N m~*
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 Gal bally (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 nr3 in temperate zones, 13 yg nr3 in
tropical areas, and 3 yg nr3 over oceanic areas. His resulting
2-37
-------
ammonia emissions estimate is 130 Tg N yr-1 for the Northern
Hemisphere. If this value were doubled to about 260 Tg N yr~* 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
m~3 whereas more recent data indicate a range from less than 0.7 yg
m~3 to around 1.4 yg nr3 (e.g., Braman and Shelly 1981). For
ocean areas Galbally used a value of 3.5 yg nr3 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~l for
the Northern Hemi sphere.
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 conclusions 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 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 m~3, and Junge
(1963) reported marine air concentrations of ammonia in Florida and
Hawaii of 5 yg nr3 and 2 yg m-3 respectively. In the Southern
Hemisphere (Tasmania), Ayers and Gras (1980) found that NH3 averaged
about 0.6 yg nr3 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.
Even with these 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 m~2
hr~l as ammonia. This converts to an annual emission density of about
0.0004 g nr2 yr"1 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, inorganic, and halogenated compounds in the marine atmosphere
and in particular a set of precursor compounds. His selection was based
on limiting consideration to those compounds that were potential natural
2-38
-------
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 modeling
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 yg nr3, probably significantly higher
than is now considered realistic. Thus, his estimated global ammonia
emission from the ocean of 3.2 Tg (NHa) yr-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 atmosphere, 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
atmospheric 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
m-2 yr-l t0 the contiguous United States (7.82 x 1012 m*) and to
the area east of the Mississippi River (2.23 x 1012 m2) results in
estimated biogenic ammonia emissions of 4.3 Tg N yr"l and 1.2 Tg N
yr"1, respectively. However, as noted above, NH3 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 Gal bally or other authors of
previous studies. A proportionate change in Galbally's estimate would
result in an indicated global emission rate of 13 Tg N yr~l 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 nr2 yr"l. Applying this emission value to the contiguous
United States (7.8 x IQl"2 m2) ancj the region east of the Mississippi
River (2.23 x 1012 m2) results in estimated annual NH3 emissions
of 1.1 Tg N yr~l and 0.3 Tg N yr-1, respectively.
This biogenic emission source can be compared to man-made 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
2-39
-------
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 nr2 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 NOX and Ammonia
Emissions—The natural 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 the 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 NOX and Ammonia—The previous
discussions have indicated that both dry and wet deposition processes
are important 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 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.
2-40
-------
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-1. This wide range is indicative of the
fact that little is known about these compounds. Because at least some
are not particulate 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 natural emissions 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 HNOa
after atmospheric reactions, may lower precipitation pH, while ammonia,
when absorbed into liquid drops as NHA+ 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 NOv 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-1. For NH3, 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 NOv 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-1.
2.2.3 Chlorine Compounds
2.2.3.1 Introduction—Part of the acidity of precipitation is
contributed oy cniorides. It is hypothesized by many investigators that
hydrochloric acid (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 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 CHjCl, volcanic emissions, and forest fires. The
sea salt processes will be shown to be dominant. This was also Cadle's
2-41
-------
(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 hydrochloric 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 potential 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~l 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 (1960) 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 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 £-i, 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
2-42
-------
TABLE 2-8. ATMOSPHERIC BACKGROUND CONCENTRATIONS
OF NATURAL CHLORINE COMPUNDS
Compound
Inorganic gaseous Cl"
Aerosol Cl"
CHaCl
Concentration
yg FIT 3
1.4-2.8
1-10
- 1.2
Reference
Cicerone (1981)
Cicerone (1981)
Rasmussen et al .
(1980)
2-43
-------
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 013 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 H2S04, produced originally in
the atmosphere from S02, 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 N02- As a result of these experiments, these authors
proposed a reaction system involving the hydrolysis of NO? 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 H2$04 and competing reactions with NH3 for
H2$04 in a mechanism that produces HC1 from the NaCl/H2$04
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~l. 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 10^ Tg yr~i or
550 Tg Cl yr-1, this 3 percent estimate gives an estimated gaseous Cl
production rate of 17 Tg yr"l. This lower value compared to the 200
to 600 Tg yr-1, quoted above for gaseous 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~l 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
2-44
-------
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~l 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 nr2 yr-1, averaged over the global
land area.
An additional natural source of atmospheric gaseous Cl2 or 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 m-3 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 CHsCl. Using this lifetime estimate with an average
concentration of 1.2 ug m-3 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 CH3C1 is presumed to be a biogenic process
(Rasmussen et al. 1980) while the formation of HC1 can result from
reactions involving CH3d 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 detailed 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 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
2-45
-------
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 Volcanism—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 emission 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 $042-^ Although flux values were
not calculated, one may infer from this and the S02 and S042" data
of Hobbs et al. (1982) that Mt. St. Helens' Cl contributions to the
atmosphere would be less than 0.15 Tg yr~l. 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 Cadle1s 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
combustion 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 significant 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 CHsCl, 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
2-46
-------
is indicated for combustion sources. This is a minor global source of
Cl and 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 natural chlorine sources are listed in terms of
global totals and in terms of the estimated deposition on land areas.
As indicated, 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 population of chlorine, 65 Tg Cl
yr-1, averages to about 0.4 g Cl nr2 yr-1 if 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 (I960)
findings that, away from coastal areas, chloride in precipitation is
generally 0.5 g nr2 yr'l 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
do 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 to land 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 percent or so of the total
anion content in precipitation at stations in the northeastern United
States. As such they would be relatively unimportant in altering
precipitation pH by themselves.
2-47
-------
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 CH3C1
Volcanos
Combustion CH3C1
Total
or approximately
Global
emission
550
17
2
10.0
2.0
581
580
Land
Depositions
55
5
0.5
3.0
2.0
65.5
65
text for details
2-48
-------
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 circumstances. 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 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 exceptional 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 Hi-Vol 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 operation. 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.,
Delaware, to a low of 18 yg nr3 in Cops Co., New Hampshire. This
average, 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 particles 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 folowing
calculation:
36 yg m-3 x 500 m x 365 = 6.6 g nr2 yr'1.
Applying this annual emission density rate of 6.6 g m-2 yr-1 to the
United States east of the Mississippi River, about 2 x 1012 m^,
gives an estimated emission of dust into the nonurban atmosphere of
about 13 Tg yr-1 or 13 x 106 mT yr-1.
Of the total natural dust loading in the atmosphere, probably the
most important constituents for precipitation chemistry is its calcium
and magnesium content (Stensland and Semonin 1982). These elements make
2-49
-------
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 composition 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 iag m~3
for an estimated average concentration and 0.5 Tg yr~l for an
estimated annual emission. For Mg the estimated values are: 0.8 yg
m-3 for the average concentration and 0.3 Tg yr~l for the annual
emi ssion.
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 Ca+2 and Mg+2 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 emissions is an important consideration for acidic deposition
situations. We will examine briefly some of the aspects of natural pH
variations in this section. Since 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 C02 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 COjj-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 has not been a considerable amount of precipita-
tion pH data from locations that could not have been influenced by
anthropogenic pollutant sources, this assumption of a ^-equilibrated
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
2-50
-------
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 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 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. Since 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 background situations
have been summarized by Sze and Ko (1980), and they conclude that
5042- concentrations in very remote, clean areas can be about 0.05
yg nr3. This very low S042~ concentration with a background
S02 value of 0.26 yg nr3 and 0.61 yg nr3 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 $042- would
lead to lower pH values, as would lower cloud water content. Situations
where HNOs 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 bogenic NH3 and dust
containing CaC03 could be expected, a higher pH than 5.4 might be
expected if the S042~ were as low as 0.05 yg nr3 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 H2S04 and HMh, and weak, probably organic,
acids. The higher acidities were primarily due to H2S04- Especially
2-51
-------
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°47'S-77°3rE) in the remote Southern
Indian Ocean, the average pH for 26 rainfall events was 4.9. Galloway
et al. (1982) speculated that some 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
contributions of the three acids, H?S04, HNOa, 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 Katherine, 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 determined 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), however, 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 Gustafson 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 CaCOa content in
soil dust could be an explanation.
2-52
-------
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 ^04; Galloway et al. (1982)
believe it was an organic acid. Believe it was an organic acid.
2-53
-------
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 drought, excessive amounts of soil dust containing alkaline salts
were present in the precipitation 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.
2.2.6 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 anthropogenic sources totaled about 11 Tg S yr~l (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 NH3
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
2-54
-------
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 NH^+ ions in precipitation, have
been estimated to be about 0.3 Tg N yr-1 for the whole United States.
Anthropogenic sources of NH$ include significant contributions from
domestic animal waste and other sources and have been estimated to be
about 3 Tg N yr~l 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% of the total acidity. 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 m-2 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. 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.
2.3 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 agricultural, aquatic,
and materials effects (see Chapters E-3, E-5, and E-7).
In addition to S02 and NO, other fossil fuel combustion products
are emitted that may influence acid precipitation formation. These
include H2$04, HC1, and particulate matter. Sulfuric acid
represents a variable fraction (0.01 to about 0.05) of the S02
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 H2SOd 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 HC1 emissions to the
atmosphere. Figure 2-4 illustrates trends in total anthropogenic
emissions of particulate matter, S02, 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
2-55
-------
ro
i
en
en
1940
1980
Figure 2-4. Total emissions of participate matter, S0
Adapted from U.S. EPA (1978).
, and NO for the United States from 1940 to 1978
x
-------
increased many-fold, a switch in fuel from coal to oil in the north-
eastern United States during the late 1960's and early 1970's has
lowered both total SO? 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 or 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.
(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, subclass!fication 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
2-57
-------
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 Stastes. 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 (^$04), chloride (HCI), volatile metals
(As, Hg), and certain key metals indigenous to residual oil combustion
(V, N1).
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. 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 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
2-58
-------
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
2-59
-------
-------
2.5 -
1950
LEGEND
MISCELLANEOUS
HIGHWAY VEHICLES
COMMERCIAL/RESIDENTIAL
INDUSTRIAL
ELECTRIC UTILITIES
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).
2-61
409-261 0 - 83 - <4
-------
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.
If electric utilities are contributing 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 have dominated 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 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 SO? 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
2-62
-------
TABLE 2-12. TEN LARGEST INCREASES IN SULFUR OXIDES EMISSION
DENSITIES BETWEEN 1950 and 1978
SOv
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
2-63
-------
TABLE 2-13. ANNUAL EMISSIONS DENSITIES OF SULFUR OXIDES
(kg knr2 yr-1)
1950
1960
1970
1978
Alabama
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
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).
2-64
-------
TABLE 2-14. ESTIMATES OF ANNUAL EMISSIONS OF SULFUR OXIDES
(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
r r
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
322.2
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
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
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
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
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
1031.9
10784.1 18852.6 23492.1 21741.6 21435.6
2-65
-------
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
emission 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 dioxides 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 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.
To estimate long-term trends in primary sulfate emissions
characteristics, historical sulfur oxides emissions estimates summarized
in Figure 2-6 were adjusted by an 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
2-66
-------
TABLE 2-15. SULFATE EMISSIONS FACTORS FOR SOURCE CATEGORIES
AND FUELDS (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
2-67
-------
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 S02 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
sul fates 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 northeast United States and/or local
low-level emissions sources. Similarly, the low-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,
2-68
-------
TABLE 2-16. ANNUAL EMISSIONS DENSITIES OF PRIMARY SULFATE
(kg km-2 yr~l)
Al abama
Arkansas
Connecticut
Del aware
District of Columbia
Florida
Georgi a
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
39
15
595 (6)
952 (5)
6419
80
32
164 (9)
154
28
28
94
25
982 (4)
2924 (2)
115
45
17
158
163 (10)
3260 (1)
292 (8)
50
210
302 (7)
1589 (3)
28
35
65
8
77
83
38
1960
89
8
649 (5)
1371 (1)
10079
111
39
332 (10)
333 (9)
41
91
79
35
428 (8)
962 (4)
158
67
15
69
56
1008 (3)
380
44
451 (6)
431 (7)
1090 (2)
44
118
57
14
53
143
82
1970
130
14
1307 (5)
1544 (2)
32608
230
73
354 (10)
344
47
179
118
62
551 (7)
1952 (1)
193
82
24
133
141
1507 (4)
555 (6)
84
470 (9)
482 (8)
1535 (3)
62
156
75
23
190
319
51
1978
116
45
590 (4)
1584 (1)
6366
214
104
255
367 (10)
50
193
151
50
494 (5)
1298 (2)
168
76
106
133
110
878 (3)
459 (8)
98
485 (7)
387 (9)
494 (6)
111
182
72
23
151
261
83
Note: Numbers in parentheses indicate numerical ranking of 10 highest
emissions densities (D.C. excluded).
2-69
-------
TABLE 2-17. ESTIMATES OF ANNUAL EMISSIONS OF PRIMARY SULFATE
(106 kg yr'1)
A1 abama
Arkansas
Connecticut
Delaware
District of Columbia
Florida
Georgia
Illinois
Indiana
Iowa
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Mi nnesota
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
491.5
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
506.4
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
704.6
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
643.0
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
496.1
2-70
-------
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.
1-1 State Sources Emissions Sources Emissions Sources Emissions Sources Emissions Sources Emissions
Ohio 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
-------
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 is estimated to be emitted at heights between 31 and 76 m.
In Florida, 55 percent of the sulfur oxide emissions from point sources
is 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. The northeastern Atlantic
coastal states, Ohio, and Pennsylvania 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.
2-72
-------
TABLE 2-19. ESTIMATED S02 AND PRIMARY SULFATE EMISSIONS FOR 1980
FROM UTILITY SOURCES IN FLORIDA AND OHIO
(106 kg yr-l)
Stack Height No. of S02 Sulfate
m point sources emissions emissions
Florida
0-30
31-76
77-152
153-305
12
23
30
0
99.9
107.9
469.6
0
3.7
4.0
17.5
0
Ohio 0-30 3 4.5 0.1
31-76 17 48.5 0.5
77-152 35 441.4 4.8
153-305 48 1722.8 18.6
2-73
-------
TABLE 2-20. ANNUAL EMISSIONS DENSITIES OF NITROGEN OXIDES
(kg knr2 yr'1)
1950
1960
1970
1978
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
Pennsyl vania
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).
2-74
-------
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 emissions 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
representation 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.
Recent chemical analyses 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 rainwater acidity through
both primary emissions and atmospheric transformations.
2.3.4 Historical Trends and Current Emissions of Hydrochloric Acid
(HC1) ~~~~
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 (Stahl 1969). Chlorine is present in coals in the form
of i-norganic 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 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).
2-75
-------
TABLE 2-21. ESTIMATES OF ANNUAL EMISSIONS OF NITROGEN OXIDES
(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
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
6386.0
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
10817.2
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
15299.6
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
17609.4
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
15059.7
2-76
-------
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
2-77
-------
CD
O
t—I
O
I—I
oo
O
I—I
GO
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).
2-78
-------
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 aalt 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 distribution 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 Combustion
As with calculated emission 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 the 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 for combustion 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 readibly available.
2-79
-------
TABLE 2-23. ANNUAL EMISSIONS DENSITIES OF CHLORIDE
(kg km-2 yr-1)
1950 1960 1970 1978
Al abama
Arkansas
Connecticut
Del aware
District of Columbia
Florida
Georgia
Illinois
Indiana
Iowa
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Mi nnesota
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).
2-80
-------
TABLE 2-24. ARSENIC, MERCURY, AND VANADIUM CONTENT OF
BITUMINOUS COAL
State
Alabama
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 Gloshbter et al. 1977,
2-81
-------
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 SOg, NOX, and primary sulfate to evaluate the
transport and deposition of emissions. As tracers, the SOv/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 SOJ2. The presence of vanadium in
atmospheric aerosols could be used in conjunction with meteorological
measurements to estimate the regional origins of the air mass contaning
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 hydrodesul furization. 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 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.
2-82
-------
TABLE 2-25. ANNUAL EMISSIONS DENSITIES OF ARSENIC
(kg kirr 2 yr~l)
Al abama
Arkansas
Connecticut
Del awre
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
Pennsyl vania
Rhode Island
South Carolina
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
2-83
-------
TABLE 2-26. ANNUAL EMISSIONS DENSITIES OF MERCURY
(kg km-2 yr-1)
Al abama
Arkansas
Connecticut
Del awre
District of Columbia
Florida
Georgia
Illinois
Indiana
Iowa
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Mi ssouri
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
2-84
-------
Estimates of vanadium emissions from coal combustion pose an
additional problem in that various levels of particulate emissions
controls have been 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 calcula-
tions, 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 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
2-85
-------
TABLE 2-27. ANNUAL EMISSIONS DENSITIES OF VANADIUM
(kg km-2 yp-1) *
Al abama
Arkansas
Connecticut
Del awre
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
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
2-86
-------
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
Suestions 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 SO? and NOx
for the years 1955, 1965, and 1976 as a contribution to trie 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 for each of the years
1955, 1965, and 1976 were given in Table 2-28. Total SO? emissions in
Canada were approximately 5.3 million metric tons for 19/6, 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 SO? 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
2-87
-------
ro
i
00
00
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 smel tersb
Power plants
Other combustion0
Transportation
Iron ore processing
Others
TOTAL
S02
2,887,420
56,246
1,210,108
83,474
109,732
189,876
4,536,856
1955
N0xa
-
10,335
227,837
323,785
-
68,065
630,022
1965
SO?
3,901,950
261,837
1,129,548
48,669
155,832
1,095,341
6,593,177
N0xa
-
57,402
247,323
511,868
-
33,778
850,371
1976
SOa
2,604,637
614,323
884,867
77,793 1
175,829
954,215
5,311,664 1
N0xa
-
206,454
445,315
,017,936
-
190,327
,860,032
aNOx expressed as N02-
blncludes emissions from pyrrhotlte roasting operations.
clncludes 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 km-2 yr~l)
Province
Year
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
2-89
-------
TABLE 2-30.
ESTIMATED OF ANNUAL EMISSIONS DENSITIES OF PRIMARY
SULFATES FOR 1978 (VENA 1982)
(kg km-2 yr-1)
Province
Total S04
(103 kg)
Density
Newfoundland
Prince Edward Island
Nova Scotia
New Brunswick
Quebec
Ontario
Manitoba
Saskatchewan
Al berta
British Columbia
Yukon & N.W.T
4,081
435
12,320
12,582
53,452
45,714
13,217
3,742
7,321
33,380
213
11
77
233
176
39
50
24
7
12
37
< 1
2-90
-------
western Canada and can be explained by 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 SO?, 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 S02 and MOX data are from Environment
Canada 1980 files and the Canadian primary sulfate data represents 1978
emissions calculated by Vena (1982). The area of highest S02
emissions in the United States is bound by Pennsylvania on the east and
Missouri on the west. Highest Canadian provincial S02 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 SO^ 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.
2-91
-------
TABLE 2-31. ESTIMATES OF ANNUAL EMISSION DENSITIES OF
NITROGEN OXIDES (VENA 1982)
(kg km-2 yr-1)
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
2-92
-------
S 48 106 kg/yr.
> 48 < 250 106 kg/yr.
> 250 < 1015 106 kg/yr.
106 kg/yr.
Figure 2-8. Annual emissions of SOp by state.
Emissions Data System 1980.
Data are from National
2-93
409-261 0-83-5
-------
2-94
-------
Figure 2-10. Annual emissions of NO by state. Data are from National
Emissions Data System i960.
2-95
-------
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. The electricity
demand growth rate is estimated to be 1.5 percent per year from 1981 to
1985 and about 2.7 percent per year from 1985 to 2000. 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 38 (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 9f 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 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
2-96
-------
TABLE 2-32. NATIONAL U.S. CURRENT AND PROJECTED S02 AND NOX
EMISSIONS (Tg yr'1)
Source category
1.
2.
3.
4.
5.
Electric utilities
Industrial boilers and
process heaters
Nonferrous smelters
Residential/commercial
Other industrial
processes
6. Transportation
TOTALS
Current
1980
S02 NOX
15.0 5.6
2.4 3.5
1.4
0.8 0.7
2.9 0.7
0.8 8.5
24.1 19.0
Projected
1990
S02
15.9
3.4
0.5
1.0
1.2
0.8
22.8
NOX
7.2
3.0
0.7
0.8
7.8
19.5
Projected
2000
S02
16.2
6.5
0.5
0.9
1.5
1.0
26.6
NOX
8.7
4.0
0.6
1.1
9.7
24.1
Summarized from: U.S./Canada Work Group 3B Draft Report (1982).
2-97
-------
emitted to the atmosphere as SC^. 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), 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 System by the 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
S02 and NOX for all area and point source categories. The inventory
will be 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
2-98
-------
TABLE 2-33. NATIONAL CANADIAN CURRENT AND PROJECTED S0£ 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
SO? 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).
2-99
-------
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 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 disegregation 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 northeast 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 for the atmosphere
was coal combustion, much of the sulfur was precipitated to the land
near the combustion source in particulate form as neutralized salts.
2-100
-------
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, 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
S02. Newman (1975) has also pointed out the inability of coal fly ash
to neutralize S02 further in the atmosphere.
Therefore, from available data, we could conclude that the roles of
SO?, 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 precipitators. Even if one could demonstrate a minimal
effect of further reaction of combustion particles with S02 at
atmospheric concentrations, asserting that eliminating all particulate
controls would enhance neutralization of the atmosphere is misleading.
The absence of controls would result in a continual 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 carbonaceous 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 mechanical collectors if the power plant was
of the type converted from coal to oil in the mid-19601s. The mean
aerodynamic particle diameter of oil ash has been measured as 3 vim,
with 30 percent weight of the ash sized less than 0.5 pm (Boldt et al.
2-101
-------
ro
i
o
ro
TABLE 2-34. APPROXIMATE LIMITS OF FLY ASH COMPOSITION FOR VARIOUS COALS
(6LOSKOTER ET AL. 1977)
Chemical analysis, weight-percent of ash
A1203 F2203 Ti02
P205
CaO
MgO
Na20 K20
British coal
S03
American coals
Anthracite
B 1 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
0
0
.0-2
.5-4
.6-2
.0-0.8
0.1-4
0.0-3
0.0-3
0.0-1
0
0
2
12
.2-4
.7-36
.2-52
.4-52
0.2-1
0.1-4
0.5-8
2.8-14
0
0.2-3.0 0.2-4 0
3
0.2-28 0.1-1.3 8
.1-1
.1-32
.0-16
.3-32
Brituminous
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
B i tun i nous
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
-------
TABLE 2-35. ANALYSIS OF A TYPICAL RESIDUAL OIL ASH
(Boldt et al. 1980)
Oil Ash Constituents
Water-soluble components
S042-
Cl-
NH4+
N03_
Metals
V
Na
Mg
Ni
Fe
K
Mn
Carbon
C
Mean
(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
deviation
(%)
9.1
1.5
0.5
0.03
1.2
1.5
1.1
0.3
0.2
0.1
0.01
6.3
2-103
-------
TABLE 2-36.
SULFURIC ACID AND SULFATE CONTENT IN PARTICIPATE MATTER COLLECTED FROM COAL- AND
OIL-FIRED BOILERS (HOMOLYA AND FORTUNE 1978)
no
i
Source of ash
A. Coal
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
B. Oil-
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
-fired boilers:
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
fired boilers:
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
Collection Sulfur content
site Wt %
ESP
Stack
ESP
ESP
ESP
ESP
ESP
ESP
ESP
ESP
Stack
Stack
Stack
Stack
Stack
Cyclone
Cyclone
Cyclone
Cyclone
Cyclone
Air heater
Air heater
ESP
1.7
1.7
2.0
3.9
3.9
3.9
3.9
3.9
1.7
0.5
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
Ash composition (dry basis)
Wt % H2S04
0.06
0.08
0.02
0.04
0.07
0.03
0.01
0.02
0.02
0.01
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
Wt %
total sulfur
0.41
0.97
0.20
1.06
4.96
1.31
1.44
0.79
0.90
0.42
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 preclpitator.
-------
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 $03 and NOX occur.
Therefore, it is conceivable that the sulfuric acid fraction of acidic
precipitation consists of a mixture of primary (particles and condensed
H2S04 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.
2.4 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:
0 Natural sources of sulfur compounds are insignificant
contributors to 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, natural sources of sulfur
compounds are estimated to total about 0.07 Tg S yr~l. 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).
° Natural emissions of nitrogen oxides (NOX) are 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).
0 NOX from natural sources in the area east of the Mississippi
River have been estimated to be in the range of 0.04 to 0.7 Tg N
yr"1 with values from the lower part of the range being the
more recent 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).
2-105
-------
Ammonia, when incorporated into precipitation, tends to
counterbalance 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).
Biogenic sources of ammonium compounds in the area east of
Mississippi River are estimated to be about 0.3 Tg N yr"S
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).
o 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 totally—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 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 S02 emissions occur in the Midwest.
0 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.
2-106
-------
Changes in fuels from coal to oil reduced emissions in New
England by 20 percent.
Estimates of Canadian $03 emissions indicate a 20 percent
increase from 1955 to 1976 (Section 2.3.6).
° Copper and nickel smelters represent the major Canadian
source category, with most point sources located in eastern
Canada.
(2) Primary Sulfate (Section 2.3.2.2)
0 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.
° 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.
0 Primary sulfate emissions increased in the Midwest in proportion
to increases in coal consumption.
(3) Nitrogen Oxides (Section 2.3.3)
° 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.
o 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.
° 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 sul fated ash, their relative
contribution to acidity is unclear.
2-107
-------
(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
0 Implementing particulate controls reduced arsenic emissions in
the eastern United States, but mercury emissions increased in
proportion to coal consumption.
0 Vanadium is emitted from residual oil combustion in varying
amounts.
o Highest vanadium emissions occur in the northeastern United
States.
2-108
-------
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. Control 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.
Tech. 15:1493-1498.
Altshuller, A. P. 1958. Natural sources of gaseous pollutants in the
atmosphere. Tellus 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. Control
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. J^n 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.
2-109
-------
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, North Carolina.
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. of Geophys. and 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
sul fate emissions from the combustion of fossil fuels. EPA-600/9-
78-020a. U. S. Environmental Protection Agency, Research Triangle Park,
North Carolina, pp. 53-63.
Cicerone, R. J. 1981. Halogens in the atmosphere. Rev. Geophys. and
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. Sandal Is. 1974. The photooxidation of hydrogen
sul fide and dimethyl sul fide in air. Atmos. Environ. 8:1269-1281.
2-110
-------
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.
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.
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. Tellus 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.
2-111
-------
Friedlander, S. K. 1981. New developments in receptor modeling theory,
pp. 1-20. In atmospheric Aerosol, Source/Air Quality Relationships. E.
S. Macias ancl P. K. Hopke, eds. ACS Symposium Series No. 167, American
Chemical Society, Washington, D.C. 359 p.
Friend, J. F. 1973. The global sulfur cycle, pp. 177-201. In Chemistry
of the Lower Atmosphere. S. I. Rasool, ed. Plenum Press, New York.
Galbally, I. E. 1975. Emission of oxides of nitrogen (NOX) and
ammonia from the earth's surface. Tellus 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 SOg 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, North Carolina.
Graedel, T. E. 1977. The homogeneous chemistry of atmospheric sulfur.
Rev. Geophys. and 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.
2-112
-------
Granat, L., R. 0. Hallberg, and H. Rodhe. 1976. The global sulfur
cycle, pp. 89-134. Ln Nitrogen, Phosphorus and Sulfur - Global Cycles.
B. H. Svensson and R. Soderlund, eds. Ecological Bulletins No. 22,
SCOPE Report 7, Royal Swedish Academy of Sciences, Stockholm.
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 teh 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 Meterological Society, Boston, Mass.
Hidy, G. M. 1982. Bridging the gap between air quality and
precipitation chemistry. Water, Air, and Soil Pollut. 18:191-198.
Hill, R. D., R. G. Rinker, and H. D. Wilson. 1980. Atmospheric
nitrogen fixation by lightning. J. Atnos. 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. Control 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
sul f uric acid and sul fate 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,
North Carolina, 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 Pollution Control 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.
2-113
-------
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. Georg11, H. Claude, and H. Malewskl. 1978.
Contribution
Geophys. 116
Contributions of H?S to the atmospheric sulfur cycle. Pure and Appl
:465-475.
Johannes, A. H., E. R. Altwlcker, and N. L. Clescerl. 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 1n
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. Meterol. 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. Meterol. 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. Fllbery, and G. P. Marx. 1979. Effect of
temperature on the gaseous nitrogen products of dentrification in a silt
loam soil. Soil Scl. 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.
2-114
-------
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.
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, G. S., G. E. Gordon, and S. W. Rheingrover. 1982.
Identification of atmospheric particulate sources in Washington, D.C.,
using chemical element balances. Environ. Sci. Tech. 16:79-90.
Lawson, D. R. and J. W. Winchester. 1979. A standard crustal aerosol
as a reference for elemental enrichment factors. Atnos. 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. 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, and 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.
2-115
-------
Logan, J. A. 1982. Nitrogen oxides in the troposphere: Global and
regional budgets. Advances in Chemistry: 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.
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. Sul fate 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. T7~J. Kneip and P. J.
Lioy, eds. Ann. NY Acad. Sci. 338-145-173.
2-116
-------
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 6. L. Hutchinson. 1972. Air
containing nitrogen-15 ammonia: Foliar absorption by corn seedlings.
Science 175:759-761.
Rasmussen, R. A., L. E. Rasmussen, M. A. Khalil, 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. In 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, and 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. Meterol.
16:53-56.
Robinson, E. and R. C. Robbins. 1970a. Gaseous sulfur pollutants from
urban and natural sources. J. Air Pollut. Control Assoc. 20:233-235.
Robinson, E. and R. C. Robbins. 1970b. Gaseous nitrogen compound
pollutants from urban and natural sources. J. Air Pollut. Control
Assoc. 20:303-306.
Rodhe, U. 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.
2-117
-------
Ryan, J. A. and N. R. Mukherjee. 1975. Sources of stratospheric
gaseous chlorine. Rev. of Geophys. and Space Phys. 13:650-658.
Shannon, J. D. 1979. The advanced statistical trajectory regional air
pollution model. Agronne 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 sul fate. EPA Project Report
IAG-AD-89-F-1-116-0.
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, Phosphrous and Sulfur-Global Cycles. SCOPE Report
7, Ecolo"gTcal 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. Meterol. 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, C$2, CH3,
SCH3, and h^S: 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.
2-118
-------
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 Work Group 3B. 1982. Emissions, costs, and engineering
assessment. Final Report. June 15, 1982.
Valach, R. 1967. The origin of the gaseous form of natural atmospheric
chlorine. Tellus 19:509-516.
Vena, F. 1982. Environment Canada, Persona 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.
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. Ir± International Conference on
Environmental Sensing and Assessment, Inst. of 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, and 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.
2-119
-------
THE ACIDIC DEPOSITION PHENOMENON AND ITS EFFECTS
A-3. TRANSPORT PROCESSES
3.1 INTRODUCTION (N. V. Glllanl)
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 transformations 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 atmospheric residence time. Transport processes
exert a major, or possibly even a controlling, influence on T and the
trajectories.
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
3-1
-------
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 deposition 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 speciation. In this chapter, the nature and significance of
the role of transport processes are explored specifically for S02
emissions, partly because S02 ^s 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 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 governed by spatial and temporal
inhomogeneities in the flow field. The dispersive capacity of the
transport layer 1s 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.
3-2
-------
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, atmospheric motions of progressively larger scales influence
its further dispersion. 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 microscale phenomena include atmospheric
turbulence.1 The meteorological mesoscale 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 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 discussion of pollutant transport
processes is divided into mesoscale transport (Section 3.4) and
continental (synoptic) and hemispheric transport (Section 3.5). The
^Atmospheric turbulence is sometimes interpreted broadly to include
vortex motions over all meteorological scales. Our use of the term
is more specific, 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 turbulence and microscale turbulence synonymously.
3-3
-------
term "long-range transport" commonly refers to transport over the
synoptic and hemispherical scales.
3.2.2 Atmospheric Motions
The energy which 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 equitoral regions would heat up. In a
fluid as mobile as air, temperature differences will immediately give
rise to currents that tend to equalize them. 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 notion 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 latitudes. The Coriolis force also
increases with wind speed, and its effect is to cause the wind to turn
to the right (in the northern hemsphere) 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 isobars).
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 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.
3-4
-------
500 - 1000 m ft»
GEOSTROPHIC WIND
Figure 3-1. The Ekman spiral of wind with height in the northern
hemisphere. Adapted from Barry and Chorley (1977).
3-5
409-261 0-83-6
-------
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 Coriolis force. This planetary flow is approximately
geostrophic horizontally. Vertically, a weak pressure gradient force
(pressure decreases with height) is nearly balanced by the gravitational
(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 (southwesterlies in the northern
hemisphere); near the surface, the flow is towards the equator
(northeaster!ies). 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.
Anticyclones 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 flew within the
cell, particularly in the core, but frequently they are also
characterized by weak or stagnant motion. When an anticyclone stagnates
for multi-day 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 km? for more than 36 hours) in the eastern United
States is presented 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 discussion 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 and cold Canadian 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.
3-6
-------
Figure 3-2. Climatology of air stagnation adversions issued over a ten
year period. Adapted form Lyons (1975).
3-7
-------
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. (R)r 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 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 dominated 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
of the order of 1 cm s-1, vertical speeds in local mesoscale systems
are typically on the order of 1 m s-i, and may even exceed 10 m s'1
in strong updrafts, especially in thunderstorms (Panofsky 1982).
Mesoscale complex flows may be terrain-induced or synopticany-
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 mechanical turbulence may sometimes also occur high in the
upper atmosphere in locally strong wind shear associated with frontal
zones (see, for example, Panofsky 1982). This type of clear-air
3-8
-------
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
clouds and precipitation are made of. 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 distribute 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
interactions 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
3-9
-------
beginning to be modeled realistically and explicitly, and has not
progressed to the point of formulating reliable parameter!zations.
Field data pertinent to mesoscale motions are very limited. Routine
monitoring of upper air winds is confined to a sparse spatial network
(stations being separated, on the average, by well over 300 km), and the
temporal resolution of the measurements is also coarse (typically at
12-hourly intervals). Such monitoring is adequate for the
reconstruction 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, particularly 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, 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. Since 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
of elevated emissions, or 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.
Since 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.
3-10
-------
3.3 POLLUTANT TRANSPORT LAYER: ITS STRUCTURE AND DYNAMICS (N. V.
Gnianl)
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, including 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 momentum field (winds), and the moisture field]
are perturbed significantly as a direct consequence of the earth's
surface. In air pollution meteorology, 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 vertically 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 thunderstorms, 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.
3-11
-------
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 transported over long
distances while remaining decoupled from the ground. Nighttime urban
releases within the shallow mixing layer 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
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 estimated 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 parameterization 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 WMO 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). Collectively, 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
3-12
-------
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
originial references and are also included and in the PBL review
articles identified 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 the relatively quiescent 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.
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 rapid and greater temperature 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 characterized chiefly by small eddies.
Consequently, the dispersion within the surface layer is relatively much
slower than in the mixed layer, and dissipation 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
parameterization of the mean flow as well as its turbulent components
are well-established and, at least over smooth terrain under relatively
3-13
-------
stationary conditions, fairly reliable. Turbulent dispersion Is
parameterized In terms of an "eddy diffuslvlty," by analogy with the
concepts of molecular diffusion. Eddy "diffusion" Is on a relatively
larger scale, however, since 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 non-linearly 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.
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, rdry), 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 upwardso 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"
3-14
-------
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 SC>2 profile). Typically, no physical or stable boundaries
exist horizontally, and the turbulence structure is more homogeneous.
Turulent 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 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 transport 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.
3-15
-------
2000
: 1500
1000
500
•so.
-TEMPERATURE X
TURBULENCE (€1/3)
0
L
10
20 30
TEMP.(°C)
5
i
10 / 15
S02 (ppb)
i i
4 6
TURB. (cm2/3 s'1)
40
50
i
20
i
25
8
10
Figure 3-3. Vertical profiles of temperature, small scale turbulence,
and S02 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).
3-16
-------
2500
2000
o
1500
GO Q3
1000
500
0
-10
•ih
JANUARY
DAYTIME
ELEVATED
INVERSION
LAYER
UNSTABLE
-ih
ELEVATED
INVERSION LAYER
NEAR NEUTRAL
UNSTABLE
NOCTURNAL
SURFACED-BASED \
INVERSION LAYER
\^
10
AMBIENT TEMPERATURE (K)
20
30
Figure 3-4. Monthly-average diurnal and seasonal variations of the vertical thermal structure of the
PEL for a rural site near St. Louis, MO based on 1976 data.
-------
2000
1500
CD
i—i
LU
:r
e>
x
i—i
s:
LU
C3
CC
LU
3»
-------
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 connective 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 mid-afternoon. 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.
Late in the afternoon, when ground level insolation has diminished
considerably, 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. Consequently, 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 inversion 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. Particularly on evenings with clear skies and 1ight-to-moderate
winds, the surface inversion layer becomes extremely stable and strongly
suppresses vertical transport of mass, momemtum, or energy. The heat
flux is now downward owing to the inverted temperature profile.
Turbulence is inhibitied 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
3-19
-------
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 nocturnal 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 oscillation 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 wind speeds frequently become supergeostrophic in these
layers. The phenomenon 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:
0 There is a nearly three-fold increase in the free stream wind
speed (at 2 km, say) from summer ( ~6 m s*1) to winter (~
18 m S"l). Wind speeds are correspondingly greater in winter
in the boundary layer below.
3-20
-------
o
o
ca
2500
2000
1500
ST. LOUIS 1976
o 1000
500
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
-------
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 particular, 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.
0 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 contrast, 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 average seasonal transport ranges. The
actual straight!ine 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.
The inertia! 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 degrees 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
3-22
-------
ST. LOUIS 1976
1000
500-
^
o
o
a:
UJ
o
CQ
SHEAR IN WIND DIRECTION (deg)
Relative to Wind Direction at Ground Level.
Figure 3-7. Monthly-average absolute change in wind direction with
height relative to wind direction at ground level. Data
are for July 1976 near St. Louis, MO.
3-23
-------
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 (Holton 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 speeds in the jets are frequently supergeostrophic. In
their analysis of ten 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 level 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, Nebraska) was
characterized by speeds of up to 36 m s-1 (130 km Ir1) 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
3-24
-------
s-1), and more pronounced in Texas (2 to 3 m s-1). Hering and
Borden (1962) observed the average amplitude based on six-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-i.
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
distributions 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. Bonner's
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 distributions) 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 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 both daily soundings at each
3-25
-------
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 (1978).
3-26
-------
ST. LOUIS
JANUARY 1976
o
o
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.
3-27
-------
site. For a comparison of average winds in Missouri and Ohio, the
annual average (1960-64) wind roses at 1000 m MSL for the Columiba, 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. Gill anil
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 characterized 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 that become stagnant over one area can produce devastating floods
such as the one in Johnston, 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 formation of cumulus clouds, like PBL growth, is related to vertical
convection (e.g., see Manton 1982). The top of the mixing layer is
3-28
-------
(a)
(b)
WARM
WARM
COLD
(c)
INVERSION
— TOP
— BASE
(d)
WIND,
— -/—. am n n n n n n n n H
RURAL
//// //// TV / / / / ///
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).
3-29
-------
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 of 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
down drafts around the clouds, large infusions of mid- and upper-
troposheric cold and clear air into the PBL, and surface mesoscale high
pressure regions. Such mesoscale vertical circulations were detailed by
Byers and Brahm (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 precipitation 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 the total removal, may have
been transported in thunderstorm updrafts to heights of 10 km or more.
3-30
-------
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 (1983a) 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 1973). 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 temporally by the existing upper air weather monitoring
network. Nor can the denser monitoring network of surface winds
adequately fill the gap, particularly 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
3-31
-------
flow systems. A number of mesoscale observational programs have 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
roughness, 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, pollutant 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 undergo 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
shoreline 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
3-32
-------
water surface (Figure 3-10c). Pollutants from elevated sources in such
cases may he transported over water over 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 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 temperature 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 corrective 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
3-33
-------
5432
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).
3-34
-------
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 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 cooling 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 state!lite imagery, estimated this
difference to be as high as 17 C for New York City--a value considerably
higher than those made using on surface-based air temperature
measurements. His explanation for the apparent discrepancy 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 observations,
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
3-35
-------
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 emissions 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 inland flow is typically about
1.0 m s~l 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
atmospheric 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).
3-36
-------
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 stratification (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 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 Fohn. Lilly and Zipser (1972)
observed wind gusts of about 50 m s-1 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
recirculating eddies represent localized stagnant flow. The atmospheric
residence time, dilution, and overall trajectory of pollutants in such
flows is significantly 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
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
3-37
409-261 0-83-7
-------
HYDRAULIC JUMP
tan'1 de/dz
CLOUDS
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).
3-38
-------
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 nocturnal 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).
3.4 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 S02 is
emitted from elevated sources, NO* 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 S02 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 than 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
3-39
-------
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
increase 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
concentration of central electrical power generating stations burning
fossil fuels, particularly coal. In a recent study of trends related to
power plant stack heights and $62 emissions in this region, Koerber
(1982) focused attention on power plants with a generating capacity
greater than 50 MWe, and located in a two county row on both sides of
the Ohio River in Illinois, Indiana, Ohio, Kentucky, W. 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
of the S02 emissions were from stacks lower than 100 m, most of the
remainder 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 $02
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, bo.ttom). 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
3-40
-------
IO
o
o
I—I
oo
CM
O
t/J
1950
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).
3-41
-------
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
elevated stable layer during winter. Such a plume has a
significant likelihood of remaining aloft over long-range
transport.
3-42
-------
00
I
CO
-------
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 excesss 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 mid-morning 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 hmir, plume touchdown
occurred on the ground, and ground removal of the pollutants by dry
deposition 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 mid-afternoon. 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
experienced 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
3-44
-------
1500
1000 -
CO
o
CO
«£
CD
UJ
3:
CUMBERLAND PLUME
AUGUST 23,1978
- (BNL and EMI DATA)
0800
1000
1200
TIME OF DAY
1400
1600
1800
i—i i-H DOWNWIND DISTANCE
80 km 110 km 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 1983a).
-------
accompanied by significant in-cloud chemistry (Gillani and Wilson 1983a,
b). 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
background, 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 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
3-46
-------
plume is diluted by the background air, and the conditions within it are
conducive to photochemically-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 stratification. 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, decoupled from the ground, until it is brought down in the
future, either by a deep enough mixing layer, or 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 problably 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 decoupled 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 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-1 (35 km
hr-1). 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.
Since 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
3-47
-------
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=clouds, T=top of mixing layer). Adapted from
Gillani and Wilson (1983a).
3-48
-------
AUG. 23, 1978
0940
0990
1020
79 Watt U 79 East
Indian Mound Rd.
U
Cook Rd.
79 Wnt U 79 East U
Indian Mound Rd. Cooper Creek
1030
1040
1130
1140
Watt
Indian Mound Rd.
79 fatf
U
Woodlawn Rd.
79 Bast
U 79 Wet t
LylevMood Rd.
1250CDT
1300
1640
79 East
X 79 West
Co. Line
X U X 79 East
79/120 79/120
3-49
-------
Figure 3-17. Lidar photographs depicting the diurnal variation of the
vertical cross-sectional structure of the Cumberland plume
on Aug 1978. All data were collected at the same distance
(about 35 km) downwind of the source (Uthe et al. 1980).
3-50
-------
ALTITUDE - km
Or
Or
OJ
i
en
ro
O
ro
O
OJ
O
CJI
2
m
I
o
p
fN3
O
00
>
o
O)
00
-------
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).
3-52
-------
ROUTE OF MOBILE LIDAR OBSERVATIONS OF THE
LABADIE PLUME ON 14/15 JULY 1976
Top
View
LOCAL TIME (CDT) —hours
1 50
2320
Missouri |—Missouri
100 340
43 59
DISTANCE (from Labadiel —km
East on US 40 -)
Side
View
East on US 70
85
4-
Cross-sectional View
at 90-100 km Downwind
'"» *>" 1376 US.NC THE SR,
3-53
-------
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 corresponding to
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 50 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 transport (< 100
km), the Johnsonville plume probably experienced considerable ground
removal, while the Cumberland plume was spared such losses. The
Johnsonville plume was also exposed to morning foq 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.
3-54
-------
1500 -
• 1000 -
o
a:
GO
cn
cn
AUGUST 27, 1978
(EMI DATA)
TIME OF DAY
DOWNWIND DISTANCE AT SAMPLING: 100 km 150 km
EMISSION SOURCE: CUM JHV
Figure 3-19.
The physical behavior of the emissions from the 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 (1983a).
-------
To investigate the diurnal and seasonal dynamics of plume
mixing-layer interactions, one must resort to a time-varying,
plume-transport-and-diffusion 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. (Detailed
description of the model will be included in another paper now under
preparation by Gillani.) The meteorological data used in the model
calculations 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 representative 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 rn 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 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,
3-56
-------
JANUARY
JULY
E
0-Q.O
CQ
^0.2
\
\
10600
PLUME
RELEASE
Figure 3-20.
\1800
\
\
I I
[0400
PLUME
RELEASE
1600
12 16 20 24
4
8 12
PLUME TRANSPORT TIME
(Hours after Plume Release)
16
20 24
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.
3-57
-------
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 simmer 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, transformation and ground ranoval modules are added to the
above plume model. Transformations 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) and Gillani
and Wilson (1983b). All transformation and removal rates are assumed to
be pseudo-first-order rates, include diurnal and seasonal variabilities,
and are based on St. Louis, MO, data for 1976. 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 mechanism) and about 0.4 percent hr-1 in January (mostly
by liquid-phase mechanism). Ground removal of S02 by dry deposition
is based on a diurnally varying deposition velocity, being 0.3 cm s~l
at night and peaking at 1.9 cm s-1 at noon in July, with corresponding
values of 0.15 cm s"l and 0.95 cm s~l in January. Deposition
velocity of sulfate is assumed to be constant (0.1 cm s"l) 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 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
3-58
-------
2000
PLUME DYNAMICS
"(Power Plant Plume)
JANUARY
o
a:
CD
o
ca
CO
•—I
UJ
3:
1000
060
PLUME RELEASE TIME
1200
00 06 12 18 24 06
TIME OF DAY
' •'>;.;'. MIXING-
12
-. -A- .'
18
24
100
oo
to
ti 50
Lu
O
PLUME SULFUR BUDGET
(Power Plant Plume)
JANUARY
Figure 3-21
DAY 1
•06 —
—18-
% AEROSOL
SULFUR FORMED!
PLUME
RELEASE
TIME
% GASEOUS SULFUR
REMAINING AIRBORNE
0
10
20
30
40
50
40
30
20
% SULFUR
DRY DEPOSITED
10
_
_L
DAY 2
12
18
24
30
36
42
48
HOURS AFTER PLUME RELEASE
(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.
3-59
-------
2000
o
o
LU
O
CQ
O
i—t
LU
1C
1000-
PLUME DYNAMICS
(Power Plant Plume)
PLUME
RELEASE
TIME
22DJ
18 24 06
TIME OF DAY
100
oo
00
00
LL.
O
O
DC
50
PLUME SULFUR BUDGET
JULY
Figure 3-22.
PLUME RELEASE TIME
04
DAY 1
% AEROSOL
SULFUR FORMED
% GASEOUS SULFUR
REMAINING AIRBORNE
% 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
(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 trans-port, in the
absence of wet deposition. Results are shown for the 0400
and 1600 hr plume releases.
3-60
-------
calculations. Ground removal is about 18 percent on each day in
January. In July, the ground loss is about 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 SO^ 1" 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 trajectories, 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 from 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 superstacks may remain airborne for longer periods.
Emissions in the coastal areas of the northeast, may experience
significant local shoreline recirculations, thereby reducing their
impact range over the land mass.
In winter, the atmospheric residence of sulfur is expected to be
significantly 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 conversion and removal parameters. While the reliability of the
absolute values of the results may be questioned, important and
3-61
-------
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 and wind speeds.
Consequently, these releases are entrained early in the day and
fumigated to ground at relatively high concentrations, leading to
substantial 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 depletion. 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 concentrations 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
area! 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 tnan
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 interaction 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.
3-62
-------
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 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
Monitoring network for the month of July 1976. The Sg data are
segregated by sectors pointing to three major local sources: the
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 shown
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, NQX 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, indicating 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 precipita
tion scavenging) 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. Liquid-phase
oxidation of S02 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
3-63
-------
GAS CONCENTRATION: Sg (yg m-3). N0 (ppb)
O
co
en
•—• &> 73 <-*•
rc 3 -;• 3-
C Q. < -J
10 fD fD
Oi <-*• -S fD
fD fD Ol
n- <-f -*><-••
Oi -1' O
O> —>3-J
—• —i fD
• I -S 10
l/i "< O
'-' c+ C
kO 01 O -s
^•J O O O
00 A- 3 fD
•—• T3 10
. -o —' ••
O fD
-S X
C"^ <~^
o> -—^r
co —«• (t>
CD ^
o o
O. —' fD
fD C 3
10 Q. rt
-•• -s
OO 3 Ol
O
C Ol O
X -••
C^c+
T3 O1*<
o o
£ Oi
IO -t:
fD O
LQ T
fD
Ol
i.lQ
C+ fD O
fD c
O. 10 3
c+ a.
O" Ol
* 23
S O c+
3 in O
CL • -5
CL
-•• -n -h
-s o o
fD -J -5
O
run c-
—*• pt J^
O IO —'
3 fD *<
O
l/» C I—'
fD IO >-O
O --J
rf 10 cn
O c •
-J —•
fD
CO
I
ro
to
?3 -*. —(
3> 3 3-
fD
r*- Q.
-S s: fD
T3 "O ••
—• O
01 s; c*
rf -S fD
(— —• O
O O> O
O 3 Q.
O c-l-
s: -i 3-
3"-. fD
O OO
Ol
• cf
T3 Oi
O r*
->• 3" ft"
3 fD -5
c+^< fD
fD
O. 01 Oi
<-r fD fD
o -s
Ol
fD
o.
1=
I S
O 3
C Ol
>. cr
fD
a>
o- o
Qt "1
fD O
Q. -h
O 10
3 C^
3 -h
o c
3 ~$
3- &
—' 3
•< Q.
01 3
fD ft
Oi O
fD fD
3
Oi O
OJ 3
o
O fD
-"^
rt T
3- 01
fD rf
—i*
O
3
O
3=
-------
2000
PLUME DYNAMICS
• (Urban Plume)
JANUARY
o
o:
O
OQ
1000
0600
1200
"N •.
•>;• PLUME RELEASE
V\ TIME
' •'•• X1"
.'o'i \ 1800 0000^
;••. V. MIXING.';:-
. .VrHEIGHT •
06 12 18 24 06
TIME OF DAY
12
18
24
lOOr-
o
i—i
oo
CO
u.
o
o
oc
50
PLUME SULFUR BUDGET
(Urban Plume)
JANUARY
\EROSOL
FORMATION
1800 .£... ^u"""'««'
-^1200.
PLUME RELEASE
TIME
1800'
-40
GROUND
LOSS
-a
m
3D
0
10
£<">
Si
30 ^<
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.
3-65
-------
2000
E
Q
O
01
CJ3
1000
O
3
m
i—i
UJ
PLUME DYNAMICS
- (Urban Plume)
JULY
- 0400
100
PLUME RELEASE TIME
' ' 1600
2200
06 12 18 24 06
TIME OF DAY
12
MIXING
HEIGHT
18
24
o
t-H
oo
o
LU
Q-
100 r—^—
\ •*•-.
• PLUME SULFUR BUDGET
(Urban Plume) ^^
50
DAY 1
2200
2200
.AEROSOL
FORMATION
1000
1000
.GROUND
LOSS
DAY 2
30
40
50
12
18
24
30
36
42
48
Figure 3-25.
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 co
50
40°
-o
30 §_
20 53
10
0
3-66
-------
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 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 emissions of acid precursors,
particularly NOX, are likely to be deposited within 500 km of the
source.
3-67
-------
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
(Sisterson 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 PRL 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 meteorological 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
3-68
-------
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 trajectories, since the sample size is much
larger, and can be extended further downstream in time. In addition,
the problem of estimating horizontal diffusion becomes easier because
over long-term regional scales, horizontal dispersion is due primarily
to the spread of plume or trajectory centerlines, 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 transformation 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 meteorological
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 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. The
importance of stack height to deposition patterns would, in general,
vary inversely with the source/receptor distance.
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).
3-69
409-261 0-83-8
-------
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 June 23 to July 7, 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 visibility, 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, since 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 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 mid-
morning 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 explain large regions of
haze emanating from just four representative source cities. The mixing
3-70
-------
JUNE 25, 1975
JUNE 27, 1975
JUNE 29, 1975
JULY 1, 1975
JULY 3, 1975
0 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).
3-71
-------
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.
3-72
-------
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. Unfortunately, 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~l over the first few days,
with 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 cl imatological examples of key meteorologi-
cal 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
3-73
-------
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.
3-74
-------
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
climatological 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 important as the seasonal averages in determining the net
transport of emissions. 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 parameterization 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 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 pollutant emitted in northeast Canada is more likely, less likely, or
equally likely than a pollutant in the southeastern U.S. 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.
3-75
-------
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.
3-76
-------
OJ
•-J
OURRTER 3,1977
FRBCTION
[Do.01-0.02
Ho. 02-0. Oil
§0.011-0.08
• >o.oe
Figure 3-30.
Statistics of hourly precipitation data during July-September of 1977. (a) Fration 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).
-------
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 deposition patterns would be
centered on the source location. Here, the deposition 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 distributed across the U.S. 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 observations beyond coastal regions
were available for model use; thus, the wet deposition maxima cannot
progress beyond the coast, although there is not 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 oxide emission inventory for the
U.S. and Canada and with meteorological 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 SO? concentra-
tions 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
3-78
-------
WET DEPOSITION
FIRST DRY DEPOSITION
DRY 'DEPOSITION
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
::>y~ y
SECOND DRY DEPOSITION
DRY'DEPOSIT ION
SECOND DRY DEPOSITION
Figure 3-32. Cumulative wet and dry total sulfur deposition patterns during the second day of transport.
-------
tep pjiqq. e^q. 6uunp stua^ed 110.14 isodap
LB;O; /Up pue
>ee-e
NOinsod3Q Aya ayiHi
\* ' ' "*
.-•"''-, "^ '; / /»—{ ""'':
'f-"".~-'~'^ -jf-—•' "'-, »'""•'•'•-._
*.
NOIlISOd3Q.
NOIlISOd3Q MQ Qy I HI
NOIlISOd3Q 13M
CO
oo
-------
00
ro
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
oo
co
WET DEPOSITION
SECOND DRY DEPOSITION
DRY- DEPOSITION
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.
-------
CO
I
00
-p-
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
i
CO
en
SULFUR DIOXIDE
SUMMER RVDWGC
TROM SOURCES HITHIN'500 KM
pG/CUBIC MCTCR
MflX - 42.2
SULFUR DIOXIDE
SUMMtR HVERHGe COtEHTRRTION '--
TRW1 SOURCES BCtONO~500 KM
M6/CU81C MCTER
fWX - 3.36
Figure 3-37.
Contribution to average summer SO;? concentrations resulting from U.S. and Canadian
anthropogenic sulfur sources within 500 km and from sources beyond 500 km.
-------
CXI
SULFflTE
SUMMER HVCRflGE CWJCENlltflTIW
FROM SOURCES HITHIN'SOO KM
pG/CUBIC METER
MRX - 10.I
SULFRTE
SUMMER flVERflSE COHCENTRRTION '
FROM SOURCES BETOND'SOO KM
pG/CUBIC METER
MflX - 4.73
Figure 3-38.
Contribution to average sulfate concentration resulting from U'.S. and Canadian anthropogenic
sulfur sources within 500km and from sources beyond 500 km.
-------
00
DRY DEPOSITION
SUMMER RCCUMULRTlON /
PROM SOURCES WITHIN'500 KM
KG SULFUR/HECTflRE
MflX - 6.83
DRY DEPOSITION
.-,
SUMMER HCCUMULOTIdM /'
rROM SOURCES BCYOMTSOO KM
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
CXI
WET DEPOSITION
SUMMER RCCUMULflTldN
rROM SOURCES WITHIN'SOO KM
KG SULruR/HECTflRE
MRX - ^.97
WET DEPOSITION
SUMMER FCCUMULHTI0N /
TROM SOURCES BETONO-500 KM
KG SULfUR/HECTflRE
MflX - 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.
-------
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. 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 simulations of the transport and deposition of all
anthropogenic sulfur emissions from the U.S. and Canada with the ASTRAP
model gave a continental budget of 28 to 32 percent dry deposition over
land, 13 to 31 percent wet deposition over land, and 37 to 54 percent
transport out to sea. The higher deposition percentages occurred during
summer; the lower values were calculated for winter. The annual totals
were 29 percent dry deposition, 24 percent wet deposition, and 47
percent transport. Wet deposition is the most variable of the three
terms, because of periodic droughts or rainy periods. Rigorously
determined confidence limits cannot be placed on the simulation results,
since 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 continent, occasionally 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),
3-89
-------
but in part the concentrations result from production/destruction
activities in the sea, greatly complicating the analysis of box-budget
studies. While the North American contribution is not the major share
in acidic deposition in Scandinavia, the multiplicity of sovereign
source regions in Europe and-the resulting fragmentation 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 man-made 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 understanding and quantifying individual processes, and
development of tractable parameterizations for use in computer
simulation models of transport and deposition. An illustrative,
although not necessarily complete, list includes 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
pollutants and thus complicate the definition of the transport
1 ayer.
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.
5) The effect of wet and dry removal cannot be separated from
transport 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 contentious matters, and the "best
estimate" can vary widely. Wet deposition has been
investigated in detail mostly on the local scale, although the
OSCAR experiment of the EPA/DOE MAP3S program in 1981 was aimed
3-90
-------
at the regional scale (Easter 1981). Wet removal
parameterizations, 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 expicit diurnally varying parameters, 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 PBL, which is responsible for pollutant
transport between 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 and 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 convection 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
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
3-91
-------
seasonally in simmer. In particular, as a result of
substantially lower mixing heights in winter than in summer, a
significant 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).
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 pollutant 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 stratification and distortion during the night followed
by vertical homogenization by day. This results in a rapid
dispersion of emissions over a regional scale (Sections 3.2,
3.3.2, and 3.4.1).
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 remain trapped
within a shallow mixing layer at night, experiencing 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. In winter, in particular, tall-stack
emissions may have long enough atmospheric residence that
substantial fractions of them in the northeastern United States
may be blown off the East Coast (Sections 3.4 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 uncertainties is limited to particular
models, at best (Sections 3.1 and 3.5).
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
3-92
-------
Eulerian measurements. Their spatial-temporal coverage cannot
provide important information concerning mesoscale flows
(Sections 3.2.2 and 3.5).
0 Deposition from a pollutant source is greatest near the source,
a substantial fraction of it occurring during the first day of
transport, on the average. Average or cumulative deposition,
particularly dry deposition, 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 ecologically sensitive regions of
eastern North America, downstream generally means toward the
east or northeast. This conclusion is based primarily on
observations and modeling of SOX. 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
3.4.1 and 3.5).
o Modeling simulations indicate that, in the upper Ohio River
Valley, sources within 500 km dominate ambient SOp
concentrations, and also contribute the greater snare of the
maxima of aerosol sulfate concentration and the total sufur wet
and dry deposition. Long-range transport may be responsible for
most of the sulfate and total sulfur deposition in upper New
England and over parts of eastern Canada. 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).
Acknowledgment: A significant amount of the material presented in this
chapter was developed under cooperative agreement be-
tween Washington University and the U.S. Environmental
Protection Agency (CR-80-9713, CR-81-0325, and
CR-81-0351).
3-93
-------
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 SOz and Ox1dants
from the atmosphere. Adv. Environ. Sci. Techno!. 8:9.
Anthes, R. A., H. A. Panofsky, J. J. Cahir and A. Rango. 1975. The
Atmosphere. Chas. E. Merrill Publ., Columbus, OH.
Arya, S. P. 1982. Atmospheric boundary layers over homogeneous
terrain, Ch. 6. j£ 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.
Bornstein, R. D. and D. S. Johnson. 1977. Urban-rural wind
differences. Atmos. Environ. 11:597.
Burpee, R. W. 1979. Peninsula-scale convergence in the south Florida
sea breeze. Mon. Weather Rev. 107:852-860.
3-94
-------
Briggs, G. A. 1969. Plume Rise. USAEC Critical Review Series,
TID-25075, Clearinghouse for Federal Scientific and Technical
Information.
Briggs, G. A. 1975. Plume rise predictions, ^n Lectures on Air
Pollution and Environmental Impact Analyses. Amer. Meteorol. Soc.,
Boston, MA.
Brown, R. A. 1974. Analytical Methods in PBL Modelling. Halsted Press
(John Wiley), New York.
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
mesoscale 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.
3-95
-------
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.
Day, S. 1953. Horizontal convergence and the occurrence of summer
precipitation at Miami, Florida. Mon. 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 Meterological Society, Boston,
MA.
Defant, F. 1950. Theorie der land-und seewind. Arch. Meterol.,
Geophys. Biokl imatol., Ser. A 2:404-425.
Defant, F. 1951. Local winds. In Compendium of Meteorology, pp.
655-672. American Meterological 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. In 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.
3-96
-------
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.
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. Meterol. 16:459-466.
Garrett, J. R. 1982. Observations in the nocturnal boundary layer.
Boundary Layer Met. 22:21-48.
Garstang, M. 1973. Proc. Mainz Symp. on Atmospheric Boundary Layer.
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. 1983a. Gas-to-particle conversion of
sulfur in power plant plumes: II. Observation of liquid-phase
conversions. Atmos. Environ. (In press).
Gillani, N. V. and W. E. Wilson. 1983b. Gas-to-particle conversion of
sulfur in power plant plumes: III. Parameterization of plume-cloud
interactions. Atmos. Environ. (In press).
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.
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 plan plume out to 300 km. Atmospheric Environment
12:589-598.
3-97
-------
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.
Hering, W. 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 variations of the PBL over land. Atmospheric Physics
Section, Argonne National Laboratory, Report ANL-RER-81-1, Argonne, II.
Holton, J. R. 1967. The diurnal boundary layer wind osciallation over
sloping terrain. Tellus 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. ^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.
3-98
-------
Izmir!, 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.
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. (In press).
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. Jj^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.
3-99
-------
Lyons, W. A. and R. H. Calby. 1983. Impact of Mesoscale Convective
Precipitation 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.
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. Amer.
Meterol. 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.
3-100
-------
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
assessments 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.
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. ^n_
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., G. J. Ferber, J. L. Heffter, K. Telegadas, J. K. Angel 1, 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. In 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 ommparison of three-dimensional and
two-dimensional numerical predictions of sea breeze. J. Atmos. Sci. 31:
1577-1585.
Pielke, R. A. 1981. Mesoscale numerical modeling. In Adv. in Geophys.
23:185-344. —
Plank, V. G. 1966. Wind conditions in situations of patternform and
non-patternform cumulus convection. Tellus 18:1-12.
Plate, E. J. 1971. Aerodynamic Characteristics of Atmospheric Boundary
Layers, U.S. AEC Critical Review Series.
3-101
409-261 0-83-9
-------
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. Tellus 28:335-372.
Price, J. C. 1979. Assessment of the urban heat island effect through
the use of satellite data. Mon. Wea 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. In 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.
3-102
-------
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.
Smith, T. B., D. L. Blumenthal, J. A. Anderson, and A. H. Vanderpol.
1978. Transport of S02 in power plant plumes: Day and night. Atmos.
Environ. 12:605-611.
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.
Thompson, D. E., P. A. Arkin, and W. D. Bonner. 1976. Diurnal
variations of the summertime winds and force field at three mideastern
locations. Monthly Weather Review 104:1012-1022.
Tong, E. Y., 6. 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.
3-103
-------
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.
3-104
-------
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
precipitation-deposition 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 proceses. 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 neutraliza-
tion. 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 S0£ by
03 and H202 are the two familiar examples.
In the Sections 4.2 and 4.3 of this chapter, gas-phase and aqueous-
phase transformations are discussed separately. The section on
4-1
-------
GAS PHASE
GASEOUS OXIDE
S02
NO; N02
A
ro
OXIDATION
(HYDRATION]^
HO, H02, 03
AQUEOUS OXIDE
AEROSOL, (WEAK ACID)
SLOUD DROPLET,
RAIN S02(aq.)=H+
N0(aq.), N02(aq.)
GASEOUS ACID
H2S04
HN03
OXIDATION
' ' 2H+,
02 + Fe, Pin... H+, N03~
AQUEOUS NEUTRALIZATION ,
STRONG ACID
NH3;MO; MC03
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).
-------
homogeneous gas-phase reactions suggests that the fundamental chemistry
is fairly well established, although there are specific areas of uncer-
tainty pertaining to the formation of acidic species. A major problem
is that field measurements have not been adequate to definitively 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, theorectical
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 (k-1). 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 fundamen-
tal 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 solute 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 nr3 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 valuable research on this topic, it is not yet possible to assess
the importance of these reactions to the acidic deposition problem. The
concensus 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)
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 (H2S) to +6
(H2S04). The chemical mechanisms and kinetics of reduced S compounds
4-3
-------
such as hydrogen sulfide (H2S) and carbonyl sulflde (COS) have not
been studied as extensively as sulfur dioxide ($02) and sulfuric acid
(H2S04) have.
The oxidation of reduced S compounds in the troposphere presumably
leads to S02 formation. Some possible reactions are listed in
Table 4-1. Except for the first reaction, HO + h^S, considerable
uncertainty surrounds the products and rate constants (Baulch et al .
1980).
The atmospheric lifetimes of these reduced S compounds with respect
to gas-phase reactions are expected to be determined by their reactions
with hydroxyl (HO) radicals. Table 4-2 lists some typical background
concentrations for the compounds (Sze and Ko 1980) and estimated
lifetimes for removal by a background HO level of 4 x 10$ 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 HO with COS, carbon disulfide (C$2), dimethyl sulfide
(CH3$CH3) and other biogenic, reduced S compounds need to be
identified.
4.2.1.2 Sulfur Dioxide—The atmospheric chemistry of SO? 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 (Moller
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 SO? ~ 0.02 percent hr*1 (Calvert et al .
1978).
Oxidation of S02 by excited oxygen (Ug, 1^g+), nitrogen dioxide
(N02), nitrogen tri oxide (N03), nitrogen pentoxide (NoOs), or ozone (03)
is unimportant in the troposphere (Calvert et al . 1978). 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 peroxy (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
4-4
-------
TABLE 4-1. REACTIONS OF REDUCED SULFUR
Reaction
Rate constant
molecule'1 s"1)
Reference
Reaction
number
HO + H2S •* US + H20
5.3 x lO'12
-14
HO + OCS + C02 + HS(?) ^ 6 x 10
1 x 10
HO + CS2 -»• ?
HS + 02 + SO + HO
uc j. n en x u
no T uo ->. oUo ^ n
SO + 0 + SO + 0
-14
-13
< 2 x 10
1.5 x 10
-15
< 10
-13
9 x 10
-18
Baulch et al. (1980) [4-1]
Baulch et al. (1980) [4-2]
Oemore et al. (1981)
Baulch et al. (1980) [4-3]
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
Molecule
H2S
COS
C$2
Typi cal /Concentrati ona
(ppb)
0.004 - 0.40
0.49
0.069 - 0.370
Lifetime for removal
by HO (s x 10-5)
1.9
1,000
6,750
and Ko (1980).
4-5
-------
TABLE 4-3. PHOTO-OXIDATION REACTIONS OF S02
Reaction
Reaction number
S02(X *Ai) + hv (340-400 nm) •* S02(3Bi) [4-6]
S02(3Bi) + 02(3V) •" S0 (* ^l) + 02{lEg+) [4-7a]
[4-7b]
S02 (Bi) + 02 (£g~) * S04
S04(cyclic) + 02 -»• S03 + 03 [4-9]
+ 02(3£g~) •* S03 + 0(3P) [4-10]
M-^03 + M [4-11]
4-6
-------
much too slow to be significant in the troposphere. An analogous
reaction is that of SO? with methyl peroxy radicals (CH302K
Although this system has received attention in recent years, the
tropospheric role of the CHjOg + SO? reaction has not been
interpreted concretely. Table 4-4 lists some recent rate constant
determinations for this reaction.
The rate constant for the S02 and methoxy radical
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 02 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 02, keeps the level of ChhO very low; probably lower than that
for OH. Thus, if [CH30] « [OH] and k(CH30 + S02) < k(OH +
S02), then oxidation of S02 by CH30 is not important.
The combined oxidation of S02 will depend on the concentration of
other reactive species (e.g., H02, (^30?, CH^O, NO, NO^). 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 CH302 will need to be
assessed.
Ozone-alkene reactions are complex and give rise to diverse
reactive radicals that may oxidize SC^. Some possible reactions are
listed in Table 4-6. Cox and Penkett (1972) observed that water
markedly inhibits SO? 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 + C4H8 •* molozonide -> CH3CHOO + CHsCHO [4-23]
03 + C4H8 -" RCHO, RCOOH, etc. [4-24]
CH3CHOO + S02 -* CH3CHO + S03 [4-25]
CH3CHOO + C4H8 -> CH3CHO + C^gO + 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]
4-7
-------
TABLE 4-4. RATE CONSTANTS FOR CH302 + S02 -»• PRODUCTS
k (cm3 molecule-1 s~l) Reference
< 5 x 10-17 Sander and Watson (1981)
8.2 x 10-15 Sanhueza et al. (1979)
5.3 x 10-15 Kan et al. (1979)
1.4 x 10-14 Kan et al. (1981)
TABLE 4-5. CH302 + S02 MECHANISM OF KAN ET AL. (1981)
Reaction Suggested rate constant Reaction
number
CH,09 + S09 -»• CHJ)9S09 1.4 x 10"14cm3 molecules"1 s"1 [4-12]
*5 £ L. O tL £
CH302S02+ CH302 + S02 < 24 s"1 ' [4-13]
CH302S02 + 02 + CH302S0202^| K14/k15 - 1.7 x 1020 [4.14]
J3 -1
cm molecule [4-15]
CH302S0202 + NO +
N02 + CH302S020 6.2 x 10"12 cm3 molecule"1 s"1 [4-16]
CH302S020 + CH30 + 02 3.3 x 1013 cm3 molecule"1 s"1 [4-17]
2 + S03 [4-18]
4-8
-------
TABLE 4-6. POSSIBLE S02 . Q3 - ALKENE REACTIONS
Reaction Reaction
number
/-••°\
R - CH CHR + S00-> 2RCHO + SO- [4-19]
o. o-o.
RCH - CHR + S02 + 2RCHO + S03 [4-20]
RCHOO + SO -> RCHO + SO [4-21]
2 3
o
RCHO- + S09 + RCHO + SO, [4-22]
4-9
-------
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 HO.
HO + S02 -> HOS02 [4-30]
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 HOS02 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, there is no strong evidence to suggest that the final product
initiated by the HO-S02 reaction is anything other than sulfuric
acid.
The fate of sulfur trioxide (S03) 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:
S03 + H20 + (S03-H20) -> H2S04 [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 [NH3]) to +5 (pernitric [H02NO?]
acid), including both bases (ammonia [NHs] and amines) and acids
(nitrous [HOMO], nitric [HN03], and pernitric [H02N02] acids).
NH3 is the most abundant form of reduced N (after molecular
nitrogen N2 and HOMO) 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
4-10
-------
TABLE 4-7. PROPOSED MECHANISMS FOR THE FATE OF HOS02
HO + S02
HOS02+ 0
HOS0200
HOSOgOO
Reaction
Mechanism of
+ (+M) + HOS02 (+M)
2 + HOS0200
+ NO % HOS020 + N02
+ N02 * HOS02OON02
- AH, kcal mole-1
Calvert et al . (1978)
-37
-16
-25
?
HOSO^OONO,, •* HOSO.O + NO. ?
£ C- C. O
HOS0200
HOS0200
2HOS0200
HOS020 +
HOS02ONO
HOS020 +
HOS020 +
HOS020 +
HOSO,0 +
+ un -> nn<;n n + un
~ nUn ^^ nuou«u ~ nu.
L C. O
+ H02 -> HOS0202H + 02
-*• 2HOS020 + 02
NO + HOS02ONO
+ h v + HOSO 0 + NO
N02 -»• HOS02ON02
H02 -*- HOSOgOH + 02
C3 Hg -> HOS02OH + 1so-C3
C,HC + HOSO.OCH.CHCH,
- 2
-43
-22
-26
-22
-57
;H7 "10
?
Reaction
number
[5-31]
[5-32]
[5-33]
[5-34]
[5-35]
[5-36]
[5-37]
[5-38]
[5-39]
[5-40]
[5-41]
[5-42]
[5-43]
[5-44]
H(?SO/,+ aerosol (H90, NH-, CH,0, CnH~ ..) •* (growing aerosol) [5-45]
£ T" c. j c, n tM
HOS02ONO + aerosol (HgO) ^aerosol (H2S04> HONOg...) [5-46]
HOS02ONO + aerosol (HgO) + aerosol (H2S04, HONO ...) [5-47]
4-11
-------
TABLE 4-7. CONTINUED
Reaction
Reaction ~AH, kcal mole-1 number
Alternative mechanisms of Davis and Klauber (1975)
HOS020 + 0£(+M) -> HOS0203(+M) [4-48]
3 + NO -> HOS0202 + N02 [4-49]
2 + NO -> HOS020 + N02 [4-50]
Mechanisms of Davis et al. (1979) for HOS02
HOSO + 0 + M -> HOSO. + M
HOSO. + H00 -»• HSOC-H00
1 f. o f.
HS05-H20 -*• HS05-(H20)2
HSOK(H90)V + NO ^ HS04(H90) NO.
b d. x <\ i. X e.
HS05(H20)X + S02 •> HS04(H20)XS03
HSO(.(H00)V + H09 •* H9SOK(H90)V + 0
4-12
-------
Levine et al. 1980). The homogeneous loss mechanism should be
dominated by reaction with HO, but the fate of the product of this
reaction, NH2, is unknown. The NH3 reaction rate with gaseous acids
(HNOa, H2S04) is not well established but should be rapid
(Seinfeld et al. 1981).
The most abundant nitrogen oxides (NOX) in the troposphere
(excluding the relatively unreactive nitrous oxide [^0]) are nitric
oxide (NO) and N02« Chemistry that is rather complex and not
completely understood interconverts these compounds (which are also
primary emissions), to NOa, N205, MONO, HNOa, and HO?N02
(Table 4-8). NO is converted to N02 and MONO 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 concen-
trations as high as 1 ppb (Perner and Platt 1979). Concentrations this
high are not readily explained from the known homogeneous reactions that
produce HONO and the photolysis rates that destroy it. Additional
homogeneous sources might exist, and the heterogeneous promotions of the
reaction of NO + N02 + H20 2 2HONO are possibilities. HONO is a
relatively weak acid (pKa 5.22) and has its greatest tropospheric
significance as a photolytic source of HO radicals.
N02 has a gas-phase removal mechanism dominated by reaction with
HO to form HNOa. With an HO concentration of 4 x IQ-^ppb, N02
would have a lifetime of ~ 17 hr. N02 also reacts with ozone to
form NOa, which can photolyze to give back N02-
like H2S04, 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 HO to
form N03. The lifetime of HN03 with respect to the HO reaction,
HO ~ 4 x ID'5 ppb, is 2 to 3 x 103 hr.
is a strong oxidizer in the atmosphere and may be removed by
oxidizing 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). N03
also exists in equilibrium with N20s 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 ^5 as ni'9n
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, N20s, OH, and 03 involves
a close interrelationship that should have a profound significance to
the acidity of precipitation, especially in remote areas where HN03
4-13
-------
TABLE 4-8. REACTIONS OF NITROGEN COMPOUNDS
Reaction
Rate constant
k (cm3 molecule"1 s"1)
Reference
Reaction
number
I
I—"
-p»
NH3 + HO + NH2 + H20
NO + N02 + H20 + 2HONO
2NO + 02 -»• 2N02
HO + NO + M + M + HONO
NO + 03 + N02 + 02
N02 + 03 -> N03 + 02
HONO + h v •* HO + NO
HO + HN03 + H20 + N03
N02 + N03 -> N20s
N03 + NO + 2N02
N20s -> N02 + N03
N02 + h v ->• NO + 0
2.3 x ID'12 exp (-800/T)
k = 1.56 atm-1
3.3 x 1039 exp (530/T)a
1 x ID'11
3 x lO'11
1.8 x 10-14
3.2 x 10-17
8.5 x 10-1J
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
N03 + h v + N02 + 0
N03 + h v -> NO + 02
HO + N02 + M -* HN03 + M
H02 + N02 -»• H02N02
H02N02 •* H02 + N02
+ N02 -»• CH3C002N02
CH,C000 + NOo
- 2 32 f-
N2°5 H2° "" 2HN03
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
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 10"14 exp (-25000/RT) Cox and Roffey (1977)
< 1 x ID'20 Hampson and Garvin (1977)
C4-71]
[4-72]
[4-73]
[4-74]
[4-75]
[4-76]
[4-77]
[4-78]
molecule~2 s~l.
-------
may dominate the pH of acidic precipitation (Seinfeld et al. 1981).
Further studies are warranted involving field measurements of NOs and
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 IQl* 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.
4.2.1.4 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 (CHaCl) and methyl bromide (CHaBr) have
tropospheric lifetimes probably dominated by aqueous-phase processes
that produce and consume hydrochloric acid (HC1). HC1 is also produced
by gas-phase reactions following the reaction of HO with a halocarbon as
the rate-limiting step. It has been suggested that rainwater's acidity
in remote areas is controlled principally by the presence of HC1 and
HN03 (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 Adds—Organic acids are expected to occur as photo-
oxldation 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—formic (HCOOH) and acetic (CH3COOH)— 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
H02 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).
CH3CHOO + H20 + CHaCOOH + H20 [4-79]
4-16
-------
TABLE 4-9. ATMOSPHERIC HALOGEN COMPOUNDSa
Compound
(Natural)
CH3C1
CH3Br
CH3I
HC1
(Anthropogenic)
CHC13
C2C14
CHCloF
CH3CCl3b
Concentration
(ppb)
0.81
0.01
0.01
0.20
0.02
0.03
0.01
0.1
Li fetimea
(s x 10?)
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.
4-17
-------
The loss mechanisms for these adds are not known but should be a
combination of reaction with HO, wet and dry deposition, and rainout.
Recent measurements (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
HO. 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 to 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., Kocmond and Yang 1976, Friedlander 1978,
Whitby 1978, McMurry and Wilson 1980). 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, McNelis 1974). As indicated earlier,
the significance of this oxidation path has been assessed by computer
simulations of the S02 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 HO, H02,
and CH302 (Kuhlman et al. 1978, Graham et al. 1979, Miller 1980).
The experimental results, aided by computer simulations of the experi-
ments, indicated that S02 is oxidized predominantly by HO under
urban-air conditions.
Chemical kinetics and smog chamber results indicate that the
HO radical is responsible for the majority of the H2S04 and HN03
formed via gas-phase reactions in the atmosphere. HO 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
4-18
-------
TABLE 4-10. TROPOSPHERIC HCOOH AND
(DAWSON ET AL. 1980)
Acid
HCOOH
CH3COOH
pKa
3.75
4.75
Remote site
(ppb)
2
1
Lifetime3
Urban site (hr)
3.5 8
6.0 48
aAssuming removal by HO - 2 x 10~4 ppb and assuming k(HO + HCOOH)
~6 x 10-12 cm3 molecule-1 s-1 and k(HO + CHaCOOH) ~ 1Q-12 cm3
molecule-1 s~l.
4-19
-------
the HC-NOX-03 cycle affects the HO population and the formation of
H2S04 and HMOs.
A series of smog experiments focused on S02 oxidation indicated
that the maximum rate of S02 conversion to H2SOA 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 S0;> conversion rate. Computer modeling of
these experimental conditions indicated that HO was primarily
responsible for S02 oxidation, and the effects of HC and NOX
concentrations on the relative levels of HO 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 HO 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 i'n 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 and Lazarus 1979) and particulate
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/HN03 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
4-20
-------
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/lf the average pseudo first-order rate for S02 oxidation
was - 0.012 hr-l, so an average S02 lifetime toward ^$04
formation would be 83 hours. For similar conditions, the pseudo first-
order rate for N02 oxidation to HN03 (given PAN/HNOa ~ 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 NOg 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 N02 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.
Many observations of S02 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 S02 oxidation rates as
large as 32 percent hr-1 have been inferred for St. Louis, MO, 13
percent hr'1 for Los Angeles, CA, and 9 percent hr-1 for Milwaukee,
4-21
-------
TABLE 4-11. S02 OXIDATION RATES (% hr'1) FROM STUDIES OF URBAN
PLUMES AND LONG RANGE TRANSPORT
Range
6-25
1.2-13
1.1
0.3-1.7
5.3-32
5
31
10-14
8-11.5
0.6-4
0-4
1-9
Average
16.6
7.1
1.1
0.7
16
5
31
12
9.8
1.7
2
4
Location/periods
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 ,
Netherlands/S &
W/D & N
St. Louis, MO/S/D
Milwaukee, WI/S/D
References
Benarie et al . (1972)b
Roberts and Friedlander (1975)b
Pratai 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.
^Higher rates possibly related to aqueous-phase reactions.
ccalculated from their half-life data.
dCalculated from their data by Alkezweeny and Powell (1977).
eBased on kinetic analysis of data by Isaksen et al. (1978).
4-22
-------
WI. 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, NOv, 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 existing knowledge of gas-phase chemistry is applied
to predict S02 oxidation in urban areas.
Information on the gas-phase transformations of NOx to aci<1
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. j^e -j^ 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-1 were observed.
Measurements by Spicer et al. (1981a) resulted in pseudo first-order
4-23
-------
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 partlculate N03~ was < 1 percent of the total
product. Given an average PAN/HN03 ratio of 1.8, the pseudo
first-order rate for NO? conversion to acid would have been 6.3
percent hr-1, and the NOX lifetime with respect to HNOa production
would be about 16 hrs. These values are similar to estimates given
earlier with respect to global HO 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 N0£
was removed by reaction with 03 and subsequent reactions of NgOs
and N0;j Into condensed products (partlculate nitrates) not amenable to
detection In their cell.
This Interpretation conflicts with the conclusion reached by the
Battelle researchers (Splcer et al. 1981a) which asserts that 95 percent
of the NOX losses in urban plumes can be accounted for as gaseous
HN03 and PAN, and that the amounts of partial!ate nitrate produced in
urban plumes are very small.
As indicated earlier, it is apparent that more research is needed
concerning the fate of PAN, N205 and N03 in the atmosphere and
their potential contributions as acidic species.
4.2.3.2 Power Plant P1umes--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 Hi 1st 1972, Lamb and Shu
1978, Shu et al. 1978).
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.
4-24
-------
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 H2S04
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 N0s~ 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 1016 particles s'1 of H?S04 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, SC«2 oxidation was
related primarily to photochemical reactions. In accord with
photochemical mechanisms, oxidation rates were low in February (< 0.5
percent hr~M and relatively high in June (1 to 3 percent hr^).
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 hf1 compared to nighttime 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 S042~ while the
remainder is lost by deposition mechanisms.
4-25
-------
TABLE 4-12. SUMMARY OF POWER PLANT PLUME STUDIES WITH EMPHASIS ON GAS-PHASE TRANSFORMATION RATES
-F»
1
ro
en
Range of SQ2
conversion rates
Plant/location Season (% hr"1)
Four Corners, NM October 2-8
GCOS/Alberta Feb. & June 0-3
Labadie/MO July 0.41 - 4.9
Labadie/MO July 0-4
Range of NOX
conversion rates
(% hr"1) Reference
Pueschel and
(1978)
Lusis et al .
Cantrell and
(1978)
Wilson (1978)
et al. (1978)
MO7Q\ rilla
Van Valin
(1978)
Whitby
, Husar
, Gillani
Four Corners, NM
Central 1a/WA
Leland-Olds/ND
Sherco/MN
Big Brown/TX
June
0.9 - 5.4
Spring & fall 0.03 - 1.4
June
June
June
0-0.7 0.2 as partlculate
0-3
0.2 as particulate
(1978)
Hobbs et al. (1979),
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
£ Labadie/MO ' ,
Sherco/MN I
Cumberland/TN !
Navajo/AZ _J
Cobb/MI "^
Andrus/MS
Breed/ IN J
Season
Summer
August
Summer & winter
July
-
August
Summer
May & Nov
May & Oct
Jun & Nov
Range of S0£ 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 -
4
0.8
5.4
14.2
7.1
2.9
11
5.9
1.5
Eatough et al . (1980)
3-12 Forrest et al . (1981)
3-10 times R$o2 Richards et al . (1981)
-
Whitby et al . (1980)
-
-
23 - 31 as NOX loss
5 - 21 as NOX loss Easter et al . (1980)
-
-------
Power plant experiments conducted by the University of Washington
(Hobbs et al. 1978, Hegg and Hobbs 1979b) employed a variety of
particle-measuring techniques. SC>2 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 N03- at three of the plants (Hegg and
Hobbs 1980) showed minimal N03~ 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.
(1980) has led to interesting differences between the chemical
composition of secondary SO^ particles, depending on regions of the
United States. In the East, where SOg conversion rates are generally
high, secondary S042~ 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 NOv transformations. Total N03~
formation (gaseous and particulate N03-) rates were 0.1 to 3 percent
hr"1 at night and 3 to 12 percent hr~* 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 N03".
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 (HN03) 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 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.
4-28
-------
Measurements of NOX transformation rates in the above study were
inconclusive. Chemical analyses indicated that transformations to
HN03 and particulate W$- were minimal, yet large NOx losses
were often calculated when NOx was compared to SFfi 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 NO? oxidation
in the atmosphere have been crude and imprecise. This relates to
analytical difficulties, extensive spacial and temporal averaging and,
particularly in the case of SO;?, a lack of distinction between
gas-phase and aqueous-phase reaction paths.
Rates of S02 oxidation measured in urban areas and plumes range
from near zero to 30 percent hr"1. The preponderance of data,
however, indicates upper-level rates of 12 percent hr"1 for midday,
summer conditions. Average daytime conversion rates are in a range of 3
to 5 percent hr"1 for summertime conditions. Systematic measurements
of seasonal and diurnal variations have not been made; peripheral data
indicates 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 N0£ with hydroxyl radicals is the principal gas-phase
route for HN03 formation, although other reactions are also important.
In general, NO;? conversion rates under daylight, summertime conditions
range from < 5 percent hr"1 to 24 percent hr"1, with at least half
of the product yield being nitric acid vapor.
There is conflicting evidence about the role of Np05 in nitrate
formation; its gas-phase reaction with water is very slow, but it hydro-
lyzes rapidly on moist surfaces. There is also considerable uncertainty
4-29
409-261 0-83-10
-------
regarding the fate of peroxyacetyl nitrate (PAN) in the atmosphere and
its potential to 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 pro-
ducts 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"1, although some exceptions exist. S02 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 HNC>3 formation rate in a range 3 to 10 times greater than
that for H2$04 formation. This result would seem likely if the
hydroxyl radical was the principal oxidant.
Overall, field studies of S02 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 HO radical and to a much lesser, but
uncertain, extent by (^302 and H02- Of the reduced forms of
sulfur gases, H£S is by far the most reactive in the atmosphere. Its
reaction with OH radicals is faster than is the rate between S02 and
HO and the product of the reaction is 502- 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 HN03 formation will be dominated by
the reaction of N02 with HO radicals. The rate for this reaction is
approximately ten times faster than the rate for S02 oxidation by HO.
As mentioned above, other products of nitrogen oxides reactions in air
are potentially important to acidic deposition, particularly NoOs
and PAN and to a lesser extent N03 and HN02» and tne fate of ™ese
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
4-30
-------
oxides (N02). The findings were qualitatively consistent with kinetic
models that predicted HO 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 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 con-
centrations 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 concentra-
tions of free radicals is largely theoretical. In polluted air, the
concentration of HO is considered to be strongly related to the con-
centrations of hydrocarbons, aldehydes, carbon monoxide and nitrogen
oxides, whereas, in relatively clean "background" air, the HO
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 "back-ground" HO 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 HO, an assessment of the
relationship between precursor concentrates 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. Postu-
lating 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,
4-31
-------
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 HpSOA in hydrometeors are evaluated
in this section. During this evaluation, the relative importance
of direct acid vapor absorption reactions and acid-precursor oxidation
reactions is considered. In addition, the importance of neutralization
in acidic hydrometeors 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 considered here.
With regard to particle scavenging, Chapter A-6 shows that scaveng-
ing 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, particulate
sulfuric acid. This procedure allows us to treat the incorporation of
H2S04 into hyrdometeors 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, HN03,
and (^$04) and of the various trace gases (Cl;?, N02, ^04, HNO?, and
$02) assumed to be the precursors of these acids in the atmosphere are
listed in Table 4-13.
4-32
-------
END OF CONDENSATIONAL GROWTH,
START OF GROWTH VIA COLLECTION
PROCESSES. FOR WARM CLOUDS,
DILUTION EFFECTS CEASE.
(
PRODUCTION OF ACIDS IN DROPLET
FROM ABSORBED PRECURSORS.
CONTINUED ABSORPTION OF GASES.
*
t
CONDENSATIONAL GROWTH OF DROPLET
AND ABSORPTION OF VARIOUS ACIDS
AND ACID PRECURSORS.
*
SOLUBLE FRACTION OF CLOUD
CONDENSATION NUCLEI DISSOLVES
IN THE DROPLET. *
t
NUCLEATION OF CLOUD DROPLET
ON A CLOUD CONDENSATION NUCLEI.
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.
i
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.
1
PRODUCTION OF ACIDS IN RAINDROPS
FROM ABSORBED PRECURSORS.
•
DEPOSITION OF
1
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
6asa
C12
(HC1)
(mol *
6.2
2.5
H
--1 atm-1)
x 10-2
x 103
Temperature
(0
25
25
Source
Whitney and Vivian
(1941)
Calculated from vapor
N02
N204
HN02
(HN03)
S02
(H2S04)
2.48 x 10-2
2.15
4.76 x 101
1.98 x 105
1.24
108
15
15
25
25
25
25
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
-------
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 JT1 atir1) shown in Table 4-13 cannot
possibly be applied to conditions in the atmosphere; they simply
indicate large deviation from Raoults' 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 Cl2 *s 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 (National Academy of Sciences
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 absoprtion via the very fast
reaction (Whitney and Vivian 1941):
Cl2 + H20 U) -> H+ + Cl- + HOC!, [4-80]
and could therefore be considered the anhydride of HC1. Chlorinated
organics, 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).
4-35
-------
TABLE 4-14. GAS-PHASE CONCENTRATIONS OF ACIDS
AND THEIR PRECURSORS IN THE ATMOSPHERE
6asa
Concentration
in "background"
air (ppb)
Concentration
in urban air
(ppb)
Source
C12
(HC1)
N02
N204
HN02
(HN03)
S02
(H2S04)
1 8 Kritz and Rancher (1980),
Okita et al. (1974)
0.1-4 10-100 Robinson and Robbins (1969),
Noxon (1975), Spicer (1977b)
Negligible Negligible No measurements available.
0.003 2-4 Crutzen (1974), Winer (1979).
0.02-5 10 Huebert and Lazrus (1978),
Kelly et al. (1979), Spicer
(1977b).
1-14 10-50 Georgii (1978), Hidy et al.
(1978)
c 1& (0.5) <_ lb(0.5,4) Commins (1963), Tanner et al.
(1977), Elshout 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 Elshout 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
-------
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 £ nr3 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 ^$04) in
solution is given by:
S02 S02
Kls Kls K2s
n f*f\ • i e
[H+]2
[4-81]
where HSO? is the Henry's law constant for S02, and KIS and
K2s are the first and second dissociation constants for S02.H20.
When appropriate values are used for those constants and a cloud water
pH of ~ 5 (Petrenchuk and Drozdova 1966, Hegg and Hobbs 1981a) is
assumed, a maximum concentration of S(IV) in urban air is found to be
7.9 x 10~5 mol £-1. 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 production.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
different. Again, acid concentration must be estimated from considera-
tions 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 HN03 concentrations of
4.1 x 10"* mol jT1 and 4.1 x 10~5 mol £-1 for urban and background
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 hydrometer activity (although direct absorption
of acids may be substantial). Attention is therefore focused on
solution reactions in cloud droplets.
4-37
-------
atmospheres, respectively. The corresponding liquid-phase N(III) (the N
species generally assumed to be the precursor of HN03 in solution)
concentrations would be ~ 8 x 10-5 moi £-l in an urban atmos-
phere and 3 x 10~8 mol £-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 102 percent hr~l and ~ 1 x 105 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 HN03 in solution
from absorbed N(III) is its production from absorbed ^05 at night
(Platt et al. 1980). However, since ^05 is an hydride of HNOa,
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 HN03 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 H?S04
(Robbins et al. 1959, Eriksson 1960). For both HNOo and H2S04,
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 NaCl to HC1 within a
5-minute reaction time; presumably, the ^$04 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 ^$04 from which it is derived.
4.3.4 Production of HMOs in Solution
The production of HN03 in solution by means of nitrite N02~
(or HN02) oxidation has been proposed as a significant atmospheric
reaction. The oxidants currently considered significant are 03
4-38
-------
(Penkett 1972) and »2Q2 (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 £-1 atm-1 f0r the Henry's law
constant of NO?, compared to the value of 2.48 x 10~2 mol JT1
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).
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,
3 N02 + H20 = NO + 2HN03, [4-82]
occurs and spontaneously forms a two-phase system consisting of HN03
vapor and droplets of dilute HN03 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) + HNQ-3U) [4-84]
N02 absorbs in a straightforward manner but then forms N204
(a), 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 nT1), 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
4-39
-------
(7.5 x 104 £ mol'1; 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
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 NO? or HN03, are increased roughly 103 hr-1 and 2 x 1(P
nr-l 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 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 equiva-
lent to gas-phase concentrations of several hundred ppm, 103 to 104
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 0? levels (Larson et al . 1978)
suggests that extrapolation of the N(IIl) 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:
[03] [N(III)J [4-85]
dt dt
with k£ = (1.60 +_ 0.13) x 105 i mol'1 s'1. Assuming that the
ambient 0? concentration at cloud level is generally at or below 50
ppb (at STP), the characteristic time? for M(III) oxidation at 283 K
and a pH of 5.9 would be ~ 2 hr, and the conversion rate (R) 50
percent hr-1.3 Clearly, this reaction will be of little importance
in HN03 production.
The oxidation of N(III) in solution by (^2 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:
- d[H2°2] = k [H202] [HN02] [H+] [4-86]
r\
The e-1 decay time.
d ( £ 1 n C1)
3R*U of hr'1) = 100 x (It , where C^ is the concentration
of the reactant under consideration. Consequently, R| (% hr'1) =
100 x k1, where k1 is the pseudo-first order rate coefficient.
4-40
-------
with k = 1.4 x 102 &2 mQ-\-2 $-lt These investigators considered HN02
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, deter-
mined 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:
. d[H202] k3 k2 [H+]2 [N02~] [H202] [4_87]
dt = k_2 + k
where the k's are rate constants as defined by Anbar and Taube, k3 =
5.8 x 106 £3 mol"3 s"1, and k3/k_2 = 2.4. For atmospheric
levels of H202, this reduces to:
- d[H2°2] = k1 [H+]2 [N02-] [H202] [4-88]
dt
with k1 = 1.4 x 107 £3 mol-3 s-l.
The rate expression of Anbar and Taube must be converted to one
with explicit HN02 dependence by means of the N02- -HN02
equilibrium to compare this value directly with that of Halfpenny and
Robinson. This results in a rate coefficient of 6.3 x l(r £2
mol-2 s-lf 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 concentrations
(>^0.05 mol £-1) and N(III) concentrations (>_ 0.017 mol £-!) far
higher than those encountered in the atmosphere. This should be
considered 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 pH. For the Ho02 concentration in hydrometeors, a value of
1.5 x 10-5 moi £-1 wli*f (je employed (based on measurements in
precipitation [Kok 1980] and a few, as yet unpublished, measurements in
clouds over the eastern United States [Kok, pers. comm.]). Inserting
these values into Anbar and Taube's rate expression yields a character-
istic time for N(III) oxidation of 1.3 x 104 hr, surprisingly
4-41
-------
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 unimportant 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 f^SO^ production. This
dependence should be considered when the influence of HN03 on acidic
deposition is assessed.
At this juncture, we conclude that HNOa 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.
4.3.5 Production of H2$04 in Solution
4.3.5.1 Evidence from Field Studies- -From analyses presented in Section
4.3.2, it appears that h^SOa 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 sulfates) 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 $042- production in
clouds is necessary to account for the acidity and S042~ levels
found in precipitation. Also, recent field results (Lazrus et al . 1982)
suggest appreciable sulfate formation in warm frontal clouds. Finally,
Gillani and Wilson (1982), 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 S02 to S042~ conversion rates observed were on the order
of 10 percent hr-1, a significant rate even in light of the analysis
in Section 4.3.2, because SO? 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 S0£ were converted to
The most direct and quantitative evidence for S042" production
in clouds has come from recent measurements of $042- concentrations in
4-42
-------
the air entering and leaving wave clouds (Hegg and Hobbs 1981a,b).
These measurements have yielded S02-to-S042- conversion rates
typically on the order of 102 percent hr*, 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:
d [s°4l = ki [H+]a [so32-] exp (EA/RT) [4-89]
cfE
where Iq = (3.3 x 105 + 6.2 x 105) a1'1 moT1'1 s'1.
a =!.!+_ 0.1, "and EA = (2.9 +_ 2.7) kJ moT1.
Section 4.3.3 shows that the value of a is similar to that
expected if the S042~ is produced in solution via 63 oxidation.
However, the Sfy2- production rates measured in these field studies
showed no significant correlations with 63 concentrations.
These field measurements dictate examination of H2S04 produc-
tion in hydrometeors in greater detail than for HC1 and HN03.
4.3.5.2 Homogeneous Aerobic Oxidation of S02*H20 to H2S04--
4.3.5.2.1 Uncatalyzed. This reactions is the most extensive 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 (S032-) were measured by
4-43
-------
10V
10
-1
CO
10
10
-3
10
-4
FULLER AND CRIST (1941) -
AS MODIFIED BY McKAY (1971)
LARSON ET AL
(1978)
.MILLER AND
de PENA (1972)'
(pH = ?)
RIMBLECOMBE AND
SPEDDING (1974)
SCHEOEDER (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 SQ^- in solution (Hegg and
Hobbs 1978).
4-44
-------
monitoring S032- (and sometimes sulfate) concentrations, and S02
degassing from solution was evaluated quantitatively. Such procedures
obviate criticisms made of other laboratory studies of $032-
oxidation rates with respect to mass-transport limitation of the
oxidation (Kaplan et al. 1981, Schwartz and Freiberg 1981). Similar
procedures were employed by Fuller and Crist (1941) and by Brimblecombe
and Spedding (1974). Hence, the disparities shown in Figure 4-3 are not
entirely due to mass-transport problems.
Because it is unlikely that the reaction is much faster than that
measured by Larson et al. (1978) (and it may be appreciably lower due to
inhibitors; Hegg and Hobbs (1978), the Larson et al. rate may be
considered an upper limit to the atmospheric oxidation rate. The rate
expression for this reaction at pH £ 7.0 is:
d[$°4 ] = (ki + k2 [H+]1/2) [S032-] [4-90]
dt
with ki = (4.8 +_ 0.6) x lO'3 s'1 and k2
= (8.9 + 1.0) £l/2 moT1/2 s-1.
Activation energies for these two coefficients are 40 +_ 10 kJ
mol"1 and 7 +_ 6 kJ mo!"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 hr1.
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
5032- to a S(IV) dependence. This change can be done by multiplying
the righthand side of Equation 4-90 by the ratio of $03*- to SUV)
in solution at the given pH. For a pH of 5.0 at 278 K, this is
essentially the ratio of S032" to bisulfite (HS03~) and equals 1
x 10-2. This ratio implies an S(IV) oxidation rate and thus an
H2S04 production rate, of 80 percent hr'l. Comparing this to the
rates calculated in Section 4.3.2 for significant ^504 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
H2S04 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
4The rate necessary to produce a sulfate concentration similar to that
obtainable by direct adsorption of H2S04-
4-45
-------
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 transport 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 Georgii1s 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 kinetics. 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 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 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:
o_
For Mn: d[S°4 ] = kMn [Mn+2] [H+]°-46[S032-] [4
dt
For Fe: d[S042"] = kFE [Fe+2] [SQ32-] [4-92]
dt
4-46
-------
For Mn dS°4" = kmix[Mn+2+Fe+2][H+]°-64 [S032-] [4-93]
and Fe: dt
with kMn = 1.6 x 108 £l-46 moil. 46 s-lf kpe= 5.8 x 106 £ mol-l
s-1, and kmix = 1-8 x 109 A1-64 mol1-64 s'1, 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 energies 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 considerable 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.
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 g
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 ^-l of Mn and ~ 3 x 10- 7 mol jr1 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 10~8 mol n~l
of Mn and 4 x 10-7 mol £-l of Fe (Tana|
-------
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
rates are ~ 100 percent hr-1 (Mn), 500 percent hr"1 (Fe), and ~
5 x 103 percent hr~l (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 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 £-1. Furthermore,
these estimates have yielded rates that produce substantial ^$04 in
solution relative to initial concentrations of SO/j. One would
therefore expect the solution pH to drop substantially. Given the
inverse square dependence on FT concentration of the SOs2" concen-
tration 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?so4--SQ2 absorbed into atmospheric hydrometeors can be oxidized
by oxidants other than 0. Indeed, recent work on H2$04 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 SOa^"
oxidation by N(III) in solution (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 SO^2- formation in hydrometeors was
first examined by Penkett (197?), who studied S032' 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 concentrations 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.
4-48
-------
TABLE 4-15. LABORATORY STUDIES OF S(IV) OXIDATION BY 03 IN AQUEOUS SOLUTION
Rate expression
Experimental
PH
Molar ratio of
reactants
Reaction rate3
(1n mol J.'1 s"1)
at 278 K, 1 ppb S02
40 ppb 63, and a
pH of 5.0
Penkett (1972)
Barrle (1975)
k![03][HS03-]
ki = 3.3 x 105 a mol'1 s'1
at 283 K
4.65
4.0
0.03 - 0.5
10-6-5 x ID'5
1.5 x 10-9
5 x 10-llb
EHckson et al .
(1977)
Larson et al .
(1978)
Penkett et al.
(1979) as
modified by
Oasgupta
(1980a)
k2[03][HS03-] + k3
[03][S032-]
k2 = 3.1 x 105 «. moT1 s'1
k3 = 2.2 x 109 a mol'1 s'1
at 298 K
k4[03][HS03-] [H+]-0-1
k4 = 4.4 x 10* j>0.9 mo1-0.9 s-l
at 298 K
k2[03][HS03-] + K3
C03][S032-]
k2 = 3.73 x 105 4 moT1 s"1
k3 = 3.12 x 108 i. moT1 s-1
at 298 K
-1.3 - 4.02 5-50 2 x 10'7
4.0-6.2 6 x 10-4 5 x 10'10
-2 x ID'3
1-5 0.1 - 0.4 6.6 x 10-9
aShows derived rates for atmospheric conditions.
bThe measured rate at pH = 4 and 283 K was converted to that at pH = 5 and 273 K by assuming that the
rate 1s proportional to [HS03-]> and changes negligibly with temperatures over 5 K.
-------
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 63 to S(IV) in
solution increases, suggesting that oxidation rates measured in the
laboratory were limited by mass transport of (h. 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 ^$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 applica-
bility 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 0? 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). Interestingly, Barrie and Georgii (1976)
reported a substantial enhancement in sulfite oxidation rate by 03
when Mn ions were present at roughly 10-5 moi £-1. 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
10-3 jnol £-1 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
enhancements were negligible for Fe and about a factor of 3 for Mn at
the high concentration of 10~5 mol £-!, this rate will be
considered indistinguishable from the uncatalyzed rate already
discussed.
6The characteristic or e-1 folding time is given by
1 _ d S(IV)"1
(S(IV) eft ; in the atmospheric pH range of ~ 3 to 6, HS04-
1 _ d S(
-S(IV) and this becomes: [HS03-J ~3t
4-50
-------
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 integrate 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 is
adequate and appealing. It is:
= k [H202] [S02.H20] [4-94]
dt
with k = 8.3 x 105 a mol'1 s"1 at 298 K and an activation energy
of ~ 28 kJ mol'1 (Martin et al . 1981).
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 H2Q2 concentrations of 1 ppb or greater (based on actual
measurements) and a value of the H202 Henry's law constant, based on
H20o 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 10$ mol
r1 at 273 K. At 273 K, 1 x 10-9 atin H2Q2 is equivalent to
4.46 x 10-8 mol nr3 of H202. For a cloud water content of 0.5 g
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 concentration 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 equilbrium. Because of this uncertainty in the value of the
4-51
-------
concentration in hydrometeors derived from gas-phase measure-
ments, values derived from direct measurements of this species in rain
and cloud water (Kok 1980, pers. comrn.) will be employed. The value
selected Is 0.5 ppn or ~ 1.5 x 10~5 mol A-l. Employing this
value in the Martin and Damschen rate expression for atmospheric
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
conversion 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 ^$04
(Duecker and West 1959, Schroeter 1966) and remains considerably
important in flue-gas scrubbing technology (Takeuchi et al. 1977).
Because NOJs and S02 commonly coexist 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
considerably more information on the conversion mechanism. Furthermore,
it was conducted 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 concentrations 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. Therefore, Martin et al.'s
expression will be employed as a significance test. This expression is:
HP ^n ~\
* = kl[H+]l/2 {[HN02] + [N02]}{[S02.H20] + [HS03] } [4-95]
dt
with ki = 142 £3/2 moT3/2 s"1 at 298 K. No activation energy
was determined by Martin et al. (nor by Oblath et al. for atmospheric
conditions); 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 characteristic time with respect to oxidation of S(IV) of 70 hr
for urban conditions. This reaction's importance to the H2S04
production in hydrometeors 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.
4-52
-------
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 63 and ^02 are highly reactive in water and can
suffer either catalytically or photochemically 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 H2S04 in Solution—Few
heterogeneous reactions in solution Tiave been proposed for H
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 S032~ occur in
relatively insoluble form in the atmosphere. Heterogeneous oxidation
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 S042~
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 H2S04 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 H2S04 production in atmospheric
hydrometeors:
= k [Cx] C02] ' g [S(IV) ] _ [4-96]
_
dt (i + &[s(iv)] + a[s(iv)]2)
where k = 1.69 x 10~5 mol -03 £°'69 g"1 s'1, a = 1.50 x 1012 £2 mol~2,
3= 3.06 x 10° x, ml"1, [Cx] = grams of carbon per liter, and [0?] and
LS(IV)] are in molar concentrations. The activation energy of the reaction
is given as 48 kJ mol~l.
It should be noted that the graphitic carbon used to derive
Equation 4-96 was Nuchar C-190, a commercial product with a well -
characterized elemental composition 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
4-53
-------
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 characteristic times that are
functions of the concentration of S(IV). Finally, use of Equation 4-96
requires an estimate of the graphitic carbon concentration in
hydrometeors. A recent direct measurement of elemental carbon in
rainwater collected in Seattle that was 2.4 x 10-4 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 mo! a~^t the
rate expression of Brodzinsky et al. yields a characteristic time for
S(IV) oxidation of ~ 103 hr. Therefore, this reaction should be of
little importance in H2S04 production in precipitation, although it
might be important in Togs of low liquid water content in urban areas.
4.3.5.5 The Relative Importance of the Various H?S04 Production
Mechanisms--!n sharp contrast to HC1 and HNOg production in
hydrometeors, numerous reactions are capable of producing significant
levels of H2S04 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 $0^2- 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 nr* (about the lowest liquid water content that can produce
precipitation in a warm cloud) is considered. HoS04 production by
0? (catalyzed and uncatalyzed), by 03, and by HgO^ oxidation of
SlIV) 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 H202 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 dependencies, 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 ^2 reaction dominates HpS04 production in polluted clouds,
with the possible 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
nr3, situated in a moderately industrial region. The parameter values
4-54
-------
TABLE 4-16. VALUES OF PARAMETERS USED TO ESTIMATE
H2S04 PRODUCTION IN A POLLUTED CLOUD
Parameter Value
Partial pressure of H2S04 1 ppb
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 mo-| £-1
Concentration of Fe 10-6 mo] $,-1
4-55
-------
-1 -1
PRODUCTION RATE OF H2$04 (Mole Is)
f
tn
O>
cu ro
3 CO
IQ O
o
-a CL
o c
— ' O
c+ O
n> 3
D-
O CD
S--K °
I" 2 -"
-s oo
S g
o
X
Q.
O>
3
C/l
o
ro
r+
3-
ro
-------
used in this case are listed in Table 4-17. The pH range is again
derived from Falconer and Falconer (1979) and the HoC^ concentrations
from rainwater measurements 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 concentrations 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
H2$04 production above a pH of ~ 5.2. Of course, this result is
quite sensitive to the concentration of Ho02 employed; further data
on this important parameter would be highly desirable. Nevertheless, it
is important to note that, on the basis of available field data and rate
studies, no one oxidant dominates H2S04 production in all
atmospheric situations.
Figures 4-4 and 4-5 show the time scale for acid produced in
solution to reach the concentration produced by direct absorption of
gases into cloud drops. This important point was approached in the
derivation of the S(IY) conversion rates necessary to produce
significant acid in solution. However, Figures 4-4 and 4-5 allow a more
precise estimate.
The maximum concentration of directly absorbed h^SOA in an
urban polluted cloud should be ~ 4.2 x 10-4 moi a-l (based on
the H2$OA and cloud water concentrations in Table 4-16, 1 ppb and
0.1 g m-3} respectively). For a mid-level cloud, the maximum
H2$04 concentration should be 4.4 x 10~6 mol £-1 (based on the
values for H^SO* and cloudwater in Table 4-17: 0.1 ppb and 1 g
nr3, respectively). These concentrations would be reached by the
Ho02 reaction alone in ~ 3 min for both urban and mid-level clouds
if the H202 were undepleted. With depletion, the time dependence of
H2S04 production is more complex, which is shown in Figures 4-6 and
4-7 for the urban and mid-level clouds. For an urban cloud (Figure
4-6), H2S04 production is dominated by H202 oxidation until the
H202 is completely depleted after about 2 min. Thereafter,
H2SO~4 production is maintained by catalyzed aerobic oxidation at a
much slower rate (solution pH is assumed to be 4.0). Indeed, it would
take roughly 41 min for the ^$04 produced in solution to reach the
concentration of the ^804 directly absorbed. In a mid-level cloud
(Figure 4-7), the ^2 concentration, even with depletion, is
sufficient to produce concentrations of H2S04 equal to those
produced by direct absorption in about 4.5 min. However, if the
solution pH is assumed to be in the upper half of the range listed in
Table 4-17, oxidation by 02 and 03 produces sufficient additional
H2S04 to reduce this time to ~ 1 min. These results suggest that
4-57
-------
TABLE 4-17. VALUES OF PARAMETERS USED TO ESTIMATE
H2S04 PRODUCTION IN A MID-LEVEL CLOUD
Parameter Value
Partial pressure of ^$04 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 H202 5.9 x 10-6 moi
Concentration of Mn 2 x 10~9 mol j
Concentration of Fe 3.3 x 10~8 mol
4-58
-------
-1 -1
PRODUCTION RATE OF H2$04 (Mole 1 s )
-p.
cn
03
CD
en
-s re
o> ro
3 on
ia o
o -a
-h -s
o
3 Q-
o. o
I r+
fD O
< 3
a>
Qi
O r+
— i (V
O
C
Q- -+i
(/) O
• -s
o>
o
E
1/5
O
X
Q.
3
<-h
(Si
O
05
-S
r+
(T)
•o
O
MD
O
I
DO
CX5
cn
C/)
o 01
i— •
c ro
cn
cn
•
CT>
cn
•
00
-------
-1
CONCENTRATION OF H2$04 (Mole 1 )
to
c
-^
cr>
o 3
c n>
CL
o.
• — - fD
o -a
— i fD
O 3
C Q-
CL fD
S O
Q> 05
r+
fD O
-a
ni
M o
33 3
• Q. 3
o
3
cu
3
-s
cr
a>
-a
o
-------
1
CONCENTRATION OF H2$04 (Mole 1 )
-------
not only the rate, but also the pH dependence of the HpSCk
production 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 (^0? oxidation) that will
not be self-limiting in the usual sense of the term. If absorbed
concentrations of H2$04, HN03, and HC1 are considered as well,
within a few minutes of cloud formation, cloud water pH'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--Probab1y the most important single
neutralization process in the atmosphere is the absorption-hydration of
NH3 by acid aerosols and hydrometeors and, in the case of
hydrometeors, the subsequent dissociation reaction:
iq = [NH4OH]
[NH4OH] = [NH4+] + [OH"] [4-97]
The preeminence of this neutralization process arises because HN3
is the only basic gas of widespread, substantial occurrence in the
atmosphere. The hydration and dissociation reactions are generally
assumed to be fast compared to acid production reactions in solution
(Scott and Hobbs 1967, Beilke and Gravenhorst 1978). Therefore, the
concentration of NH^ (and consequently OH~) 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 mo! s,"1 s'1 of ^$04, or
9.6 x 10"1° mol s~l in a 10 ym radius droplet. If a background
concentration of NH^ of 1 ppb (Levine et al. 1980) is assumed, the
rate of NH3 scavenging due to collisions with a 10 ym droplet will
be 8.25 x 10-15 mol s-le
Recent work by Huntzicker et al. (1980) suggests that the reaction
coefficient 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
4-62
-------
hydrated virtually instantly in solution, and the product ammonium
hydroxide (NH40H) dissociates with a rate constant of kd = 6 x 10
s-1 (Eigen 1967). Thus, after ~ ICT6 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
NHA+ 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, NHAOH
dissociation and water dissociation, the concentration of NH^ in
solution is given by:
[NH4+] = Ha pa Ka [H+] [4-98]
Kw
where Pa is the partial pressure of NH3, Ha the Henry's Law
constant for NH3, and Ka and Kw the equilibrium constants for
NfyOH and H20 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~L atnr1 at 298 K is too high by about a factor of ~ 5 for
atmospheric hydrometeors (due to interaction between dissolved NH3 and
C02 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+3. [4-99]
is yields a range of NH4+ concentrations from 1.65 x 10~4 to 0.8
l r1. Thus, 1.65 x 10'4 to 0.8 equivalent of acid could be
Thi
mol
neutralized by NH3 alone. However, a word of caution is in order.
While concentrations 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 hydrometeors 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 (CaCOs'MgCOs) are often
4-63
-------
substantial components of the atmospheric aerosol have led to the
assertion (Winkler 1976) that these minerals will act to neutralize
H2S04 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 significant quantities in urban atmospheres (Sumi et al.
1959, Kasina 1980), and Ca2+ 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 precipitation (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 SOd2" 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, threfore, act as
hydrometeor buffers in the atmosphere, much like N03- The absolute
amount of Ca and Mg available for such buffering is highly variable,
with Ca ranging from lO"'' to 10~4 mol £-1 and Mg fairly
uniformly a factor of 5 to 10 lower in both rainwater and cloud water
(Petrenchuk and Orozdova 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
H2$04, 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
4-64
-------
context of acidification of lakes, vegetation and soil, however, the
chemical form in which the deposition arrives at the surface is of
little significance, because precursor depositions are rapidly converted
to the secondary forms following deposition. The real significance of
atmospheric transformations in this case lies in the fact that the rate
of the deposition 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
suppositions of individually-calculated concentrations of multiple
plumes. Individual plumes may be referred to 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
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 parameterizations. 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
4-65
-------
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 pro-
cesses and plume interacti-ons 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 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 (S02, NO), and to any other chemical and meteorolog-
ical 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 catalysts. 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, atmospheric 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 coemitted
species (Schwartz and Newman 1978), but also to promote reactions of
primary emissions with background 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
4-66
-------
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.
4.4.2 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 counter-
parts. In theory, the explicit mechanisms method involves considering
of all significant reactants and their elementary reactions involved in
each mechanism of sulfate or nitrate formation. Concentration changes
by all chemical reactions 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, H02, 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., ^$04 and HMO^) but also catalysts and
significant intermediate species involved in the mechanisms. Of par-
ticular 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 forma-
tion. In a spatially homogeneous system (well-mixed plume) consisting
of n species, a total of 2n first-order, nonlinear, ordinary differen-
tial 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 inter-
actions 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.
Implemention of the explicit mechanisms method has many associated
problems. 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.
4-67
-------
Also, the spatial-temporal resolution of the current area source emis-
sions 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 ^0?). Detailed HC and
aldehyde measurements in the atmosphere'are not common. Input specifi-
cations 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 Demerjian 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. Despite this model's comprehensiveness, the
authors warn that it may be an oversimplification of the real atmos-
phere, which undoubtedly contains hundreds of organic compounds. 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 repre-
sentation 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
effectively include the role of the omitted species (Levine and Shwartz
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, temperature, 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
4-68
-------
meteorological variables such as turbulence and inhomogeneous mixing
generally is not simulated in laboratory experiments. 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 sulfates 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 (Gillani and Wilson 1980). Appropriate back-
ground 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 transformatins, as 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
4-69
-------
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.
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
NOX, 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 S02 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 nature and amounts of the acid precursors, other coemitted
species, the reactivity of the airmass 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
atmospheric 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 reactions governing S02 oxidation have the general form:
S02 + Ox + (M) ->• products •> S042', [4-101]
4-70
-------
where Ox represents the principal oxidizing agents; i.e., OH and
possibly H02 and R02 for gas-phase oxidation (Calvert et al. 1978),
and H20?, 03, and 02 for liquid-phase oxidation (Penkett et al.
1979); (M) represents 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 = _1_ (S042-) = ks • (S02), [4-102]
3t
where the fractional conversion rate, ks, depends on Ox»
oxidizing species. Parameterization of ks which is the goal of
empirical transformation models, is thus a representation of the
weighted contributions of factors which effectively determine the Ox
concentrations. It may be broken down by mechanisms into:
ks = ksG + kSL + kSHET> [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 parameterizations
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 k§G, and efforts to parameterize kSL have just begun. These
are discussed in the next section.
Similarly, the formation of the two principle secondary nitrates
(HN03 and PAN) are largely governed by the reactions
N02 + OH + HNOs [4-104a]
and N02 + RC002 + PAN. [4-104b]
Hence, their formation rates may be expressed as:
""HNOs = kHN03 • (N02) [4-105a]
rPAN = kPAN • (N02), [4-105b]
where the fractional conversion rates, k(j (N = HN03» PAN), depend on
the concentrations of OH and RCOO?, respectively. The parameterizations
of k|»j would represent the weighted contributions of the factors which
effectively determine these free radical concentrations. Empirical
parameterizations of k^ based on field data have not been formulated
or tested. Sensitivity of kN 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.
4-71
-------
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 of relationships between rs and SOg, and r*N and NOX.
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 which includes
concentrated emissions sources may give erroneous transformation
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 nonlinear 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 is independent of
S02. Clearly, ks is variable through its dependence on species such
as the OH free radical which are responsible ultimately for the
oxidation of S02. Therefore, the critical question is whether these
oxidizing agents are themselves dependent on S02« 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.
Subsequently, if there are no further fresh injections of S02 into
this plume, the formation of OH will be governed by the NOX-HC
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 NOX-HC chemistry on local S02 concentra-
tion 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 particu-
lar plume release. In the broader context of modeling of longer-term
averages or continuous emissions, possibly varying with time, and 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 S02 but
also of co-emitted species is implicit in the works of Cahir et al. 1982
and Hidy 1982.
4-72
-------
The significance of the role of the co-emitted species 1s illus-
trated in the following practical example. Suppose we wish to answer
the following question: "Will a 50 percent reduction of S02 emission
from source A for 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 in turn, will cause k« to be different. The answer to
the question, therefore, is "no", the cause of this apparent or ef-
fective nonlinearity is the indirect dependence of ks on S02 through
the correlation between co-emitted SO? and NOX. The 50 percent
reduction in S02 emission could also have been achieved by the use of
fuel of 50 percent lower sulfur content or by scrubbing S02 from the
combustion products prior to stack emission. To the extent that these
latter procedures may not have changed NOX emissions, k$ 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 which will profoundly influence downwind sulfate
formation is the composition of the air which the plume encounters
during mesoscale and long range transport. There is field evidence to
suggest 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
comparable 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 entrapment 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 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.
Based on theoretical considerations, the relationship between rN
and NOX is expected to be nonlinear, since kw depends on OH. for
example, which depends directly on the NOX chemistry. Results of
4-73
-------
fD
CO
CONCENTRATION (ppb)
c -a
-s —•
cr c:
a> 3
3 fD
co
•a •
fD
a. 3
CD O
-a a>
<-+ -••
fD CL
Q.
-a
ro
o>
o fD
3 -S
a>
3
3
Q-
i
INi
O
O
O
O
-s —• -s
O; OJ O
< 3 to
fD r+ CO
-S S
to "O —'•
fD —• 3
E D.
r+ 3
3- fD -O
O 0> O
d 3 -h
CO Q_ —I.
3" 1
-J. fD
r+ 3 to
3"
fD c~f O
3- -+,
T3 fD
C C -S
3 ~S ~J*
fD cr 3
to O) O)
=3 -S
OJ <<
O) C CO
cr 3 o
o
fD
OJ
•
1 — »
00
ume represents
O
cr
to
fD
T
<
fD
3"
fD
O
O
tn
O
3
|—
— 1
O
-h
CO
r+
•
( —
0
c
CO
and secondary
QJ
O)
rl-
ro
<<*
-s
fD
to
•O fD 3- O
O> 0> CO
-s ~a -—
Q) —' O-
cr c Q) ~a
—' 3 r+ O
fD fD OJ —'
to —i
< fD -S O)
fD -S fD 3
—' fD <-+
to O CO
-S O
O fD —' -"•
fD fD
3 X
-•• o
to fD
to CO
-j. co
O
3 O
N
-i. o
3 3
r+ fD
O
O) O r+
to r+ 3"
fD fD fD
Q- Q.
I —
Q) Q. fll
-SCO"
O -S 01
c -•• a.
3 3 -"•
Q.CQ fD
-•• 3 O
Q. S
Q. OJ fD
ai -•• -s
<< -s
». n
-s
O)
ro
o
o
CO
o
oo
CTi
O
00
O
O
O
t—O
VO
tn
o
o
tn
o
ro
o
o
ro
o
o
o
ro
o
o
tn
I
ro
o
ro
o
tn
SB
-------
recent smog chamber experiments suggest, however, that the nonlinearity
of rjy| 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 kjg may also need to reflect the make-up of the
air which an emission encounters during transport.
4.4.4 Some Specific Models and Their Applications
4.4.4.1 Detailed Chemical Simulations—Detailed 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. Therefore, the simulation is continued in
the Eulerian frame.
On a global basis, the troposphere is presumed to be clean and the
organic species most relevant to smog formation are carbon and monoxide
(CO) and methane (CH/j). Recently, a two-dimensional global model was
employed by Fishman and Crutzen (1978) to predict the global distribu-
tion of OH, H02, and CH302 radical concentrations. Predicted OH
concentrations were reasonably comparable with recent, measured
atmospheric concentrations (Sheppard et al. 1978). Altshuller (1979)
used this model for OH to investigate the variability 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 latitudes,
e.g., in the northeastern United States, Canada, and northern Europe,
large seasonal differences in sulfate formation by this mechanism
4-75
-------
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, ^03, 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 important 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 S02 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 1980,
Hov and Isaksen 1981, Stewart and Liu 1981). In these caculations,
proper simulations of the changing background air and of plume-
background interactions were necessary for at least qualitative
agreement with field observations. Levine (1980) 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
inhomogeneities 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 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, and that OH and peroxy radical (H02» R02)
concentrations, which play an important role in 03 formation, peak at
midafternoon in polluted regions.
4-76
-------
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 H?02 in the presence of "clpuds" to a highly lumped
gas-phase chemistry model. ^Og 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 precipitation. 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.
Overall, HN03 formation proceeded rapidly, principally by the
gas-phase mechanism, peaking at 13 percent hr-1 after 15 hr.
H2SC>4 formation rate during 90 hr of simulation ranged between 0.1
and 1 percent hr-1 by the gas-phase mechanism and between 0 and 1.8
percent hr'1 by the liquid-phase mechanism.
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 S02. Documentation of sunlight
as a dominant environmental factor governing sulfate formation in power
plant plumes (Gillani et al. 1978) has since been verified and widely
accepted and used. In particular, 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 (DOE 1979).
A sunlight-dependent model of the form ks « RT, 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 rate with sunlight, or
4-77
-------
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 parameter!'zations 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 ROg that oxidize
gas-phase S02 would be facilitated by reactions involving NOX
emissions and 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, airmass reactivity. Vertical plume spread, Azn, was chosen as a
measure of the extent of plume dilution. The resulting gas-phase
parameterization is:
kSG-(.03 +_ .ODRy • (Az)p • (03)0, [4-106]
where k$G is in percent hr"l, Rj is in kW m~2, (Az)n 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 ksg« The measured values were for dry
(relative humidity < 75 percent), cloudles 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.
4-78
-------
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 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 kS£
as shown in Figure 4-9. The magnitudes and temporal variations snown
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 S02 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 1983a). It
may be inferred, therefore, that liquid-phase mechanisms may convert
about 5 to 15 percent of the S02 to S042~ per day during summer in
the Midwest.
Gillani and Wilson (,1983b) 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, ksi . depends not only
on the chemical reaction rate within cloud droplets, KS. , 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 • KSL [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 pH, and
concentrations of dissolved S, oxidizing agents (^03, 03, and
03), 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$, under typical summertime fair-weather
4-79
-------
to
fD
CD
O
CO rh O O
c+ O O -••
3 C
&> o < -s
—' O> fD 3
O to —i
to CU 3 3
CO rt- Q.
H-J _*. o
-—-O -h to
• 3 fD
GO CO CU
C to
O- —' O
CU -h 3
to c Cu
fD -S —•
Q.
O
3
CQ . .
rt CU fD
3- oo
fD
-5
T3 to
T3 3"
CU CU O
~$ to —t}
CU fD
3 r+
fD 3 3-
r+ fD fD
fD O
• CU Cu
3 to
N
O>
to
3
-••30
O oo
3 -. T3
10 cu
Cu -S
a- n c+
"< n -••
o n
CD ~S —'
-"• Q. fD
—i 3
CU (Q
3
AVERAGE SULFUR CONVERSION RATE (% hr"1)
ro
CO
CO
O CD
O 3>
Z 00
«: i
m -o
73 OC
co 3>
t—I CO
o m
5?
CD -<
CD
>-• m
IX)
-~J o
CTl O
CO <
—I m
» 73
CO
I— I—I
o o
co
-------
convective cloud conditions in the Kentucky-Tennessee area. The
inferred value of K$. (summer daytime average conversion rate within
clouds) was 12 percent hr"1. This value compares with values of 0 to
104 percent hr"1 estimated by Hegg et al . (1980), based on ambient
S02 and SO^- measurements in wave cloud situations and with
predicted values ranging from 10 to 20 percent hr"1 in large storm
cloud systems in the summer based on an indirect mass balance technique
(Scott 1981). Also, the value of P averaged over 24 hr is expected to
be significantly less than 0.1 during summer as well as winter. In
other words, the average bulk plume conversion rate by liquid-phase
mechanisms is likely to be less than the local droplet-phase conversion
rate by more than an order of magnitude. All of these estimates involve
several assumptions and approximations and must be used with caution.
Values of K$. 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
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 parameteriza-
tion of ks which highlights the role of absolute humidity and
temperature:
ks - (PH20)3'°8 (P^O.sat)1'213.
where P^o denotes the partial pressure of water vapor, and
PHpO.sat denotes the saturation vapor pressure of water vapor (a
measure of temperature) .
No comparable parameterizations of NOX transformations have
been formulated. Summertime plume measurements suggest that N03~
formation is primarily in the form of HN03 vapor (Forrest et al. 1979,
1981; Hegg and Hobbs 1979b; Richards et al . 1981) and that oxidation of
N02 to HN03 may proceed about three times faster than does oxidation
of SO? 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
4-81
-------
one simmer 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 S042- formation presented by White et al.
(1976) and estimated peak HgSC^ and HMOs 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-1.
Spicer (1977a) estimated the N02-to-Products transformation rate
in the Los Angeles urban plume as 10 + 5% hr-1. jn more recent
measurements downwind of Los Angeles TSpicer et al. 1979), the observed
lower limit of NOX conversion rates ranged from 1 to 16% hr-1, with
typical rates in the 5 to 10 percent hr-1 range. Spicer (1980)
estimated NOX 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., HN03), 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. Concentrations of the oxidizing species (e.g., OH,
and H202) 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 ^02 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 ^02 reaction rate was enhanced by 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.
4-82
-------
Detailed chemical models are not yet practical for application in
regional models to predict acidic product formation and deposition.
Many individual pieces of information--microphysical 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 sul fate and nitrate formation (e.g.,
^2®? 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 models is
questionable. Furthermore, their application requires substantial
computational resources.
It appears that, for the foreseeable future, empirical parameteri-
zations will serve as transformation modules in regional models.
Preliminary parameter! zations have been developed only for $04
formation in power plant plumes, and will undoubtedly continue to be
improved. No practical parameter!' zations exist yet for N03~
formation or for urban plumes. Adherence to mechanistic considerations
is recommended in formulating the parameter!' zations. More, and more
reliable, measurements of such important variables as the atmospheric
concentrations of OH, H202, NHs, HC's, SQ^~ and NOs" and
of cloud dimensions and cloud chemical composition are needed direly.
and HN03 formation apparently peaks during daytime and
in summer. Gas-phase mechanisms are considered contribute a larger
share, on the average, to these secondary formations under warm, sunny
conditions. Typically, 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 S042~. 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
$04^" formation rate by gas-phase mechanisms. Aerosol N03~
formation rate is apparently very small, at least in the summer. Both
S042" and NOs" formation are faster in urban plumes.
The time has arrived to abandon the use of constant conversion
rates in regional models, at least for different seasons. In short-term
models, diurnal variabilities can also be resolved. We may not be able
to apportion secondary formations to different formation mechanisms
confidently, but we are at least reasonably comfortable with overall
conversion rates for average seasonal and diurnal conditions, at least
for S compounds in the summer. More atmospheric measurements are needed
of SOX transformations in the other seasons and of NOX transforma-
tions in all seasons.
4.5 CONCLUSIONS
The discussion of homogeneous gas-phase reactions has led to the
following conclusions:
4-83
-------
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 dissoci-
ation constants. More information is needed for assessment
(Section 4.2.1).
Acids (HX) produced from gas-phase reactions of halocarbons are
also expected to make insignificant contributions to regional dis-
position problems; their effects on global precipitation chemistry
is more plausible but uncertain. Direct anthropogenic emissions of
HX are potentially important (Section 4.2.1; Chapter A-2).
Oxidation of reduced forms of sulfur in the atmosphere generally
leads to sulfur dioxide (S02) formation (Section 4.2.1).
SOo oxidation in air is dominated by reaction with hydroxyl (HO)
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).
The average lifetime of S02 with respect to this reaction is
approximately 3-4 days (Section 4.2.2).
Of the remaining free-radical processes for S02 oxidation, only
the reaction by peroxy'alkyl radicals appears to have possible
atmospheric significance; additional information is needed for
assessment (Section 4.2.1).
Gas-phase oxidation of nitrogen dixoide (N02) leads to a variety
of products; nitric acid, dinitrogen pertoxide (^05) and
peroxyacetyl nitrate (PAN) are in greatest abundance. Nitrogen
trioxide 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 N20s is direly needed (Section
4.2.1).
The average lifetime of N02 with 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 meteorological effects. Gas-phase trans-
formation rates in power plant plumes are usually smaller than in
urban plumes because of imperfect mixing an 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 HO concentrations are not
satisfactorily defined except, perhaps, on a global scale. In
4-84
-------
polluted air, the ration of hydrocarbons (HC) to nitrogen oxides
(NOX) is expected to be dominant variable for the HO 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 che-
mistry affectign acidic species are understood, albiet 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 HN03 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).
° Production of H2S04 in solution within hydrometeors, by any of
several different mechanisms, can rival or even suppress direct
absorption of H2S04 by hydrometeors (Section 4.3.5).
° Of _the Carious production mechanisms for H2S04 in solution,
oxidation by H202 and by catalyzed and uncatalyzed aerobic
oxidation appear to be most important (Section 4.3.5).
0 While oxidation by H202 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).
0 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 CaC03, MgCOa 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:
° 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
4-85
-------
representations of the hydrocarbons, even though details of the
chemical mechanisms are not fully understood (Section 4.4.4).
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).
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 reli-
able 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, ^Oo, etc.) is not
understood at the present time (Section 4.4.4).
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 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 parameterizations will continue to predominate in
operational regional models. All models, to varying degrees use
prameterizations 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 esti-
mates 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 expres-
sions, which are constantly being improved. The state-of-the-art
of such prameterizations 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 residence
times of pollutants are probably associated with transport and
deposition processes rather than with transformation processes
(Sections 4.4.2 through 4.4.4).
4-86
-------
4.6 REFERENCES
Alkezweeny, A. J. 1978. Measurements of aerosol particles and trace
gases in METROMEX. J. Appl . Meteor. 17:609-614.
Alkezweeny, A. J. and D. C. Powell. 1977. Estimation of transformation
rate of S02 to 864 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 S0£ 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 CsHe and NOs 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. S02 removal by cloud
and fog drops as affected by ammonia and heavy metals. j£ Precipitation
Scavenging, Proceedings of a Symposium held at Champaign, 111., 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.
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.
4-87
-------
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-oxidation 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.
Car-michael, 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.
4-88
-------
Castleman, A. W., R. E. Davis, H. R. Munkelwltz, I. N. Tang, and W. P.
Wood. 1975. Kinetics of association reactions pertaining to sulfurlc
acid aerosol formation. Int. J. Chem. Klnet. Symp. 1:629.
Chang, S. G., R. Toorsl, and T. Novakov. 1981. The Importance of soot
particles and nitrous acid In oxidizing SO? In atmospheric aqueous
droplets. Atmos. Environ. 15:1287-1292.
Clark, W. W., D. A. Landis, and A. B. Marker. 1976. Measurements of
the photochemical production of aerosols in ambient air near a freeway
for a range of SC»2 concentrations. Atmos. Environ. 10:637-644.
Commins, B. T. 1963. Determination of parti oil ate acid in town air.
Analyst 88:364-367.
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 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.
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.
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.
4-89
-------
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. In Advances in Environ. Sci. and
Tech., Vol. 4, 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. Tech.
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. Met. 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.
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.
4-90
-------
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. In^
Fast Reactions and Primary Processes in Chemical Kinetics. Interscience
Publishers, New York, NY.
Eliassen, A. and J. Saltbones. 1975. Decay and transformation rates of
S02 as estimated from emission data, trajectories and measured air
concentrations. Atmos. Environ. 5:425-429.
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. Technol.
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.
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.
4-91
-------
Flack, W. W. and M. J. Matteson. 1979. Mass transfer of gases to
growing water droplets. In Polluted Rain, T. Y. Ton'bara, 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. Envir.
13:1287-1297.
Forrest, J., R. Garber, and L. Newman. 1981. Conversion rates ifi power
plant plumes based on filter pack data - Part I: The coal-fired
Cumberland plume. Atmos. Envir. 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. Technol. 8:731-734.
Frieberg, 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.
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.
Gillani, N. V. 1978. Project MISTT: Mesoscale plume modeling of the
dispersion, transformation and ground removal of $03. 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. 1983a. Gas-to-particle conversion of
sulfur in power plant plumes: II. Observations of liquid phase
conversions. Atmos. Environ. In press.
4-92
-------
Gillani, N. V. and W. E. Wilson. 1983b. Gas-to-particle conversion of
sulfur in power plant plumes: III. Parameterization of plume-cloud
interactions. Atmos. Environ. In press.
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. Chen. 82:254-268.
Graham, R. A., A. M. Winer, R. Atkinson, and J. N. Pitts. 1979. Rate
constants for the reaction of H02, S02, CO, ^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. Jh^ Chemical Reactions in
Urban Atmospheres. C. S. Tuesday, ed. American Elsevier, New York.
Hales, J. M. and D. R. Drewes. 1978. 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.
Hampson, R. F., Jr. and D. Garvin. 1977. Reaction rate and
photochemical data for atmospheric chemistry. NBS Special Publication,
U.S. Dept. of Commerce.
4-93
409-261 0-83-12
-------
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. In press.
Hecht, T. A., J. H. Seinfeld, and M. C. Dodge. 1974. Further
development of generalized kinetic mechanism for photochemical 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
sulfur 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
SO* in clouds. Quarterly Progress Report for April 1-June 30. EPA
Grant R805263010, Dept. of Atmospheric Sciences, University of
Washington, Seattle, WA.
Hegg, D. A. and P. V. Hobbs, and L. F. Radke. 1980. A preliminary
study of cloud chemistry, pp. 7-10. Jji Preprint Volume, Eighth 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.
Hidy, G. M., P. K. Mueller, and E. Y. Tong. 1978. Spatial and temporal
distributions of airborne sulfate in parts of the United States. Atmos.
Environ. 12:735-752.
4-94
-------
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., 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, jji Nitrogeneous Air
Pollutants. D. Grosjean, ed. Ann Arbor Science, Ann Arbor, 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.
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.
4-95
-------
Isaksen, I. S. A., E. Hesstvedt, and 0. Hov. 1978. A chemical model
for urban plumes: Test for ozone and parti cul ate 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 Techno!. 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. Chemi luminescence 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.
Tellus 26:206-212.
Junge, C. and T. G. Ryan. 1958. Study of SO? 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 methyl peroxy radicals. J. Phys. Chem. 85:1126-1132.
Kan, C. S., R. D. McQuigg, M. R. Whitbeck, andJ. Calvert. 1979.
Kinetic flash spectroscopic study of the CHsOe and CHa02 - S02
reactions. Int. J. Chem. Kinetics 11:921-933.
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 HN03 in rural air. Geophys. Res. Lett. 6:375-378.
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.
4-96
-------
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, and
Soil Pollut. 4:319-328.
Larson, T., N. Horiko, 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.
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.
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. C~hap. 1TT ln_ Trace Atmospheric Constituents, S. E.
Schwartz, ed. Vol. 12 in Advance's in Environmental Science and
Technology. John Wiley & Sons, Inc., New York.
4-97
-------
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. Techno!.
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 SO? 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. Met. 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.
McMurry, P. H., and J. C. Wilson. 1980. Growth laws for formation of
secondary ambient aerosols: Implications for chemical conversion
mechanisms. Atmos. Environ. In Press.
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.
McNelis, 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.
4-98
-------
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. Precurser 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. NY 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 PressfTTew York, NY.
Mills, E. L., C. B. Murphy Jr., and J. A. Bloomfield. 1979. Oxidants
in precipitation. lr± 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 SOg 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 Science. 1976. Chlorine and hydrogen chloride. In
Medical and Biologic Effects of Environmental Pollutants.
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.
4-99
-------
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. N02 in the stratosphere and troposphere by ground
based absorption spectroscopy. Science 189:547-549.
Noxon, J. F., R. B. Norton, and E. Marovich. 1980. N03 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 hydroxylamine disulfonate by reactions of
nitrites with sulfites. J. Phys. Chem. 85:1017-1021.
OECD. 1977. The OECD programme on long range transport of air
pollutants: Measurements and findings. Final Report. Organization for
Economic Co-operation 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 S02 and other atmospheric gases by
ozone in aqueous solution. Nature Physical Science 240:105-106.
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. Tellus 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.
Peuschel, R. V. and C. C. Van Valin. 1978. Cloud nucleus formation in
a power plant plume. Atmos. Environ. 12:307-312.
Platt, U., D. Perner, J. Schroder, C. Kessler, and A. Toennissen. 1981.
The diurnal variation of NOa. J. Geophys. Res. 86:11965-11970.
4-100
-------
Platt, U., D. Perner, A. M. Winer, 6. 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.
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.
Rasmussen, R. A., L. E. Rasmussen, M. A. Khalil, and R. W. Dalluge.
1980. Concentration distribution of methyl chloride in the atmosphere.
J. Geophys. Res. 85:7350-7356.
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. Met.,
16:53-56.
Roberts, P. T. and S. K. Friedlander. 1975. Conversion of S02 to
sulfur particulate in 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.
4-101
-------
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. Chem. Kinet. 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 U 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. Techno!. 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.
Schwartz, S. E. and W. H. White. 1982. Kinetics of reactive
dissolution of nitrogen oxides into aqueous solution. In Advanc.
Enviro. Sci. Technol., 12, S. E. Schwartz, ed. Wiley andTSons, Inc., New
York.
Scott, B. C. 1981. Predictions of in-cloud conversion rates of -S02
to $04 based upon a simple chemical and dynamical model. Atmos.
Environ. In press.
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.
4-102
-------
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 l^C 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. and Tech. 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. "M7 M. Benarie, ed. Elsevier Scientific Publishing Company,
Amsterdam.
Spicer, C. W. 1983. Smog chamber studies of NOX transformation rate
and nitrate-precursor relationship. Environ. Sci. and Tech. 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.
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. Poll. Control 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.
4-103
-------
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 C^-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$2,
CHsSCHs and H2$: 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, andJ. T. Parcher. 1968. Static
studies of sulfur dioxide reactions in air. Environ. Sci. Technol.,
2:611-618.
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 SO? 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.
104
-------
White, W. H., J. A. Anderson, D. L. Blumenthal, R. B. Husar, N. V.
Gill am', 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 Tecnical Report, EPA-6003-77-011.
Whitten, G. Z., H. Hogo, and J. P. Killus. 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 CSe- 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.
_I_n Proceedings of the Workshop on the Formation and Fate of Atmospheric
Nitrates. EPA, Research Triangle Park, NC.
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 N02S03 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 also can have toxic
effects in lakes and streams at sufficiently high concentrations
(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 substantial 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
5-1
-------
particularly important in 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 surface. These near-surface, emissions
plumes would have impacted on the adjacent countryside resulting in high
sulfur oxide concentrations in and near urban centers.
Coal usage declined immediately after World War II in the United
States. By the late 1940's and 1950's, the use of coal 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.
Increases in sulfur oxide emissions were seen 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 I9601s and early 1970's,
regulations were promulgated 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.
5-2
-------
The spread of cities Into suburban areas after World War II
resulted in more diffuse sources of urban plumes, although emission
sources in surburban 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 remote areas is unclear, although long-range transport is more likely
in winter (Chapter A-3, Section 3.4.2) because of unique 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
subsequent 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 (503). Sulfur
dioxide was monitored in various large cities in earlier years, but no
nationwide monitoring network existed until the I960'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 concentrations in winter than in summer. The diurnal
profiles showed midmorning and late afternoon peaks or early morning
peaks in $03 concentrations. Jacobs reported hourly S02
concentrations as high as 2500 to 3000 pg 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 pg 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 pg nr3, 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-19601s. This is not unexpected because fuel sulfur content was not
restricted during this time.
5-3
-------
In the 1963-72 period the decreasing order of annual average
concentrations 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 $02 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 $02 with time have been evaluated by geographical region and
in specific urban areas (Altshuller 1980). Between 1963-65 and 1971-73,
S02 concentrations (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 S02 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 had been reduced to relatively low levels. In 1976 the composite
annual average (and 90th percentile) concentrations were: United
States--20 yg nr3 (40 yg nr3), New England--25 yg m'3 (40
yg nr3); Great Lakes--28 yg nr3 (50 yg nr3) (U.S. EPA 1977a,
1978). 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.
5-4
-------
o -H
tn <<
3" fD
C CU
cn
ro
o
ro
cn
—• fD
fD -b Cu
-S O -S
I—> C C
VD -S 3
GO O" 3
O Cu -"•
v s —^ -^
to
CU
3 CU
Q- <
fD
3 -S
O QJ
3
co h-
i
CT>
cn
vo
01
cn
i
CT>
O
-1- X
3 -"•
Q.
rl- fD
3"
fD CU
3
3 Q.
O
""5 tn
fD -ti
CO d>
tn rt
rh fD
fD
-S O
3 O
3
d O
3 fD
-"• 3
rt rt-
fD -S
Q- CU
CO -"•
rl- O
CU 3
rl- tn
fD
to Q.
J= 3
0-tQ
O>
-O tn
rt- ft)
CD O
Q. O
3
-h Q-
O -Q
3 c
fD
-s
70
m
m
m
o
VO
CT>
cn
vo
CTl
VO
- o
VO
CO
vo
CO
I
tn
vo
*vl
cn
- O
o
- o
ro
cn
cn
O
cn
O
O
ro
cn
SULFUR DIOXIDE CONCENTRATIONS (yg m3)
-------
9-S
IQ
C
fP
cn
i
ro
I— > (T> 3"
WD Q) -S
00 -S ft)
o n>
-S CD
QJ
C ~i
-s
cr -s
QJ C
3 3
3
Q) _1.
3 3
CLCQ
3 0>
O <
3 CD
C -S
-S 0»
SULFATE CONCENTRATIONS (yg m3)
01 (V
3
to
to C
n> c
to -5
-•• Q.
3 -"•
O
3" -"•
n> o.
rp
3
O O>
-S 3
rt- 0-
3"
CD tn
O) C.
rt- -h
fD O>
-S r+
3 CD
C O
3 O
-"• 3
c-f O
CD CD
Q. 3
CO -S
0) O
CO 3
10
Q.
> C
CL -S
(D _i.
-a 3
.
CD
a.
-5 -S
O O-
> c
— ' O>
r+ -S
to c-t-
3- m
c -s
ro
-s
TO
m
en
m
CO
ro
o
vo
en
i
CT>
tn
vo
tn
i
I
av
en
co
CO
»vl
en
en
i
vj
vj
I
vj
IO
-0
o
•»«•
I
o
CO
O
X
en
en
O
VJ
en
O
O
SULFUR DIOXIDE CONCENTRATIONS (yg m3)
-------
z-s
SULFATE CONCENTRATIONS (yg m3)
on
OJ
vo cu -5
co -s ro
o CD
-s n>
Cu
c: -s
cr -s
cu c:
Q.CQ
3 Cu
O <
3 O>
-s
c: cu
-s (O
cr ft)
cu
3 (/>
CD -S
Q.
3 O
X
3- Q-
fD ro
3 CU
O 3
-S Q.
3" to
fD C
Cu —'
to -t,
r+ Cu
CD d-
-S fD
O
C O
3 3
-<• O
c+ 0>
CD 3
Q. c-h
OO CU
Cu -"•
rt- O
fD 3
t/> to
3=- -S
Q. --•
Qi 3
-a to
Q- O
c
-h -S
-S c+
O 3-
— i CU
r+ -S
3" ft)
c: -s
ro
o
PO
tn
30
m
73
«:
m
O
m
co
en
co
en
en
CTl
en
i
CTi
VO
CTi
^4
I
CTi
VO
CT>
VO
VO
»-J
t—'
I
^J
co
VO
CO
I
en
vo
«»j
en
ro
en
en
o
en
O
O
SULFUR DIOXIDE CONCENTRATIONS (yg m3)
ro
en
-------
8-g
ANNUAL SULFUR DIOXIDE EMISSIONS FROM COAL- AND OIL-FIRED POWER PLANTS
CQ
ro
en
ro >
Q> 3
to 3
rt- C
(T> OJ
&>
a> <
3 ro
Q- -s
o>
3 to
-j- ro
QL
ro -s
to cr
rt- O>
ro 3
-s
3 O)
cr cr
3 -i-
-i. ro
rt- 3
ro rt-
Q.
o>
00 -"•
rt- -S
o>
rt- O
ro o
CO 3
• o
ro
J= rt-
CL -s
O> O)
"O rt-
rt- -"•
ro o
CL 3
to
Q>
3
Q-
ro
rt- -••
to to
3" to
c -1-
— • O
— ' 3
ro to
o
'"^ -h
i— >
to to
00 C
o — •
-s
a.
o
X
Q.
ro
3
O
-s
rt-
•<
m
>
ANNUAL AVERAGE SULFUR DIOXIDE CONCENTRATION (yg
-------
TABLE 5-1. SULFUR DIOXIDE CONCENTRATIONS AT NONURBAN SITES
IN THE EASTERN UNITED STATES (in yg nr3)
(ADAPTED FROM NASN DATA BANK)
Site
First
quarter
Second
quarter
Third
quarter
Fourth
quarter
Annual
average
Acadia National Park, MA
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
ND
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
6 7 7
ND ND
12 8 -
799
499
ND ND
10 18
8 9 13
6 § 7
ND 8 -
6 11 10
9 11 11
11 8 11
7 10 11
5 19 9
6 7 9
5-9
-------
TABLE 5-1. CONTINUED
Site
First
quarter
Second
quarter
Third
quarter
Fourth
quarter
Annual
average
Jefferson County, NY
1970
1971
1972
1973
ND
8
3
8
ND
5
5
19
16
6
5
ND
ND
7
9
25
7
6
Monroe County, IN
1967
1968
1969
1970
1971
1972
1973
19
13
19
13
11
15
30
5
7
10
8
8
10
11
6
7
8
16
7
7
10
33
12
18
10
14
15
10
11
10
14
12
11
11
15
ND = not detectable.
5-10
-------
The annual average concentrations range near 10 yg m-3. 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 concentrations 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 between urban and nonurban S02
concentrations narrowed substantially in the 1970's.
Mueller et al. (1980) reported measurements from the Sul fate
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 S02 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 yg m-3 were shown
for small areas in August 1977 and January-February 1978. The highest
S02 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 S02 concentrations of 18 yg nr3
(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 m-3 during summer months to as high as 30 to 40 yg m-3 during
the winter months (Shaw and Paur 1982).
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 collected 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
concentrations 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,
which 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).
5-11
-------
Within Europe, annual mean S02 concentrations range from about 20
yg m-3 in rural areas of the United Kingdom, the Netherlands, and
the Federal Republic of Germany to concentrations of 2 yg m-3 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 SO? decreased from
about 5 yg m-3 at 2 to 3 km altitude down to 1 yg nr3 at 5 km
altitude. From other aircraft flights, Georgii and Meixner (1980)
obtained a mean concentration of 1.3 yg m-3 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 $03 concentrations as a function
of latitude over the Atlantic Ocean. Concentrations of SO? ranging
from 0.1 to 0.2 yg m-3 were observed at latitudes above 60bN and
below 10°N in the northern hemisphere as well as in the southern
hemisphere. Between latitudes of 10°N and 60°N over the Atlantic Ocean
S02 concentrations increase to 1 yg m-3 at 25°N and at 55°N
latitude and peak at about 3 yg m-3 at 40°N latitude. These large
increases in S02 concentrations at midlatitude were attributed to
continental emission sources. Other investigations resulted in
concentrations of S02 averaging 0.3 yq nr3 over the Pacific Ocean
and 0.2 yg m-3 over the 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 southern Pacific Ocean and the western
section of the United States and Canada. The average S02
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
trophosphere, 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 m-3 over the tropical Indian
Ocean to 0.3 yg nr3 over the Peruvian upwelling. A relationship was
identified between the atmospheric S02 concentrations and the
biological activity in sea surface waters (Bonsang et al. 1980).
The SO? 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).
5-12
-------
A similar impact of large plumes from strong source areas has been
observed at several rural Canadian sites (Barrie et al. 1983).
5.2.3 Sulfate
5.2.3.1 Urban Concentration Measurements—In 1963 the National Air
Sampling Network collected partlculate matter on high-volume (h1-vol)
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 S02 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 S02 concentration, the
air volume per unit area of filter surface, temperature, and other
parameters (Coutant 1977, Mesorole et al. 1976). The conversion of
S02 to sulfate on clean glass-fiber filter surfaces was sensitive to
temperature but showed little dependency on humidity. 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 filters
under usual flow conditions in hi-vol 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 substantial, 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
increases 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 yg 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 S02
5-13
-------
were small compared to urban concentration levels, regional background
levels of sulfate have been substantial 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 sulfute 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.
The urban excess, the difference between the average urban and the
average regional (nonurban) sulfate concentration in a region, decreased
substantially 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 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
sul fates were developed.
5-14
-------
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 Philadelphia, PA, Chicago, IL, Charleston, WV, and Secaucus,
NJ, the sulfate appeared to be in the form of ammonium sulfate
[(NH4)2$04] (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 (NH/^SCty in winter (February 1977), but to
have the more acidic average composition of letoricite,
(NH4)3H(S04)2» in the summer (August 1976). These investigators
also found sulfate to be highly correlated with ammonium in both summer
and winter aerosols. Lioy et al. (1980) during a high sulfate episode
in the east on August 3 to 9, 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 suggesting higher ammonia fluxes
in and near New York City.
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 ^4)2804.
In earlier measurements in the Los Angeles area during 1972 and
1973, sufficient ammonium ion appeared to be present to neutralize the
sulfate to (NH4)2$04 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
H2$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 ^4)2804, but that its composition is
variable with more of a tendency toward acidic species in the summer.
5.2.3.3 Nonurban Concentration Measurements--Althsuller (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 m-3, and those at
western sites averaging 2.6 yg nr3. At 10 percent of western sites,
annual average concentrations 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
5-15
-------
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 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 consistently for nonurban sites in the Ohio Valley
area.
Although urban sites showed decreases in sulfate concentration
during the winter quarters, presumably owing 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 fr$m 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 between
1970 and 1978 of sulfur emissions in most midwestern states (Chapter
A-2, Table 2-14).
A plot of the regional distributions of nonurban sulfate concen-
trations averaged from months in 1977 and 1978 are shown in Figure 5-5
(Hilst 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 concentra-
tions 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
5-16
-------
1-HOUR
S02 (ppb)
24-HOUR
2" hig I"'3
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)-
5-17
-------
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 (1982) reported monthly average
concentrations ranging from as low as 3 yg m-3 in some winter months
up to 12 to 15 yg m-3 in the summer months. The seasonal 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 midafternoon 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
yg m-3, and occurred after passage of a cold front.
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 southeastern 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 SO? 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
5-18
-------
the 24-hr average sul fate concentrations 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 S02 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 (Barrle et al. 1980).
Concentrations of sulfate as a function of percentage cumulative
frequency are plotted In Figure 5-6 (Barrle et al. 1983). Results from
Canadian sites from the period November 1978 to December 1979 are
compared with those obtained 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—Charl son et al. (1974)
reported evidence obtained from 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 (Char!son et al. 1974, 1978a). The sul fate 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 sul fate aerosol (Charlson et al.
1974).
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 [NH4+]
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 November, which were
predominantly in the form of (NH/i^SOA. 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 for 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 portion 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 July 29-30, 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.
5-19
-------
100
50
m 10
i
e
oo
LU
-------
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 m"3 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 NH4HS04. 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+)/(s042~) molar ratios ranging from 0.5 to 2.0 with strong
diurnal variations. The particles were most acidic in mid-afternoon 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 July 24 and August 11, 1977 (Pierson
et al. 1980a). On the average, the [H+] was slightly in excess of
[NH4+], corresponding to a composition near that of NH4HS04.
The concentrations of the other cations were so low that [H+] and
[NH4+] were the predominant cations associated with [S042"], and
the sum of [H+] and [NH4+] was essentially stoichiometric with
[SO*2-]. For sulfate concentrations above 15 yg m~3 the [H+]
to [SOd ] mole ratio was between 1:1 and 2:1 and approached 2:1 for
several samples. Therefore, appreciable amounts of ^$04 must have
been present at the high sulfate 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).
Evidence of strong acidity comes from samples collected at various
rural sites in the northeastern and midwestern United States between May
and September, 1977. In several investigations, the tendency was for
the higher ratios of [H+] to [NH4+] to occur concurrently with the
higher sulfate concentrations (Stevens et al. 1978, 1980; Pierson et al.
1980a).
5-21
-------
The only samples collected at Brookhaven National Laboratory in
October and November 1975 were of low acidity (Tanner et al. 1977). In
samples collected on the Swedish Coast, Brosset (1978) also obtained low
[H+]-to-[NH4+] ratios for winter samples. During "white" winter
episodes, the [H+]-to-[NH4+] ratios rarely exceeded 1:1 and
frequently were well below 1:1. The species observed by X-ray
diffraction included (NH4)2S04, (NfahHtSO/ib, and
NH4HS04 (Brosset 1978).
In general, there appears to be substantially more evidence for
strong acidic species at rural sites than at urban sites, and the
highest acidities were those measured at mountain sites.
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 nr3 over the Atlantic Ocean. The sulfate concentration as a
function of latitude have two maxima. One of these occurs near 40°N
latitude where SC^ also has a maximum concentration and the other
occurs south of the equator. Around 40°N the sulfate concentration is
2 yg nr3, but decreases below 1 yg m'3 above 50°N. He estimated
sulfate concentrations of about 0.3 yg m"3 over clean areas in the
Northern Hemisphere.
Gravenhorst (1978) obtained an average sulfate concentration of
excess sulfate (excluding the contribution of sea salt) of 0.9 yg
nT3 ± 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 yg m~3. During another
period when air masses had passed over the British Isles upwind, the
excess sulfate averaged 1.07 yg m~3.
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 troposhere, < 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.
5-22
-------
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 1n
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, Philadelphia, PA, and New York, NY. The
sulfite content of the samples did not exceed the minimum detection
limit of 8 ng nr-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 pro)
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 particulate 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 ym, with nearly 90 percent of the sulfate below 3.5 ym.
Wagman et al. (1967) obtained sulfate size distributions at sites
in Chicago, IL, Cincinnati, OH, and Philadelphia, PA, during 1965. Lee
and Patterson (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 ym.
The higher MMD in Philadelphia was attributed in part to dust generated
from road construction near the site. Eighty percent of the sulfate was
below 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 below
0.2 ym at lower humidities to 0.6 to 0.8 ym at higher humidities at
three midwestern 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
5-23
-------
iodimetric nricrocoulometric tltration for the hydrogen sulflde. Average
MMD's were computed from measurements at several sites 1n 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 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 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 particles 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 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 September 29 and October 10,
1970. Seven-stage Andersen cascade impactors were used at 28 a
min-1, with either Gelman 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
5-24
-------
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 ammonium. 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 hi-vol
sampling. The diffusion battery was used to classify particles by size
below 0.25 ym before filter sampling and analysis. During the summer
month, about 50 percent of the sulfur-containing aerosols were below
0.25 ym; during the winter month only 25 percent were below 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
below 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 sul fates 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 sul fates, these estimates would form upper
1 imits.
Separation of particles into two fractions with a fine fraction
consisting of particles below 3.5 ym involves use of a virtual
impactor or dichotomous sampler (Stevens et al . 1978). The percentages
of sulfur found in the size range below 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 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
particle 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 (NH4)2S04 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
5-25
-------
nephelometer with humidity control (humldograph). Because the
nephelometer would respond to particles predominantly in the optical
range, 0.1 to 1 m, the technique associates acidity with
subm1cron-s1ze acid sulfate particles. In subsequent work in the St.
Louis area, well over 90 percent of sulfur 1n particles measured at
rural sites near St. Louis were found to be in the fine particle
size range with little, 1f 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 below 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 b1sulfate 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 distribution curves for sulfate and hydrogen ion were
almost identical, with an average MMD of 0.8 ym. About 90 percent of
the sulfate and hydrogen ion content was below 3 ym. The
[H+]-to-[S042-] ratios were somewhat higher for particles between
0.7 and 1.1 ym than for those below 0.7 ym, or between 1 and 2 ym.
Acidity was measured in even larger particles but the [H+] to
[SOA^-] ratio was lower than for particles below 2 ym (Pierson et
al. 1980a).
Aircraft outfitted with particle sizing equipment were flown across
portions of Arizona, Utah, Colorado, and New Mexico on October 5 and 9,
1977 (Madas et al. 1980). The MMD for sulfur In the collected
particles was not reported, but can be approximated as below 0.5 ym.
Sulfur particles below 1 ym constituted 92 percent of the sulfur
content.
5.2.4.3 Measurements at Remote Locations—Gravenhorst (1978) found the
excess sulfate in marine aerosols to be present In submicron-size
particles. The sulfate associated with sea salt was present In
supermicron particles. Meinert and Winchester (1977) also found the
excess sulfate to be present 1n submlcron-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).
5-26
-------
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, particularly 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 systans 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 atmosphere or, if present, occur at parts-
per-trill ion concentrations. The exceptions are the peroxyacetyl
nitrates (PAN), which are present at significant concentration levels
relative to the other nitrogen oxides and their acids. Because the
peroxyacetyl nitrates and their precursors are in reversible
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 (NO;?)- 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 NOg 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 limitations, it is not useful to separate historical trends
from current measurement results.
5-27
-------
5.3.2.2 Measurements Techniques-Nitrogen Ox1des--Most of the nitrogen
oxide measurements made during the 1970' s involved use of chemi 1 umi -
nescent analyzers. While the chemi luminescent technique can be used to
analyze nitric oxide directly and specifically, analysis of nitrogen
dioxide or nitrogen oxides (NO + N02) 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 chemi luminescent 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 chemi luminescent 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 quantitative, but somewhat variable chemi luminescent 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 converters, 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
chemi luminescent 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 N02-
Trends in NO? concentrations at the six CAMP sites in Chicago,
IL, Cincinnati, OH, Denver, CO, Philadelphia, PA, 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 NOa at the sites ranged from 50
to 150 yg m~3 with the higher concentrations occurring at the sites
5-28
-------
in downtown Los Angeles and in Chicago. The maximum 1-hr N02
concentrations at these sites ranged from 200 to 1500 yg m-3. Peak
1-hr concentrations above 750 pg 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 m-3.
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 concentrations 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 m-3 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
5-29
-------
TABLE 5-2. MEASUREMENTS OF CONCENTRATIONS OF NITROGEN OXIDES AT SUBURBAN AND RURAL SITES
GO
o
Site (Type)
Montague, MA (R)
Ipswhich, MA (R)
Scranton, PA (S)
DuBois, PA (R)
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
Ni trie
yg in-
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
yg
Mean
7
2.6
11
19
5.1
11
5
4
4
10.6
14.3
dioxide,
nr3
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
co
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.)
June-Aug. 1974
(chemilumin.)
June-Aug. 1974
(chemilumin.)
June-Aug. 1974
(chemilumin.)
July-Aug. 1974
(chemilumin.)
May-Dec. 1980
Nitric
yg in-
Mean
10
2.7
ND
1.9
ND
ND
ND
6.0
4.3
oxide,
Max.
249
NA
ND
17
ND
ND
ND
64
NA
Nitrogen
Mean
13
6.4
1.8
4.9
13
12
13
27
15.6
dioxide,
m~3
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.)
-------
TABLE 5-2. CONTINUED
CO
ro
Si
Franklin
(R)
Union Co
Giles Co
Creston,
Wolf Poi
Pierre,
site 40
Pierre
Jetmore,
te (Type)
Co., IN
., KY (R)
., TN (R)
IA (R)
nt, MT (R)
SD (R),
km WNW of
KA (R)
Period of
measurement
(method)
May-Dec. 1980
(chemilumin.)
May-Dec. 1980
(chemilumin.)
Aug. -Dec. 1977
(chemilumin.)
June-Sept. 1975
(chemilumin.)
June-Sept. 1975
(chervil umln.)
July-Sept. 1978
(chemilumin.)
April-May 1978
(chemilunrfn.)
Nitric
Mean
3
2
5
4
< 1
< 0
1
.0
.5
.7
.0
.25
.2
Oxide,
Max
NA
NA
96
28
NA
NA
NA
Ni trogen
yg
Mean
14
12
11
4
1
2
7
.3
.3
.3
.5
.3
.5
Dioxide,
nr3
Max
NA
NA
55
25
NA
NA
NA
Reference
Shaw et al . 1981
Martinez and
1979
Martinez and
1979
Decker et al.
Decker et al .
Kelly et al.
Martinez and
1979
Singh
Singh
1976
1976
1982
Singh
R = Rural.
S = Surburban.
ND = Not determined.
NA = Not available.
-------
St. Louis area (U.S. EPA 1982). During an air pollution episode in St.
Louis during October 1 and 2, 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 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
yg nr3, while the concurrent concentrations of nitrogen dioxide at
the outermost rural sites, 40 km from the center of the city, ranged
from 20 to 40 yg nr3. Similarly the 24-hr average concentrations of
nitrogen dioxide ranged from 200 to 265 yg nr3 at urban sites but
averaged only 20 yg nr3 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
concentrations 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 cocentrations were infrequently above 60
yg nr3. The highest nitrogen dioxide concentrations 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 concentrations of nitric oxide, and (3) the ratio of nitric
oxide to nitrogen oxides (NO + N02). 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 considered 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 concentrations 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 concentrations 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 measurements for nitrogen oxides from a number of studies carried out
5-33
-------
TABLE 5-3. CUMULATIVE FREQUENCY DISTRIBUTION OF HOURLY CONCENTRATIONS OF
NITROGEN DIOXIDE AT RURAL AND SUBURBAN LOCATIONS
Site/reference
DuBois,PA
Research Triangle
Institute 1975
Bradford, PA
Decker et al. 1976
McHenry, MD
Research Triangle
Institute 1975
en
do Wooster, OH
4:1 Research Triangle
Institute 1975
Measurement
period
June-Aug. 1974
July-Sept. 1975
June-Aug. 1974
June-Aug. 1974
Percent of hourly average concentrations
greater than stated
20 yg m~3 40 yg nr3
13.2 1.0
2.1 0.1
6.9 0.2
23.8 6.9
concentrations
60 yg m~3 80 yg m"3
0.2 0.0
0.0 0.0
0.1 0.0
1.9 0.3
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
June-Aug. 1974
June-Aug. 1974
July-Sept. 1975
July-Sept. 1975
July-Sept. 1975
5.6
14.9
0.5
2.6
0.1
1.1
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
-------
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
yg 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.
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
nr3 in clean air, while Bellinger et al. (1982) reported nitrogen
oxide concentrations in a number of clear air masses passing this site
below 0.1 yg nr3. in contrast, Kelly et al. (1980) observed
nitrogen oxide concentrations up to 40 g 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 concentrations of
nitrogen dioxide of 10 yg m~3 and a maximum daily average value of
about 3 yg nr3. similarly at Loop Head the concentrations of
nitrogen dioxide measured in maritime air by Platt and Perner (1980)
were below 0.3 ug m"3, in other air masses they measured nitrogen
dioxide concentrations from 4 to 5 yg m~3. 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
Volz (1982) measured nitrogen dioxide concentrations in the 0.1 to 1
yg m"3 range. Kley et al. (1981) measured nitrogen oxide
concentrations at 7 km over the vicinity of Wheatland, WY, as low as
0.1 yg m-3. During the 1977 and 1978 GAMETAG flights, nitric oxide
concentrations equal to or below 0.1 yg 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 contribute significantly to
the nitrogen oxide concentration levels in eastern North America.
5.3.3 Nitric Acid
5.3.3.1 Urban Concentration Measurements--Mitrie acid (HN03)
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 and Fourier transform infrared
spectroscopy (FTIR) with a detection limit of 6 ppb (Tuazon et al.
5-35
-------
TABLE 5-4. CONCENTRATIONS OF NITROGEN OXIDtS MEASURED AT REMOTE LOCATIONS
co
CTi
Sites
Colorado, USA
Niwot Ridge
Colorado, USA
Niwot Ridge
Colorada, USA
Fritz Park
Island of Hawaii
Mauna Kea
La ramie, 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
Concentratic
in yg nr>
N02
NA
NA
< 0.2
NA
2
06 NA
0.8
0.3
0.16
)ns
N0xa
0.4-0.5
< 0.1
NA
0.2-0.5
ND
0.2-0.8
NA
ND
NA
Remarks Reference
Kelly et al . 1980
Bol linger 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
Sites
Concentrations
Measurement
period
(method)
1n yg n
3
NO
N02 N0xa
Remarks
Reference
tn
i
CO
Tropical Areas
1965-1966
(colorlmetrlc)
0.1-0.6 0.4-0.8
0.3-0.5 0.6-0.9
0.3-0.8 0.6-0.1
0.3-0.8 0.6-0.9
Under
canopy of
forest
Above
canopy of
forest
Rlverbank
Seashore
and
maritime
Lodge and Pate
1966, Lodge et al
1974
*NOX = NO + N02.
-------
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 on Claremont, CA, during an 8-day period in August and September
1979 (Forest et al. 1982; Spicer et al. 1982a). The methods compared
included chemiluminescence, infrared, diffusion denuder, and filtration
techniques. The nitric acid concentrations ranged from 1.85 to 37.05
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 nr3 (Table 5-5). The Riverside site where
the highest ammonia 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 m~3 and Dayton, OH, 134 yg m~3 (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 afternoon 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 Npnurban Concentration Measurements--Measurements of nitric
acid at suburban and rural sites 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 coulometry. The average concentrations of nitric
acid were in the 2.6 to 5.2 yg m~3 range. The maximum concentration
of 116.1 yg nr3 reported at New Carlisle appears to be too high.
5-38
-------
TABLE 5-5. CONCENTRATIONS OF NITRIC ACID, PEROXYACETYL NITRATE
NITRATE AND AMMONIA AT URBAN SITES IN THE UNITED SJATES
Concentrations, ug
Site
West Los Angeles, CA
(Cal. State Univ.)
West Covina, CA
Claremont, CA
en (Harvey Mudd College)
GO
<£> Claremont, CA
(Harvey Mudd College)
Riverside, CA
(UC Riverside)
Riverside, CA
(UC Riverside)
St. Louis, MO
Dayton, OH
Period of
year
June 1980
Aug-Sept. 1973
Oct. 1978
Aug-Sept. 1979
Oct. 1976
July-Oct.
July-Aug 1973
July-Aug 1974
HN03
Avg
18.1
7.7
41.3
20.6
5.2-12.93
12.9-18. la
7.7
15.5
Max
30.0
103.2
126.4
56.8
20.6
51.6
206. 4b
139. 3b
Avg
35
10
25
20
45
30
10
ND
PAN
Max
80
95
185
55
90
90
95
ND
m-3
NH3
Avg
2.1
2.1
5.6
0.7-2.83
14.0
14.7
2.8
ND
Max
5.6
9.1
21.0
8.4
42.0
92.4
11.2
ND
References
Hanst et al. 1982
Spicer 1977
Tuazon et al .
1981b
Tuazon et al .
1981a
Tuazon et al .
1978
Tuazon et al.
1980, 1981a
Spicer 1977
Spicer et al.
1976a
ND = Not determined.
aMany individual values were below detectability limits (OL); lower concentrations listed based on
assuming values below DL equaled zero; upper concentration values listed based on assuming values below
DL equaled following concentrations: HN03, 12.9 g nr3; PAN, 10 g nr3; NH3, 2.1 g nr3.
bThese values appear unusally high when compared with NOx. PAN and 03 concentrations reported as
present during same time periods.
-------
TABLE 5-6. MEASUREMENTS OF CONCENTRATIONS OF NITRIC ACID, PEROXYACETYL
NITRATE AND AMMONIA AT SURBURBAN AND RURAL LOCATIONS
Concentrations, yg m~3
Site
Beverly Airport,
MA (S)
Van HI Seville, NJ
(R)
en Luray, VA (R)
0 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- June
. 1979
1980
1980
HNOa
Avg
2.6
< 2.1
1.0
2.1
2.1
5.2
1.8
0.8
1.3
2.4
Max
1 5
11
2
2
38
116
9
< 2
~ 5.2
~15.5
.2
.6
.1
.4
.7
.1 .
.8
.6
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
Avg
ND
ND
1.3
0.4
< 0.7
ND
0.4
0.8
0.6
0.9
NH3
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
en
i
Site
Abbeville, LA (R)
Commerce City, CO
(S)
Thurber Ranch, AZ
(35 mi. SE Tucson)
Pittsburg, CA (S)
Concentrations, ng -3
Period of HN03 PAN NHa
measurement Avg Max Avg Max Avg Max References
June-Aug. 1979 1.8 NA ND ND 2.1 NA Cadle et al
Nov. -Dec. 1978 2.1 NA ND ND 1.3 2.9 Cadle et al
July-Aug. 1981 1.6 5.2 ND ND 0.8 1.5 Farmer and
1982
February 1979 2.1 4.1 ND ND 0.4 0.8 Appel et al
. 1982
. 1982
Dawson
. 1980
ND = Not determined.
NA = Not available.
-------
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
use of either a nylon or Nad-impregnated filter to collect HMOs-
Positive interference problems are known to occur because of loss of the
nitrate from the particulate collected on the prefilter owing to
volatilization onto the filter used to collect HN03. The range of
nitric acid concentrations was from 0.7 to 3.9 yg m~3 (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 (5.2 yg m-3) 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.
In this same study (Spicer et al. 1982c), aircraft flights were
made following 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 (HN03 and
N03~) was determined by use of the Teflon filter prefilter and
nylon backup filter collection method. These authors suggested that the
potential for loss of nitrate off the Teflon filter 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
inorganic 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 the
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 (0.2 yg m-3). Nitric
acid is effectively adsorbed 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
5-42
-------
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, show 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
concentrations of nitric acid and ammonia at their measuring sites were
too low to result in the formation of ammonium nitrate in particulate
matter.
Kelly and Stedman (1979b) measured nitric acid by a
chemiluminescent technique at a rural site about 15 miles east of
Boulder, CO. The nitric acid concentrations during February 1978
usually were in the 1.3 to 12.9 yg nr3 range with many of the
concentrations of nitric acid in the 2.6 to 5.2 pg nr^ 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 occurring 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 concentrations 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 concentrations 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 relatively 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
5-43
-------
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 m-3 (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.
5.3.4 Peroxyacetyl Nitrates
Peroxyacetyl nitrates can be determined by electron capture gas
chromatography down to the 0.1 ppb 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
concentrations 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 concentrations 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 concentrations. Two peaks were observed at the
site in Riverside, CA. The earlier peak was associated with formation
of peroxyacetyl nitrate from local emissions while the later peak was
associated with formation of peroxyacetyl nitrate from emissions in air
5-44
-------
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
yg m~3
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
5-45
-------
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 Reference
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
Lonneman et al. 1976
Lonneman et al. 1976
Westberg et al. 1978a
HAOS 1979
Martinez et al. 1982
Brennen 1980
Singh et al. 1979
Singh et al. 1981
Singh et al . 1981
Singh et al. 1982
Singh et al. 1982
Singh et al . 1982
Singh et al . 1982
Singh et al. 1982
5-47
-------
rural and remote locations 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, NY, Badger Pass, CA, and Point Arena,
CA, are about 0.5 yg m~3.
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, measureable
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.
5.3.5 Ammonia
Unlike nitric acid and peroxyacetyl nitrate, which are formed
through atmospheric 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 interferences
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
5-48
-------
TABLE 5-9. PEROXYACETYL NITRATE MEASUREMENTS AT RURAL AND REMOTE SITES IN THE UNITED STATES
tn
Site
Wilmington, OH
H'jntington Lake,
IN
East Central
Missouri
Sheldon Wildlife
Preserve, TX
Jetmore, KA
Reese River, NV
Rodger 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 -marl time
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, vg n~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 et
1976
Splcer et al
Splcer et al
Westberg et
1978a
Singh et al.
Singh et al.
Singh et al.
Singh et al.
Singh et al .
al.
. 1983
. 1983
al.
1979
1979
1979
1979
1979
ND = Not determined.
NA = Not available.
-------
(McClenny and Bennett 1980). Harvard et al. (1982) also used the
acoustic detector. The tungstic acid technique was used by McClenny et
al. (1982) to measure ammonia. Gaseous 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) owing 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
substantially different from those measured at locations in the Los
Angeles Basin area other than the Riverside area. Concentrations of
ammonia remain 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 immediately 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 southeastern Virginia
at a site bordering the Great Dismal Swamp (Harward et al. 1982). The
ammonia concentrations obtained in August and September 1979 ranged from
5-50
-------
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 observed 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
be explained by a 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 particulate nitrates. During recent years
substantial positive and negative artifacts have been identified as
occurring 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.
(2) Negative artifacts derived from
(a) reactions of particulate nitrate in the collected matter
with strong acids in the particulate 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 measurements reported in the literature are likely to be
questionable, if not erroneous.
Most of the early measurements of particulate nitrate involved
analysis for nitrates on samples collected on glass fiber filters in
high volume (HIVOL) samplers (National Academy of Sciences 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
5-51
-------
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, Forest 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 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
(Harker 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 filters, 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 vg m"3. 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).
5-52
-------
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
investigate the losses of nitrate off Teflon filters loaded with
submicron (<_ 0.2 ym) 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 ammonium nitrate. Losses of
about 90 percent of the nitrate occurred when the filters were exposed
to 17 to 23 ppb of hydrochloric acid. Forest et al. (1980) observed
losses of preloaded nitrate from Pall flex Tissuquartz exposed 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 satisfactorily 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 demonstrated 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, Forest 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
(National Academy of Sciences 1977, U.S. EPA 1982). The same problem
probably applies to the measurements on cellulose filters used to
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
5-53
409-261 0-83-14
-------
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 diffusion 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 July 23 and July 27, 1979 at
Claremont, CA (Harvey Mudd College). The particulate nitrate peaked in
concentration during the late morning hours. Particle nitrate
concentrations exceeded nitric acid concentrations between 2200 and 1200
hours. The average particle nitrate concentration during this period
was 25 yg m~3. The average particle nitrate concentration
moderately exceeded the average nitric acid concentration.
Forest 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
August 27 and September 3, 1979. The particulate nitrate concentrations
tended to peak in the morning hours. The 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.
5-54
-------
5.3.6.2 Nonurban Concentration Measurements—Discussion earlier in this
section notes that the nitrate cocentrations obtained at nonurban sites
using glass fiber filters 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
m~3 of nitrate at rural sites, and during the spring and summer months
the nitrate concentration 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~3, 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 within the Research Triangle Park, NC during 16 days
in June, July, and August 1980. The assembly used contained a cyclone
to remove coarse particles. 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, Forest 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 m~3 and the particulate nitrate was
37 percent of the total inorganic nitrate. The average particulate
nitrate concentration at this nonurban site was 4 percent (Appel et al.
1981a) and 7 percent (Forest et al. 1982) of the average particulate
nitrate concentrations measured in Claremont, CA.
Tanner (1982) used the same diffusion denuder assembly arrangement
as Forest et al. (1982) at a site within Brookhaven National Laboratory
on Long Island, NY. Measurements of nitrates were made several hours
each day on November 7,8, and 9, 1979. The average particulate nitrate
concentration was 1.7 yg nr3 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, Forest et al. 1982).
5.3.6.3 Concentration Measurements at Remote Locations—Huebert (1980)
and Huebert and Lazrus (1978, 198UD) used a tamden niter 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, Spricer and Sverdrup 1981,
Forest et al. 1982). Teflon filters alone were used to collect
5-55
-------
participate 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 character-
istics 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 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
distributions of particulate nitrate in the South Coast air basin. The
5-56
-------
TABLE 5-10. MASS MEDIAN DIAMETERS REPORTED FOR NITRATE FROM PARTICLE
SIZING WITH CASCADE IMPACTORS
Site
Measurement
period
Reference
Mass median
diameter in ym
for nitrate
Cincinnati, OH
(CAMP Site)
Fairfax, OH
Riverside, CA
U. Cal. Campus
Secaucus, NJ
3/14-23/66
Lee and Patterson (1969) 0.23 (est)
Dominquez Hills,
CA
West Covina, CA
Pomona, CA
Rubidoux, CA
3/25-4/21/66 Lee and Patterson (1969) 0.59
11/1-15/68 Lundgren (1970) 0.8
9/29-10/10/66
Background
Level A
Level B
Level C
10/4-5/73
10/10-11/73
7/23-24/73
7/26/73
8/16-17/73
9/5-6/73
9/18-19/73
Patterson and Wagman
(1977)
Appel et al. (1978)
Appel et al. (1978)
Appel et al. (1978)
Appel et al. (1978)
0.20
2.6
0.38
0.37
1.64
0.72
1.13
0.62
0.68
0.33
0.34
5-57
-------
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 ym was an arbitrary upper cutoff). At Hermosa Beach, CA at
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 Chino, CA, well inland, a large part
of the particulate nitrate was in the submicron range. Coarse mode
nitrate was still present. Chino 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 distributions 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 to 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 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.
5-58
-------
Transformations of sulfur dioxide to sulfate 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 measurements 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 production 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). Peroxyacyl
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 sulfates.
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
troposphere 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 transported to these
altitudes eventually will be transported to the surface by vertical
movements, depending on the lifetime of ozone under these circumstances.
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 concentrations 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
5-59
-------
significant ambient air concentrations, to react would have to mix with
anthropogenic nitrogen oxides. 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
(Alt shul! er 1983).
Ozone formed in homogeneous photochemical reactions in the
atmosphere from anthropogenic precursors can be present at elevated
concentration levels at rural locatins 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 generated 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
TPEET
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 300 to 400 ppb 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 kilometers downwind. Measurements have been made on
the flow of the New York metropolitan area plume into southern New
England (Cleveland et al. 1976, 1977, Si pie 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
5-60
-------
occur shorter distances downwind of small cities such as Springfield, IL
(Spicer et al. 1982b).
Impacts of urban plumes from large or medium-si zed cities within
several hundred kilometers on elevated ozone concentration levels at
specific 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 pg nr3.
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 July 9, 1976. The ozone concentration within the plume at this
distance downwind was 220 yg nr3, about 100 yg nr3 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 associated 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 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, MO, 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
5-61
-------
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 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 betwen 200 and 500 yg m-3 at the Pennsylvania
sites, 150 yg m-3 at Creston, IA and less than 100 yg m-3 at
Wolf Point, MO. 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 concentrations 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 quarter 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 yq m-3; McConnelsville, OH, 117 yg m-3; Wooster, OH, 119
yg m-3, McHenry, MD, 116 yg m"3; DuBois, PA, 132 yg nr3.
In some of the studies discussed above, either sulfate measurements
or visibility measurements as a surrogate for fine particles are
available (Decker et al. 1976, Husar et al 1977). The sulfate
concentrations (in yg m-3) and the sulfate as a percentage of total
suspended particulate from west to east were as follows: Wolf Point,
MO, 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
5-62
-------
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 yg 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.
5.4.2 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 yg nr3 during the spring months, and as
low as 40 to 60 yg m-3 during the fall months. While the seasonal
patterns tend to be consistent, 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 aircraft 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 yg fir3 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 yg nr3 at 3
km, 210 yg m-3 at 5 km and 330 yg 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 unlikely 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 peroxide concentrations are of considerable
interest.
Several chemical methods for measuring of hydrogen peroxide in
ambient air and in rainwater are in use. Both the reaction of titanium
sulfate and 8-quinolinol 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 Cu(II) 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 containing 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
5-63
-------
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 yg nr3 (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 solutions. These results suggest that methods involving
collection in aqueous solutions may not provide useful measurements of
ambient air hydrogen peroxide concentrations. Both groups found
hydrogen peroxide to be generated within the aqueous collecting
solutions when ozone in oxygen-nitrogen 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 m-3 in
Hoboken, NJ and 251 yg nr3 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 nr3 with a maximum concentration
near 140 yg nr3 (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, CA,
Claremont, CA and Palo Verde, CA (Kok 1982). In Los Angeles at Cal.
State University, the hydrogen peroxide concentrations on June 18 and
19, 1980 ranged between about 0.7 and 3.5 yg nr3. 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 pg 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 September 11 and 12, 1980 the hydrogen
peroxide concentrations peaked at 8 to 11 yg nr3. The maximum
hydrogen peroxide concentrations compared to the maximum ozone
concentrations show no distinct relationship (Kok 1982).
5-64
-------
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. The about 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 pg nr3 during this period.
Hydrogen peroxide was measured in water condensate by the luminol
chemiluminescence technique at rural sites near Tucson, AZ (Farmer and
Dawson 1982). In more remote areas around Tucson the hydrogen peroxide
concentration were about 1.4 yg nr3, while at a Thurber Ranch site
the hydrogen peroxide ranged up to 6 yg nr3. The hydrogen peroxide
concentration was observed to drop off drastically when high sulfur
dioxide concentrations were measured. 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 concentrations also frequently were below 100 yg
i~ • The lower concentrations could be accounted for by the
absorption of less than 0.14 yg m~3 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
5-65
-------
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 atmosphere. 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 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 hydrogen and form hydrogen chloride (Hanst 1981). Phosgene
(C^CO) has 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
ug m'3: Florida--!.6, Ipswich, MA--4.4, and Hawaii—I.9. Gaseous
chlorine compounds were measured by the same technique by Duce et al.
(1965) on the island of Hawaii. The concentrations of gaseous chlorine
compounds ranged from less than 0.3 yg m~3 to 218 yg m"3
although the gaseous chlorine concentrations were at or below 10 yg
5-66
-------
m-3 in most 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
tropospheric and stratospheric measurements at the ground and from
aircraft. The tropospheric mixing ratio at ground level was 10~y
corresponding to 1.5 yg m-3, with the mixing ratio decreasing to
10-10 in the upper troposphere. 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 chloride 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. (1965) measured particulate chloride on a
four-stage cascade impactor. The total chloride concentrations ranged
from 0.5 to 137 yg m-5. Three of the nine samples had total
chloride concentrations of 39, 95 and 137 yg m-3; 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 m-3 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 concentrations of 5 to 10 yg m'-3
above background were observed, particularly in the industrialized
Pasadena-Houston Ship Channel area.
At urban and nonurban locations somewhat inland, atmospheric
chloride concentrations 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 particles. 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 below 1 ym were
associated with processes occurring over land.
5-67
-------
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 predominantly 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 below 0.5 ym were associated with lead
aerosols from automobile exhaust, the particles 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 pm 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 chloride to about 9 urn.
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 above 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 below 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 at sites in
Philadelphia, PA, Cincinnati, OH (Fairfax), and Chicago, IL, in the
summer and fall by Lee and Patterson (1969). 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 vehicle
emissions rather than sea salt aerosols.
5.7 METALLIC ELEMENTS
The various interests and possible concerns related to metallic
elements have been discussed briefly in the introduction. Alkaline
earth elements such as calcium and magnesium can help neutralize acidic
materials either during precipitation events or as a result of dry
deposition. Manganese and iron are possibly of consequence in the
5-68
-------
chemical transformations of sulfur dioxide to sul f ate (Chapter A-4,
Section 4.3.5). Aluminum, manganese, nickel, zinc, lead and mercury are
discussed elsewhere in this document (Effects Chapters) in relationship
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 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 concentration levels of these elements. This study
is useful in also providing the percentages of these elements in
particles below and above 3.5 ym at these urban sites. Sulfur is the
most abundant element, followed by calcium, aluminum, iron and lead.
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 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 rrr3, 38; 1000 to
1999 ng m-3 45; 2000 to 3999 ng m-3, 8; 4000 to 5300 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
compilation 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 m~3. 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 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 ym) compared to coarse
particles (greater than 3.5 ym) 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 associated 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
5-69
-------
TABLE 5-11. CONCENTRATIONS AND PERCENTAGES OF ELEMENTS PRESENT AS FINE
PARTICLES IN PARTICIPATE MATTER AT SITES IN THE UNITED STATES
Site
Period of measurement
New York City, NYa
February 1977
Philadelphia, PAa
Feb. -March 1977
Charleston, W VAa
April -Aug. 1976
and January 1977
St. Louis, M0a
en December 1975
0 Portland, ORa
February 1977
Glendora, CAa
March 1977
Smokey Mt. , PAd
July-Aug. 1977
Parameter
Cone, ng m~3
% Fineb
Cone, ng nr3
% Fine
Cone, ng m~3
% Fine
Cone, ng m~3
% Fine
Cone, ng nr3
% Fine
Cone, ng m~3
% Fine
Cone, ng nr3
% Fine
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
NO
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
Z?5
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 m.
cConcentrations reported not consistent with other Al measurements at site.
dStevens et al. 1980.
NA = not available.
ND = not determined.
-------
the zinc measured in more polluted air on particles below 1.5 ym with
most of the zinc associated with particles between 0.5 and 1.5 urn.
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 nr3 was measured during February 1972 (Foote
1972). Dams et al. (1970) reported mercury concentrations of 4.8 ng
m-3 on particulate matter collected 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 (Williston 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).
5.7.2 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 Smokey Mountains, TN site,
as at the urban sites, are found to a large extent in the coarse
particles (Tables 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 fold
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 reductions in concentrations 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 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~3 and frequently were below 500 ng nr3 on the average at
these sites. Lead concentrations were in the 30 to 80 ng m-3 range,
but on the average were below 50 ng nr3 at almost all of the sites.
5-71
-------
TABLE 5-12. CONCENTRATIONS OF ELEMENTS IN PARTICIPATE MATTER AT NONURBAN SITES
IN THE UNITED STATES AND IN CANADA
ro
Site
Period of measurement
Alleghany Mountain, PA
July-August 1977
Smokey 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 nT3
Fe
320
146
ND
410
25.3
13.1
71
Ni Zn Cd Pb
ND 20 3 90
2 <12 ND 114
ND 16 0.6 45
0.6 6.5 ND 14
0.1 4.2 ND 1.9
0.1 11.3 ND 1.8
ND 3.8 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.
-------
The overall mean concentration of coarse particles was 8000 ng nr3
with 60 percent associated with soil elements and their associated
oxides. The percentage of iron in fine particles (less than 2.5
was given for the sites in the study area. The percentage of iron in
fine particles 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 Smokey
Mountains site (Table 5-11).
Dams and Dejonge (1976) measured aerosol composition from August
1973 and April 1975 at Jungfraujoch (3752 m above sea level) in
Switzerland and also tabulated unpublished results by K. A. Rahn
obtained at Lakelv in marine air at North Cape, Norway during the winter
of 1971-72. The concentrations in ng nr3 of the elements considered
above were as follows: Jungfrau, Al, 51; Mn, 1.5; Fe, 36; Zn, 9.9; Pb,
4.4; Lakelv, 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 concentrations of soil derived elements should be
substantially higher. The much higher concentrations of aluminum at
Chadron, NB and Colstrip, MT (Table 5-12) than at mountainous and marine
sites are consistent with this expectation.
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 nr3. Over areas
containing mercury minerals, ambient air concentrations of mercury vapor
were in the 7 to 53 ng nr3, 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 nr3 have been reported (Foote
1972). Williston (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, Niles, MI, a
mercury concentration of 1.9 ng m-3 was measured in partial!ate 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).
5-73
-------
5.8 RELATIONSHIP OF LIGHT EXTINCTION AND VISUAL RANGE MEASUREMENTS TO
AEROSOL COMPOSITION
Visual range measurements can be influenced by a rubber of natural
and manmade factors. Visual range can be reduced substantially on an
episodic basis by rain, fog, snow, and by wind blown 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 significantly 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 ym 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.
Sul fates 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 extent that visual range and light extinction are accounted for to a
substantial extent by sulfates and nitrate concentrations in the
atmosphere, these visibility measurements can serve as surrogates for
concentration measurements in geographical areas where measurements are
not available. Because aerosol concentrations are related to deposition
rates, the visibility measurements also can be related to deposition or
the potential for deposition.
5.8.1 Fine Particle Concentration and Light Scattering Coefficients--A
number of investigators have demonstrated a proportionality between fine
particle concentration and light scattering coefficient. Sulfates and
nitrates, in some locations, are major components of the fine particle
concentration.
Waggoner and Weiss (1980) obtained a ratio of fine particle
concentration to the light scattering coefficient, bsc, of 0.36 g
m-2 (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 m-2. 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 correlation coefficient of 0.94 for the measurements of fine particle
concentration as related to bsp and a ratio of 0.24 g nr2. A
cyclone was used to eliminate particles above 1 vm from measurement as
fine particles. Ferman et al. (1981) made measurements at the same site
during the same period. These workers obtained a correlation
5-74
-------
coefficient of 0.91 for the measurements of fine particle concentration
as related to b§p and a ratio of 0.14 g nr2. However, a substanti-
ally higher particulate size cutoff was used by Ferman et al. than by
Wei ss 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).
HIVOL sampler measurements, not fine particle measurements, were made.
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: sul fate, 47; nitrate, 39; organics,
14. Except at high humidities, the contribution, on a unit mass basis,
of sul fate 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.
In November and December 1978 at a location in Denver concurrent
measurements were made of both light scattering and absorption of
nitrogen dioxide, and of ammonium, sul fate, 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 extinction were as follows:
sulfate as ammonium sul fate, 20; nitrate as ammonium nitrate, 17;
organic carbon, 12; elemental carbon, 38 (scattering, 6.5, absorption,
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
5-75
-------
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,
absorption, 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
predominately 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, Dzuabay 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 measurements. 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 possible
sources of error. These sources of error include some related to
airport visibility measurements (1) inadequate sets of markers (2)
changes in markers (3) changing environment in vicinity of airports.
The differences in the locations where the visibility and the air
quality measurements are taken can also result in differences also in
aerosol concentration and composition at these locations. The lack of
compositional measurements on some significant species can result in
overestimations of the contributions of measured species. Such
5-76
-------
overestimations can occur when there are good correlatons 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 Mountain, 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 Tound for bsp with the product of
sulfate and humidity than with sulfate 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 composition 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 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
5-77
-------
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
measurements, 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 substantially lower percentage of the fine particle
mass at the Great Smoky Mountain 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 fourth 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
measurements of light extinction and of aerosol composition have been
made (Macias et al. 1980). From samples obtained in flights over the
Southwest the average percentge 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 measurements
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 sites make a much smaller contribution to the
5-78
-------
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 sul fates
(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 visibility 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 concentrations. 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 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).
5-79
-------
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
Nitrogen dioxide
Nitrate aerosols
Nitric acid
Peroxyacyl nitrates
Ammonia
Hydrogen chloride
Chloride aerosols
Maritime
Inland
East
10-20a
5-15a
10-20&
1C
0.3-1.3
0.5-1C
0.5-2<*
1-10C
1-10C
<_ 1C
West
NA
l-3a
12C
NA
1 lc
0.1-0.3C
0.5-2C
1-10°
1-10C
1 lc
aAnnual average.
bSummer months: August to December averages.
cLimited number of measurements.
NA= Not available.
5-80
-------
Sulfate concentrations decreased in eastern cities during the
1960's 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 decreased appreciably throughout the year and sulfates
have increased in concentration 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 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 ym. 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 summer months than in
winter months and 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 shown to be in the form of
strong acid species in the eastern mountains of the United States
during 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 (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 eastern United States than in the western United
States (Section 5.3.2.4).
5-81
-------
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
measurements 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 concentrations 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) stratospheric extrusions reaching ground
level (Section 5.4).
5-82
-------
Rural locations within urban plumes may experience ozone
concentrations in the range of 300 to 500 yg m"3. Within high
pressure systems, ozone concentrations at rural locations can range
from 150 to 250 yg nr3 (Section 5.4.1).
At remote elevated sites, hourly ozone concentrations are as high
as 140 to 160 yg nr3 during the spring months and as low as 40
to 60 yg m~3 in the fall months. Occasional observations of
ozone concentrations in excess of 200 yg nr3 attributed to
stratospheric air extrusions at remote sites appear too high
compared to aircraft measurements 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
correspond to those resulting from the absorption of less than 1
yg nr3 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
concentrations 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).
5-83
-------
At eastern rural sites sulfate accounts for a large part of the
fine particle mass and the light extinction (Section 5.8.3).
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
concentrations (Section 5.8.3).
5-84
-------
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. Natural volatile organic substances and their
effect on air quality in the United States. Atmos. Environ, in press.
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.
5-85
409-261 0-83-15
-------
Barrie, L. A., K. G. Aulsuf, H. A. Wiebe, and P. FeTHu. 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.
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. VI. 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.
5-86
-------
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.
Cleveland, W. S. and B. Kleiner. 1975. The transport of photochemical
air pollution from the Camden - Philadelphia urban complex. Environ.
Sci. Technol. 9:869-872.
Cleveland, W. S., B. Kleiner, J. E. McRae and J. L. Warner. 1976.
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 andTts Control. B. Dimitviodes, ed.
EPA-600/3-77-0012. Environmental Sciences Research Laboratory, Research
Triangle Park, NC.
Coburn, W. G., R. B. Husar, andj. 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, NO?,
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 Col strip,
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. Dejonge. 1976. Chemical Composition of Swiss aerosols
from the Jungfranjoch. Atmos. Environ. 10:1079-1084 additional results
obtained from K. A. Rahn, Ph.D. thesis, U. Michigan. 1971.
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.
5-87
-------
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.
Davis, D. D., 6. 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,
Laramie.
Drummond, J. W. and A. Volz. 1982. Simultaneous measurements of NO and
N02 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. Anal. 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. Technol. 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.
5-f
-------
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.
Flocchlni, R. G., T. A. Cahill, D. J. Shadoan, S. J. Lange, R. A.
Eldred, P. J. Feeney, G. W. Wolfe, D. C. Slmmeroth, and J. K. Suder.
1976. Monitoring California's aerosols by size and elemental
composition. Environ. Scl. Techno!. 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.
Forest, J., R. L. Tanner, D. Spandau, T. D'Ottario, and L. Newman.
1980. Determination of total inorganic nitrate utilizing collection of
nitric acid on NaCl-impregnated filters. Atmos. Environ. 14:137-144.
Forest, 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 SOo 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.
5-89
-------
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. Techno!. 8:551-557.
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 nitropeucous 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. AirPollut. 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. Brogin. 1982. A long path infrared
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.
Marker, 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. In Vol. 1,
International Conference on Photochemical Oxidant Pollution and its
Control. B. Dimitviodes, ed. EPA-60013-77-0012. Environmental
Sciences Research Laboratory, Research Triangle Park, NC.
5-90
-------
Heikes, B. G., A. L. Lazrus, G. L. Kok, S. M. Kunen, B. W. Gandrud, S.
N. GUI in and P. D. Sperry. 1982. Evidence of aqueous phase hudrogen
peroxide synthesis in the troposphere. J. Geophys. Res. 87:3045-3051.
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. Jhi Vol. 1, International
Conference on Photochemical Oxidant Pollution and its Control. B.
Dimitviodes, 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. Atmos.
Environ. 12:735-752.
Hilst, G. R., P. K. Mueller, G. M. Hidy, T. F. Lavery, andJ. 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., 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. Jji Proceedings of the Fourth Symposium on Atmospheric
Turbulence, Diffusion and Air Pollution, American Meteorological
Society, Reno, NV.
5-91
-------
Husar, R. B., D. E. Patterson, C. C. Paley and N. V. Gill am'. 1977.
Ozone in hazy air masses, pp. 275-282. Ir± 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 oT"~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. ^ni 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. Tellus
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 ^2 in
rural air. Geophys. Res. Lett. 6:375-378.
Kelly, T. J. and D. H. Stedman. 1979b. Chemiluminescence measurements
of HN03 in air. ^n 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.
5-92
-------
Kley, D., J. W. Drummond, M. McFarland, and S. C. Liu. 1981.
Tropospheric profiles of NOX- J- Geophys. Res. 86:3156-3161.
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. Darnell, 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. Technol. 12:1077-1080.
Kok, G. L., T. Hallen, M. Lopez, H. Nectreib, and M. Yuau. 1978b.
Chemileminescent method for determination of hydrogen peroxide in the
ambient air. Environ. Sci. Technol. 12:1072-1076.
Laird, A. R. and R. W. Miksad. 1978. Observations on the particulate
chlorine distribution in the Houston-Galveston 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. Minnick, 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.
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.
5-93
-------
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 hemisphere air at Quillayute, VIA. 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-.
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. Zwidker, J. R. Quintette, 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. Aerosols
chemical composition. Atm. Environ. 15:1971-1986.
Maroulis, P. J., A. L. Torres, A. B. Goldberg, and A. R. Bandy. 1980.
Atmospheric SOo measurements on Project Gametag. J. Geophys. Res.
85:7345-7349.
Martens, C. S., J. J. Weslowski, R Kaifer, and W. John. 1973. Lead
and bromine size distribution in the San Francisco Bay areas. Atmos.
Environ. 7:905-914.
Martinez, 0. R. and H. B. Singh. 1979. Survey of the Role of NOv in
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.
5-94
-------
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. K. A.
Rahn. 1971. An approach to clean air. Ph.D. Thesis Univ. of Michigan.
McClenny, W. A., P. C. Gailey, R. S. Bramam, and T. J. Shelley. 1982.
Tunastic acid technique for monitorian 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-600/3-77-053. 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 Biological
Effects of Atmospheric Pollutants. Div. of Medical Service 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. Hoffau, and A. H.
Bockiau. 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.
5-95
-------
Paciga, J. J. and R. E. Jervis. 1976. Multielement size
characterization of urban aerosols. Environ. Sci. Technol.
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. VI. 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. L. Truex, J. W. Butler, and T. L.
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. L. Truex, and J. W.
Bulter. 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 bestimmung kleinster mengen von
wasser stoff peroxyd in luft. Int. J. Environ. Anal. Chem. 3:257-270.
Pitts, J. N., Jr. and 0. Grosjean. 1979. Detailed characteristics of
gaseous and size-resolved partial!ate pollutants at a south coast air
basin smog receptor site. PB-302, 294 National Technical Information
Service, Springfield, VA.
Platt, V. and D. Perner. 1980. Direct measurements of atmospheric
CH20, HN02, 03, N02, and SO? by differential optical
absorption in the near Uv. J. Geophys. Res. 85:7453-7458.
Prahm, L. P., V. 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-60017-77-056.
Environmental Sciences Research Laboratory. Research Triangle Park, NC.
Renzetti, N. A. and R. J. Bryan. 1961. Atmospheric sampling for
oldehydes 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.
5-96
-------
Ripperton, L. A., L. Kornreich, and 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.
Rohlach, L. A., W. C. Hawn, K. R. Williams, and T. P. Parsons. 1979.
Nitrogen oxide interferences in the measurement of atmospheric
particulate nitrate. EPRI EA-1031, Research Project 801-1. Electric
Power Research Institute, 3412 Hillview Ave., Palo Alto, CA.
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.
Schiff, H. I., D. Pepper and B. A. Ridley. 1979. Tropospheric NO
measurements up to 7 km. J. Geophys. Res. 84:7895-7897.
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
downwind of the Chicago-Gary Urban complex. J. Air Pollut. Control
Assoc. 30:911-914.
Shaw, R. W. and R. J. Paur. 1982. Long term measurements of sulfur in
gases and particles in the Ohio Valley. Environmental Sciences Research
Laboratory. Research Triangle Pk, NC.
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-60012-79-051. 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.
5-97
-------
Singh, H. B., L. J. Salas, H. Shigeishi, and A. Crawford. 1977b.
Urban-non-urban relationships of halocarbons, SF5, ^0 and other
atmospheric trace constituents. Atmos. Environ. 11:819-828.
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 NC^. 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.
Sing, 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.
Si pie, 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
Dimitviodes, 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. Atnos. 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 R. 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 60012-79-051.
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.
5-98
-------
Spicer, C. W. and P. M. Schumacher. 1979. Participate nitrate;
laboratory and field studies of major sampling interferences. Atmos.
Environ. 13:543-552.
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. In press.
Spicer, C. W., J. E. Howes, Jr., T. A. Bishop, L. H. Arnold, and R. K.
Stevens. 1982a. Nitric acid measurement methods: An intercomparison.
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. Techno!. 13:975-984.
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. LTKlimisoh, eds. Plenum Press, New York.
5-99
-------
Struempler, A. W. 1975. Trace elanent composition in atmospheric
particulates during 1973 and the summer of 1974 at Chadron, Neb.
Environ. Sci. Techno!. 9:1164-1168.
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+, NHy+, SO^2, N03-
-NH3(g), HN03(g). Atnos. 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.
Atnos. Environ. 11:1443-1150.
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. Atnos. 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, to be published.
Tesche, T. W., J. A. Ogren and D. L. Blumenthal. 1977. Ozone
concentrations in power plant plumes, pp. 157-171. lr± Comparison of
Models and Sampling Data. Vol. II, International CoTTFerence on
Photochemical Oxidant Pollution and Its Control. B. Dimitviodes, ed.
EPA-60013-77-0012. 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,r ""'<:-.
\ ,' >Ct- A'\t
Trijonis, J. and S. Mortimer. 1982. ValidatftVof 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.
5-100
-------
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.
Tuazon, E. C., R. A. Graham, A. M. Winer, R. R. Easton, and J. N. Pitts,
Jr. 1978. A kilimeter 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. PiTHfs 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, andJ. 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.
Superintendant 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.
5-101
-------
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.
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. W. 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. Influence of some
atmospheric variables on the concentration and particle size
distribution of sulfate in urban air. Atmos. Environ. 1:479-489.
Weiss, R. E., T. V. Larson, and A. P. Waggoner, 1982. In site
rapid-response measurement of HoSOd/(NH4)2S04 aerosols in rural
Virginia. Environ. Sci. and TeChnSl. 16:525-532.
Weiss, R. E., A. P. Waggoner, R. J. Charlson, and N. C. Ahlquist. 1977.
Sulfate aerosol: Its geographical extent in the mideastern and southern
United States. Science 195:979-981.
Westberg, H., K. Allwine, 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., J. A. Anderson, D. L. Blumenthal, R. B. Husar, N. V.
Gill am', J. D. Husar, and W. E. Wilson, Jr. 1976. Formation and
transport of secondary air pollutants: ozone and aerosols in the St.
Louis urban plume. Science 194:187-189.
5-102
-------
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 andTts Control. B. Dimitviodes, ed. EPA-600/3-77-0012,
Environmental Sciences Research Laboratory, Research Triangle Park, NC.
White, W. H. and P. T. Roberts. 1977. On the nature and origins of
visibility-reducing aerosols in the Los Angeles Basin. Atmos. Environ.
11:803-812.
Williston, S. H. 1968. Mercury in the atmosphere. J. Geophys. Res.
73:7051-7055.
Winer, A. M., J. W. Petous, J. P. Smith, and J. N. Pitts, Jr. 1974.
Response of commercial chemiluninescent NO-N02 analyzers to other
nitrogen-containing compounds. Environ. Sci. Technol. 8:1188-1121.
Witz, S. and R. D. MacPhee. 1977. Effect of different types of glass
filters on total suspended particulates and their chemical composition.
J. Air Pollut. Control 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. 1982. A comparison of
mass, lead, sulfate, and nitrate concentrations in a field study using
diebotomous, size-selective, and standard Hi-Vol samplers. J. Air
Pollut. Control Assoc. 32-276-278.
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. Manson, and M. A. Ferman. 1979. On the nature of
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. 1982. Interaction of ozone and
hydrogen peroxide in water. Implications for anlaysis of ^2 in
air. Geophys. Res. Letts. 9:231-234.
Zika, R., E. Saltzman, W. L. Chameides, and D. D. Davis. 1982.
H202 levels in rainwater collected in South Florida and the Bahama
Islands. J. Geophys. Res. 87:5015-5017.
5-103
-------
THE ACIDIC DEPOSITION PHENOMENON AND ITS EFFECTS
A-6. PRECIPITATION SCAVENGING PROCESSES
(J. M. Hales)
6.1 INTRODUCTION
The complex process of precipitation scavenging can be subdivided
into a number of distinct steps, which occur interactively within a com-
posite storm system. These are itemized as follows:
0 intermixing of pollutant and condensed water within
the same airspace,
° attachment of pollutant to the condensed water
elements,
0 chemical reaction of pollutant within the aqueous
phase,
0 delivery of pollutant-laden water elements to the
surface via the process.
Each of these steps can be associated with a corresponding
processing time that depends upon the pollutant, synoptic circumstances,
and storm type. In the simplest sense, the scavenging process occurs as
a forward progression through these steps; reverse processes are common,
however, and a pollution element may experience several cycles through
segments of this process before its ultimate wet deposition to the
Earth's surface. This chapter examines the several steps as they relate
to the problem of wet deposition of acidic substances.
Pollutant condensed-water intermixing, the process that introduces
pollutant to the immediate vicinity of cloud and precipitation systems,
can involve considerable time lags between a pollutant's emission and
its subsequent processing by the storm. Usually it is not cloudy or
raining in the vicinity of a pollutant's release point, and often
several days may occur before a storm is encountered. During this
period the pollutant may become involved in a variety of processes
(e.g., dry deposition, chemical reaction) that may alter its
concentration and physical state, and consequently alter its scavenging
characteristics once a storm is encountered. Thus, while the
storm-pollutant intermixing process is not considered totally within the
realm of wet removal, it is a highly important determinant of scavenging
time and distance scales and the resulting chemical composition of
precipitation.
The actual physical attachment of pollutant to condensed water
elements (ice, cloud droplets, rain) greatly depends upon both the
6-1
-------
physical and chemical states of the pollutant. For aerosol particles,
any or all of the following collection mechanisms may be active:
o nucleation of cloud droplets on the pollutant particles
o electrical attachment
° diffusiophoretic and thermophoretic attachment
o Brownian motion
° inertia! attachment
All mechanisms in the above list depend upon particle size, and usually
several mechanisms operate simultaneously to provide a composite capture
process in given situations.
Diffusional and convective transport are the primary attachment
mechanisms for gaseous pollutants. Gas scavenging differs from aerosol
scavenging in the important respect that gases may desorb from, as well
as absorb, in cloud particles and hydrometors. Thus relative rates of
absorption and desorption often determine to a large extent the net
efficiency of attachment, and for this reason gas solubility emerges as
an important factor in the scavenging process.
This chapter deals only briefly with the aqueous-phase reaction
step, owing to the fact that it is treated elsewhere within this
document (Chapter A-4). It should be stressed, however, that although
reaction is not necessary for scavenging to occur, it often emerges as
an important rate-limiting step. This importance stems primarily from
chemical conversion's capability, in some circumstances, to devolatilize
absorbed gaseous pollutants and thus inhibit their tendency for
desorption noted earlier. The conversion of dissolved S02 to sulfate
is an important example.
The final stage of the composite scavenging process is the actual
wet delivery of pollutant to the ground. This step is linked closely to
rain formation and precipitation processes and thus depends strongly
upon the variety of cloud-physics phenomena commonly associated with
water extraction. These include autoconversion of cloud elements to
form precipitation, accretion and condensation processes, and a host of
ice-formation phenomena. The kinetics of such processes often cast a
significant rate influencing influence on the overall scavenging
process.
Area! deposition by storm systems strongly depends on
climatological features of the storms themselves. Although a detailed
treatise on North American storm climatology is well beyond the scope of
this work, some limited insight in this regard may be gained by a
partial classification of storm types and a climatological analysis of
storm tracks.
6-2
-------
Much of what is known presently with regard to precipitation
scavenging has been learned as a consequence of field studies.
Pertinent field experiments are summarized in tabular form in
Section 6.4.
Mathematical models of precipitation scavenging tend to reflect the
stepwise sequence suggested above. Based upon conservation equations
for pollutant material, these models are similar in many respects to
typical air pollutant models, but differ in the sense that they must
account for gas-liquid exchange and wet delivery. A profusion of
different wet removal models is currently available and is presented in
tabular form in Section 6.5.
6.2 STEPS IN THE SCAVENGING SEQUENCE
6.2.1 Introduction
Precipitation scavenging is defined generally as the composite
process by which airborne pollutant gases and particles attach to
precipitation elements and thus deposit to the Earth's surface. 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 used consistently throughout this text. When
specific reference to the alternative situation is made, the terms
"attachment" and "capture" will be employed essentially
interchangeably.
This scavenging process typically contains many parallel and
consecutive steps, so 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 pollutant
moleculei 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.
^•Initial 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 SOg, HN03, NH3, sulfate, chloride, metallic cations,
and so forth.
6-3
-------
MIXING
1
UNREACTED POLLUTANT
REACTED POLLUTANT
CONDENSED WATER
PRECIPITATION
Figure 6-1. Steps in the scavenging sequence: Pictorial representation.
6-4
-------
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 portrayal 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 pollutant 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 physio-
chemical reaction within the aqueous-phase is potentially an important
step in the scavenging process, it is not essentialI. 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 the devolatilization of dissolved
sulfur dioxide via wet oxidation to sulfate. 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 scavenging 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 (SI inn 1974a). Finally, it is important to note
6-5
-------
00
OO
LU
O
O
o:
Q.
t
00
o
i—i
00
0.
oo
oo
LU
POLLUTANT
IN
CLEAN AIR
EVAPORATION
SEPARATION
o;
o
o.
Q.
Oi
O
00
TYPE 1 OR
TYPE 2
MIXING PROCESSES
POLLUTANT
AND
CONDENSED WATER
INTERMIXED IN
COMMON AIRSPACE
EVAPORATION
DESORPTION
CSL
O
n.
«c
ATTACHMENT
POLLUTANT
ATTACHED TO
CONDENSED-WATER
ELEMENTS
REACTION
REACTION
ATTACHED POLLUTANT
MODIFIED BY
AQUEOUS-PHASE
PHYSIOCHEMICAL REACTIONS
DEPOSITION
POLLUTANT DEPOSITED
ON
EARTH'S SURFACE
O
70
-O
30
O
O
m
Figure 6-2. Scavenging sequence: Interaction diagram.
6-6
-------
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
co-mingle. 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 Grandel 1 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 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. J/f a particular pollutant is
easily scavengable 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 (SI inn 1983) it is not unreasonable to
6-7
-------
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 precipitation. 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, Davles 1966, Dingle and Lee 1973, Pruppacher
and Klett 1978, Hales 1983, Slinn 1983, Slinn and Hales 1983). This
process (Figure 6-1) 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 comixing 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 particles provide such surfaces within the air parcel,
and the consequence 1s a cloud of droplets (or ice crystals)2 contain-
ing attached pollutant material.
Different types of aerosol particles possess different capabilities
to nucleate cloud elements and grow by the condensation process. As a
consequence, typically competition for water molecules exists among the
aerosol and associated cloud particles. Some will capture water with
high efficiency and grow substantially in size. Others will acquire
2At 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.
6-8
-------
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 characteristics, and its chemical composi-
tion. 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 thermodynamic
equilibrium. Because of diffusional constraints on delivering water
molecules from the surrounding atmosphere, the growth in droplet
diameter slows appreciably as droplet size increases (SIinn 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
atmospheres, for example, contain large sea-salt particles as a primary
component of their aerosol burden. Warm marine clouds are noted for
their wide drop spectra containing large drop sizes and their
corresponding capability to form precipitation easily. Continental
clouds, on the other hand, are typically composed of larger populations
of smaller droplets. Figure 6-3, 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 continental air-mass cloud
exhibits a distinct tendency toward smaller drop 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.
6-9
-------
CTt
OO
01-9
J=- O
Q. — "
Q) O
-a c:
ct- Q.
rt>
Q. Q.
-b O
-s "a
O — '
GO
-Q t/5
c: T3
-i. fD
-5 O
Ol
13
Q. :
CLOUD DROPLET CONCENTRATION (Droplets cm3}
o
o
o
cn
o
a
§
•n
o
c:
Q.
CO
. o
—'•
OJ
-5
O>
3
Q-
o
o
3
r-h
OJ
to
CO
(D
-------
There are numerous mechanisms by which pollutants can attach to
cloud and precipitation elements after the elements already exist, 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.
Diffuslonal attachment, as Its name Implies, results from
dlffusional 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) dlffuslvlty;
because dlffuslvlty Is Inversely related to particle size, this
mechanism becomes less Important as pollutant elements become large.
Dlffuslonal 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 precipitation
element, through some sort of outgassing mechanism, supplies pollutant
to the local air, then the concentration gradient will be reversed and
diffusion 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. 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 1983, Slinn 1983).
Inertia! attachment processes directly depend 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 particles and scavenging elements relative to the surrounding
air, which arise because both have finite volume and mass. The most
important example of inertia! attachment is the impactlon of aerosols on
falling hydrometeors. Here the hydrometeor (because of its mass and
6-11
-------
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 1983, SI inn 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 inertia! 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 inertia! 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 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
6-12
-------
tQ
c:
fD
cr>
i
91 - £8 - 0 192-60*1
ei-9
O d- oo —I
3" O -•• 3-
CU N fD
-S O fD O
(O O • -S
fD 3 fD
O -S 3> -"•
O -"• O. O
3 CT Cu Cu
CL C T3 —•
CAPTURE EFFICIENCY
fD
-•• fD 00
• O Q- O
13 O1
00 -h <
-S fD
CT O 3
-<< 3 to
fD O 3
fD -S
O < ..
d- fD -h
-S oo -h
fD
• O
fD
O
—• < fo'
' fD 3
CU 3 O
Q. cr
fD
O
fD
-h
O
-S
Q-
fD
rt-
Cu
O -S Cu
-S C
fD "O —h
O O —'
3" ->•
fD fD 3
-h -S 10
-h
fD Cu -s
n 3 Cu
c-h O. -••
• 00 3
' 7^ O-
'3 fD O
Q- r+ "O
fD c+
-S CU
^-^ oo
n i—1
3"
-------
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 direction 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.3
Thermophoresis depends directly upon the temperature gradient in
the vicinity of the capturing element. In cloud and precipitation
systems local temperature gradients are caused most often by
evaporation/condensation effects; thus, thermophoresis is usually
strongly associated with diffusiophoresis,4 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
(SI inn 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
knowledge of this area satisfactory.
Electrical attachment of aerosol particles to cloud and precipita-
tion elements 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 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.
30ne should note that the precise mechanisms of thermal transport
differ radically, depending upon particle size (cf., Cadle 1965).
^As 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.
6-14
-------
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 in it 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 mathe-
matics of the scavenging process, and lends an additional degree of
difficulty to the problem of scavenging calculations. This complicity
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 (Step 4-5)
Although a variety of mechanisms exist (e.g., impaction of fog on
vegetation), 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, which depend in turn upon hydrometeor size. Thus, the
processes by which the pollutant-laden cloud droplets grow to
precipitation elements emerge as major determining factors of the final
stage of the scavenging sequence.
Once attached to condensed water, a pollutant molecule has several
alternative 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.
6-15
-------
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 we consider 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. Because clouds contain varieties of makeup elements having
different thermodynamic characteristics, 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 association 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 precipitation 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)5 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 Bergerson-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.
5Coagulation is often referred to as autoconversion in the cloud
physics literature. It is interesting to notice in this context that,
while coagulation tends to accumulate nucleated pollutants, the
Bergeron-Findeisen process tends to re-liberate nucleated pollutants to
the air.
6-16
-------
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 accretion process tends to accumulate
the pollutant burden of all collected elanents.
Accretion can occur via drop-drop, drop-crystal, and crystal-
crystal interactions. 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"1. 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 gravitational
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 projected object. The common observations of rime-ice
accumulation on alpine forests and on power-transmission lines give
direct testimony to the effectiveness of this process.
Chemical deposition by fogs is directly proportional to fog-bound
pollutant concentration, and this fact often acts to enhance
substantially the pathway's overall effectiveness. Owing to their
proxmity to the Earth's surface, 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
6-17
-------
find evidence to support this point. Scott and Laulainen (1979), for
example, reported sulfate and nitrate concentrations 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, New York observatory.
Recently, Waldman et al. (1982) have reported nitrate and sulfate
concentrations in Los Angeles fogs ranging up to and beyond 5000 ym
£-1. This compares with typical precipitation-borne concentrations
of about 35 ym £-1 for the northeastern United States.
Recently 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 area! basis. This requires substantial 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 appropriate 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 framework 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 occurrences, 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
6-18
-------
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 precipitation,
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, 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 cooling 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
elevations of 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 mid-latitude 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 mid-latitude
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
6-19
-------
model of frontal-storm development and behavior. 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
atmosphere'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
depression in atmospheric pressure occurs at the vertex of the warm-air
intrusion; as a consequence a general counterclockwise or "cyclonic"
circulation pattern emerges. Because of this feature, Bjerknes1
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 consequence 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 structure 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 facilities have become available.
Certainly nonidealities 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 assessment 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--!t is important to note that the plane 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
6-20
-------
CM
I
r\>
HIGH
LOW
HIGH
HIGH
Figure 6-5. Cyclonic storm development according to Bjerkne's conceptual model.
-------
en
ro
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).
-------
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
6-23
-------
o>
I
ro
0
A
200
FLOATING ICE NEEDLES
FALLING ICE NEEDLES
FLOATING FOG DROPS
"ICE NUCLEI LEVEL"
FALLING SNOW
400
Km
600 800
FALLING RAIN
::':!":::. FALLING DRIZZLE
0°C ISOTHERM
RELATIVE VELOCITY OF WARM AIR
RELATIVE VELOCITY OF COLD AIR
A'
Figure 6-7. Vertical cross section of a typical warm front (Section A-A1 on Figure 6-6) Adapted
from Godske et al. (1957).
-------
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 precipitation, 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.
Warm-front storms usually can be expected to be rather effective as
scavengers 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 pollutants 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
substantially 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 situation composes an important class of convective
storms, which will be discussed 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
6-25
-------
CT>
ro
CTI
2 -
0
^sWw^^
0 200 400 600 800
B Km B
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
Figure 6-8. Schematic vertical cross section of a typical cold front (Sction B-B1 on Figure 6-6)
Adapted from Godske et al. (1957).
-------
ro
-j
4 -
0
C
200
s\^Av^swc>^^
400 600 800
Km
FLOATING ICE NEEDLES
FALLING ICE NEEDLES
C'
FLOATING FOG DROPS
*-*• "ICE NUCLEI LEVEL"
*|*|*
*l*l*
Figure 6-9.
FALLING SNOW
FALLING RAIN
FALLING DRIZZLE
0°C ISOTHERM
RELATIVE VELOCITY OF WARM AIR
RELATIVE VELOCITY OF COLD AIR
RELATIVE VELOCITY OF COLDEST AIR
Schematic vertical cross section of a typical occluded front (Section C-C' on Figure 6-6)
Adapted from Godske et al. (1957).
-------
vertical cross section (Section C-C1 on Figure 6-6) of an occluded
system, demonstrates this point. Typically the easterly flow of warm
air aloft maintains a relatively stable environment 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 necessary vertical motions and concurrent cooling and
condensation processes and are therefore 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 are such that violent updraft velocities capable of
lifting entrained air, water vapor, and pollution to extremely high
elevations (sometimes breaching the stratosphere) often occur. 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: Nom'deal 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 typically lose much of their
6-28
-------
ro
cr>
i
D-
n>
cu
6Z-9
HEIGHT ABOVE GROUND (m)
o
o
(SI
(si
in
n>
o
o
-h
01
3
I/)
O
o>
<-h
fD
D-
O
O
3
ro
o
o
-s
TEMPERATURE (°C)
-------
original identity after impacting with the western mountainous regions.
In addition to the physical distortion of flow patterns, 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 counterparts.
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. Orographic effects also are suspected to
influence storm behavior over substantial 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 nonideal
phenomenon superimposed with more major meterological patterns.
Typically such storms occur during fall and early winter, when land
surfaces tend to be cooler than their adjoining water bodies. Con-
sidering an air parcel moving on an easterly course across Lake
Michigan, for example, we note the warm lake surface tends to 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 the 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, Michigan, and Buffalo, New York. 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
6-30
-------
70
30
v
iOUTH BEND
Figure 6-11.
Average annual snowfall pattern (inches) over Lake Michigan
and environs. Adapted from Changnon (1968).
6-31
-------
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 cl imatological
averages of monthly precipitation amounts at various stations throughout
the United States. This figure, taken directly from the U.S.
Cl imatological 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 western and arid stations, whose
seasonal variabilities tend to be pronounced. 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. Successful interpretation of such information in the
context of source-receptor analyses requires careful and skilled
meteorological guidance.
Several additional points should be emphasized in the context of
Figure 6-13. First, 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. Second, 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
6-32
-------
NORMAL MONTHLY TOTAL PRECIPITATION (Inches)
I
CO
co
Figure 6-12. Cl Imatological Summary of U.S. Precipitation. From U.S. Climatological Atlas (1968).
-------
Figure 6-13.
Major climatological storm tracks for the North American
continent. Adapted from Haurwitz and Austin (1944). Dashed
lines denote tropical cyclone centers, and solid lines denote
those of extratropical cyclones.
6-34
-------
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 characterization 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 differentiate them accordingly.
The primary distinguishing feature 9f a scavenging field
investigation is that the study usually is designed around the basis of
some sort of conceptual 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 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 correcTT
Regardless of whether positive or negative 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 began in earnest during
the early 1950's to gain an understanding of radioactive fallout.
Pioneering studies in this area were performed in England by Chamberlain
(1953); they pertained to radioactive pollutant releases from point
6-35
-------
9£-9
PRECIPITATION PER STORM DURATION CLASS AS A FRACTION OF
GRAND SUM SEASONAL PRECIPITATION
cu n
o
e
ro
CD
O H-
• •
CO o
e
ro
CUMULATIVE FRACTION OF NORMALIZED TOTAL REGIONAL PRECIPITATION
-------
sources in anticipation of reactor accidents and related phenomena.
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.
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 "Scavenging 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 importance 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 have been
conducted. These are summarized in Table 6-?., 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 information 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 areal 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 illustrated by
considering the fate of specific acidic precipitatin precursors (SOX,
NOX, and HC1) as they are transported over these increasing scales of
time and distance.
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
6-37
-------
TABLE 6-2. SUMMARIES OF SOME PRECIPITATION SCAVENGING FIELD INVESTIGATIONS
General source type
Specific source type
Selected references
I
GO
00
Continuous Point
Source
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 plumes
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)
Mil Ian et al. (1982), Chan et al. (1982)
"Instantaneous" and/ Aircraft releases of rare-
or Moving Sources earth tracers
Dingle et al. (1969), SI inn (1973), Young et al.
(1976), Gatza (1977), Changnon et al. (1981)
Rocket releases of radioactive Shopauskas et al. (1969), Burtsev et al. (1976),
tracers
-------
TABLE 6-2. CONTINUED
General Source Type
Specific Source Type
Selected References
Urban Sources
General and Regional
Sources
I
CO
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.
-------
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, have 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
intentionally 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 made the interpretation of plume scavenging a
difficult task. Typically the sulfate and nitrate concentations 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.
6-40
-------
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) relatively few data come from
intensive field 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 suggestion 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) win 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 SOX 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
6-41
-------
150
100
50
TOTAL SULFUR
'
AV- .7X*»* y\'
:/-.A . *A'i .•/•A.
100
50
NITRATE
0 0.5 1.0 1.5 2.0 2.5
YEARS SINCE JULY 1976
3.0 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).
6-42
-------
measurements for a northeastern U.S. site. The decidedly periodic6
behavior of sulfate-ion concentrations 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.
An noted above, most past field experiments have 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
60ne 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 n2
r = p linear regression - q periodic fit
a2linear regression
where thea2's pertain to variances of the data points over the
three and one-half period. For sulfate in Figure 6-15 r2 equals
0.22, indicating a significant reduction in variance; the corresponding
r2 value for nitrate is 0.01, suggesting that no significant
annual periodicity exists in this case.
6-43
-------
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.
The data collected and assembled by the OSCAR project are summarixed 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 1n 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 indicating 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 significant 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 conjunction with the Indiana-Ohio concentrated network. One early
analysis, conducted for the April 22,24, 1981 storm is presented in
Figure 6-19. Back trajectories of this type are currently being
combined in diagnostic scavenging models with aircraft and surface data
to evaluate source-receptor relationships in greater detail (Easter and
Hales 1983a,b).
6-44
-------
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
o North American hourly precipitation amount data
o Trajectory forecast data (Limited Fine Mesh and Global
Spectral Models)
Gridded forecast data (Limited Fine Mesh Model)
Satellite observations
PRECIPITATION-CHEMISTRY DATA
0 OSCAR network: Sequential measurements of rainfall,
field pH. lab pH, conductivity, NOa", N02~, $042-, S0s2-, Cl",
NH4+, Ca2+, Mg2+, K , Na*, A13+, po4x-, total Pb
0 Additional networks: Time-averaged data as available
from sources such as NADP, CANSAP, CCIW, and APN
° Special rainborne ^02 measurements
AIRCRAFT DATA
Trace gases: 03, NO/NOX, S02, HNOa, NH3
° Aerosol parameters: scattering coefficient (b^t). Aitken
nuclei, aerosol sulfur, sulfate size distribution, aerosol
size distribution, aerosol acidity
o Cloud water chemistry: N03", NO?", S042~, S0a2-, pH, NH4+,
conductivity, CT, Ca2+, 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
6-45
409-261 0-83-17
-------
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, $042-, 03, NO/NOX, HNOs, aerosol composition
vs particle size, aerosol acidity
0 Selected air quality data from specific surface monitoring
sites throughout eastern North America
EMISSIONS
0 MAP3S/RAINE standard inventory
6-46
-------
CT>
. EXISTING
MAP3S SITES
SUPPLEMENTAL
REGIONAL SITES
'I | NE INDIANA GRID
r____
Figure 6-16. General layout of OSCAR sequential precipitation chemistry network, showing hypothetical
"design-basis" cyclonic system.
-------
CO
V
Figure 6-17. pH distribution for initial precipitation sampled during OSCAR storm of 22-24 April 1981.
-------
.£>
VO
1 T '4'5
Figure 6-18. pH distribution for final precipitation sampled during OSCAR storm of 22-24 April 1981.
-------
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-Z, In a manner which allows their expression In mathematical form.
Often such models take the form of submodels or "modules" within a
larger calculatlonal framework, such as a composite regional pollution
code. When considered 1n 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 clusters of boxes)
can be Identified with the modules, themselves. This modular
relationship Is described 1n somewhat more detail In Chapter A-9, where
composite regional models are discussed.
Scavenging models are currently rapidly evolving, and a profusion
of associated computer codes and computational formulae Is currently
available. Indeed, one of the major problems 1n precipitation-
scavenging assessment Is determining precisely which model to select
from the large number of available candidates. A major aim of the
present subsection 1s 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:
0 Predicting the Impact on precipitation chemistry of proposed
new sources, source modifications, and alternate emission-
control strategies;
0 Predicting long-range precipitation chemistry trends;
0 Estimating relative contributions of specific sources to
precipitation chemistry at a chosen receptor point;
o Estimating transport of acidic precipitation precursors
across political borders;
0 Estimating and predicting a1r-qual1ty Improvements occurring
as a consequence of the scavenging process;
o Selecting sites for precipitation-chemistry network sampling
stations;
0 Designing field studies of precipitation scavenging; and
0 Elucidating mechanistic behavior of the scavenging process on
the basis of field measurements.
6-51
-------
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 configuration, precipitation type, and the 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, so 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+] = E Anions - ( I 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 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 empirically. Currently, newer models tend to
accommodate larger numbers of these species; but complete modeling
coverage of them will not be achieved in the foreseeable future.
Mechanistic detail is another important feature determining the
basic composition of a scavenging model. A comprehensive mathematical
description of the scavenging process can rapidly become 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 resulting simplified expressions
are termed parameterizations. Consolidating the effects of non-modeled
species in empirical form, described in the preceding 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.7 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
7For the reader interested in more detailed pursuit of this area, the
works by Hales (1983) and Slinn (1983) are recommended. The Hales
reference is something of a beginner's primer, while SI inn's treatment
delves substantially deeper into mechanistic detail. Together they
constitute a reasonable starting point for understanding and modeling
basic scavenging phenomena.
6-52
-------
TABLE 6-4. SOME INORGANIC IONS IMPORTANT
IN PRECIPITATION CHEMISTRY3
Cations Anions
H+
NH4+ CT
Na+ N03~
K+ S032-
Ca2+ S042'
Mg2+ P043'
C032-
3A11 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~).
6-53
-------
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 on such endeavors.
6.5.2 Elements of a Scavenging Model
6.5.2.1 Material Balances—In Figure 6-3 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 include material balances, which form the underlying
structure for all scavenging models. To formulate a material balance,
one simply visualizes some chosen volume of atmosphere, summing overall
inputs and outputs of the substance in question.
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 equations
(cf., Hales 1983)
• WA + rAy (9as phase) [6-2]
ctt
and
^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 media) 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.
6-54
-------
CAx= -V.CAX^AX + WA + rAx (aqueous phase). [6-3]
Here CA» and CAX denote concentrations of pollutant 1n the gaseous
and condensed-water phases, respectively. The time rate of change of
these concentrations within the differential volume element 1s related
to the sum of inputs by 1) flow through the walls of the element, 2)
Interphase transport between the gaseous and condensed phases, and 3)
chemical (and/or physical) reaction within the element. The ^
terms in Equations 6-2 and 6-3 denote velocity vectors, while v. 1s
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
and CAX- A simple example of this procedure is given in Section
6.5.4.
6.5.2.2 Energy Balances—Many terms in Equations 6-2 and 6-3,
especially yAx,*A» and rAx, depend strongly upon the amount,
state, and interconversion 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 including an
energy-balance equation 1s mandatory. This need arises because the
evaporation-condensation process Influences, and is Influenced by, a
variety of energy-related considerations. These include temperature
influences on vapor pressure and latent-heat effects, which 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
pv 3j_ = _ 7afj . pv.v + r _ D . [6-4]
3t
Here the time rate of change of temperature relates to the sum of inputs
by 1) flow through the walls of the element and 2) generation via a)
compression work, b) latent heat effects, and c) frictional dis-
sipation. The vector terms h and v denote sensible heat flux and fluid
velocity, respectively, while r and D pertain to latent heat and
6-55
-------
dissipation; P and Cy denote fluid density and specific heat in the
usual sense. A straightforward example of the incorporation of Equation
6-4 for scavenging modeling purposes is given by Hales (1982).
6.5.2.3 Momentum Balances—Solutions to Equations 6-2 to 6-4 depend
upon the existence of some previous description of fluid velocity v
(or VAV in the case of Equation 6-2). As was the case for the
preceding parameters associated with the energy balance, velocity may be
specified for the model on the basis of previous measurements or
assumptions. Flow patterns in storm systems may be sufficiently complex
to defy empirical specification, however, and the modeler may wish to
compute the associated fields on the basis of a modeling approach. If
this is to be done, a momentum-balance equation must be employed. In
microscopic form the general momentum balance may be expressed (cf.,
Bird et al. 1960) as
3pv = -v.pvv -vp - FV + Pg- [6-5]
"at"
Here the time rate of change of momentum (PV) is expressed as
the sum of inputs by 1) flow through the walls of the element, 2)
pressure forces, 3) viscous drag forces, and 4) gravitational forces.
To apply Equation 6-5 for modeling purposes, one specifies frictional,
pressure, and gravitational 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 exist. 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 considering a hydrometeor falling through a volume of polluted
air space, as shown in Figure 6-20. This hydrometeor sweeps out a
6-56
-------
o
Figure 6-20.
Schematic of a scavenging hydrometeor falling through a
volume element.
6-57
-------
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, we know attachment efficiency
accounts for a multitude of processes. Usually the efficiency is less
than 1; but mechanisms such as diffusion, electrical effects, and
interception 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, such as in the case of pollutant-gas desorption. Typical
efficiencies for aerosol particles collected 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
_ [6-6]
_
CAy
where (in a manner consistent with Equations 6-2 and 6-3) w/^ is the
rate of 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 expression 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)
A(a) = - 7rNT A2vz(R)E(R,a)fR(R)dR , [6-7]
0
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 parameteriza-
tion over the total spectra 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 corresponding 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.
6-58
-------
10
i
FRONTAL RAIN SPECTRUM
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 Hales9(1976).
-------
Including reversible attachment processes in a scavenging model
usually involves using the mass-transfer coefficient. This parameter can
be defined in terms of the flux of pollutant moving from the scavenging
element as
Flux = - . (cAy - h'cA) . [6-8]
Here Ky is the mass-transfer coefficient and 6A is the concentration,
within the scavenging element, of collected pollutant; h1 is essentially a
solubility coefficient which, when multiplied by cA, 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
has been defined. Thus, the flux can be either to the drop or away
rom it, depending upon the relative magnitude of the parenthetical
terms. Equation 6-8 can be integrated 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 = _l!±L_ /°VfR(R)Ky(R) (cAy-h' CA) dR
The final scavenging parameter to be described here is the
scavenging ratio. This entity is usually the result of a model
calculation, rather than an input, and is defined by the form
C,
5 = JL [6-10]
cAy
where CA is the concentration of pollutant contained in a
collected precipitation sample. 5 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
washout 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
6-60
-------
removal. A= 1 hr-1, for example, would imply that the scavenging
process will cleanse 100 (1-1/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-i and beyond have been
reported in the literature (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 congnizant 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. S's ranging from below 100 up through
ID** 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 magnitude or more. Errors in
these efficiency estimates will of course be compounded 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 mechanisms and the complex environments involved. Typical
uncertainties in both A and 5 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 5 is given in Figure 6-23.
In all cases involving reactive gases, the values of E, A, and
£ 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.
6-61
-------
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 macroscopic 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 Approaches'
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 formulations that demonstrate the
microscopic and macroscopic approaches to the problem. The present
subsection 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 that contains a well-mixed, nonreactive pollutant with
concentration CAy The 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 1s well-mixed, no concentration gradients
occur; thus, the divergence term 1n Equation 6-2 is zero. Because of
nonreactlvity 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 1s, then,
= - A cAy . [6-2a]
at
Given some initial pollutant concentration CAyo» Equation 6-2a can be
integrated to obtain the form
CAy (t) = 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.
6-62
-------
The reader is cautioned to consider this treatment as an example
only and to recognize that actual atmospheric conditions seldom conform
to the idealizations invoked above. Gas-phase concentrations are
usually not uniformly distributed in space, raindrop characteristics are
usually not 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 pollutant often is not
unreactive. Examples of existing models where these constraints are
relaxed in various ways are presented in the following subsection.
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
CAy = airborne concentration of pollutant
H = airborne concentration of water vapor into cloud
CA = concentration of scavenged pollutant in rainwater
w = density of condensed water
w-jn = flow rate of water vapor into the storm
wout = fl°w rate °f water vapor out of the storm
fjn = flow rate of pollutant into the storm
fout = flow rate °f 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
£P = W [6-12]
and e = . [6-13]
nn
If one further performs material balances over this storm system for
pollutant and water vapor, and then combines the two, the following form
is obtained:
5=fA = epPw [6-14]
cAy "FFT
where the scavenging ratio, £ , is as defined earlier in Section 6.5.3.
6-63
-------
CONDENSATION,
PRECIPITATION FORMATION,
POLLUTANT ATTACHMENT
FLOW RATE OF WATER VAPOR OUT = w
FLOW RATE OF POLLUTANT OUT = fQut
x^
/v
FLOW RATE OF WATER VAPOR IN = w,.
FLOW RATE OF POLLUTANT ... -1n
^IBHI
3OR IN = win\ \\V\\MV\^V\\\v
riN = f.n \m\\A
FLOW RATE OF PRECIPITATION OUT = W
FLOW RATE OF SCAVENGED POLLUTANT OUT = F
DEFINITIONS OF EFFICIENCIES:
WATER REMOVAL
E_ «
POLLUTANT REMOVAL
Figure 6-22. Schematic of a typical macroscopic material balance.
6-64
-------
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 efficiency^
(ep = e ) for example, then
5 -fj -10-5 „ 10-6- [6-151
Experimentally-measured scavenging ratios often fall in this range,
although wide variability often may be observed.
Using a rather involved series of arguments pertaining to
cloud-physics processes 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 that 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
homogeneous systems, are discussed in references given in the following
section.
6.5.5 Systematic Selection of Scavenging Models: A Flow-Chart
Approach
Hales (1983) 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, the temporal and geographical scales, the
pollutant characteristics, the choice between macroscopic and micro-
scopic material balances, and the type of conservation (i.e., material,
energy, momentum) equations involved. Various pathways through this
decision tree are discussed in the original reference.
^There is no direct reason to expect that ep should be similar to
ein magnitude. In the absurd circumstance where all the pollutants
were concentrated into one particle, for example, then scavenging of
that pollutant by a very light rainfgall would yield e=1.0»ep.
Conversely a large storm processing an insoluble gaseous pollutant
(SFs, say) would provide e=0« p. For practical conditions involving
acid-forming aerosols, however, the scavenging of vapor and water
pollutant appears to be sufficiently related to allow en^eto be
employed as an approximate rule-of-thumb.
6-65
-------
99-9
to
c
CD
cn
ro
u>
U> CO
i o
o o
O c+
— ' rb
Q. -
o o
-S Q»
3 <
to (D
» 3
IQ
"
n> -s
o>
DO r-t-
fD ->•
-s o
CO
ro o
-5 C
o -s
3 <
i ro
TI (/>
3
Q.
fD 1— >
in o
fD O
3 3
T3 fD
-S O
O c+
O ->•
fD <
l/l fD
to
1/5
-•• c+
l/i O
O to
rt »•
< INi
, -- QJ
c/> -s
O 3
o »
n-
r+ 3
O
(-• 3
U3 O
*~J O
00 3
*"•— ' ** <
• fD
O
n-
fD
to
r+
O
SCAVENGING RATIO (VOLUME BASIS) (5)
°oo
i i r \ 1111 j
TO
m
o
I—I
•o
-------
PRECIPITATION
CHARACTERISTICS
AND
CONCENTRATION
FIFID
PRECIPITATION
CHARACTERISTICS ,
/AND CONCENTRATION^
FIELD OR
SOURCE STRENGTH
PRECIPITATION
CHARACTERISTICS
AND
CONCENTRATION
FIELD
PLUME MODEL
COMPUTE GASEOUS-
ANO AQUEOUS-PHASE
CONCENTRATIONS
1
PRECIPITATION /
CHARACTERISTICS /
NO CONCENTRATION/
I ELD OR SOURCE /
STRENGTH /
COMPUTE WASHOUT
COEFFICIENTS
'
PARAMATERIALIZE
AEROSOL SIZE
DISTRIBUTION
J
J~
IS CONDENSE
OF WATER
PRIMARY AE*
SIGNIFICAI
|NO
COMPUTE SI
01STRIBUTIO
SECONDARY Al
TION
ON
OSOL
-------
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 application. 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 inadvertent use of inappropriate physical constraints.
For a more definitive description of this model selection process, the
reader is referred to Hales' original reference.
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 that 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
scavenging 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 9ne
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 relatively 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 perf9rmance
for these applications. 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).
6-68
-------
TABLE 6-5. PERTINENT LITERATURE REFERENCES FOR WET-REMOVAL MODELS
Model
Type of Balance
Equation! s)
Mechanism(s)
Typical Application
Pertinent References
en
i
cr>
1. Classical Washout
Coefficient
2. Distributed Washout
Coefficient
3. "Two-Stage" Nuclca-
tion-Accretion
Nonreactive Gas
Scavenging
Reactive Gas
Scavenging
6. In-Cloud Aerosol
Scavenging
7. In-Cloud Aerosol
Scavenging
8. In-Cloud Reactive
Gas and Aerosol
Scavenging
9. In-Cloud Reactive
Gas and Aerosol
Scavenging
Material
(Differential)
Material
(Differential)
Material
(Differential)
Material
(Differential)
Material
(Differential)
Material
(Differential)
Irreversible Attachment
Irreversible Attachment
Irreversible Attachment
Reversible Attachment
Reversible Attachment
with Aqueous-Phase
Reaction
Irreversible Attachment
Material (Integral) Irreversible or
Reversible Attachment
Below-cloud scavenging
of aerosols and reactive
gases
Below-cloud scavenging of
size-distributed aerosols
Condensation-enhanced
below-cloud scavenging of
aerosols
Below-cloud scavenging of
nonreactive gases
Below-cloud scavenging of
reactive gases
Scavenging in storm systems
(nonreactive)
Scavenging in storm systems
Material
(Differential)
Material (Integral)
Transport, Reaction and Scoping studies
Deposition
Irreversible or
Reversible Attachment
with Chemical Reaction
Interpretation of field
study data
Chamberlain (1953), Engelmann (1968), Fisher
(1975), Scriven and Fisher (1975), Wangen and
Williams (1978)
Dana and Hales (1976), SI inn (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), Slinn (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)
Mechanlsm(s)
Typical Application
Pertinent References
10. Composite Analytical
Material Transport, Reaction and Regional scale deposition
(Differential) Deposition
Astarlta et al. (1979), Fay and Rosenzwelg
(1980)
11. Composite Trajectory Material
(Differential)
en
i 12. Composite Grid
~-j
o
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
Material Energy and Irreversible Attachment, In-cln"<1 scavenging analysis
Momentum Honreactive
(Differential)
Material and Energy All modes of scavenging In-cloud scavenging analysis
(Differential) Including chemical
reaction
Bolln and Persson (1975), Hales (1977),
Ellassen (1978), Fisher (1975), Bass (1980),
Heffter (1980), Henml (1980), Sampson (1980),
Bhumralkar et al. (1980), Klelnman et al.
(1980), Shannon (1981), McNaughton et al.
(1981) Patterson et al. (1981); Voldner (1982)
Liu and Durran (1977), Prahm and Christensen
(1977), W1lken1ng and Ragland (1980), Lavery
(1980), Lee (1981), Carmichael and Peters
(1981), Lamb (1981)
Rodhe and Grandell (1972, 1981)
Molenkamp (1974), Hane (1978), Kreltzburg and
Leach (1978)
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.
The predictive capability of a scavenging model depends upon the
choice of model .
At first sight this appears to be a self-evident and trivial
statement. A profusion of scavenging models exist, however, and it
is not at all difficult to choose an inappropriate candidate
inadvertently. Such inappropriate selections have on occasion
resulted in reported calculations that have been in error by
several orders of magnitude (Section 6.5.1).
This component of error may of course be totally eliminated by
selecting the most appropriate model for the intended application.
The flow chart presented in Figure 6-24 is a useful guide for this
purpose, especially for those only casually familiar with the
field.
The predictive capability of a scavenging model depends strongly
upon the processes model ecT
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.
The predictive capability of a scavenging model depends upon its
areal 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
Owing to the propensity of stochastic phenomena to average out to
mean values, the predictive capabilities of (especially regional)
6-71
-------
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 necessary (again requiring a precise definition
of intended model application) at this juncture.10
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 $03
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 and 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
contributing to wet removal of pollutants and has sumamrized 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:
*°This 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.
6-72
-------
0 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?
o What must be accomplished before the present capabilities can
be improved?
The answers to these questions are somewhat mixed. Certainly the
techniques discussed in this section, if used appropriately, are capable
of order-of-magnitude determinations in many circumstances; and under
restricted conditions 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 satis-
factory 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 section against the backdrop
of these deficits, several research needs become apparent. The most
important of these are itemized in the following paragraphs:
0 Much more definitive information is needed with regard to the
scavenging 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. Since 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).
6-73
-------
0 Much more extensive understanding of the competitive
nucleation capability of aerosols In in-cloud environments Is
needed, especially for those substances that do not compete
particularly well In the nucleatlon process. The Influence of
aerosol-particle composition—especially for "Internally-
mixed" aerosols (those containing individual 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 accomplished capability to describe
long-range pollution transport. Progress in this area during
recent years has been encouraging, but much more remains to be
achieved before we are sufficiently proficient for reliable
source-receptor analysis (Section 6.4).
o We still need to enhance our understanding of the detailed
microphyslcal and dynamic processes that occur in storm
systems. Besides providing required knowledge of basic
physical phenomena, such research is important in providing
valid parameterizations 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 scavenging 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.
6-74
-------
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 G. lorio. 1979. Theory of dispersion
transformation 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.
Tellus 31:1-39.
Barrie, L. A. 1978. An improved model of reversible S02 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. Met. 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. Met. 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. Met. 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.
6-75
-------
Carmichael, G. 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. November.
Changnon, S. A. 1968. Precipitation scavenging of Lake Michigan Basin.
Illinois State Water Survey Report, Bull. 52. 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. Met. 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. j£
Precipitation Scavenging 1970. R. J. Engelmann and W. G. N. Slinn, 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.
Precipitation 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.
Battelle-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.
6-76
-------
Dana, M. T., N. A. Wogrnan, 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
participate 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. Met. 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. Met.
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.
Easter, R C. and J. M. Hales. 1983a. Interpretations of the OSCAR
data for reactive gas scavenging. Proc. Fourth International Conference
on Precipitation Scavenging, Dry Deposition and Resuspension, Santa
Monica.
Easter, R. C. and J. M. Hales. 19835. Mechanistic evaluation of
precipitation-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.
Ir± 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. Met. 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.
6-77
1+09-261 0-83-18
-------
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. R. 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. In 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. Atm. 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. I.., T. Bergeron, J. Bjerkness, and R. E. Bundgaard. 1957.
Dynamic Meteorology and Weather Forecasting. American Met. Soc.,
Boston, MA.
Graedel, T. E. and J. P. Franey. 1977. Field measurements of submicron
aerosol washout by rain. J^ 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. Atom. Environ.
12:691.
6-78
-------
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. Atm. Env. 16(7):1775-1783.
Hales, J. M. 1983. Precipitation chemistry: Its behavior and its
calculation. In Air Pollutants and Their Effects on the Terrestrial
Ecosystem. S.T. Krupa and A. H. Legge, eds. John M. Wilev, New York,
NY.
Hales, J. M. and M. T. Dana. 1979a. Precipitation scavenging of urban
pollutants by convective storm systems. J. Appl. Met. 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
Journal 19:292-297.
Hane, C. E. 1978. Scavenging of urban pollutants by thunderstorm
rainfall: Numerical experimentation. J. Appl. Met. 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 $03 and sulfate and
its application to the eastern United States. J.G.R. 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.
6-79
-------
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. 1964. The modification of aerosol size distribution in
the atmosphere. Final Tech. Report, Meteor. Geophys. Inst., Johannes
Gutenberg Universetat. U.S. Army Contract DA-91-591-EVC2979.
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. Met. 15:98-102.
Kleinman, L. J., J. G. Carney, and R. E. Meyers. 1980. Time dependence
on average regional sulfur oxide concentrations. Proc. Second Conf. on
App. Air Pol. Met. 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, EPA/ESRL,
Research Triangle Park, NC.
Lange, R. and J. B. Knox. 1977. Adaptation of a three-dimensional
atmospheric 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,
and Soil Pol. 4:319-328.
6-80
-------
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 Poll. Met., New Orleans, LA.
Lee, H. N. 1981. An Alternate Psuedospectral 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. Battene-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
deposition 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:
Statistical 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.
6-81
-------
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.
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.
Met. 20:404-420.
Prahm, L. P. and 0. Christensen. 1977. Long-range transmission of
pollutants simulated by a two-dimensional psuedospectral dispersion
model. J. Appl. Met. 16:896-910.
Pruppacher, H. R. and J. D. Klett. 1978. Microphysics of Clouds and
Precipitation. D. Reidel Pub. Co., Boston, MA.
Radke, L. F., M. W. Eltgroth, and P. V. Hobbs. 1978. Precipitation
scavenging of aerosol particles. Proc. Cloud Physics and Atmospheric
Electricity. American Met. Soc., Boston, MA.
Raynor, G. S. 1981. Design and preliminary results of the intermediate
density precipitation-chemistry experiment. Report BNL 29992. For
presentation 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
turbulent clouds. J. Fluid Mech. 1:16-30.
Sampson, P. J. 1980. Trajectory analysis of summertime sulfate
concentrations in the northeastern United States. J. Appl. Met.
19:1382-1394.
Scott, B. C. 1978. Parameterization of sulfate removal by
precipitation. J. Appl. Met. 17:1375-1389.
Scott, B. C. 1981. Sulfate washout ratios in winter storms. J. Appl.
Met. 20:619-625.
6-82
-------
Scott, B. C. 1982. Predictions of in-clpud 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. App. Met. 18:138-147.
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.
Slinn, 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
reversible 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. J. Water, Air and Soil
Poll. 7:513-543.
Slinn, W. G. N. 1983. Precipitation scavenging. In Atmospheric
Sciences and Power Production. D. Randerson, ed. TJTS. DOE.
Slinn, W. G. N. and J. M. Hales. 1971. A revaluation of the role of
thermophoresis 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
particles. EPA MONOGRAPH SERIES.
Squires, P. and S. Twomey. 1960. The relation between cloud droplet
spectra and the spectrum of cloud nuclei. Jji 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.
6-83
-------
Summers, P. W. and B. Hi tenon. 1973. Source and budget of sulfate in
precipitation from Central Alberta, Canada. JAPCA 23:194-199.
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. Atm. 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. 1982. Comparison
between measured and computed concentrations of sulfur compounds in
eastern North America. J.G.R.
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. Ragland. 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. Won. Wea. Rev. 108:387-401.
6-84
-------
THE ACIDIC DEPOSITION PHENOMENON AND ITS EFFECTS
A-7. DRY DEPOSITION PROCESSES
(B. B. Hicks)
7.1 INTRODUCTION
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 discussed in the previous chapter.
These substances also impact on various receptors in the form of dry
depositions. 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, NOx, 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 biological factors
that play an important role in the processes controlling the rate of dry
deposition as a function of time and space. These 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 controlled 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; nevertheless, a "deposition velocity", v
-------
Particles larger than about 20 vm diameter will be deposited at a
rate controlled by Stokes1 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 considerable barrier to all mass fluxes
and especially to gases with very low molecular diffusivity. The
concept of a gravitational settling velocity is inappropriate in the
case of gases, but transfer is still often limited by diffusive
properties very near the receptor surface. 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 constructed on the basis of SehmeVs 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 distribution) is the
mechanically generated particle range for which gravity causes most of
the deposition. In most literature, the 2 ym 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
7-2
-------
AIRBORNE SOURCE
LARGE
PARTICLES
GASES
AERODYNAMIC
FACTORS 1
NEAR-SURFACE
PHORETIC
EFFECTS
QUASI-LAMINAR
LAYER
FACTORS
SETTLING
I
TURBULENCE
THERMOPHORESIS
I
ELECTROPHORESIS
DIFFUSIOPHORESIS
-, mr4
and
STEFAN FLOW
IMPACTION
i
INTERCEPTION
BROWNIAN DIFFUSION
1
IMCll C
rlULt
1
TURBULENCE
STEFAN FLOW
CULAR DIFFUSION
SURFACE
PROPERTIES
IFLEXIBILITYI | WAX i NESS
|STOMATA| | WETNESS j
i
CHEMISTRY |
SMOOTHNESS | { VESTITURE
j EMISSIONS
MOTION] | EXUDATESI
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
-------
15
X
CO
10
0.001
(a)
0.01
0.1 1
DIAMETER (yin)
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
-------
01
E
a.
a
en
o
3
-------
and deposited acids. Moreover, it has been suggested that some of these
larger particles may provide sites for the 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 advocated. 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
inertia! 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
surface, most turbulence energy will be associated with time scales
longer than 0.01 seconds, so that even 100 pm diameter particles would
follow most turbulent 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.
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
(ym)
Diffusivity
(cm2 s-1)
Stopping Time
(s)
Settling Speed
(cm s"1)
0.001 1.28 x 10'^ 1.33 x 10"^ 1.30 x 10"^
0.002 3.23 x 10"^ 2.67 x 10"^ 2.62 x 10"?
0.005 5.24 x 10"; 6.76 x 10"^ 6.62 x 10~°
0.01 1.35 x 10"? 1.40 x 10"° 1.37 x 10~j?
0.02 3.59 x 10~l 2.97 x 10"° 2.91 x 10"^
0.05 6.82 x 10~£ 8.81 x 10"° 8.63 x 10"J
0.1 2.21 x 10"° 2.28 x 10"; 2.23 x 10"^
0.2 8.32 x 10"; 6.87 x 10"' 6.73 x 10~5
0.5 2.74 x 10"; 3.54 x 10"° 3.47 x 10",
1.0 1.27 x 10"' 1.31 x 10"^ 1.28 x 10"^
2.0 6.10 x ID"" 5.03 x 10"J 4.93 x 10"f
5.0 2.38 x 10'° 3.08 x 10"J 3.02 x 10"1
10.0 1.38 x 10"b 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 ? 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
-------
Thus, trace gases and small particles are carried by atmospheric
turbulence as if they were integral components of the air itself,
whereas large particles 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.
In daytime, the lower atmosphere is usually well mixed up to a
height typically 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 possible mechanisms. Atmospheric transfer
does not usually limit the rate of delivery of pollutants to the surface
boundary layer in which direct deposition 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 contentious. 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 quantity such as
heat or momentum is effectively lost. There is even considerable
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 associated 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.
Inertial impaction is a process that augments gravitational
settling for particles in the size range typically between 2 and 20 ym
(SI inn 1976b). Larger particles tend to bounce, and capture is
therefore less efficient, while smaller particles experience difficulty
7-9
-------
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 Inertlal impaction. However,
from the viewpoint of acidic particles, Inertia! Impactlon 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
informative 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 over (and under) the plate, a laminar layer
develops, of thickness <5 = c(vx/ul/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, vj_, across the quasi-laminar layer is proportional
to the friction velocity u*:
vx = 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
represented 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
7-10
-------
10~F I—i i i i i IN
- • '
i I I I I II
1 I I I MM
O
10
-4
4
LEGEND
O HARRIOT and HAMILTON (1965)
A HUBBARD and LIGHTFOOT (1966)
• MIZUSHINA et al. (1971)
10
-5
J I I I I Mi
J I I I I I II
J I I I I III
10*
10'
Figure 7-4. Laboratory verification of Schmidt-number scaling for
particle transfer to a smooth surface. The quantity plotted
is BEVd/u*, evaluated for transfer across a quasi-laminar
layer of molecular diffusion immediately adjacent to a smooth
surface. Data are reported by Lewellen and Sheng (1980).
The line drawn through the data is Equation 7-1, with
exponent a = -2/3 and constant of proportionality A = 0.06.
7-11
-------
span the value of A - 0.2 recommended for the case of sulfur dioxide
flux to fibrous, vegetated surfaces (Shepherd 1974, Wesely and Hicks
1977).
Laminar boundary layer theory imposes the expectation that particle
deposition to exposed surfaces will be strongly influenced by the size
of the particle, 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 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
urn diameter, according to Davies 1967), this "thermophoresis" 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 thermophorectic velocity of very small particles (< 0.03
viti diameter) is likely to be about 0.03 cm s~l (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 ym 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 complicated that they constitute specialities in their
7-12
-------
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. (Photgraph by R. L. Hart, Argonne National Laboratory)
7-13
-------
own right. Excellent discussions have been given by Friedlander (1977)
and Twomey (1977). These phoretic forces are generally snail, and their
influence on dry deposition can usually be disregarded.
Many workers include Stefan flow in general discussion 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 evaporating 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 evaporates from a 1
m2 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~r
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 than the deposition velocity appropriate for many small
particles to aerodynamically smooth surfaces.
Electrical forces have often been mentioned as possible mechanisms
for promoting 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 (1960) 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~l, experienced considerably 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 electrostatic forces to assist
the transfer of partial!ate pollutants to vegetative surfaces were
conducted by Langer (1965) and Rosinski and Nagomoto (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~l. 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 reentrainment could take place above ~ 2 m s-1 wind
speed. Tests on branches of cedar and fir by Rosinski and Nagamoto
(1965) suggested similar results as for single needles." 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~* 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).
7-14
-------
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 possibility of bounce and
resuspension.
The role of electrostatic attraction in binding deposited particles
to substrate surfaces remains something of a mystery. The process by
which particles become charged and set up mirror-charges on the
underlying surface is fairly well accepted. For smaller particles, the
principle charging mechanism 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 particles, 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 relationships for
evaluating the electrical energy of a particle on the basis of its size,
shape, dielectric constant, and the strength of the surrounding
electrical field.
Condensation of water reduces the effectiveness of electrostatic
adhesion forces, since 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 surface 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
7-15
-------
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 considerable attention. Chamberlain (1967) tested the roles of
leaf stickiness and hairiness in his wind-tunnel tests. He concluded
that "with the large particles (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 21°Pb and 2l°Po
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.
7.2.7 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. Thus, 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 considered the role of Henry's law constant and
chemical reactivity in controlling 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
7-16
-------
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 $03 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 SI inn et al. (1978) argue
that it is better to introduce molecular diffusivity through a term
analogous to the Schmidt (or Prandtl) number of Equation 7-1, with the
exponent a - -2/3. (In contrast, the linear assumption used by Hicks
and Liss implies a = -1.0). 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
ramifications of these alternative assumptions is not warranted.
In the limiting case of a trace gas of low solubility, the
deposition velocity 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 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 velocites 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
7-17
-------
DEPOSITION VELOCITY (cm s"1)
00
(0
c
r o
™ 2
<-(• S JO
3- 0» fD
fD rt- (Si
fD C
c+ -S — '
fD ft-
-S to to
_j. -S O
3 -h -*
ftl pi
—•OS O
fD -••
10 to 3 !-•
fD • Q-
<~t* ~t3
rf rt- >
—•-HE 33
->• 3" 3 — I
3 fD 3 l-i
tQ fD O
Q. — ' 1—
in a> m
T3 trt to
fD 3- r+ O
fD fD C i— i
Q. Q. Q. ~»
O -j. to —( *
-S3 m °
fD o ya
>-> -h
tn c+XJ "C
O> 3
IQ r+ -S ^*
3- c-t-
O fD -J'
3 0
1 -S — •
CQ
T3 3* CL
O» c+ fD
-s x> J^
rt -S 0 O
-•• a> v>
O T3 -<•
"^ ~5 c™t"
fD fD -••
o q 0 50
I -H c
i ^"*^ "^
v\ ^ 55
— \ A ^y
\ ^ ^™x ^^
\ ^™*
Xw ^ "^ ^
-^ ^^ \ o -^ 15
^^ A ^L A ^^
^^.^ x. ^^^
^ ^^X^
^^ ^^
*^^^^_
^^*^^^
^^^^
O^^O
v^
^^
— x^
-fc
^J ^^^
^^
— ^
1 II II
1— »
o
1 1
—
^_
1—
m
en —
m
o c
^^
—i
m _
__
^^ ^^
^^
^Sw^
c^^
1 *^ 1
-------
Figure 7-6 might indicate, but they present an alternative hypothesis
for this more rapid transfer: that hygroscopic particles 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)
summarize 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 deposition 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. 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.
7-19
-------
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, similar 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
processes 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. 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 droplets that in other circumstances would grow and finally
precipitate in substantially 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. Most of the net
deposition of acidic species is by cloud droplet interception.
7-20
-------
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 redeposited. 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 particular size to rise from the surface. Bagnold
(1954) identifies 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 resuspending surface particles and carrying
them to considerable altitudes. 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 (1982) 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
7-21
-------
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 reactions 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 considering 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
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 vc 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.
The surface boundary resistance, r^, (separated further in Figure
7-7 between components rbf and rbs, associated with foliage and
soil, respectively) 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"1 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
7-22
-------
rbs.
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.
7-23
-------
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
7-24
-------
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(£n(z/z0) + £n(z0/zoc) - yc
= ra + (ku*)-l . £n(z0/zoc) [7-3]
where ZQC is a roughness length scale appropriate for the transfer of
the pollutant. The residual boundary-layer resistance, rb = R - ra,
is then
rb = (ku*H . n(z0/zoc), [7-4]
which alternatively is written as
rb = (u*B)-l. [7-5]
B is, therefore, a measure of the non-dimensional ized 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 resistance and use the symbol ra for it, (e.g., O'Dell et
al. 1977).
Shepherd (1974) recommends using a constant value kB-1 =
£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 Equation 7-1, so that surface boundary
layer resistance would then be written as
rb - 5 Sc23/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 diagramatically 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 (or a forest floor), rcs.
In concept, it is also appropriate to differentiate between boundary
layer resistances rbf and rbs for transfer to foliage and soil,
respectively, as is shown in the diagram. Many other resistances can be
7-25
409-261 0-83-19
-------
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 r^f on the assumption that S02
uptake is via stomata alone (although this may indeed be a sufficient
approximation in most circumstances). Instead, it is more realistic to
estimate rcf from its component parts via
rcf - (rst-l + rcut-l)-l/(LAI) [7-7]
(following Chamberlain 1980), where rst is the stomatal resistance,
and r~U£ is the 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, sometimes 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, rst and rm) or
through the epidermis (involving 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 particle transfer.
Studies of leaves show that stomata are typically slits of the order of
2 to 20 ym long. For stomatal uptake of particles to be a controlling
7-26
-------
EPIDERMIS
SPONGY,
MESOPHYLLIC
CELLS
PALISADE
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, and transfer through stomatal pores (via r$t) into
substomatal cavities, with subsequent transfer to mesophyllic
tissue (via rm). The way in which the various resistances
are combined to provide the best visualization of the overall
transfer process in not clear-cut.
7-27
-------
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 geochemical 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, gravitationally-sett!ing 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 contribute 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.,
phorectic processes, Stefan flow, etc.), many of which will be
influenced by the detailed structure of the surface involved.
7-28
-------
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. Furthermore, 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
B-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
7-29
-------
results of a number of studies of 35S02 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. 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, 1971, 1972, 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~*, depending on the assumption
made regarding the contribution by gaseous S02 deposition. Similar
work by Barrie and Walmsley (1978) yielded average sulfur dioxide
deposition velocities 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 conceptually-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
7-30
-------
permit apportionment between gaseous and participate 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
deposition. 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 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 suggested by Hill's work
appears to be supported, providing some justification for extending the
ordering to CO, H2S, and (CH^^S 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
7-31
-------
TABLE 7-4. RESISTANCES TO DEPOSITION (S CM-1) 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
H?S
(CHo)-S
NO C
C09
°a
Nu9
so|
HF
62.0
3.6
7.7
-
-
1.3
1.1
••
67.0
16.0
5.3
-
-
1.7
1.7
mm
oo
m.
_
10.0
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)
7-32
-------
values representative 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 percent less than the values evident in
Figure 7-9 (for particles in the 0.2 to 1.0 ym 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 parti-
cle deposition to rough surfaces. Sehmel (1970) conducted similar wind
tunnel experiments, employing monodisperse particles ranging from about
0.5 to 20 ym 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~l; Sehmel et
al., u* - 19 cm s"1). 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 v
-------
CO
S=
DEPOSITION VELOCITY (cm s"1)
~~l
co
I— > • fD
VO CTi
--J C
OO O — '
o> 3 v>
. Q. o
_j. -(,
QJ
3 S
ro -j-
c+ 3
ro Q.
-s
fD
tQ — •
Oi CO
< el-
n> c:
— ' Q.
Q- O
fa -t,
T3
rt-TJ
CD O>
Q- -S
rh
-h -••
-S O
O — •
3 rt>
00 Q.
fD fD
3 O
fD CO
e-t-
0> -"•
e-t O
TJ
73
t—i
O
73 o
-------
co
CJ1
fD
COIQ TO
fD -S fD
3" a> co
CO
fD CU
rt- CO O
—• fD S.
. -a -"•
O 3
-—- ~S Q.
i—> ct-
tX> fD c-i-
~-J Q. C
00 3
CT CT 3
-—<< fD
I O
3" CO
O 01 r+
c 3 c
-S cr CL
fD -5 fD
—i co
-"• O
3 -(,
Q)
-S
1 fD
CL CL
O fD
DEPOSITION VELOCITY (cm s"1)
CO o
»• CO
Of rj-'
O- o'
3
-o
3>
O
m
a
m
-------
methods that impose no surface or environmental modification. In
concept, if an area is sufficiently homogeneous, flat, and contains no
areas of strong sources or sinks, pollutant fluxes can be assumed to be
constant with height. Therefore, questions regarding dry deposition can
be addressed by measuring the flux of material through a horizontal
layer of air at some more convenient level above the surface. The
intent of any such study is to investigate dry deposition fluxes in
carefully-documented natural situations to identify and quantify
controlling properties. The results of these investigations are formu-
lations of surface mechanisms, surface boundary layer resistances,
stomatal resistances, etc. The demanding site criteria are required to
enable these results to be obtained from the experiments; the surface
parameterizations that are derived are far more widely applicable.
Several micrometeorological methods are suitable for measuring dry
deposition fluxes in intensive case studies. The flux can be measured
directly by eddy-correlation, a process that evaluates instantaneous
products of the vertical wind speed, w, and pollutant concentration, C,
to derive the time-average vertical flux Fc as
Fc = pw'C' [7-8]
where Pis the air density and the primes denote deviations from mean
values. The over-bar indicates a time average. This is an extremely
demanding task and constitutes a specialized field of micrometeorology
in its own right. 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) considerably 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 (3C/3z), and its
corresponding eddy diffusivity (K). Knowledge of any two of these
quantities will permit the third to be evaluated. Often, when sensors
7-36
-------
-~J
oo
330
C02 (ppm) 319
308'
0.8n
0.2-1
27.0-
T (°C) 25.5-
24.0J
12:35
12:36
TIME (hr:min)
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.
-------
suitable for direct measurement of pollutant fluxes are not available,
assumptions regarding the eddy diffusivity are made to provide a method
for estimating fluxes from measurements of vertical concentration
gradients:
Fc =PK(9C/9z). [7-9]
Hicks and Wesely (1978) and Droppo (1980) have summarized a number of
critical 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 (v,j). In unstable
(daytime) conditions, smaller gradients would be expected for the same
V(j; 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
S02 (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 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 unidirectional 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 experimental care, the use of sensitive
equipment, and fairly complicated data analysis. They are more suitable
for investigating the processes that control dry deposition than for
monitoring the flux itself.
Nevertheless, some new techniques for dry deposition measurement
are presently under development. A "modified Bowen ratio" method is
being developed in the hope that it might permit an accurate
7-38
-------
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
response of the pollutant sensor is of long-standing interest (e.g.,
Oesjardins 1977) but has yet to be applied to the pollutant flux problem
with significant success.
7.4 FIELD INVESTIGATIONS OF DRY DEPOSITION
7.4.1 Gaseous Pollutants
Table 7-5 summarizes a numoer of recent field experiments on trace
gas deposition 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
deposition rates of most pollutants. These observations were made over
a pine 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
meterological properties, 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).
ra = 0.25 s cm-1
rfo = 0.25 s cm~l
rst = 1.0 s cm-1
rcut = 2.5 s cm'*
For deposition to dry soil, Fowler suggests using rcs = 10.0 s cm"*,
and rcs = 0 when the soil is wet.
Aerodynamic resistance, ra, influences the deposition of all
non-sedimenting pollutants. It is not possible for any trace gas to
have a deposition velocity greater than l/ra, i.e., about 4 cm s"1
in the daytime conditions of Fowler's experiment. Because of stability
7-39
-------
TABLE 7-5. RECENT EXPERIENCE ON TRACE GAS DEPOSITION TO NATURAL SURFACES
1
-F*
O
Worker
S02
Hill (1971)
Garland et al.
(1973)
Owers and Powell
(1974)
Shepherd (1974)
Method
35
S02 with stable S02 carrier
over alfalfa
35
S02 over pasture
35
S02 over pasture
S02 gradients over grass
Results and Comments
Vd = 2.3 cm s" (daytime)
Implies rc - 0.4 s cm~
Vd - 1.2 cm s (daytime)
rc = 0.6 s cm"
Vd - 1.3 cm s~ (daytime)
Vd - 1.3 cm s (daytime)
Whelpdale and Shaw
(1974)
Garland (1977)
Fowler (1978)
Dannevik et al.
(1976)
Garland and Branson
(1977)
S02 gradients over snow, water, and
grass
S02 gradients, calcareous soils
S02 gradients, over - wheat
- soybean
S02 gradients over wheat
35
S02 over a pine plantation
0.3 cm s (autumn)
- 1 cm s (daytime for
grass, water, and snow)
- 1.2 cm s~
rc - 0.01 s cm
-1
-1
- 0.4 cm s"
- 1.3 cm s
- 0.4 cm s
-1
-1
- 0.1 - 0.6 cm s
-1
-------
TABLE 7-5 CONTINUED
Worker
Method
Results and Comments
Belot (1975) (as
summarized by
Chamberlain 1980)
Gal bally et al. (1979)
Dovland and Eliassen
(1976)
Barrie and Walmsley
(1978)
34
over a pine plantation
Eddy correlation over pine forest
Accumulation to snow
Accumulation to snow
< 1 cm s
-1
= 0.2 cm s
- 0.1 cm s
-1
-1
- 0.2 cm s
-1
. NO,
Wesely et al. (1982b)
Eddy correlation
-soybeans
Vd - 0.6 cm s'1 (daytime)
rc = 1.3 s cm (daytime)
= 15 s cm" (night)
Gal bally and Roy
(1980)
Wesely et al. (1978,
1982b)
Gradients over wheat
Eddy correlation over a range of
natural surfaces
- 0.7 cm s
-1
Implies rc - 1.4 s cm
rc = 0.8 s cm~ (daytime)
- 1.8 s cm"1 (night)
-1
-------
SULFUR DEPOSITION (yg ni2 s1)
SENSIBLE HEAT (W rii2) FRICTION VELOCITY (cm s'1)
fD
^J
ro
O r+
O O
3 c+
<-+ OJ
-s —•
CT to
c c
CH- —'
-•• -h
O C
3 -S
CO
c:
OJ
DJ
Q.
ro
o
-h
Id
QJ
to
ro
o
OJ
3
Q.
O
DJ
d-
ro
i
H-1
ro
to -a rt- < 73
c. o o ro ro
—• -S —' O
-h rt- a> o o
c --• o -s
-s o -a -"• Q.
3 -"• rt- to
O to 3 "<
o ro o
O rt- -+)
Q.
to
to
ro
3
i/>
-J.
cr
— '
ro
3-
ro
3 3" 3 C —'
o ro rt-ca -t)
rt- &> 3" C
to rt- -S
cr c ->• oo
ro —' o -h
-h 3 a. —•
a. c cu c
ro -s ^^^ x
r+ 31 to -
ro a_ -<•
o a> o o
rt- r-h 7T -h
ro o to
r~> QJ
. -i. Q) 3
3 3
CL Q.
J» -<• 3
c-h O . .
a> ro ro 3-
Qi r+ to 3
—i ro ro to
__i
TD
c+ ro
-1- -s
3 -". vr> to c
ro o oo rt- x
to a_ o c -
u to ^ ^ Q_
• *s< QJ
r+ S 3
3-3- 00.
ro ro —i -h
33- -h
a. ro a. -s
OJ tO ~S —'•
rt- QJ Q.^ O
OJ to Q) rt-
ro -s Q- -••
-s o TT ro o
ro c ro -a 3
-fi to -s O
ro to
-s -"•
rt-
rt- -"•
O O
3
ro
o
o
to
73
O
o
o
o
o
-------
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 tabulation 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 v^. 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 cm~l during midday periods. At night,
however, vj can decrease to 0.1 cm s~* on infrequent occasions but
often is less than 2.0 cm s~l. Fowler's recommendations are probably
representative of the long-term average.
The importance of diurnal cycles in pollutant deposition and the
close relationship 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
rc for ozone and water vapor, with the former exceeding the latter by
a small amount, on the average. As the day progresses, rc increases
gradually, presumably as a consequence of increasing water stress and
eventual stomatal closure. The lower data sequence has two features of
considerable interest. First, the gradual initial decrease of rc for
03 corresponded to a period of evaporation of dewfall (note the rela-
tively low value of rf 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 H20). 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 expected with reduced insolation.
The proceeding discussion of both S02 and 03 deposition
confirms the generalization made by Chamberlain (1980) that the
deposition of such quantities might be modelled 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"1 during daytime, with rc for N0£ still greater by
a similar amount.
7-43
-------
W-L
IQ
(D
MAXIMUM DEPOSITION VELOCITY (cm s"1)
I
I— >
oo
3 3 QJ
a. < — •
fD C
z: -s fD
fD f t/>
in fO
(D O
— ' O -h
«< -h
<-h
*— »c+ 3"
h- > 3" (D
UD a>
CO 3
O O) Q>
-"-I'D X
. -s _i.
O 3
a. c
3 o
_i. CO
O to
-S CT
n> — •
in n>
«i.
w) c.
r+ fD
O) -Q
3 O
O t/>
n> -••
» rt-
-s o'
Qi 3
V
-h n>
o — i
-s o
n
a> ^<
•a o
-•• -h
3
O) r+
T3 Q)
—" O
CD 0)
3
e-HQ
-j. (D
o co
3 »
n> a.
x ro
T3 c+
ro n>
-s -s
fD 3
3 fD
c+ Q.
O CU
31 C+
->• 3"
O fD
O
•
o
o
o
O I-*
o
•
o
o
o
o
o
o
o
(D
to
O
O
O
O
00
*
00
-------
u
.p.
on
o
8
10
12 14
HOUR (CST)
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).
-------
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 applications 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 vj, 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 smoothness 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
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 cnr1 (with a standard
error margin of about + 15 percent) for 17 July, and about 1.1 s cm-1
(+_ 25 percent) for 18 Tuly.
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
v
-------
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 ym 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; vj too small
be determined.
to
Gradients highly variable.
Range of vj typically 0.2
- 1.0 cm s"1 in magnitude.
Including reversed gradients
in long-term average reduces
average v^ to near zero.
(See Hicks and Williams
1979).
Preliminary indications
only: vd very small, 95%
certainty < 0.05 cm s"*.
Surface frequently a
source: v
-------
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 ym
- 0.03 ym parti-
cles, gradients over
wild oats
0.05-0.1 ym parti-
cles, eddy correla-
tion
Everett et al. (1979) Particulate lead and
sulfur, gradients
- 0.04 cm s-1
Average vj = 0.9 cm
Direction of flux sometimes
changes. During deposition
periods, v^ - 0.8 cm
s~l, but much lower on the
average
vj greater for sulfur ( ~ 1
cm s-1) than for lead from
more local sources
Si even'ng (1982)
Hicks et al. (1982)
0.15-0.3 ym parti-
cle gradients over
mature rye and wheat
Sulfate by eddy
correlation
Wesely et al. (1982a) Sulfate by eddy
correlation
Crops
Droppo (1980)
Wesely and Hicks
(1979)
Particulate trace
metals, gradients:
senescent maize
Vd averaged 0.4 +_ 0.3 cm
s"l in light winds, unstable
stratification
Vd as high as 0.7 cm s-1
in daytime, about 0.2 cm
s'1 as a long-term average
vd largest for daytime lush
grass (- 0.5 cm s"1), much
less for short dry grass (~
0.2 cm s"1), strongly stable
conditions
varying widely with
fement, ranging up to about 1
cm s-1
0.05-0.1 ym parti- Strong diurnal variation in
cles, eddy correla- the direction of the flux.
tion: senescent maize Long-term average vd - 0.1
cm s~l
7-48
-------
TABLE 7-6. CONTINUED
Surface
Size and Method
Results and Comments
Trees
Hicks and Wesely
(1978, 1980)
Wesely and Hicks
(1979)
Sulfate particles,
eddy correlation,
Loblolly pine
0.05-0.1 ym parti-
cles, eddy correla-
tion
Strong diurnal variability
but less marked than for small
particles: average vj =
0.7 cm s"l
Very strong diurnal
variation with the canopy a
net source. During
deposition periods, vd
probably greater than 0.6 cm
Lindberg et al.
(1979)
Pb, Cd, S, etc. par- v,j > 0.1 cm s'1 for all
tides foliar washing quantities on the average
Wesely et al. (1982a) sulfate particles,
eddy-correlation
v
-------
ro
DEPOSITION VELOCITY (cm s"1)
-vj
I
en
o
i
i—»
en
->. ro i—' o cr o
3 x UD OJ << ro
CL rt- ~-J ~5 "O
—'• ro >*D o ro o
n 3 1 a. en
OJ CL • -j. Q. _j.
rt- ro 3 << rt-
ro CL OJ ->•
CL "Z. no
T3 O -"• O 3
CT ro r-t- 3 -S
<< -s ro -s <
—'• i—' ro ro
rt- O rt- UD —' —'
3" CL 3" ~-J Q) O
ro co ro *-j rt- o
3 o
ro -h
to
QJ ro
co
.—- o
re 3
< co
ro co
.. o n o>
3 3 7T CT
CQ to O
CL oj ro
• —'. 3
C CL O)
-s
-a -s oj ro ->•
O OJ —' CO 3
to r-t- ro ro
-"• 3" O —'
rt- ro ^ rt-T3
-j. -s n —
1 H-• O>
ro vo 3
= o
CL 3
ro
o
3
rt-
O
to
OJ
-o
o>
-s
rt-
O
ro
00
< OJ CO OJ
ro 3 s: •» <-(••
O Q-
n ro
-J.T3
rt- O
-J. 00
rt- s: o o
3- ro 3 •
00 i—•
-h ro ->.
-S —' 3 C
ro «< 3
ro -••
co rl--Q
= -•• C OJ O
— o ro 3 -s
• 3 3 CL , •
OJ
to
OJ
00
c:
ro
D_
O
c:
oo
-------
the nature of the surface present in the gradient studies is taken into
account.
Results of an extensive series of eddy correlation measurements of
particulate 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 case summarized by Wesely et al., normalization of
surface transfer conductances by u* appears to reduce the residual
variance. Hicks et alI. (1982) present supporting data from another
study of the same series, also over grassland.
Considerable controversy remains concerning the value of v^
appropriate for formulating the deposition of sulfate aerosol (and
presumably all similar particles). Garland (1978) advocates the
continued use of values of 0.1 cm s~l or less, because experiments
conducted over grass in England failed to detect a significant gradient.
However, some of the experiments listed in Table 7-6 indicate quite high
deposition velocities for sulfate particles. The possibility of a
strong contribution by particles much larger than the usual accumulation
size mode has been discussed (Garland 1978), and different deposition
velocities (0.025 and 0.56 cm s-1) have been postulated for the
submicron and larger particles, respectively.
There are great uncertainties about results obtained by deposition
plates or other surrogate collection surfaces. Workers sometimes assume
that the collection characteristics of some artificial surface are the
same as those of the natural surface of interest. Clearly, this
assumption will be valid when particles are sufficiently large that
gravity is the controlling factor. However, small particles are
transferred predominantly by turbulence, with subsequent impaction on
the surface of microscale surface roughness elements; these features of
the collecting surface are not easily reproduced by commonly-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 v^ 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
7-51
-------
completeness a brief summary of present capabilities to monitor dry
deposition should be given here.
It is important to recognize dry deposition for what it is: a
highly variable 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 deposition 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 meteorological
quantities such as sensible heat, moisture, and momentum. Even in these
cases, micro meteorological methods such as eddy correlation and
gradient 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
deposition 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.
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
7-52
-------
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 processes that control the uptake of various gases:
$02: 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 cuticular 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 SO?
flux.
03: Behavior is like S02 but with significant cuticular uptake at
night (rcut ~ 2 to 2.5 s cnr1 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.
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 NO?
is insoluble in water in low concentrations (see Chapter A-4),
deposition to water surfaces might be quite efficient. Chamber
studies (Table 7-4) indicat similar overall surface resistances
for S02 and N02.
NO: Typical canopy resistances are in the range 5 to 20 s cm-1, as
indicated 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 NO? deposition and of the
intimate linkage with ozone concentrations (Galbally and Roy
1980).
HNOs: No direct information is available; however, on the basis of its
high solubility and chemical reactivity, substantial similarity
to 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
similarity with S02 appears likely. Natural surfaces may be
emitters of NH3 because of a number of biological processes
occurring in and on soil.
7-53
-------
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 evaluations given in Figure 7-12 are especially
informative, because even over a pine forest whose surface roughness
operates to maximize v
-------
7.5.2 Particles
Modeling of particle deposition is complicated by three major
factors: (1) gravitational settling, which causes particles to fall
through the atmospheric turbulence that provides the conceptual basis
for conventional micrometeorological models (Yudine 1959); (2) particle
inertia, which permits particles to be projected through the near-
surface laminar layer by turbulence, 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 combination 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 Friedlander 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 Slinn
(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
resistance 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).
The above models are based largely on observations and theory
regarding the deposition of particles to smooth surfaces, usually of
7-55
-------
pipes. More micrometeorologically-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 concerning
evapotranspiration. A recent development of this genre is the canopy
model of Lewellen and Sheng (1980), which uses recent techniques in
turbulence 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 potentially critical mechanisms, especially
electrical migration, coagulation, evolution of particle size
distributions, diffusiophoresis, and thermophoresis. 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 confirmed wind tunnel results that point to a surface limitation.
Studies of the rate of deposition of particles to the internal walls of
pipes and investigations of fluxes to surfaces more characteristic of
nature, exposed in wind tunnels, tend to confirm theoretical
7-56
-------
expectations that surface uptake is controlled by the ability of
particles to penetrate a quasi-laminar layer 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 surfaces are fairly well understood. Studies conducted so far
support the theoretical expectation that particles smaller than about
0.1 ym in diameter 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, particles 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"1.
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 considerably 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
artificially-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 unexpectedly high deposition velocities for atmospheric
7-57
409-261 0-83-20
-------
sulfate particles (sometimes 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-1, while North
American experience has generated larger values.
It is informative to consider the flux of any airborne quantity to
the surface 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 associated 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 cnrl in strongly unstable, daytime conditions, to more than 10 s
cnrl 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 cavities to 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 NOg
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 cnrl. it appears probable that substantially larger
values for residual surface resistance will be appropriate for non-
vegetated surfaces, especially to snow, for which the values are more
likely to be approximately 15 s cnrl. At this time, an exceedingly
limited quantity of field information is available; however, it appears
7-58
-------
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.
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 particle uptake by snow appear to be of the order 15 s
cnrl. 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 S02, 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
exceedingly 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 surface. Once again
field observations are lacking.
While large deposition velocities of soluble trace gases to open
water surfaces might appear quite likely, water bodies are frequently
sufficiently small that an air-surface thermal equilibrium cannot be
achieved. Air blowing 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
resulting convective activity over small water bodies will induce the
air to come into fairly rapid equilibrium with the water, and rather
7-59
-------
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 accelerate the net transfer of trace gases and particles in some
circumstances. The velocities of deposition involved are small;
however, they permit an accumulation 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 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 micro-
meteorological means, without interfering with the natural processes
involved. These studies, and laboratory and wind tunnel investigations,
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
generalization 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:
o 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
conditions. The complexity of each individual process makes it
unlikely that a comprehensive simulation will be developed in
the near future (Section 7.2).
° 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 simplicity of the deposition velocity
7-60
-------
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).
Sufficient information is known about the processes controlling
the deposition of trace gases that in many instances deposition
velocities can be considered to be known functions of properties
such as wind speed, atmospheric stability, surface roughness,
and biological factors such as stomatal aperture. Impr«*t< nt
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
straightforward Stokes-Cunningham relationship. Smaller
particles are also 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 results of laboratory and wind tunnel studies, at
least for test surfaces that are usually smoother than pasture,
but field experiments 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
deposition 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
eventually 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).
7-61
-------
For both case-study measurement purposes and for long-term
monitoring, accurate measurements of pollutant air
concentrations are necessary. 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 velocities 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 applications is not yet possible (Section 7.3.3).
7-62
-------
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 - 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
deposition velocities to snow in northern Canada. Atmos. Environ.
12:2321-2332.
Batchelor, G. 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
analysis 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 scalars. 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.
7-63
-------
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. Pro. Roy. Soc. London, A, 296:
45-70.
Chamberlain, A. C. 1975. The movement of particles in plant
communities, pp. 155-203. Iji Vegetation and the Atmosphere, Volume 1,
Principles. J. L. Monteith, ed. Academic Press, London.
Chamberlain, A. C. 1980. Dry deposition of sulfur dioxide, pp.
185-197. J^ Atmospheric Sulfur Deposition. D. S. Shriner, C. R.
Richmond, and S. E. Lindberg, eds. Ann Arbor Science, Ann Arbor, MI.
Chamberlain, A. C. 1982. Roughness length of sea, sand and snow.
Boundary-Layer Meteorol. In press.
Chang, S. G., R. Toossi, and T. Novakov. 1981. The importance of soot
particles and nitroces acid in oxidizing S02 in atmospheric aqueous
droplets. Atmos. Environ. 15:1287-1292.
Gofer, W. R., D. R. Schryer, and R. S. Rogowski. 1981. The oxidation
of S02 on carbon particles in the presence of 02, N02 and N20.
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.
Danneyik, W. P., S. Frisella, L. Granat, and R. B. Husar. 1976. S02
deposition measurements in the St. Louis regions, pp. 506-511.
lr± Proceedings, Third Symposium on Atmospheric Turbulence, Diffusion,
and Air 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, California, 29 November - 3 December. In press.
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. j£
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. J^n Surface Contamination. B. R. Fish, ed.
Pergamon Press, New York. 415 pp.
7-64
-------
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. BrocF7 eds. 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-
correlation 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
calibrated lakes and watersheds for estimating atmospheric deposition
near a large point source. Water, Air, and Soil Pollut. In press.
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. In press.
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
measurements, pp. 209-221. In Atmospheric Sulfur Deposition, D. S.
Shriner, C. R. Richmond and "ST 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. Borman. 1978. The input of
gaseous and particulate sulfur to a forest ecosystem, Tellus 30:
546-551.
Englemann, R. J., and G. A. Sehmel. 1976. Atmosphere-Surface Exchange
of Particulate and Gaseous Pollutants. U.S. ERDA CONF-7409Z1, 988 pp.
7-65
-------
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. 13:931-943.
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 S02 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. 105:767-784.
Fried!ander, 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, Boulder,
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 Stratospheric/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.
7-66
-------
Garratt, J. R. 1978. Flux-profile relationships above tall vegetation.
Quart. 0. Roy. Meteorol. Soc. 104:199-211.
Gauri, K. L. 1978. The preservation of stone. Scientific American
238:196-202.
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
turbulent pipe flow. Chem. Eng. Sci. 20:1073.
Haynie, F. H. and J. B. Upham. 1974. Correlation between corrosion
behavior of steel and atmospheric pollution data, pp. 33-51. Ir±
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. Tellus 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. lr± 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~~Materials Advisory Board, National
Academy Press, Washington, 365 pp.
Hicks, B. B. and P. S. Liss. 1976. Transfer of S02 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
micrometeorological 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. In Atmospheric
Sulfur Deposition. D. S. Shriner, C. R. Richmond, and "57 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. J^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.
7-67
-------
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. In press.
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, California, 29 November - 3 December. In press.
Hidy, G. M. 1973. Removal processes of gaseous and particulate
pollutants, pp. 121-176. lr^ 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. In Recent Developments in Applied Aerosol Technology.
D. Shaw, ed. John ¥Tley 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.
Hudson, J. G. and P. Squires. 1978. Continental surface measurements
on CCN flux. J. A1mos. Sci. 35:1289-1295.
Jones, E. P. and S. D. Smith. 1977. A first measurement of sea-air
C02 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 ^S and dimethyl
sulfide on selected soil materials. Atmos. Environ. 11:1221-1224.
Judeikis, H. S. and A. G. Wren. 1978. Laboratory measurements of S02
and NO? 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 Llyer.
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. In
Modification of the Aerial Environment of Plants. B. J. BarfieldTnd J.
F. Gerber, eds. ASAE Monograph No. 2.
7-68
-------
Kanwisher, J. 1963. On the exchange of gases between the atmosphere
and the sea. Deep-Sea Res. 10:195-207.
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
deposition in subalpine balsam fir forests: Hydrological and chemical
inputs. Science 218:1303-1304.
Martell, 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
atmosphere to the earth's surface. J. Geophys. Res. 75:3013-3019.
7-69
-------
Monteith, J. L. 1963. Dew, facts and fallacies, pp. 37-56. In The
Water Relations of Plants. A. J. Rutter and F. H. Whitehead, ecTs. John
Wiley and Sons, New York.
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, G. S., J. V. Hayes, and E. C. Ogden. 1974. Particulate
dispersion into and within a forest. Boundary-Layer Meteorol.
7:429-456.
Raynor, G. S., E. C. Ogden, and J. V. Hayes. 1970. Dispersion and
deposition of ragweed pollen from experimental sources. J. Appl.
Meteorol. 9:885-895.
Raynor, G. S., E. C. Ogden, and J. V. Hayes. 1971. Dispersion and
deposition of Timothy pollen from experimental sources. J. Appl.
Meteorol. 9:347-366.
Raynor, G. S., E. C. Ogden, and J. V. Hayes. 1972. Dispersion and
deposition of corn pollen from experimental sources. Agronomy J. 64:
420-427.
Rosinski, J. and C. T. Nagomoto. 1965. Particle deposition on and
reentrainment from coniferous trees, Part I. Kolloid A. 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. 75:1966-1781.
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.
7-70
-------
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.
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
processes. 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. In 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 inertia! 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,
deposition, resuspension problem. Atmos. Environ. 10:763-768.
7-71
-------
SI inn, W. G. N. 1976b. Dry deposition and resuspension of aerosol
particles - a new look at some old problems; pp. 1-40 of
Atmospheric-Surface Exchange of Participate and Gaseous Pollutants -
1974, R. J. Englemann and G. A. Sehmel, (Coord.); available as ERDA
CONF-740921 from NTIS, Springfield, VA.
SI inn, S. A. and W. G. N. Si inn. 1980. Predictions for particle
deposition on natural waters. Atnos. 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.
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, California,
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 N02 flux to vegetation and
comparison to 03 flux. Atmos. Environ. 16:815-820.
7-72
-------
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.
Winkler, E. M., and E. J. Wilhelm. 1970. Saltburst by hydration
pressures in architectural stone in urban atmosphere. Geol. Soc.
America Bull. 81:576-572.
Yudine, M. I. 1959. Physical considerations on heavy-particle
diffusion, pp. 185-191. lr± Advances in Geophysics, Volume 6. H. E.
Landsberg and J. van Mieghem, eds. Academic Press, New York.
7-73
-------
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 receptors. Wet deposition in the form of rain, fog, and snow
and dry deposition of gases and partiuclate matter have been addressed.
This chapter considers both wet deposition monitoring during
periods of precipitation and dry deposition monitoring during periods of
no precipitation. 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 deposition 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 and 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 are probably sufficient for studies of wet deposition
emphasizing the acid precipitation phenomena.
How good are those current network data? Are the networks
adequately distributed and operated to provide a good evaluation of the
8-1
-------
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 presents deposition models to predict exposure of
receptors to concentrations of specific pollutants. Such models are
needed to predict deposition over required 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 hydrochloric 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.
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 at
the network
4. Time density—The frequency of sample collection (the sampling
interval).
Networks have been of all spatial and temporal sizes and densities,
ranging from 1 site operated for only a few days to more than 50 sites
distributed over several countries and operated for over 30 years.
The time and space configurations of networks are dictated by
scientific objectives and available financial resources. Networks are
8-2
-------
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 eventually used for research purposes, and the
data from single site research networks are frequently used to monitor
the changes in time of wet deposition. 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:
j>H - For typical rain and melted snow solutions the pH ranges from
3.0 to 8.0. The pH indicates the acidity, i.e., the free hydrogen-ion
concentration, and mathematically pH = -logjo[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, rain water 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. However, 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.
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 constituent 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
the sample) divided by the sum of the five volumes. The precipitation-
weigh ted-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
conceptually 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
8-3
-------
the hydrogen ion would be conserved in the mixing process. The pH
calculated from the volume- or precipitation-weighted-mean hydrogen
concentration will be referred to in this chapter as the weighted pH.
Precipitation - The term will 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.
Acidic rain - A popular term with many meanings; generally used to
describe precipitation with a pH of less than 5.6.
Acidic precipitation - Water from the atmosphere in the form of
rain, sleet, snow, and hail, with a pH of less than 5.6. (This is how
scientists in the past have used the term.)
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) all materials deposited by the
atmosphere in the absence of precipitation; and (2) the process of such
deposition.
Total atmospheric deposition - Transfer from the atmosphere to the
ground of gases, particles, and precipitation, i.e., the sum of wet and
dry deposition. Atmospheric deposition includes many different types of
substances, nonacidic as well as acidic.
Acidic deposition - The transfer from the atmosphere to the ground
of acidic substances, via wet or dry deposition.
8.2.3 Methods, Procedures, and Equipment for Wet Deposition Networks
For data comparability, it would be ideal if all wet deposition
networks used 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 user should check for data comparability.
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 collection containers are currently used in most
acidic wet deposition networks. However, they are unacceptable for
monitoring pesticides in precipitation. Glass collection containers are
8-4
-------
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 these washing procedures.
Sampling mode - There are three sampling modes. In bulk sampling
the collection 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 reasons. 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, bulk data pH must be
used with great caution (only in conjunction with comprehensive system
blank data). For wet deposition sites that will be operated for a long
time (more than 1 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.
For both wet-only and wet-dry sampling the automatic device has
been sometimes replaced by an observer making manual container changes,
an undesirable alternative except in very special situations.
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 sampling, 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.
In sequential sampling, a series of containers are exposed to the
atmosphere to collect wet deposition samples, with consecutive advances
to new containers 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 volumes,
and rainfall amount are frequently measured at field laboratories.
8-5
-------
Making these additional measurements requires that site operators have
greater 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 assessment of the fraction of the
precipitation captured by the precipitation chemistry stamp!er.
Sample handling - Chemical changes with time in the sample are
decreased through the addition of preservatives to prevent biological
change, refrigeration, aliquoting, and filtering. Peden and Skowron
(1978) have reported that filtering is more effective than refrigeration
for stabilizing Illinois samples. When the filtering procedure is used,
it is important to obtain frequent filter blank samples, because the
chemistry of relatively clean rain samples can be easily altered.
Analytical methods - Appropriate analytical methods are available
to measure the major ions found in precipitation, but special
precautions are necessary because the concentrations are low; thus, the
samples are easily contaminated. 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 malfunctioning 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.
Quality control 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 reports. Quality
control reports are often considered analytical laboratory reports that
document the methods used to measure chemical parameters and the bias
and precision of the analytical 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 resulting in the data
base. While quality control reports can be easily produced for the
analytical methods, some of the greatest uncertainties in comparing data
from different networks involves estimating the bias resulting from
differences in sampling mode, sample handling, and related aspects.
Quality assurance programs are very 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 1 year, for example, generally cannot
afford to produce an array of written documents to describe all the
quality control procedures and data.
8-6
-------
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 all the information needed to
assess data bias and precision for that network. However, the use of
colocated sites for various networks is one of the most direct ways to
assess network design differences. Several colocated sites sites are
necessary to evaluate network data differences at sites having different
meteorological and pollution environments. The operation of co-located
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
summarized by many authors (e.g., Eriksson 1952, Niemann et al. 1979,
Miller 1981, Wisniewski 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
from the review by Eriksson (1952).
Table 8-2 summarizes some regional- and national-scale wet
deposition networks in Canada and the United States that have begun
operation since 1955. These networks were generally not established to
monitor acidic precipitation. 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 the
major inorganic ions that account for most of sample conductance.
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 85 sites concentrated in a rather small area (Gatz 1980);
8-7
-------
TABLE 8-1. AGRICULTURAL DATA BASES (1875-1955)
(ADAPTED FROM ERIKSSON 1952)
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
8-8
-------
TABLE 8-2. SOME NORTH AMERICAN WET DEPOSITION DATA BASES (1955-PRESENT)
00
APPROXIMATE
NETWORK
National
Junge
PHS/NCARb
WMO/EPA/NOAAC
Canadian
CANSAP0
NADPe
PERIOD
1955-1956
1959-1966
1972-P resent
1977-P resent
1978-Present
NUMBER OF
SITES
60
35
10
59
110
SAMPLING
MODEa
W-M
W
W
W
W-D
SAILING
INTERVAL
Daily, (with monthly
compositing)
Monthly
Monthly (Joined NADP in
1980)
Daily, (with monthly compos
ing) (monthly before 1980
Weekly
Regional
US Geological
Survey Eastern
(USGS)
Canadian Centre
for Inland
Waters (CCIW)
Tennessee Valley
Authority
MAP3sf
1964-Present 18
1969-Present 15
1971-Present 11
1976-Present 9
W
W-D
W
Monthly
Biweekly
Daily
-------
TABLE 8-2. CONTINUED
00
1— >
o
NETWORK
Canadian APN9
EML-DOEh
UAPS1
U.S. EPAj
Great Lakes
NUMBER OF
PERIOD SITES
1978-Present 6
1976-Present 7
1978-Present 19
1977-Present 30
SAMPLING SAMPLING
MODE3 INTERVAL
W Daily
B, W-D Monthly
W Daily
B, W Monthly and Weekly
aB 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.
cWorld Meteorological Organization/U.S. Environmental Protection Agency/National and Oceanic
and Atmospheric Administration. These sites are now part of NADP.
dCanadian Network for Sampling Precipitation.
National Atmospheric Deposition Program.
^Multistate Atmospheric Power Production Pollution Study.
^Canadian Air and Precipitation Network.
Environmental Measurements Laboratory of the U.S. Department of Energy.
Utility Acid Precipitation Study.
JUnited States Environmental Protection Agency.
-------
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 (Li and
Landsberg 1975, Dana et al. 1975).
Some of the local network data have been very useful in interpretating
time trends of chemical concentrations in precipitation.
Wisniewski and Kinsman (1982) have prepared a detailed tabultation
of national, regional, and state or province networks currently in
operation in the United States and Canada, and Mexico. A total of 69
networks are described.
Whelpdale (1979) has prepared a tabulation of seven major wet
desposition 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 analyses 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 Wall en (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. Neverthe-
less, assessments such as by Galloway and Whelpdale (1980) and by
Shannon (1981) suggest that wet and dry deposition processes are of
8-11
-------
TABLE 8-3. CHARACTERISTICS OF THE WORLD METEOROLOGICAL ORGANIZATION
(WHO) AIR POLLUTION NETWORK (WHELPDALE 1979)
Program name: WMO BACKGROUND AIR POLLUTION NETWORK.
Organizatlon/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, 50*2-, ci~, NH^, Ca2+, Mg2+ Na2+, K+,
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 World', Environmental Data Service,
NCC, Asheville (annually).
8-12
-------
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.
Orgam'zation/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, S042"; optional - H+, N03", NH4+, Mg2+,
Na+, CT, Ca2+
aerosol: S042-; Opt1onal _ 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 modelling, 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 Lillestrj6m, Norway.
8-13
-------
TABLE 8-4. CONTINUED
Data/Reports/References: ECE 1977, Cooperative Program for Monitoring
and Evaluation of the Long-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.
8-14
-------
TABLE 8-5. CHARACTERISTICS OF THE EUROPEAN ATMOSPHERIC CHEMISTRY
NETWORK (EACN) (WHELPDALE 1979x
Program name: EUROPEAN ATMOSPHERIC CHEMISTRY NETWORK (EACN)
Organlzation/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
( ~ 1 station/105 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, S042~, 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
sampling systems, primarily to investigate anthropogenic-acidity related
phenomena.
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.
8-15
409-261 0-83-21
-------
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.
8-16
-------
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 considerably 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;
consequently, 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.
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 provide information on the processes that
contribute to the net transfer of pollutants to the surface, and usually
designed to help formulate the deposition 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
application of the deposition parameter!'zations 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 collection vessels have obvious applicability.
Furthermore, they provide samples in a manner suitable for chemical
analysis using fairly conventional techniques. 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 measuranent 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 pollutants, species-specific
surrogate surface techniques have been used to evaluate air
concentrations rather than deposition fluxes. Standards exist concerning
8-17
-------
sulfation plates used to monitor sulfur dioxide concentrations (ASTM D
2010-65), and once again technique intercomparisons have been conducted
(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 relevant 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 concentration 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.
8.3.2.1 Direct Collection Procedures—There is no question that the
deposition of large particles is adequately monitored by collection
devices exposed carefully over the surface of interest. Deposit gauges
and dustbuckets 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 depostion, 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 deposition 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 detected and to cover it and expose a dry collection
8-18
-------
bucket at all other times. The convenience and relative simplicity of
these devices has 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 col lection-vessel data is not only
recent. Hewson (1951) comments that the limitations of deposit gauges
are like those of rain gauges. Deposit guages 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 evidence of bird droppings and locally suspended soil particles
(Hicks 1982). The results of chemical analyses of two-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 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 conceptual 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
8-19
-------
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-
surface 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 measurements 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 formulated and quantified, it will no longer be
necessary to measure dry deposition fluxes directly since 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 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 micrometerologi-
cal 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.
8-20
-------
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 characteristic 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 measure-
ments of the variance of their concentration have been developed and
applied primarily in studies of the transfer of sensible heat, moisture,
and momentum. Techniques of this kind might be especially attractive for
some pollutants, but once again a successful system has not been
demonstrated. These three micrometeorological 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
differences on the order of 1 percent.
Attempts to improve sampling by surrogate-surface methods are
continuing. Recent comparisons between different kinds of surface 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 improved, and considerable emphasis is being given to the role
of microscale 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 collection 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 situations, 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 landscapes 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
8-21
-------
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.1 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), 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
resistances 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 plantation case and about 5 s cm-1 for grassland. It
should be emphasized, however, that considerable disagreement about these
lit is appreciated that these data might be influenced by sampling
difficulties, 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
0.9T 1 0.5
0.8
0.7
1.4
1.2
J I
0.4
0.3
0.7
0.6
AMMONIUM NITRATE
0.2
0.1
0
0.1
1 1 1
i I I
1.0 I I I I J 0.5 I. I I I
1.6 i — 1 0.3
1 1
^V
1 t 1
1
2
1 1 i
1 1
1.2
0.8
1.2
0.8
0.4
0.9
0.7
1 1 1
0.2
0.1
0.2
0.1
0
0.2
0.1
1 1
1 1
1 t t
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 nf 3
8-23
-------
c
u
oo
UJ
CJ
GO
t — I
GO
UJ
a:
Q
O
a:
0
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.
8-24
-------
values remains, with many workers preferring to continue with the
approximation 0.1 cm s"1 for the deposition velocity, regardless of the
nature of the surface or the atmosphere. 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 variation, 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 understanding to permit a better assumption,
notwithstanding the evidence of Figure 7-15 (Chapter A-7). Once again,
it is clear that surfaces of different 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 concentrations 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.
As mentioned above, biological factors play an important role in
determining deposition velocities appropriate for the deposition of trace
gases. Stomatal 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
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). THE PARTICLE SIZE RANGE MEASURED WAS 0.3 TO 1.0 MICROMETER
DIAMETER. FLUXES TO FORESTS ARE GIVEN IN BRACKETS. UNITS ARE KG
HA-1 YR-1 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 OF THE ORDER OF A FACTOR OF
TWO MUST BE EXPECTED.
Sulfur Nitrogen Nitrogen
Location (S04 - S) (N03 - N) (NH4 - N)
Raquette Lake (NY) 0.7 0.01 0.2
(0.5) (0.03) (0.6)
Upton, Long Island (NY) 0.2 0.01 0.3
(0.8) (0.03) (1.0)
Rockport (IN) 0.4 0.02 0.6
(1.3) (0.07) (2.0)
Charlottesville (VA) 0.3 0.01 0.3
(0.9) (0.03) (1.2)
State College (PA) 0.2 0.02 0.3
(0.8) (0.05) (1.0)
8-26
-------
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
concentrations 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
concentrations 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, particularly 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 accurate 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 seens 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.
8.4 WET DEPOSITION NETWORK DATA WITH APPLICATIONS TO SELECTED PROBLEMS
(G. 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
8-28
-------
DRY DEPOSITION VELOCITY OF S02 FOR APRIL (cm
r
0.1 - 0.3
0.4 - 0.5
0.6 -0.7
0.8 -1.0
Figure 8-4. Caculated deposition velocities appropriate for sulfur
8-29
-------
1-HOUR
$02 (ppb)
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
-------
were used. 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 acid precipitation
problem is generally perceived to occur.
Concentration and deposition maps will be presented, with the
contours drawn by hand instead of by computer. All objective analysis
and computer plotting packages will not produce identical maps.
Likewise hand-drawn maps will be somewhat subjective. As data values
will be shown on the contour maps in the 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 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
concentrations will be included to allow for a comparison of these two
measures of central tendency. For sites with a rather 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
from the U.S./Canada memorandum of Intent (MOD report (which is nearing
completion) were used. In the MOI report only 1980 data are 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 in this chapter for specific use in effects
studies, an example being the nitrogen deposition 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 in of liquid
equivalent precipitation was collected {NADP 1978, 1979, 1980). The
data were screened at the NADP Central Analytical Laboratory to remove
8-31
-------
Figure 8-6. Map of National Atmospheric Deposition Program site
locations and number of wet deposition samples for
each site thru approximately December 1980 (using
data from NADP, 1978, 1979, and 1980)
8-32
-------
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 certain 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 mgA"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 northeastern Minnesota site. However from Figure 8-6 one notes
that the 2.2 mg «,-! value is the median of only six values and thus
can not be considered very reliable. The 2.0 mg £-1 contour line
passes through the north-central Wisconsin site having a median value of
1.3 mg £-1 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 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 6-16 in Chapter A-6 shows the
seasonal pattern for sulfate and also indicates the great variability
among event samples for sulfate and nitrate.
Consistent with the known emission pattern for sulfur dioxide, the
maximum 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 may be the major source for this maximum. The arid
site at Bishop, CA, also has an extremely large sulfate value, but only
six samples are available. The sample-volume-weighted-average sulfate
values shown in Figure 8-8 are generally similar to those for the
median values.
Pack (1980) found the MAP3S and EPRI 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 «,-! 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
volume-weighted average concentrations of 3.3, 3.5, 3.6, 3.7, and
8-33
-------
2/0
,-1
so42-)
Figure 8-7. Map of median sulfate concentrations (mg a " as
for NADP wet deposition samples through approximately
December 1980 (using data from NADP 1978, 1979, and
1980).
8-34
-------
1.0
2.0
Figure 8-8. Map of volume-weighted average sulfate concentrations
(mg SL~L as SO^") for NADP wet deposition samples
through approximately December 1980 (using data from
NADP 1978, 1979, and 1980).
8-35
-------
4.0 mg £-1 for the data record indicated in Figure 8-8. These
values are very similar to those Pack reported.
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 8-8. 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). Possible sample evaporation after
collection or enhanced raindrop evaporation must be considered as
partial explanations for the high concentrations of all the ions in the
precipitation of the Southwest. 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 have become
increasingly important.
Figure 8-10 shows the contour pattern for the ammonium ion. The
general pattern is very similar 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. Two sites
in New York have elevated values for both ammonium and nitrate but the
site just east of Lake Ontario had only 17 samples (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 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.
Figure 8-12 shows the chloride concentration pattern. Sites closer
to the major chloride source, the sea, have higher levels.
In addition to the ions displayed in Figures 8-7 through 8-12,
ammonium, magnesium, potassium, and sodium are measured in NADP and most
other networks. The data in Table 8-7 demonstrates the relative
importance of all the ions at three NADP sites. The concentrations in
8-36
-------
Figure 8-9. Map of median nitrate concentrations (mg £-1 as NOg")
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 &~1 as
NH4+) for NADP wet deposition samples through approximately
December 1980 (using data from NADP 1978, 1979, and 1980).
8-38
-------
Figure 8-11.
Map of median calcium concentrations (mg ft,'*-) 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 JT1) for NADP
wet deposition samples through approximately December
1980 (using data from NADP 1978, 1979, and 1980).
8-40
-------
TABLE 8-7. MEDIAN ION CONCENTRATIONS FOR 1979 FOR THREE
NAOP SITES (yeq JT1)
No. Samples
S042-
N03~
Cl"
HC03~ (calculated)
An ions
NH4+
Ca2+
Mg2+
K+
Na+
H+
Cations
Median pH
42
38.9
11.6
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
aThe Georgia Station site in west central Georgia.
The Lamberton site in southwest Minnesota.
cThe Huntington Wildlife site in northeastern New York.
8-41
-------
cation sum. If all ions are measured and if there is 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
for the time and space patterns of pH to be fully understood. Currently
sites in Ohio, Pennsylvania, New York and West Virginia have the feature
shown for the New York site in Table 8-7 where H+, S042~, and
N(h- are the dominant ions. For the New York site, the acidity
(H*) could be 100 percent accounted for if all the SO^2' had been
sulfuric acid while nitrate as nitric acid could have accounted for
about 50 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 1980 from
NADP, MAP3S, and CANSAP (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, precipitation-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 representative. 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 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
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
-------
United States. Thus the greater density of weather sites that measure
precipitation amount results 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) show
the units of millimoles per liter and millimoles per square meter. For
this chapter, sites values were converted to the units shown but the
published contour lines are used. [Note: These maps have been redrawn
and the lines have not been verified].
Figure 8-14 shows data for sulfates. The Canadian sulfate data
were corrected for sea salt but the U.S. data were not. Corrections for
sulfate are generally negligible (< 5 percent) except at locations
within 5 km of open ocean areas (Barrie et al., 1982). The general
pattern for sulfates in the Northeast is similar in Figures 8-8 and
8-14. However, by comparing the location of the 1,9 and 2.9 mg £-1
contours in Figure 8-14 with the 2.0 and 3.0 mg &"1 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. In
western New York the two sampling locations are several miles apart. In
the MAP3S program, very small rainfall samples, which generally have
high ion concentrations, are not analyzed. The actual reasons for the
rather large differences in 1980 sulfate ion concentrations 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 &"1 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 are
unreliable. The uncertainty in the location of the contour lines for
different areas, averaging times, averaging procedures, site densities,
and networks have not been determined. The correlative 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
distances 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.
Figure 8-17 and 8-18 show the weighted pH and hydrogen ion
concentrations. The lowest pH values are found in Ohio, Pennsylvania,
and New York. The 5.0 contour line through the central United States is
peculiar to the weighted-averaging procedure as was discussed in rela-
tion to Figure 8-13. The area in the United States enclosed by 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
8-44
-------
UNITED STATES
• NADP
• MAP3S
Figure 8-14. Weighted average sulfate ion concentrations for 1980,
for wet deposition samples (mg sr1). Adapted from
Barrie et al. (1982).
8-45
-------
LEGEND
CANADA UNITED STATES
• CANSAP • NADP
• APN • MAP3S
A OHE
Figure 8-15. Weighted average nitrate ion concentrations for 1980,
for wet deposition samples (mg £~1). Adapted from
Barrie et al. (1982).
8-46
-------
LEGEND
CANADA UNITED STATES
• CANSAP • NADP
• APN • MAP3S
A ONE
Figure 8-16. Weighted average ammonium ion concentrations for 1980,
for wet deposition samples (mg JT1). Adapted from
Barrie et al. (1982).
8-47
1*09-261 0-83-22
-------
5.5
LEGEND
CANADA UNITED STATES
• CANSAP
• APN
A ONE
1980 pH
Figure 8-17. pH from weighted average hydrogen concentration for
1980 for wet deposition samples (reproduced from
Barrie et al. 1982)
8-48
-------
LEGEND
CANADA UNITED STATES
• CANSAP • NADP
Figure 8-18. Weighted average hydrogen concentrations for 1980, for
wet deposition samples (yeq jr*). Adapted from Barrie
et al. (1982).
8-49
-------
Figure 8-17 are lower than those in Figure 8-13. The data in Table 8-8
allow a comparison between 1979 and 1980 and between median and weighted
pH values. The weighted pH values for these sites 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 area of intense acidity in the United States in
1980.
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
events 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 value divided by the
10 percentile value ranges from about 10 to 15 for calcium and
magnesium; 10 to 40 for potassium, sodium, and ammonium; and 5 to 10 for
nitrate, chloride, and sulfate. Rather interesting is the fact that the
90 percentile pH value minus the 10 percentile pH value is very nearly
the same at the eight sites; about 1.7 +_ 10 percent.
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 samples were
collected within 24 hours after a storm 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 deposition
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 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. (1980) 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 untreated aliquots were subject to pH changes during storage
and shipment, with the acidity decreasing. This evidence, combined with
preliminary results from ion chroma trograph measurements, indicated that
the sample changes were associated with degredation 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 Table
8-10. The importance of organic acids is clearly site dependent. This
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.
8-50
-------
TABLE 8-8. NUMBER OF WEEKLY SAMPLES (N) AND
AVERAGE pH VALUES FOR 1979 AND 1980
1979 1980
MedianWeighted N Median Weighted
pH pH pH pH
Bondville, IL 32 4.34 4.35 38 4.29 4.17
Salem, IL 23 4.33 4.20
Delaware, OH 49 4.34 4.25 45 4.15 4.11
Caldwell, OH 44 4.22 4.15 44 4.15 4.08
Wooster, OH 45 4.29 4.25 44 4.21 4.17
8-51
-------
LEGEND
CANADA UNITED STATES
• CANSAP • NADP
• APN • MAP3S
OME
Figure 8-19. Sulfate ion deposition for 1980 for wet deposition
samples (kg ha"1). Adapted from Barrie et al. (1982).
8-52
-------
LEGEND
CANADA UNITED STATES
• CANSAP • NADP
• APN • MAP3S
OHE
Figure 8-20. Hydrogen ion deposition for 1980 for wet deposition
samples (meq m~2). Adapted from Barrie et al. (1982),
8-53
-------
LEGEND
CANADA UNITED STATES
• CANSAP « NADP
• APN • MAP3S
A OWE
1980 DN03-
Figure 8-21. Nitrate ion deposition for 1980 for wet deposition
samples (kg ha'1). Adapted from Barried et al. (1982)
8-54
-------
LEGEND
CANADA UNITED STATES
• CANSAP • NADP
• APN • MAP3S
A OME
Figure 8-22. Ammonium ion deposition for 1980 for 1980 for wet
deposition samples (kg ha"-'-). Adapted from Barrie et al
(1982).
8-55
-------
LEGEND
CANADA UNITED STATES
• CANSAP • NADP
• APN • MAP3S
A OME
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 (rug a~l),
pH, AND CONDUCTANCE FOR EIGHT NADP SITES9
Sites Percentlles (ea«+) (Mga+) (K+) (Na+) (NH4+) (M°3") (C1") (S042-) pH
00
C71
NE-ME
NE-NY
WV
GA
Central IL
N-MN
NE-CO
NW-OR
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
-.002
.015
.049
.005
.018
.050
.014
.035
.084
.005
.027
.124
.007
.027
.094
.017
.044
.154
.009
.076
.391
.010
.033
.144
.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
-.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
.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
.19
.57
.20
.41
1,68
.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
.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
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
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
31
100
128
91
72
94
42
99
All 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
-------
TABLE 8-10. pH AND CONTRIBUTIONS TO FREE ACIDITY (%) FOR FIVE REMOTE SITES
(ADAPTED FROM GALLOWAY ET AL. 1982)
Collector Type
No. Sample sb
Average pHc
pH Ranged
H2S04
HN03
oo HXe
en
no , ... . .. -.,., .
St. Georges,
Bermuda
W/Da and Bulk
67
4.79
3.8-6.2
< HI
< 35
> 0
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
5 18
< 17
> 65
aW/D refers to an automatic sampler which collects a wet only sample in one container and a
dry fall sample in the second container.
cAverage pH here refers to the pH corresponding to the weighted-average hydrogen ion
concentration.
eThe authors indicate that HX could be HC1, organic acids, or ^04 but they believe it
was organic acid.
dThis range is for pH measurements made at the Virginia laboratory, on the samples treated
with chloroform.
^These 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.
-------
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
(Stensland 1981). 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 £-1. 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 contribution 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 COg, 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 perturbations. They conclude that 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, in maritime
areas where basic constituents such as ammonia gas and CaC03 have low
concentrations. 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 indicator of manmade
acidification, but instead the natural elemental cycles must be studied
in order that manmade influences on these cycles can be recognized and
quantified.
8.4.3 Precipitation Chemistry Variations Over Time
8.4.3.1 Nitrate Variation Since 1950's--Likens (1976) reported
significant 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 fertilizer use has not been rigorously assessed.
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. Significantly, the approximate magnitude of
the increase is consistent with the reported increase in combustion-
related NOX emissions over the same time period. However, it would be
8-59
-------
Figure 8-24. Map of precipitation-weighted average nitrate
concentrations (mgj~^ as NC^Z-) for the 1955-56
Junge data (adapted from Stensland 1979)
8-60
-------
inappropriate to infer a quantitative relationship between NOX
emissions and increases in precipitation nitrate concentrations because
error bars for the emission and precipitation data are not yet
available.
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. However, 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 values
in Figure 8-25, averaged for the 78 sites, are 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 pH Variation Since 1950's--Cogbi11 and Likens (1974) and Likens
and Butler (1981) have published eastern U.S. maps of precipitation pH
for the mid-1950's, 1960's, and 1970's. Likens and Butler have
concluded that 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. As noted, these conclusions were based on trends shown on the pH
maps, but trends in emissions and precipitation concentrations of acidic
species were also used.
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 adjustment 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
'change1 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
8-61
-------
1.
1.0
Figure 8-25.
Map of volume-weighted average nitrate concentrations
(mg JT1 as N0s~) for NADP wet deposition samples through
approximately December 1980 (using data from NADP 1978,
1979, and 1980).
8-62
-------
TABLE 8-11. WEIGHTED AVERAGE CONCENTRATIONS9 (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
7/1/55-
6/30/56 £-1)
Change
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.
J
+=6.3
15.6
20.9
3.6
5.05
83.5
+=54.0 -47.7
-19.4
-3.0
+8.2
S042-
N03"
CT
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
PH
Measured,
Weighted pH
Number of
Samples
4.19
4.15
55
4.17 4.18
4.16
80
4.67
MAP3S data are sample volume-weigh ted averages and Junge data are
precipitation amount weighted averages.
8-63
-------
TABLE 8-12. MEDIAN PRECIPITATION CONCENTRATIONS (yeq £-1) AT
CHAMPAIGN, ILLINOIS (ADAPTED FROM STENSLAND AND SEMONIN 1982)
Cations
,,.
Na+
K+
NH4+
Sum
An ions
S042-
N03"
cr
Sum
Calculated
pH
Measured,
Weighted pH
Number of
Samples
5/21/77-
1/16/78
10.5
+=12.9
2 A'
1.9
0.5
17.7
33.0
78.9
29.8
4.8
113.5
4.09
4.02
63
10/26/53 Change
8/12/54 (yeq A-D
84.59 _71.6
7.1 -5.2
2.2 -1.7
18.6 -0.9
112.4
64.5 +14.4
20.2 + 9.6
7.3 - 2.5
92.0
6.52
-
30
Measured hardness.
8-64
-------
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 as the Larson and
Hettick study. The decrease in calcium plus magnesium 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 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 et al. (1981) 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 clear increase of nitrate in precipitation and of NOv and SOx
emissions suggests but does not prove that the acidity of precipitation
has increased in the last 25 years. However, the historical pH data,
measured or calculated, do not allow quantification of an acidity
increase.
8.4.3.3 Calcium Variation Since the 1950's--Tab1e 8-13 shows 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 their higher calcium levels. However,
urban areas such as Albany, NY, a U.S. Geological Survey (USGS) site,
can also produce relatively high atmospheric dust levels, thus, high
calcium levels. 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. Thus, dust probably leaked
in during nonprecipitation periods, producing the relatively high
calcium concentrations.
8-65
-------
TABLE 8-13. CALCIUM CONCENTRATIONS (mg £-1) FOR VARIOUS NETWORKS,
SITES, AND TIME PERIODS (FROM HANSEN ET AL. 1981)
Si tes
Jungea
1955-56
NCARb
1960-66
WMOC
1974-76
USGSd
1966-78
MAP3S6
1978-79
NADPf
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 0.15
Georgia Station, GA 0.10
Greenville, SC
Raleigh, NC
Roanoke, VA
Sterling, VA
0.31
0.32
0.30
0.67
0.20
aWeighted averages, manual wet-only sampling, July 1955-June 1956,
Weighted averages, wet-only sampler, NCAR/Public Health Service.
^Medians, wet-only.
"Medians, bulk sampling.
eMedians, wet-only sampler, July 1978-June 1979.
^Medians, wet-only sampler.
8-66
-------
If dust leaks into the sample 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. For a given collector, the problem will be most
severe in arid regions. The data in Table 8-13 suggest this problem may
also occur in the eastern United States. The magnitude of this dust
leakage effect should be continuously evaluated at all sampling sites
through collection, analysis, and reporting of appropriate blank
samples. These steps have been taken at very few networks in the past,
and they are only rarely taken now.
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) 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 into 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 relatively
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
precipitation 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 ions in liquid precipitation but did correlate with hydrogen
ions in snow and frozen precipitation.
8-67
-------
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 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 ercent 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. 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 to move the air parcel. Recently, many investigators have
calculated trajectories with the National Oceanic and Atmospheric
Administration (NOAA) Air Resources Laboratory (ARL) model, which uses
surface layer wind observations (Miller et al. 1978, Wilson et al.
1980, Miller et al. 1981). With the ARL model, an average wind through
a surface layer, such as that 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.
Some conclusions from recent trajectory studies are as follows.
Forland (1973) found that, for a site at the southwestern tip of Norway,
the precipitation 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 $03 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.
8-68
-------
Miller et al. (1981) 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 greater than 5.0 for trajectories originating
southwest to southeast of Bermuda.
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 precipitation 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. Because showers and thunderstorms are usually
associated with southwesterly flow, whether the low pH detected by this
study was more strongly related to source direction or to
characteristics of the scavenging processes taking place in these types
of precipitation events must be questioned.
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.
The question of the importance of atmospheric transformation and
scavenging processes in explaining the observed association between
southwest trajectories 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 receptors with the meteorological,
physical and chemical information available today.
Wilson et al. (1981) emphasize the importance of recognizing the
relation between precipitation amount and ion concentration. When they
normalized the MAP3S data for precipitation amount they found that the
sulfate deposition per centimeter of precipitation is greater at the
MAP3S Illinois site than at the Pennsylvania and New York sites. Stated
another way, more sulfate is deposited annually at the Pennsylvania site
than at the Illinois site, mainly because of the greater precipitation
amounts.
8-69
-------
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 500 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,
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 anthropo-
genically 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 significant 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.
8-70
-------
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 phenomenon.
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 measurements 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 characteristically 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 glaciochemical 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 indicators of man's effect on the
environment.
8.5.1.1 Sulfate - Polar Glaciers—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 2. 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 Del mas 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 S042- 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
S042~ on a seasonal 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 S042~ enrichments in precipitation need
not be all due to anthropogenic emissions. Recent work by Hammer et al.
(1980) indicates that Greenland ice concentrations of $04^" are
greatly affected by world-wide volcanism. The active volcano Mt. Erebus
may be a major sulfate source to the Antarctic continent (Radke 1982).
8-71
-------
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 $042- concentration in-the ice and snow
(Boutron and Delmas 1980), and in Antarctica gas derived $642- as
well as N03" and Cl~ becomes very important (Delmas et al. 1982).
In addition to the possible volcanic input of $03 into the
atmosphere, biogenic emission, particularly in lower latitude regions
may also be an important contributor of SO? (Lawson and Winchester
1979, Stallard and Edmond 1982, 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 adsorption onto particles may
also be an important source of $042- 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 S02 during transport (Rahn and McCaffrey 1980).
8.5.1.2 Nitrate - Polar Glaciers—The work of Parker et al. (1977,
1982) shows downhole variations in the N03- concentration of snow
ice. Parker et al. (1977) have suggested that this historic and
variation is due to changes in sunspot, auroral, and/or cosmic ray
activities and not due to 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 anthropogenic N03~. Kyle et al. (1982)
have observed Volcanically introduced 1*103- in Antarctic ice. How-
ever, Aristarain (1980) has observed on James Ross Island, Antarctica,
no variation in NOg-, on at least the seasonal level. Risbo et al.
(1981) and Herron (1982), on the other hand, observed no relationship of
N03~ with solar activity in Greenland. Herron (1982) did note a
seasonal variation of N03~ in Greenland ice; however, the highest
values were associated with the summer season. He also observed an
anthropogenic 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
"background" 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
8-72
-------
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 £-!.
Much of the earlier pH work on glacier snow and ice is unusable due
to possible 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 acidity, pH, and acid anion concentrations suggest
there has been a negligible 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 SCty2- 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 SO^- 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 addition, their mean N03~ value was approximately
15 times lower than that observed in central Alberta snows close to
populated areas. High elevation surface samples from Kashmir, India
demonstrate values as high as 1.3 yM in snow from pristine air masses
(Mayewski et alI. 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
airmass 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 yeq &-1 for 1963 to above 10 yeq £-1 in 1976. It
should be pointed out, however, that this increase from 1963 to 1976 is
only represented 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
8-73
-------
relationship for Himalayan surface snows. These authors have suggested
that the majority of the pH vs elevation trend observed is a function of
increased COg 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 CO?. If these data and the
interpretations are correct, detailed ionic balance studies must be
undertaken 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 CO?
"contamination" should increase with depth in glacial ice. If this is
true, the pH of snow and ice, especially downhole, may have little
relevance to the acid precipitation phenomenon. The measurement of
acidity via titration eliminates this contribution of C0£ to pH from
the ice as well as any contribution from the ambient atmosphere upon
melting. The newly developed acid titration technique 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 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
uncertainties 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 nq kg'1 in Greenland prior to 800 BC to
values greater than 200 ng kg-I 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
8-74
-------
precipitation. Unpublished work by Boutron and Patterson now indicates
little if any (possibly a factor of 2) increase (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 posssible
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 enrichment 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
yrs ago, again suggesting little to no anthropogenic input (Boutron
1979a). However, man-made 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 man-made 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 inventory of current back-
8-75
-------
ground 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 other anthropogenically intro-
duced trace metals beside Pb can be discerned in Greenland snow and ice
is still controversial (Herron et al. 1977, Boutron 1979a,b; Boutron and
Del mas 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 to 1972 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); and 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 in 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 glaciochemical 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 glaciochemical studies have produced
time-series of only a few years.
In conclusion, the alpine glacier data available could be con-
sidered 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, $042-, and N03~ 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 concentra-
tions 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
8-76
-------
and present anthropogenic emissions. It is apparent, however, that
Antarctic data do provide information concerning background concentra-
tions 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. 1979, 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 information is available at this time to eludicate
long-term historic trends in regions where they should be easily
detected (i.e., mid-latitude 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 pertinent to resolution of chemical time-series in
polar regions have been retrieved for periods on the order of 10° 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-1) 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 y*""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 (1978) has demonstrated, through
direct measurement of net short- and net 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 understanding 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.
8-77
-------
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 spectrophometry, and ion chromatography 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~, S042~ as well as Cl~ and
Nfy . If possible Na+, K+, Ca2+, 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 measurements.
In addition to the glaciochemical studies, more information is
needed on possible aerosol-snow fractional on 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 compositions 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 deposition) 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).
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
8-78
-------
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 summarizes 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 understood.
8.6 CONCLUSIONS
The following conclusions may be drawn from the preceding
discussion of deposition monitoring.
0 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).
0 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).
0 Automatic devices designed to exclude dry deposition canproduce 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).
0 Most precipitation chemistry networks have only measured the
soluable fraction of the major ions. This procedure is reasonable
for acidic wet deposition studies because major ions generally can
be used to predict a pH that is close to the measured pH (Section
8.2.4).
0 Understanding reasons for pH changes sometimes observed during
handling and storage requires consideration of other chemical
constituents and measurement of both the soluable and insoluable
fractions (Section 8.2.4).
8-79
409-261 0-83-23
-------
TABLE 8-14. POTENTIAL SOURCES FOR CHEMICAL SPECIES FOUND
IN SAMPLES OF GLACIER ICE
Chemical
Species
*1,4,5
h'ogenic
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-
genic
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)
8-80
-------
Sampling networks should be operated for periods of many years to
determine 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 (Section 8.2.4).
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).
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 (Section 8.4.1).
Maximum sulfate, nitrate, and hydrogen ion concentrations in
precipitation are observed in the northeast quadrant of the United
States. Levels decrease to the west, south, and farther north in
New England. Elevated levels extend into southeastern Canada
(Section 8.4.1).
Highest calcium concentrations occur in the central regions of the
United States (Section 8.4.1).
Highest chloride concentrations occur along the coasts (Section
8.4.1).
Patterns for each of these ions are consistent with the known
source regions (Section 8.4.1).
Nitrate in U.S. precipitation has increased since the 1950's
(Section 8.4.3.1).
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 8.4.5).
Sulfate and hydrogen ion are much higher in warm season
precipitation in the eastern United States than in cold season
precipitation. 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).
8-81
-------
Remote site pH data suggest that the common reference to C02
atmospheric equilibrium value of pH 5.6 is not very useful. Recent
measurements in Hawaii and other locations not strongly influenced
by alkaline dust, indicate that the precipitation is less than pH
5.0. Samples at some remote sites have been found to be unstable,
with pH rising with time, presumably due to organic acid loss.
These relatively acid samples at remote sites meed to be explained
to better understand the acidic samples in areas with strong
anthropogenic influences (Section 8.4.2).
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., mid-latitude alpine regions both close to
and remote from emission sites) (Section 8.5.3).
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
vessels, 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
monitoring techniques for surface fluxes. These include eddy-
accumulation, modified Bowen ratio, and variance (Section 8.3.2.2)
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)
8-82
-------
8.7 REFERENCES
Allen, L. tf., 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, and 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. Environ.
Quality 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. Fell in. 1980. The
Canadian air and precipitation monitoring network APN, pp. 355-365. In
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 bigenic gaseous sulfur compounds.
J. Geophysical Research 85:7410-7416.
8-83
-------
Boutron, C. 1979a. Past and present day troposherlc fallout fluxes of
Pb, Cd, Cu, Zn and Ag in Antartica and Greenland. Geophysical Research
Letters 6:159-162.
Boutron, C. 1979b. Trace element of snows of Greenland along an
east-west transect. Geochimica et Cosmochimica 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 C. Lorius. 1979. Trace metals in Antarctica snows
since 1914. Nature 277:551-554.
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. Lorius. 1977. Trace element content in East
Antarctic snow samples. I.A.H.S. Publ. 118:164-171.
Boutron, C. and S. Martin. 1979. Preconcentration of dilute solutions
at the 10-12 g/g ieve-| by nonboiling evaporation with variable
variance calibration curves. Analytical Chem. 51:140-145.
Boutron, C., M. Echevin, and C. Lorius. 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. In
Proceedings of the 74th Annual Meeting, Air Pollution ControT"
Association, Philadelphia, PA. June 21-26. Paper No. 81-6.1.
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. Geophysical Research 84:1705-1708.
8-84
-------
Butler, D., B. Lyons, J. Hassinger, and P. A. Mayewski. 1980. Shallow
core snow chemistry of Athabaska Glacier, Alberta. Canadian J. Earth
Siciences 17:278-281.
Chamberlain, A. C. 1980. Dry deposition of sulfur dioxide, pp.
185-197. ^Atmospheric Sulfur Deposition. D. S. Shriner, C. R.
Richmond, and S. E. Lindberg, eds. Ann Arbor Science, Ann Arbor, MI.
568 pp.
Charlson, R. J. and H. Rodhe. 1982. Factors controlling the acidity of
nature rainwater. Nature 295:683-685.
Clark, H. L., K. E. Clark, and B. L. Haines. 1980. Acid rain in
Venezuelan Amazon. lr^ Tropical Ecology and Development, p. 683,
International Society of Tropical 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 and Alpine Environments
205-213.
Cogbill, C. V. and G. E. Likens. 1974. Precipitation in the
northeastern United States. Water Resour. Res. 10:1133-1137.
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
S02 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, in press.
Davidson, C. I., J. M. Miller, and M. A. Pleskow. 1981. The influence
of surface structure on predicted particle dry deposition to natural
grass canopies, Water, Air, and Soil Pollut. 18:25-43.
8-85
-------
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. Presented at
Inter. Conf. on the Ecological Impact of Acid Precipitation.
Sandefjord, Norway.
Delmas, R., J. M. Ascencio, and M. Legrand. 1980b. Polar ice evidence
that atmospheric CO? 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, pp. 57.
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.
8-86
-------
Dovland, H., and A. Eliassen. 1976. Dry deposition on a snow surface,
Atmos. Environ. 10:783-785.
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.
Elgmork, K., A. Hagen, and A. Langfland. 1973. Polluted snow in
southern Norway during the winters, 1968-1971. Environ. Pollut.
4:41-52.
Eriksson 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). AGU:L041A-15.
Forland, E. J. 1973. A study of the acidity in the precipitation in
southwestern Norway. Tellus 25:291-298.
Foster, J. F., G. H. Beatty, andJ. E. Howes Jr. 1974a. Final report
on interlaboratory 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.
Frohliger, J. 0. and R. Kane. 1975. Precipitation: it's acidic nature.
Science 189:455-457.
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 G. E. Likens. 1981. Acid precipitation: The
importance of nitric acid. Atmos. Environ. 15:1081-1085.
Galloway, J. N. and D. M. Whelpdale. 1980. An atmospheric sulfur
budget for eastern North America. Atmos. Environ. 14:409-417.
8-87
-------
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., W. C. Keene, and G. E. Likens. 1981. Global
precipitation chemistry network. Progress Report. Dept. of Environ.
Sciences, University of Virginia, Charlottesville. 80 pp.
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.
GCA Corporation. 1980. Acid Rain Information Book - Draft Final
Report. DOE Contract AC02-79EV10273-1, GCA Corp., Bedford, MA.
Gorharn, 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. Tellus 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 40: In press.
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., 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., West!ake Village, CA. February.
8-f
-------
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.
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. Tellus
9:180-183.
Herrara, R. A. 1979. Nutrient distribution and cycling in an Amazonian
coating a forest on spodosols in Southern Venezuela. Ph.D. Thesis,
Dept. of Soil Science, Univ. of Reading, England.
Herron, M. M. 1982. Impurities of F~, C1-, NOs", and S042'
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, andj. H. Cragin. 1977.
Atmospheric trace metals and sulfate in the Greenland ice sheet.
Geochimica et Cosmochimica Acta 41:915-920.
Hewson, E. W. 1951. Atmospheric pollution, pp. 1139-1157. In
Compendium of Meteorology. American Meteorological Society, "Boston,
MA., 1334 pp.
Hicks, 8. B. 1981. An analysis of Wangara micrometerology: 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 Assessnent 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, amPS. E. Lindberg,
eds.). Ann Arbor Science, Ann Arbor, MI. 568 pp.
Hicks, B. B., E. B. Cowling, H. Feely, J. N. Galloway, J. H. Gibson, R.
Graham, A. J. Johannes, V. C. Kennedy, J. M. Miller, J. M. Prospero, and
G. Stensland. 1981. On monitoring dry deposition, draft report to the
National Atmospheric Deposition Program (Atmospheric Turbulence and
Diffusion Laboratory Report # ATDL-81/22).
8-89
-------
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, and Soil Pollut. 18:191-198.
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.
Jaworoski, Z., M. Bysiek, and L. Kownacka. 1981. Flow of metals into
the global atmosphere. Geochimica et Cosmochimica Acta 45:2185-2199.
Jaworowski, Z., J. Bilkiewicz, E. Dobosz, and L. Wodkiewicz. 1975.
Stable and radioactive pollutants in a Scandinavian glacier. Environ.
Poll. 9:305-315.
Jeffries, D. S. and W. R. Snyder. 1981. Atmospheric deposition of
heavy metals in central Ontario. Water, Air and Soil Pollut.
15:127-152.
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.
Johnson, B. B. and J. M. Chamberlain. 1981. Sodium, magnesium,
potassium and calcium concentrations in ice cores from the Law Dome,
Antarctica. Geochimica et Cosmochimica 45:771-776.
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.
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. Geophysical Research 76:6589-6595.
8-90
-------
Koide, M. K., R. Michel, E. D. Goldberg, M. M. Herron, and C. C.
Langway, Jr. 1979. Depositional History of artificial radionuclides in
the Ross Ice Shelf, Antarctica. Earth and Planetary Science Letters
44:205-223.
Koide, M., E. 0. Goldberg, M. M. Herron, and C. C. Langway. 1977.
Transuranic depositional history in South greenland firn layers. Nature
269:137-139.
Kyle, P., J. Palais, and R. Delroas. 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.
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. Geochimica et Cosmochimica
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. Geophysical Research 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.
8-91
-------
Leuning, R., M. H. Unsworth, H. N. Newman, and K. M. King. 1979. Ozone
fluxes to tobacco and soil under field conditions. Atnos. Environ. 13:
1155-1163.
Lewis, W. M., Jr. and M. C. Grant. 1980. Acid precepitation 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. and 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.
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,
in press.
Lindberg, S. E. and R. C. Harriss. 1981. The role of atmospheric
deposition in an eastern U.S. deciduous forest. Water, Air, and Soil
Pollut. 16:13-31.
Lindberg, S. E., R. C. Harriss, 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.
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, and
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. In press.
Mayewski, P. A. and W. B. Lyons. 1982. Merserve Glacier ice core: iron
and reactive silicate concentrations, Antarctic J. Geophys. Res. Lehr.
9:190-192.
8-92
-------
Mayewski, P. A., W. B. Lyons, and N. Ahmad. 1983a. Time-series
analysis of glacio-chemical data from the Sentik Glacier core, Ladakh,
India: mass balance and paleoclimiate implications. In press.
Mayewski, P. A., W. B. Lyons, and N. Ahmad. 1983b. Reconniassance
Glacio-chemical studies in the Indian Himalayas, 38th Eastern Snow Conf.
In press.
Mayewski, P. A., W. B. Lyons, and N. Ahmad. 1983c. Chemical
composition of a high altitude fresh snowfall in the Ladakh Himalayas.
Geophys. Res. Letters. 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.
Geophysical Research Letters 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. and A. M. Yoshinaga. 1981. The pH of Hawaiian
precipitation preliminary report. Geophys. Res. Letters 7:779-782.
Miller, 0. M., A. H. Knap, T. M. Church, and J. N. Galloway. 1981.
Interpreting precipitation chemistry events at Bermuda using ARL back
trajectories. Presented at the Third Scientific Assembly of the
International Association of Meteorology and Atmospheric Physics,
Hamburg, F. R. Germany, August 24-26.
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.
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. Geochimica et
Cosmochimica Acta 33:1247-1294.
NADP. 1978. National Atmopsheric Program Data Reports. Vol. I (1-2).
Available from NADP Coordinator's Office, Natural Resource Ecology
Laboratory, Colorado State University, Fort Collins, CO.
NADP. 1979. National Atmopsheric Program Data Reports. Vol. II (1-4).
Available from NADP Coordinator's Office, Natural Resource Ecology
Laboratory, Colorado State University, Fort Collins, CO.
8-93
-------
NADP. 1980. National Atmopsheric 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. 243 pp.
National Academy of Sciences. 1980. The Atmospheric Sciences:
National Objectives for the 1980's. 130 pp.
Ng, A. and C. C. Patterson. 1981. Natural concentrations of lead in
ancient Arctic and Antarctic ice. Geochimica et Cosmochimica Acta
45:2109-2121.
Niemann, B. L., .1. 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 CO? content in the past deduced from ice-core
analysis. Annals of Glaciology 3:227-232.
Ottar, B. 1978. An essessment of the OECD study on long range
transport of air pollutants (LFAP), Atmos. Env. 12:445-454.
Owens, J. S. 1918. The measurement of atmospheric pollution. Quart.
J. Roy. Meteorol. Soc. 44:149-170.
Pack, D. H. 1980. Precipitaion 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.
Nitrogenin South Polar ice and snow: tool to measure past solar,
auroral and cosmic ray activities. Antarctic J. 133-134.
8-94
-------
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. In Aerosols: Anthropogenic and Natural,
Sources and Transport. AnnaTs N.Y. Acad. Sci. 338:486-503.
Rahn, K. A., E. Joranger, A. Semb, and T. J. Conway. 1980. High winter
concentration of SO? 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 CO? 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, and 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.
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, andj. K. Cochran.
1982. Lead precipitation fluxes at tropical oceanic sites determined
from Pb210 measurements. J. Geophys. Res. 87(C2):1239-1245.
8-95
-------
Sequeria, R. 1981. Acid Rain: some preliminary results from global
data analysis. Geophysical Research Letters 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. In press.
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.
Stallard, R. F. and J. M. Edmond. 1982. Geochemistry of the Amazon I:
Precipitation chemistry and the marine contribution to the dissolved
load at the time of peak discharge. J. Geophys. Res. In press.
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-fl2~-1199. Illinois 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. PlenumTress, New York.
8-96
-------
Stensland, 6. J. 1981. Characteristics and Interpretation of the Ion
concentrations of sequential rain samples from Hilo, Hawaii. Presented
at The Third Scientific Assembly of the International Association of
Meteorological and Atmospheric Physics, Hamburg, F. RE. Germany, August
24-26.
Stensland, G. J. and R. G. Semonln. 1982. Another Interpretaion of the
pH trend in the United States. Bull. Amer. Meteorol. Soc. 63:1277-1284.
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.
Summers, P. W. 1981. The global distribution of acidic precipitation
and the implications for eastern North America. Work Group 2, Report
2-14, United States-Canada, Memorandum of Intent. July. Chapter IV.
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.
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.
Vosters, M., F. Hanappe, and P. Buat-Menard. 1970. Determination of
Cl, Na, Mg, K, and Ca, in firn sample 66-A-2 from 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 J. Glacioloqy
20:149-162.
8-97
-------
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. Geochimica et Cosmochimica
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. ^Ecological Effects of Acid Precipitation, Workshop
Proceedings, EPRI EA-79-6-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, and Soil Pollut. 12:343-353.
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.
8-98
-------
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
fundamental 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 relationships 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-knowl edge 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.-Canadian Working Group 1982) established
under auspices of the Memorandum 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 concen-
tration of a particular chemical species, say C-j, this equation is
9-1
-------
. = Si - Ri + kiV2Ci [9-1]
at
where:
t = velocity vector,
Si = sources of species 1,
Ri = sinks of species i, and
ki = molecular diffusivity of species 1.
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 = Ci + Ci'. [9-2]
where Ci is the average concentration and Ci1 is the instantaneous
deviation from the average.
Equation 9-1 is then averaged using mean values to give:
1 + ^ • v Ci = Si - Ri + kiV2Ci - v • cV [9-3]
at
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 • KiVCi [9-4]
which formally defines Ki, the eddy diffusivity of the 1 species.
Because the eddy diffusivity Ki is much larger than the molecular
diffusivity ki, the latter term can be neglected in Equation 9-3.
Thus the equation
3—L + V -V Ci = Si - Ri + V • KiV Ci [9-5]
at
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 the turbulent velocity field. Most
considerations in atmospheric transport and diffusion modeling are based
9-2
-------
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
° Average temporal and spatial scales
0 Treatment of turbulence
° 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
emissions from point sources or area sources or both, perhaps
distinguishing between elevated and ground emissions. Table 9-1 shows
some of the significant 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 non-linear processes such as those
governing reactive pollutants.
9.1.2.1 Spatial and Temporal Scales—Atmospheric motions span a range
of spatial scales from 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 southeastern 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 precursors. However, it appears that insufficient interaction
has occurred among the modelers and effects researchers on this subject.
9-3
-------
TABLE 9-1. CHARACTERISTICS OF POLLUTION TRANSPORT MODELS BY FRAME OF REFERENCE1
Model class
( frame of
reference)
Eulerlan
Lagranglan
Hybrid
(mixed
Eulerian-
Langranglan
Types
of models
Rollback
Statistical
Gaussian plume
and puff
Box and mul ti-
box
Grid
Gaussian plume
and puff
Trajectory
Box and
mul t1 box
Statistical
trajectory
Trajectory
Particle-
In-cell
Puff-on-cell
Physical
Space
Si te-
spedfic/
local
Regional
Site-
specific/
local
Regional
Local
and
regional
Time
Dally
(Episodic)
Daily or
long-term
(monthly
seasonal
annual)
Daily or
long-term
(monthly
seasonal
annual)
Treatment
of
turbulence
Implicit
Eddy
diffusivities
Complex formu-
lation (higher
moment theory)
Well -mixed
vol ume
Eddy
diffusivities
Implicit
Eddy
diffusivities
Complex formu-
lation (not
applicable to
physical models)
Reaction
mechanism
Nonreactlve
Monli nearly
reactive
Nonreactlve
Heavily
parameterized,
linearly
reactive
Nonreactlve
Nonli nearly
reactive
Removal
mechanism
Implicit
Dry and
wet
Dry and
wet
Dry and
wet
Ability
to quantify
source-receptor
relationship
Possible hut
difficult to
implement
Yes
Yes
^Adapted from Drake et al. (1979) and Hosker (19RO).
-------
9.1.2.2 Treatment of Turbulence—Atmospheric turbulence dilutes and
mixes gaseous and particulate pollutants as they are transported by the
mean wind. 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 diffusion 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 planetary
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 statistics 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 (photochemical 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 S02/sulfate and photochemical models have gas-to-
particle components to account for the production of particles directly
from gases via gaseous reactions or condensation. In LRT models this
treatment most frequently is limited to the conversion of sulfur dioxide
to sulfate. Other acidic precursors (e.g., NC^) 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 coagulation, 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 transfer from the air to exposed surfaces.
9-5
409-261 0-83-24
-------
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 processes 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 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:
o Estimate contributions of given sources to receptors of
interest.
0 Estimate consequences of projected emissions changes.
0 Fill gaps between observations.
0 Assist in field study planning, determining such factors as
which variables to measure and where to site stations.
o 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. However, in hilly terrain the Gaussian model
is inadequate even at short distances. In such cases, a trajectory
model is perhaps more suitable. For intermediate ranges, a Gaussian
plume model is sound if uncertainty about dispersion coefficients at
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
-------
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 applications. Some modelers (e.g., Bhumralkar et al. 1981)
have also developed a 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 interest into a two- or three-dimensional array of grid cells.
Advection, diffusion, 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:
o Eulerian grid models are capable of sophisticated
three-dimensional physical treatments.
0 The approach can handle nonlinear chemistry.
0 Data input is simplified on the Eulerian grid.
The disadvantages of the Eulerian grid approach are:
0 Such models usually require large amounts of computer time,
computer storage, and input data.
o These models, when they incorporate non-linear modules,
are cumbersome to use to determine contributions from
individual sources.
o Artificial (computational) dispersion can be significant.
9-8
-------
TABLE 9-3. SHORT-TERM AND LONG-TERM MODELS
Temporal Range
Model Type
Short term
(hourly, daily)
Long term
(monthly, seasonal,
and annual)
Gaussian puff
Lagrangian trajectory
Particle-in-cell
Puff-on-cell
Grid
Gaussian plume and puff
Lagrangian trajectory
Statistical trajectory
9-9
-------
TABLE 9-4. LONG AND INTERMEDIATE RANGE TRANSPORT MODELS
Model Type
and Method
Investigator
Eulerlan
Finite Differencing
Pseudospectral method
Lagrangian
Statistical trajectory
Receptor orienteda
Source oriented
Hybrid; Mixed
Lagrangi an-Euleri an
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); Nordo
(1976, 1974); 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 (1978, 1975);
Mills and Hirata (1978); Sheih
(1977); McMahon et al. (1976);
Bolin and Persson (1975); Scriven
and Fisher (1975); Rodhe (1974,
1972)
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
-------
9.2.2 Lagrangi'an 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 receptor oriented, in which trajectories are calculated backward
in time from the arrival of an air parcel at a receptor of interest, or
source oriented, 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 pollutant 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:
o The extension to three dimensions is not straightforward.
o 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.
9-11
-------
CONTINUOUS PLUME MODEL
SEGMENTED PLUME MODEL
m
PUFF SUPERPOSITION MODEL
"SQUARE PUFF" MODEL
Figure 9-1. Trajectory modeling approaches. Adapted from Bass (1980).
9-12
-------
9.2.2.2 Statistical Trajectory Models--As shown In Table 9-4, several
Lagrangian models are characterized as 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.
Statistical trajectory models have the following advantages:
° Computational requirements are modest.
° The models are cost efficient for repeated runs using
alternative emissions scenarios.
0 The models do not suffer from computational dispersion.
0 The models may be used to estimate contributions from
individual sources.
° Pollutant mass balances can be estimated.
Disadvantages of statistical trajectory models are:
o Most types are not adaptable to short averaging times (i.e.,
episodes).
0 Dispersion and other processes are usually highly
parameterized.
o 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 distribution 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 nonlinear reactions between the
9-13
-------
compounds of Interest (In the Eulerian framework) and the source-
receptor relationship (in the Lagrangian framework). One of the main
disadvantages of the hybrid approach (especially the particle-in-cell
method) is that to simulate spatial distribution of pollution satis-
factorily, 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 M0x-hydrocarbon reactions and
sulfates from SO^ oxidation reactions. The compounds that appear in
rainwater are mainly 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 com-
pounds, 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 transformation. 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:
0 By approximation with simplified first-order reactions. As
described in Chapter A-4, the conversion of S02 to sulfate
is usually treated this way.
0 With more complex sets of reactions describing transformations
among many compounds. However, only a few developmental
models (e.g., Carmichael and Peters 1979) employ non-linear
mechanisms.
9-14
-------
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
multlreaction 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 Simp!ified Modules—Currently, many models treat transforma-
tions 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, HO) 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
transformations.
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 $63 is
converted to $042- each hour. Many models have yet to consider
dependence on temperature, relative humidity, 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 transforma-
tion rate a function of zenith angle and of source type. This resulted
in a variation of 5 to 10 percent in predicted $03 and S042"
concentrations in comparison with results from the same model using a
fixed transformation rate.
9.3.2.2 Multireaction Modules--Although 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 multireaction sets to the real world is often
difficult because of the wide variety of ambient conditions 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.
9-15
-------
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:
o Oxygen
0 Iron
o Manganese
0 Carbon (soot)
o Ozone
0 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 different 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 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 compromise 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 applications 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
9-16
-------
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 NOjj chemistry, such coarse
spatial and temporal resolution cannot be justified, thereby making
model application impractial.
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 approximately 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 incorporate 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« 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 constants 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 N02 is:
= -a(kn[N02])+b(kp[PAN]). [9-6]
dt
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 conversion 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-
9-17
-------
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
NO •? N02a
*n
N02 •* PAN + HN03 + N03" 0.1 0.02
kd
PAN + PAN + HN03 + N03~ 0.1 0
"P
PAN •* N02 0 0.02
aThe ratio N02/N0 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
-------
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 nonspecific 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 pollution 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 influenced by a number of factors. Some of the factors
known to have significant effects on wet and dry depositions are:
Wet deposition:
0 Atmospheric properties
- Precipitation rate and type
- Cloud type and size
- Storm intensity
- Temperature and humidity.
o Pollutant properties
- Form (and size distribution if particulate)
- Solubility and reactivity
- Concentration vertical profile
- Location relative to clouds.
Dry deposition:
0 Atmospheric properties
- Solar radiation
- Wind speed
- Atmospheric stability
- Surface aerodynamic roughness
- Humidity.
0 Pollutant properties
- Form (and size distribution if particulate)
- Concentration vertical profile
- Solubility and reactivity.
9-20
-------
0 Vegetation properties
- Type, size, leaf area, density
- Stomatal condition
- Growth stage
- Stress condition
- Wetness.
0 Other surface (non-vegetation) properties
The current models account for wet and dry deposition with highly
parameterized 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 coefficients. In general, however,
much additional fundamental knowledge about the deposition processes is
needed to facilitate further progress in developing 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 in terms of rainfall rate and
characteristic scavenging efficiency. These offer at least the
capability to describe wet deposition formally. Precipitation rates can
be highly variable, and spatially 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 = CQ exp (- At) [9-7]
where:
C^ = atmospheric concentration at time t,
CQ = atmospheric concentration at initial time, and
A = scavenging coefficient (in units of time-*).
9-21
-------
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 = % [9-8]
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 = W [9-9]
where:
R = the precipitation intensity.
The values of washout coefficients, at least for S02 and
SO/}2', 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
calculated 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 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 calculated under simplifying assumptions of steady state, negli-
gible 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
9-22
-------
_ 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
process. 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
exponential decay term, exp(- At), 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 3 is an "expected" overall scavenging rate that takes
into account the probability of rainfall, its likely duration and
intensity, and the actual scavenging rate 3 expected for such
precipitation (Rodhe and Grande!! 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, precipitation 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 parameter!'zations of only partially understood
9-23
-------
processes and (like most deposition models) is still unvalidated. Such
research, while potentially useful, is presently unsuitable for
practical application.
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 aerosols/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
differentiate 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.
° The scavenging coefficient is a function of raindrop and
snowflake size distribution and effective scavenging area.
0 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
available 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.
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), NAOP (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
V
-------
Vd = F/C [9-13]
where
F = the flux of material,
C = the ambient concentration at a particular height, and
V(j (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.
However, 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 atmospheric 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 grasslands, the grass dies and and is replaced by a snow
cover. The reason for including atmospheric stability as part of the
categorization scheme is that dry deposition depends on the concen-
tration 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
9-25
-------
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 the vertically integrated concentration of puffs is
depleted by an exponential factor
[9-14]
where:
k = dry deposition flux
vertically integratedconcentration
Most of these models compute the dimension!ess value for kd from
Vd-C
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 redistribu-
tion 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
Vd are often used, probably for simplicity, although some grid models
(Durran et al. 1979) include an algorithm that allows Vd to vary in
time and space, reflecting changes in terrain, ground cover, and
atmospheric conditions.
9.4.3.3 Dry Deposition Modules for N0x--As stated previously, most
models treat the sulfur oxTde-sulfate 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 experimentally 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
9-26
-------
field experiments. 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 is the only means by which
the efficiency 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 sulfates 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 transport 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 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 Selection Criteria
Ideally, the choice of a particular model as an operational tool is
based on the specifications of the particular application at hand; how
9-27
-------
well the model has performed in comparable applications; 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 transport (< 100 km), or combinations of all three. The
discussion here has focussed 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 limitation in almost
all existing 1RT 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 designed 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
understanding of LRT model design and application can be obtained by
examining 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-hourly values. ENAMAP-1, Eastern North American Model of Air
9-28
-------
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 S042~ 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 S02
and S042- concentrations
0 Graphical displays of the distributions of S02
and S042~ wet and dry depositions
o 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 re-
lationships 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 assessment of air pollution
impacts for multiple scenarios of emissions. The transport matrix
9-29
-------
SOUTH QUEBEC
(a) EPA Regions used in this study
33
30
20
10
1
.VIII-NORTH
V
0
I]
Ul
l-
0
VII
.VI-EAS1
1 1
V-NORTh
1
ON
LV-SOUTH
J
SOUTH
TAR 1C
1 1 1
SOUTH Q
IV-NORT
V-SOU7
H
,
II
II
UE
3EC
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 Bhumralkar et al.
(1980).
9-30
-------
Local maximum values shown apply at points marked by plus signs.
Figure 9-3. Calculated SO? and S042- concentrations for August 1977.
Adapted from Bhumralkar et al. (1980).
9-31
-------
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 SOzp (10 mg
for 1977. Adapted from Bhumralkar et al. (1980).
9-32
-------
TABLE 9-7. ANNUAL INTERREGIONAL EXCHANGES OF SULFUR DEPOSITION FOR 1977
AS CALCULATED BY THE ENAMAP - 1 MODEL (BHUMRALKAR ET AL. 1980)
I
CO
CO
TOTAL CONTRIBUTION TO S DEPOSITIONS WITHIN RECEPTOR REGIONS
EMITTER
REGION
1 VIII - NORTH
2 V - 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 - NORTH
2 V - 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
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.
4fi.
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.
0.
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.
(Mlotons)
10
0.
78.
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.
40.
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.
21.
14.
22.
204.
407.
13
0.
6.
21.
0.
0.
0.
6.
1.
1.
5.
3.
5.
50.
-------
concept is based on the assumption that the average concentra-
tion/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 sulfates 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 sulfate 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 nr3.
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. Chemi-
cal 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 diagnostic in that it does not attempt to generate
meteorological information from dynamic principles but instead makes
maximum use of available meteorological observations. Two other cate-
gories 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 and distributed
sources. Input data requirements are shown in column 1 of the figure.
9-34
-------
TABLE 9-8. INTERREGIONAL CONTRIBUTIONS TO SULFATE CONCENTRATIONS
AMONG FEDERAL REGIONS (BALL 1981)
10
i
CO
en
Emitter
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 (S%)
U.461 (92%)
5.914
0.059
1.199
2.212
0.934
4.120
0.058
0.295
0.014
0.019
0.000
1.199 (13%)
7.712 (87%)
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 (66%)
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 (42%)
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
o.oni
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 BML AIRSOX model for average of January and July 1974 meteorology; units are mlcrograms per cubic meter.
-------
COLUMN 1
COLUMN 2
COLUMN 3
CO
CTI
PRIMARY DATA
TIME-VARYING FIELDS
METEORLOGICAL DATA
• SURFACE (HOURLY,
3 HOURLY)
• UPPER AIR [6-,
12-HOURLf)
• SYNTHESIZED FROM
NUMERICAL WEATHER
GEOGRAPHICAL INFORMATION
• TOPOGRAPHY
• SURFACE CHARACTER-
ISTICS (LAND USE)
EMISSIONS
• 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
)IATION
3-d WIND
- HORIZONTAL
- VERTICLE
3-d TURBULENT
DIFFUSION
CHARACTERISTICS
2-d SURFACE
UPTAKE
CHARACTERISTICS
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
Figure 9-5. Interaction among the data sources and components of a pollution transport model.
-------
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 the
figure. Some of the input data required in transport model simula-
tions, such as rainfall rate (used in calculating wet deposition) and
humidity (used in chemical transformations), 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--!t
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 1n 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 un-
certainty is the importance of primary sulfates (e.g., SOX emitted
from the stacks already in the form of sulfate). This has become a con-
troversial 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 an 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 inventories 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 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
9-37
409-261 0-83-25
-------
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
diagnostic 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 interpolated 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 km?) is insufficient 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 estimation 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 calculating 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 precipitation data are required.
9.5.4 Model Performance and Uncertainties
9.5.4.1 General--The evaluation of model performance must consider
accuracies inherent in:
9-38
-------
0 the model itself--!.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;
° 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 compared
with the model calculations.
A major limitation in most assessments of model performance is that the
cause of disagreements between calculations and measurenents cannot be
isolated among the four items mentioned above. Normally, the four items
are considered 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 physical 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:
° Technical evaluation: "Does the model perform as intended and
is it scientifically sound?"
0 Operational evaluation: "Does the model compute the correct
values?"
o 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.
o Diagnostic analysis—identification of conditions associated
with accuracies and inaccuracies in the model's performance.
0 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.
0 Scientific Evaluation—a comprehensive technical evaluation of
the model's conformity with the appropriate physical and
computational principles.
9-39
-------
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 set indicated that, while it is adequate for initial evaluation of
sulfur pollution transport models and perhaps wet sulfur deposition, it
is inadequate for substantially refining the current generation of
models.
The years 1978 and 1980 are most frequently used for LRT model
evaluation. The former corresponds to the second year of SURE, which
collected the most comprehensive air quality data base. 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 greatest 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) sulfate
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
procedures for evaluating models (e.g., Martinez et al. 1980, Ruff 1980,
United States/Canadian Working Group 1981).
Many of the widely used performance measures require data bases
from relatively dense networks of ground stations. Data bases for
evaluating performance 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:
o How closely does a model calculation match the corresponding
observed value?
0 How well do the fluctuations In the predictions follow the
fluctuations of the measured parameter in time and space?
9-40
-------
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:
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 scatterplots with regression coefficients.
These statistics and graphical displays of scatterplots (and sometimes
frequency distribution comparisons) traditionally 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 Represent!'vity of Measurements—The evaluation of model
performance 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 acquisition 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 measure-
ments (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 theuncertainty can be thought of as "reducible" by means
of improvements 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 behavior, 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.
9-41
-------
9.5.4.6 Selected Results—Numerous examples of LRT model evaluation
exercises exist in the open literature. However, most of these are
presented In a qualitative manner or with very minimal statistical
evidence. Research programs underway will greatly enhance existing
information on the subject. The MOI, EPRI, EPA, and National Park
Service are all sponsoring such studies, and results will appear in the
literature within the next year.
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. 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 calculated 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 (t), the bias (cf), and
the root-mean-square error (sd). it was noted that the number of
locations used in the evaluation did vary among models. The MOI group
also noted that the models appeared to perform better for wet deposition
than for the ambient concentration. They found this surprising because
wet deposition is episodic in nature, whereas the model results were
presented as non-episodic or longer term. No consideration was given to
S02 concentrations because they were considered to be always affected
by local sources. A major conclusion of the MOI is that it is not
possible to recommend a "best" model (among the eight compared) because
of the uncertainties in the emissions and precipitation data and in the
measurement data used for evaluation.
9-42
-------
(-Q
c:
CALCULATED S042" AIR CONCENTRATION (yg m"3) CALCULATED S042" AIR CONCENTRATION (yg nf 3)
c
to
I
CTl
3 oo
o o
3 CU
ID
CO
r-H —< fD
*< -s
VO < D.
^J Co —'•
•*v| —I £u
• C Cu
3» 3
Q- O
Cu -h O
•O -h
fD O O
f*y p>- ^y
row
-h i n>
-s -5
o o <
3 i S.
2 O
CU fD 3
"< 3 O
fD r+ 3
-S -S rt-
3" CU 3"
O c-l- —•
-h -••<<
fD O
-S 3 <
to cu
fD —'
r+ -h c
O fD
Cu -S C/l
CU to
i—' c n
VO Cu Cu
°°^0
c
CU —•
3 CU
Q. c+
fD
Q.
oo
cr>
ro
o
ro
oo
co
ro
co
O
DO
co
m
O
CO
oo
ro
ro
i
73
o
O
ro
g K
ro
oo
j
co
co
CM
-------
ro
i
05
3.
LU
CJ
O
o
CM
O
o
CJ
co
I
o
o
o
a:
CM
O
o
LU
<
3
O
_l
O
30 40 50 60 70 80 90 100
OBSERVED S02 AIR CONCENTRATION (ug nf°)
(a) JANUARY
54
48
42
36
30
24
18
12
6
0
0 6 12 18 24 30 36 42 48 54 60
OBSERVED S02 AIR CONCENTRATION (yg m"3)
(b) AUGUST
Figure 9-7. Scatter diagram of observed monthly values vs calculated
monthly values of SOo concentrations for January and
August 1977. Adapted from Mayerhofer et al. (1981).
9-44
-------
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
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
ENAMPA-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.
9-45
-------
Model
TABLE 9-10. MOI PRELIMINARY COMPARISON BETWEEN MODELED
AND MEASURED SULFATE CONCENTRATIONS AND DEPOSITIONS
January
July
Annual
(a) Sulfate Concentrations (yg nr3), 1978
AES
MOEa
MEP
ENAMAP
UMACID
CAPITA
RCDM
ASTRAP
AESb
MOEb
MEpb
ENAMAP
UMACID
CAPITA
RCDM
ASTRAP
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
-0.7
_
2.6
1.3
-0.4
-0.7
3.7
0.8
Sulfate
-0.1
_
0.0
0.0
0.3
0.2
0.3
0.0
2.4
-
1.7
1.8
0.4
1.8
1.6
3.0
8.5
-
11.9
8.1
10.8
11.9
9.0
8.9
Depositions
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
2.3
_
2.1
3.6
2.4
3.2
2.7
3.2
10.0
7.0
8.3
_
_
10.3
7.2
7.2
-0.9
2.1
0.8
-
_
-1.2
1.9
1.9
1.7
2.3
0.9
_
_
1.6
1.6
2.8
period-1), 1978
0.4
—
0.4
0.4
0.3
0.2
0.5
0.3
9.7
10.1
6.5
_
-
6.4
6.6
8.1
1.2
0.8
4.4
_.
-
4.5
4.2
2.7
4.0
2.8
2.5
_
-
3.5
4.1
4.3
Background of 2 yg m~3 added to the calculation.
bBackground of 2 kg ha"1 added to the annual calculations only,
9-46
-------
The major point here is that, in a limited number of studies, monthly
concentration and deposition concentrations are often moderately
correlated (in a statistical sense) to measured values and often agree
within a factor of 2. Hence, LRT model results may provide a useful
estimate of reality. However, the accuracy and uncertainties of these
estimates must be quantified more thoroughyy. Also, the evaluation
studies are limited strictly to comparisons of sulfur dioxide and
sulfate concentrations.
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 supporting policy decisions about acidic
deposition because the validity of the modeled
source-to-receptor relationships has not been established
(Sections 9.4.1 and 9.5.4).
0 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 compounds only, ignoring
other compounds germane to acidic deposition (e.g., nitrogen
oxides) (Section 9.5.4).
0 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 treatments of:
9-47
-------
- 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.
o Previous and existing measurement programs have not provided
sufficient data to evaluate models or model components to the
extent 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.
9-48
-------
9.7 REFERENCES
Altshuller, A. P. 1979. Model predictions of the rates of homogeneous
oxidation of sulfur dioxide to sulfate in the troposphere. Atmos. Env.
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 PollutTon 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, E.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. Tellus
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. Tellus 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 Po luTTon Meteorology. American Meteorological Society, Boston, MA.
Carmichael, G. R. and L. K. Peters. 1979. Numerical simulation of the
regional transport of S02 and sulfate in the eastern United States,
pp. 337-344. In Proceedings of the 4th Symposium on Turbulence,
Diffusion, and~A"ir Pollution, Reno, NV, January 15-18, 1979. American
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.
9-49
-------
Dana, M. T. 1979. Overview of wet deposition and scavenging, pp.
263-274. j£ Atmospheric Sulfur Deposition. D. S. Skinner, 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 established from emission data, trajectories, and measured air
concentrations. Atmos. Environ. 9:425-429.
Fay, 0. 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.
Atnos. 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. 6. 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.
Frieberg, 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.
9-50
-------
Hane, C. E. 1978. Scavenging of urban pollutants by thunderstorm
rainfall: numerical experimentation. J. Appl. Meteorol.
17(5):699-710.
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-5Q. 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. Ir± 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. To appear in 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 transfrontier 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. J^n Proceedings of the AMS/APCA 2nd Joint Conference on
Applications of Air Pollution Meteorology, New Orleans, LA, March 24-27,
1980. American Meteorological Society, Boston, MA.
MacCracken, M. C. 1979. Simulation of regional precipitation chemis-
try. j[n 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.
9-51
-------
Mann, C. 0. 1980. NEDS Information. EPA Report 450/4-80-013.
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, Jji 2nd Joint Conference on Applications of Air
Pollution Meteorology. American 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/sul fate
concentrations—formulation and initial evaluation. In 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. ^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.
9-52
-------
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 Association, Paper 81-46.4.
Niemann, B. I.., 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. j£ Proceedings of the 2nd Joint Conference on Applications
of Air Pollution Meteorology, New Orleans, LA, March 24-27, 1980.
American Meteorological Society, Boston, MA.
NordjJ, J. 1974. Quantitative estimates of long-range transport of
sulphur pollutants in Europe. Ann Meteor. 9:71-77.
Nordtf, J. 1976. Long-range transport of air pollutants in Europe and
acid precipitation in Norway. Water, Air, and Soil Pollut. 6:199-217.
OECD. 1977. The OECD programme on long-range transport or air
pollutants: Measurements and findings. Organization for Economic
Cooperation and Development (OECD), Director of Information OECD, Paris,
France.
Olson, M. P., E. C. Voldner, K. K. Oikawa, and A. W. 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. 20:70-86.
Pedersen, L. B. and L. P. Prahm. 1974. A method for numerical solution
of the advection equation. Tellus 26:594-602.
Peterson, K. R. and T. V. Crawford. 1970. Precipitation scavenging in
a large-cloud diffusion code, pp. 425-431. J^n Precipitation Scavenging,
Proceedings of Symposium, Richland, WA. R. J. Engelmann and W. G. N.
SI inn, 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.
9-53
-------
Prahm, L. P. and 0. Christensen. 1977. Long-range transmission of
pollutants simulated by the 2-D pseudospectral dispersion model. J.
Appl. Meteor. 16:896-910.
Rodhe, H. 1972. A study of the sulfur budget for the atmosphere over
northern Europe. Tellus 24:128-138.
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.
17:1375-1389.
Scriven, R. A., and B. E. A. Fisher. 1975. The long-range transport of
airborne material and its removal by deposition and washout, I and II.
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. Jji 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 wet 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., Wesley, and B. B. Hicks. 1979. Estimated dry deposition
velocities of sulfur over the eastern United States and surrounding
regions. Atmos. Environ. 13:1361-1368.
9-54
-------
Sklarew, R. C., A. J. Fabrik, and J. E. Prager. 1971. A
particle-iit-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.
SI inn, W. G. N. 1978. Parameterizations 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 NOX chemistry in power plant plumes. Battelle-Columbus
Laboratories, Columbus, Ohio.
Szepesi, D. J. 1978. Transmission of sulfur dioxide on local, regional
and continental scales. Atmos. Environ. 12:529-535.
United States/Canadian Working Group. 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. Loisell. 1981. Comparison
between measured and computed concentrations of sulphur compounds in
eastern North America. JGR. Vol. 86. NO. C6.
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
S0£ and sulfates from elevated and ground sources, ^jn 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.
9-55
-------
REVIEW COMMENT GUIDELINES
-------
Review Comment Guidelines
Reviewers are encouraged to submit comments regarding the material
presented in this document. To facilitate the revision process, we
request that certain guidelines be followed. Both general and specific
comments on each of the chapters will be useful to us. However, we
cannot emphasize enough the- importance of making your comments specific,
to the extent of indicating the section and page numbers of the material
in question. This will greatly increase the likelihood that the editors
and authors will be able to adequately address your comments within a
reasonable period of time. We also hope your comments will include your
concurrence with the information as well as any disagreements.
Please carefully review the following guidelines prior to providing
your written comments. Since these comments may be included, as
provided, possibly in a separate volume at a later date, please have
your comments neatly presented; typing is preferred.
Guidelines to Reviewers -
1. Chapters should be reviewed using a separate form for each.
2. General comments, impressions, or opinions of the chapter should be
presented on the comment form, separate from specific comments
referencing sections, paragraphs, or statements.
3. Comments regarding concepts or discussions presented should be
addressed by noting the section at the beginning of the reviewer's
comments. Specific comments on the text within sections should be
identified with page number, paragraph (P) and line (L), if
necessary.
Example -
Section 2.2.3
15 P2 Discussion on ... does not
consider studies by ... in
making conclusions
20 P3L5 "Susceptible to change" would be
more appropriate than "sensitive"
4. When commenting on specific sections/discussion and paragraphs/
statements, some suggested reviewers responses that may be useful
include (with additional appropriate details by reviewer):
- information/statement inaccurate
- information/statement incorrectly interprets available
data and/or literature
- information/statement is based on insufficient data and/or
literature
-------
Commonly, reviewer's comments note "this information is not correct!"
You should be aware that unless specific evidence is provided
demonstrating that the information is, in fact, inaccurate, no action
can be taken.
During your review of each chapter, it will be necessary to keep in
mind the intended purpose and scope of this document as outline,, in the
Preface. This is not a criteria document and is not designed to address
policy option issues. The Acidic Deposition Phenomenon and Its Effects:
Critical Assessment RevTewTapers Ts an authored document. Comments
will be addressed by the authors in cooperation with the editors as
deemed appropriate. These authors are charged with producing a
scientifically sound and accurate document and are not expected to make
recommendations for the use of the material in future decision-making
processes. They will therefore address comments on technical accuracy
only.
Only one copy of the review form follows. This form should be
copied and used for all written comments.
-------
MANUSCRIPT REVIEW FORM
Title/Draft
The Acidic Deposition Phenomenon and Its
Effects: Critical Assessment Review Papers
Public Review, Draft, May 1983
Review Coordinator/Return to:
Ms. Betsy A. Hood
CAD Coordinator
NCSU Acid Precipitation Program
1509 Varsity Drive
Raleigh, NC 27606
Reviewer/Organization/Address
Chapter No./Title
SUMMARY RATING
Please rate the manuscript as follows: Satisfactory Unsatisfactory
Cont^it "nd sr' p» ,
QrganiTatinn and prp^ontation
Soundness of conclusions based on
material presented
Technical accuracy _.--—,
Crwsrago of extant literature
Editorial quality
Other (specify)
RECOMMENDATIONS
D (1) Acceptable as is
D (2) Acceptable after minor revision
D (3) Acceptable after major revision
If you have checked either 3 or 4,
please specifically state reason(s) in
the comments space below.
Reviewer's Signature Date
General Comments (Use extra pages if needed but please be concise)
U.S. GOVERNMENT PRINTING OFFICE : 1983 0 - 409-261
-------
Page
Specific Comments (by Section with P and L)
-------
U.S. Env':
Region V
230 So,.-
Chicago
Agency
-------
United States
Environmental Protection
Agency
Washington OC 20460
Official Business Special
Penalty for Private Use $300 Fourth-Class
Rate
Book
------- |