vvEPA
           United States
           Environmental Protection
           Agency
          Office of Erii/iro'itTK?Mia!
          Piocesses and Effects Fesearcii
          Washington DC 20460
           Research and Development
Proceedings of the
Workshop on
Transport and Fate of
Toxic Chemicals in the
Environment

Norfolk, Virginia
December 17-20,1978

-------
                                          EPA 600/9-81-024
                                          May 1981
             PROCEEDINGS OF THE WORKSHOP ON
TRANSPORT AND FATE OF TOXIC CHEMICALS IN THE ENVIRONMENT
                   Norfolk, Virginia
                 December 17 - 20, 1978
           Rizwanul Haque, Workshop Chairman
 Office of Environmental Processes and Effects Research
           Office of Research and Development
          U.S. Environmental Protection Agency
                    Washington, D.C.
 OFFICE OF ENVIRONMENTAL PROCESSES AND EFFECTS RESEARCH
           OFFICE OF RESEARCH AND DEVELOPMENT
          U.S. ENVIRONMENTAL PROTECTION AGENCY
                WASHINGTON, D.C.  20460

-------
                                    DISCLAIMER
     This report has been reviewed by the Office of Environmental Processes and
Effects Research, U.S. Environmental Protection Agency, and approved for
publication.  Mention of trade names or commercial products does not constitute
endorsement or recommendation for use.
                                         11

-------
                                   FOREWORD
     One significant parameter which determines environmental exposure to toxic
substances is chemical transport and transformation.  Knowledge of chemical
transport and transformation is a key factor in specifying protective measures
and is becoming an important tool in regulating chemicals under various legis-
lation.  The Office of Research and Development of the U.S. Environmental
Protection Agency held a workshop in December, 1978, to delineate research
priorities in this essential area, resulting in this document as a product.
The workshop addressed several broad categories including water, air, soil/
sediment, biota, modeling, and regulatory support.

     This document is a functional summary of chemical transport and transfor-
mation state-of-the-art and research needs.  This document will provide
guidance to the Agency in designing research programs as well as defining the
role of transport and fate parameters in testing and regulating chemicals.  We
hope that it will also be utilized by organizations outside the EPA in develop-
ing new ideas and techniques for defining environmental exposure of toxic
chemicals.
                                       Allan Hirsch
                                       Deputy Assistant Administrator
                                       for Environmental Processes and
                                       Effects Research
                                      111

-------
                                  PREFACE
     Expansion of our technological society has resulted in an increased pro-
duction and use of synthetic chemicals.  Concern for the hazards posed by
these chemicals has resulted in the enactment of legislation by the U.S.
Congress mandated to the U.S. Environmental Protection Agency.  The Toxic
Substances Control Act of 1976 is one of the recent acts of legislation in
this regard.  Because use of synthetic chemicals is critical for the
maintenance of the national economy, the disadvantages versus the benefits
of using potentially hazardous chemicals must be carefully scrutinized.  Since
thousands of chemicals are currently in use and many more are being developed,
the task of a cost/benefit analysis of each of these compounds is monumental.
To expedite the regulation process, tests for the screening of potentially
hazardous chemicals are being developed.  These tests are intended to provide
rapid, dependable detection of potentially hazardous chemicals.

     In order to determine hazard, knowledge of exposure level—the extent to
which susceptible individuals or biosystems will be in contact with the
chemical—is requisite.  Since the extent of hazard is usually dose dependent,
accurate assessments of hazards must also include an expected concentration
range.

     Determination of exposure level in relation to ambient concentrations of
chemicals in the environment requires transport and transformation data.
Similarly, information concerning a particular chemical's mode of circulation
and transformation, and where it will form sinks after release into the
environment, is necessary to determine the safe levels of effluent discharge
for that chemical.  Based on the use of this transport and transformation
information, first the exposure level then the risk posed at given levels of
release of the toxicant can be determined.  Thus, transport and transformation
data are vital tools in chemical hazard assessment, and this important
potential necessitates maximum development of this area.  Many organizations
are currently compiling information regarding transport and transformation of
toxicants in natural systems, and increased communication among those working
in this area can promote efficient development of research programs.

     In pursuit of these goals, the Office of Environmental Processes and
Effects Research (formerly Office of Air, Land, and Water Use) sponsored a
workshop to promote communication and to delineate research needs in the area
of transport and transformation of toxic substances.  The workshop was held in
Norfork, Virginia, December 17-20, 1978.  Initial planning took place at the
EPA with the involvement of the Office of Research and Development, Office of
Pesticides and Toxic Substances, Office of Water and Waste Management and the
Office of Air, Noise and Radiation.  The workshop was divided into six (6)
groups with the workscope of each group defined prior to the beginning of the
workshop.  Each group was directed by a non-EPA scientist serving as chairman

                                     iv

-------
and an EPA scientist as a lead person.  Scientists active in the study of
transport and transformation within industry, academia, and government
agencies were selected and invited to participate in each group.  The six
areas consisted of (1) water, (2) air, (3) soil/sediment, (4) biota, (5)
modeling and exposure, and (6) the regulatory aspect of transport and fate.
The chairman, EPA lead person, and group members cooperated to produce a
document concerning research priorities in the area of transport and
transformation of toxic chemicals.  The chairmen subsequently reconvened  to
discuss revisions of the document.

     The objectives of this workshop were to address the state-of-the-art and
to identify research needs in the area of transport and fate of toxic
chemicals in the environment as related to (1) the testing and screening  of
toxic chemicals and (2) the development of a predictive technique for
estimating exposure to toxic chemicals in the multimedia environments. These
two factors are important considerations in defining the risks and hazards
imposed by toxic chemicals to humans and the environment.  Thus, these factors
should be used in regulation by the U.S. Environmental Protection Agency  of
chemical pollutants under such legislative authorities as Toxics Substances
Control Act, Federal Insecticide, Fungicide and Rodenticide Act, Resource Con-
servation and Recovery Act, Clean Air Act, and Safe Drinking Water Act, and
Clean Water Act.  The workshop did not address the issues of health and
environmental effects of chemicals.

     This document is intended to aid in the understanding of current work in
the area of transport and fate of pollutants in the environment and to help
clarify future research directions.  We hope it will be useful to anyone
interested in the processes of transport and transformation, and especially to
those who use this information to assess hazards posed by introduction of
toxicants into the environment.

                                                Rizwanul Haque
                                                Workshop Chairman
                                      v

-------
                                     ABSTRACT
     Presentations at the U.S. Environmental Protection Agency sponsored workshop
on Transport and Fate of Toxic Chemicals in the Environment, conducted in Norfolk,
Virginia, December 17-20, 1978, are documented.  The six sections, corresponding
to the sections of the workshop, derive from the efforts of ninety-three
scientists from government, academia, and industry, and represent state-of-the-art
understanding in transport and fate research in water, air and soil/sediments
compartments; effects of biota on toxic substances; exposure assessment and
modeling; and regulatory aspects of transport and fate research.

     Within each section current methodologies are reviewed, research needs are
presented, and priorities for future research are discussed.  In the final
section, federal legislation for regulation of toxic chemicals in the environment
is summarized and research efforts which will aid in formulation and implementa-
tion of laws and regulations are suggested.
                                         vi

-------
                               TABLE OF CONTENTS



Number                                                                     Page


Section 1.0  Transport and Fate in the Aquatic Environment

  1.1  INTRODUCTION 	     1
1.2.1
1.2.2
1.2.3
1.2.4
1.2.5
1.2.6
1.2.7
1 .2.8

Octanol Water Partition Coefficient




Accuracv and Phvsical Pronertv Data








                                                                             2
                                                                             2
                                                                             2
                                                                             3
                                                                             3
                                                                             3
                                                                             3
                                                                             3
                                                                             4

  1 .3  TRANSPORT PROPERTIES	     4
     1.3.1  Bulk Flow or Hydraulics	     4
     1.3.2  Transport in Sorbed State (Vertical and Horizontal)  	     4
     1.3.3  Release Rates from Sediments	     5
     1.3.4  Air-Water Exchange	     5
     1.3.5  Biotic Transport	     5

  1.4  TRANSFORMATION PROCESSES 	     5
     1.4.1  Transformations of Organic Chemicals	     6
     1.4.2  Metal Ion Catalysis of Oxidation and Hydrolysis 	     6
     1.4.3  Microbially Mediated Transformation 	     7
     1.4.4  Sediments	     7
     1.4.5  Transformations of Inorganic Chemicals	     7
        1.4.5.1  Oxidation/Reduction	     >3
        1.4.5.2  Complexation 	     B
        1.4.5.3  Precipitation and Dissolution	     9
        1.4.5.4  Bioconcentration of Metals 	     9
        1.4.5.5  Generalizations on Characterizing Inorganics
                   and Their Fate	     9
        1.4.5.6  Organometal Synthesis	     9
        1.4.5.7  Photochemistry 	     9
        1.4.5.8  Adsorption to Particulate Matter 	    10
     1.4.6  Transformation Processes Summary and Recommendations	    10
                                      Vll

-------
                        TABLE OF CONTENTS (Continued)
Number                                                                     Page
  1.5  TRANSPORT AND PERSISTENCE IN MARINE AND GROUNDWATERS 	   11
     1.5.1  Marine Environment	   11
     1.5.2  Groundwater	   13
     1.5.3  Microbial Interactions with Toxic Substances
              in Marine Environments. ............ 	   14
     1.5.4  Recommendations for Further Study of Transport and
              Persistence in Marine Groundwaters	   16

  1.6  SOURCES OF TOXIC CHEMICALS IN AQUATIC ENVIRONMENTS 	   17
     1.6.1  Introduction	   17
        1.6.1.1  Flow and Mas Balance Concepts	   17
        1.6.1.2  Sources and Exposure Assessment	   18
     1.6.2  Source Identification 	   18
     1.6.3  Source Characterization  	   19
        1.6.3.1  Fluxes 	   19
        1.6.3.2  Dissolved Versus Particulate Fractions 	   19
        1.6.3.3  Implications for Chemical Analysis 	   20
        1.6.3.4  Particle-Size Distributions and Toxic Chemicals	   20
        1.6.3.5  The Need for Chemical Species Information	   21
     1.6.4  Recommendations for Improved Source Characterization
              Methods	   22

  1.7  MODELING IN AQUATIC SYSTEMS	   22
     1.7.1  Inputs and Outputs	   23
     1.7.2  Reactions	   24
     1.7.3  Modeling Recommendations	   25

  1.8  SUMMARY OF HIGH PRIORITY RECOMMENDATIONS FOR AQUATIC RESEARCH. .  .   26

REFERENCES	             31

Section 2.0  Transport and Fate in the Atmospheric Environment

  2.1  INTRODUCTION	   37

  2.2  SOURCE CHARACTERIZATION	   39

  2.3  ATMOSPHERIC TRANSFORMATION PROCESSES  	   41
     2.3.1  Atmospheric Lifetimes 	   41
     2.3.2  Species and Processes Important  in the Chemical
              Conversion of Organics	   43

                                     viii

-------
                         TABLE OF CONTENTS (Continued)
Number                                                                     Page


     2.3.3  Predictive Modeling for Reactive Species	   44
     2.3.4  Hierarchy of Evaluation .	   44
     2.3.5  Atmospheric Transformation Processes Recommendations	   45

  2.4  SOURCE RESOLUTION:  CHEMICAL SPECIES BALANCES	   46
     2.4.1  Sources Resolution Recommendations	   48

  2.5  AIR QUALITY CHARACTERIZATION 	   49
     2.5.1  Air Quality Characterization Recommendations	   49

  2.6  MASS BALANCES	   51
     2.6.1  Mass Balances Recommendations  	   56

REFERENCES	   58

Section 3.0  Transport and Fate in Soil and Sediments

  3.1  INTRODUCTION	   63

  3.2  SORPTION-DESORPTION PROCESSES	   64
     3.2.1  Sorption-Desorption Recommendations  	   66

  3.3  LEACHING AND EXCHANGE PROCESSES	   67
     3.3.1  Leaching	   69
     3.4.2  Volatilization	   73

  3.4  PHOTODEGRADATION AND VOLATILIZATION	   74
     3.4.1  Photodegradation.	   74
     3.4.2  Volatilization	   75

  3.5  PHYSICOCHEMICAL AND BIOCHEMICAL TRANSFORMATIONS	   80

  3.6  CHARACTERIZATION OF SOILS AND SEDIMENTS	   86

  3.7  LAND DISPOSAL OF WASTES AND CHEMICALS	   88

  3.8  THE USE OF MICROCOSMS IN TRANSPORT AND TRANSFORMATION
         RESEARCH	   90

  3.9  SUMMARY OF RESEARCH NEEDS	   94

REFERENCES	   100

                                      IX

-------
                         TABLE OF CONTENTS (Continued)
Number                                                                     Page


Section 4.0  Effects Of Biota on Toxic Substances

  4.1  INTRODUCTION	107

  4.2  BIOTRANSFORMATION	107
     4.2.1  Microbial Transformations 	  «...   108
     4.2.2  The Fate of Toxic Materials in Higher Organisms	109
        4.2.2.1  Plants 	   109
        4.2.2.2  Animals	   110
     4.2.3  In Vitro Activation/Detoxification Models for
              Whole Animal Biotransformations	111
        4.2.3.1  Biochemical Investigations 	   111
        4.2.3.2  Other Detoxification Systems . 	   111
        4.2.3.3  The Use of Laboratory Obtained Parameters in
                   Estimating the Distribution of Organic
                   Compounds in Environmental Compartments	112
        4.2.3.4  The Utility of Biochemical Processes in
                   the Macroscopic Models 	   113
        4.2.4  Biotransformation Recommendations	114

  4.3  BIOLOGIC TRANSPORT 	   115
     4.3.1  Existing Case Study	115
     4.3.2  Information System	116
     4.3.3  Anthropogenic Influences	117
     4.3.4  Transport Recommendations 	   117

  4.4  MODELS:  MATHEMATICAL AND PHYSICAL ANALOGS 	   118
     4.4.1  Further Studies in Model Development	118
     4.4.2  Validity and Reliability of Ecosystem Models	119
     4.4.3  Validation of Ecosystem Models	121
     4.4.4  Microcosms	123
     4.4.5  Mathematical and Physical Model Recommendations 	   124

  4.5  MONITORING	    125
     4.5.1  Use of Biological Monitors as Transport
               and Fate Indicators	125
     4.5.2  Monitoring for Unknowns	125
     4.5.3  The Use of Chemical Surrogates in Transport and Fate	   126
     4.5.4  Monitoring Recommendations	   126

-------
                         TABLE OF CONTENTS (Continued)
Number
Page
  4.6  METHODOLOGIES	  127
     4.6.1  Software for Analyzing Monitoring Data	127
     4.6.2  Speciation of Metals	  127
     4.6.3  Analytical Techniques 	  128
     4.6.4  Exposure Determination	128
     4.6.5  Methodology Recommendations 	  129

  4.7  EXISTING DATA REQUIREMENTS	129
     4.7.1  A General Overview of the Types of Information
              Currently Reviewed by EPA in Considering the
              Registration of Pesticides in the Area of Fate
              and Transport	  129
     4.7.2  A General Overviuew of the Types of Information
              to be Reviewed for Toxic Substances	131
     4.7.3  Parameters Required by PEST Model (Park et al. 1978)	  132
        4.7.3.1  Data/Model Recommendation	133

REFERENCES	  135

Section 5.0  Exposure Assessment and Modeling in Transport
               and Fate Research

  5.1  INTRODUCTION	  138

  5.2  OBJECTIVES OF MULTIMEDIA MODELS	  139

  5.3  CHARACTERISTICS OF MULTIMEDIA MODELS 	  140

  5.4  SOURCE TERMS	142

  5.5  MATHEMATICAL MODELS	  145
     5.5.1  Available Models for Radiological Assessments 	  145
     5.5.2  Atmospheric Transport 	  152
     5.5.3  Aquatic Transport 	  152
     5.5.4  Food-Chain Transfer 	  152
     5.5.5  External Dosimetry  	  153
     5.5.6  Internal Dosimetry  	  153

5.6  AVAILABLE MODELS FOR EXPOSURE TO CHEMICAL POLLUTANTS 	  154
                                      XI

-------
                         TABLE OF CONTENTS (Continued)
Number                                                                     Page


  5.7  MICROCOSM MODELS 	  156
     5.7.1  The Microcosm as a Screening Tool	156
     5.7.2  Relating Environmental Behavior to Microcosm Data 	  161

  5.8  A COMBINED MATHEMATICAL/MICROCOSM APPROACH 	  162

  5.9  EXPOSURE ASSESSMENT	162
     5.9.1  Definition of Exposure	162
     5.9.2  Types of Exposure Models	163

  5.10.0  MULTIMEDIA EXPOSURES	164

  5.11.0  APPLICATION OF MODELS FROM ENVIRONMENTAL EXPOSURE 	  166

  5.12.0  RESEARCH NEEDS	168

  5.13.0  SUMMARY AND RECOMMENDATIONS 	  169
     5.13.1  Utility of Models	  169
     5.13.2  Present Capabilities 	  170
     5.13.3  Recommendations for Use of Models	170
     5.13.4  Research Needs 	  171

ACKNOWLEDGEMENTS	173

REFERENCES	  174

Section 6.0  Regulatory Aspects of Transport and Fate Research

  6.1  INTRODUCTION	187

  6.2  LEGAL REQUIREMENTS FOR CHEMICAL TRANSPORT AND FATE DATA
         AND THEIR USE IN REGULATION	188
     6.2.1  Toxic Substances Control Act (TSCA) PL 94-469 	  188
     6.2.2  Federal Insecticide, Fungicide and Rodenticide
              Act (FIFRA)	  190
     6.2.3  Regulatory Aspects of Transport and Fate of Chemicals
              in the Environment as Related to Resource Conservation
              and Recovery Act (RCRA), 1976	  191
     6.2.4  Safe Drinking Water Act of 1974 (PL93-523)	193
                                      Xll

-------
                         TABLE OF CONTENTS (Continued)
Number                                                                     Page


     6.2.5  Clean Water Act of 1977 (PL95-217)	193
        6.2.5.1  Section 307 - Toxic and Pretreatment Effluent
                   Standards	   193
        6.2.5.2  Section 304 - Information and Guidelines 	   194
        6.2.5.3  Section 403 - Ocean Discharge Criteria 	   194

  6.3  NECESSARY DATA FOR ESTIMATING EXPECTED ENVIRONMENTAL CONCENTRATIONS
         AND/OR EXPOSURES 	   195
     6.3.1  Introduction	   195
     6.3.2  Quantity of Chemical	   195
     6.3.3  Physical Dilutions of Chemicals Applied and Transported
              to Various Segments of the Environment	196
     6.3.4  Degradation	   198
     6.3.5  Estimation of Chemical	   198
     6.3.6  Useful Measurements 	   198
     6.3.7  The Role of Model Ecosphere in EPA Regulations. .......   200

  6.4  THE EXTENT TO WHICH THE SAME CHEMICAL FATE PARAMETERS CAN BE USED
         BY DIFFERENT EPA OFFICES	201

  6.5  QUALITY ASSURANCE CONCEPTS 	   202
     6.5.1  Introduction	202
     6.5.2  Quality Assurance Components - For Analyses 	   203
        6.5.2.1  Personnel Requirements 	   203
        6.5.2.2  Procedures 	   203
        6.5.2.3  Reagents and Standards 	   205
        6.5.2.4  Laboratory Apparatus and Instruments 	   206
        6.5.2.5  Sample Collection and Preservation 	   206
        6.5.2.6  Analytical Quality Control 	   207
        6.5.2.7  Definitions (Mandel 1977)	208

  6.6  CONCEPT OF BENCHMARK CHEMICALS 	   209
     6.6.1  Utilization of the Benchmark Approach 	   211
     6.6.2  Limitations of Benchmark Chemical Concept 	   214

  6.7  DISCUSSION	215

  6.8  RECOMMENDATIONS	216

REFERENCES	220

REGULATORY ACTS	221

                                     xiii

-------
                                    FIGURES



Number                                                                     Page

 2-1    Flow of Toxic Substances Through the Atmosphere 	    38

 2-2    The Flow of Automobile Emitted Lead Through the
          Los Angeles Basin ..... 	    54

 5-1    Schematic of the Structure of a Deterministic Model 	   141

 5-2    Frequency of Occurrence of Various Features Among
          Eighty-Three Computer Codes for Radiological Exposure 	   151

 5-3    Pathways for Toxic Substances (TXSB) in the Physicochemical
          Environment System	   155

 5-4    Hierarchy of Model Development.  Comparison of Estimated
          Environmental Concentration (EEC) with Biological Test
          Data (Effects Concentration, CR.) 	   167
                                      xiv

-------
                                    TABLES
Number                                                                     Page

 1-1    Modeling Research Priorities	       26

 1-2    Aquatic Environment Group Participants	    29-30

 2-1    Typical Ranges of Atmospheric Residence Times and the
          Boundaries of Mass Balance System 	       42

 2-2    Types of Information Necessary for Constructing a
          Mass Balance	       52

 2-3    Mass Balances for Chemicals in the Atmosphere	       53

 2-4    Atmospheric Environment Group Participants	       57

 3-1    Soil and Sediment Group Participants	       98

 4-1    Biota Group Participants	      134

 5-1    A Listing of Computer Codes and Their Characteristics 	  146-150

 5-2    Multimedia Models:  References and Characteristics	157-158

 5-3    Chief Characteristics of Terrestrial Microcosms Used to
          Study Fate and Effects of Chemicals in the Environment as
          Summarized in Gillett and Witt (1979) 	  159-160

 5-4    Estimated Intake of Chloroform and Carbon Tetrachloride
          from Environmental Sources	      165

 5-5    Exposure/Modeling Group Participants	      172

 6-1    Types of Data of Value in Assessing Transport and Fate
          of Chemicals	      199

 6-2    Suggested Benchmark Chemicals 	      212

 6-3    Suggested Classification Matrix for Chemicals 	      213

 6-4    Regulatory Group Participants 	 ..      219
                                      xv

-------
                                 ACKNOWLEDGMENTS
     The Office of Environmental Processes and Effects Research would like to
thank the participants for their cooperation and contributions to the success of
the workshop.  Also, we thank Linda Beahm, Laura Bennett, William Vaughn, and Lisa
Yost for their assistance in the preparation of this volume.
                                       xvi

-------
                             LIST OF PARTICIPANTS
NAME
                                                  AFFILIATION
Martin Alexander
Herbert Allen

John Bachman
Kris Bandal
George Baughman
Joseph Behar
Robert T. Belly
Barry Bochner
Joseph Breen
Robert Brink
Lawrence Burns
James N. Butler

Michael Callahan
Russell F. Christraan
Jess Cohen
Stuart Cohen
Carroll Collier
William Cooper
Allan Crocket
Cornell University
Illinois Institute of Technology

USEPA
3M Company
USEPA
USEPA
Eastman Kodak Company
Atlantic Chemical Corporation
USEPA
USEPA
USEPA
Harvard University

USEPA
University of North Carolina
USEPA
USEPA
USEPA
Michigan State University
USEPA
Marcia Dodge
James Dragun

Allan Eschenroeder

Walter J. Farmer
Farley Fischer
Virgil Freed
Sheldon K. Friedlander

Walter Galloway
Michael Gilbertson
James Gillett
D. Golomb
Glen F. Gordon

Corwin Hansch
Rizwanul Hague
Charles Helling
John Herron
USEPA
USEPA

A. D. Little Company         ,

University of California
National Science Foundation
Oregon State University
California Institute of Technology

USEPA
Environment Canada
USEPA
USEPA
University of Maryland

Pomona College
USEPA
USDA
National Bureau of Standards
                                     xvii

-------
                       LIST OF PARTICIPANTS (Continued)
Robert Hitch
Robert Hodson
Michael Hoffman
Robert Huggett
J. Huntzicker
Casimer Jackimonski
Jay Jacobson
Robert Jamieson
Richard Johnson
USEPA
University of Georgia
University of Minnesota
Virginia Institute of Marine Science
Oregon Graduate Center
NASA
Cornell University
Proctor and Gamble Company
USEPA
Samuel Karickhoff
Robert Kellam
Gene Kenega
Shahamat Khan
Stanley Kopcznski
Martin Kovacs
USEPA
USEPA
Dow Chemical USA
Agriculture Canada
USEPA
USEPA
Raymond Lassiter
G. Fred Lee
Asa Liefer
Haines Lockhart

Donald Mackay
David Mage
Ben Mason
John Matheson
Paul Michael
Theodore Mill
John Milliken
James J. Morgan

W. Brock Neely
Ronald Ney
Ian Nisbet
Melvin Nolan

Charles O'Melia

Richard Park
Joseph Perkowski
James N. Pitts/ Jr.
Hap Pritchard

Michael Roulier
USEPA
Colorado State University
USEPA
Eastman Kodak Company

University of Toronto
USEPA
Geomet Corporation
USFDA
Monsanto Chemical Corporation
SRI International
USEPA
California Institute of Technology

USEPA
Dow Chemical USA
Massachusetts Audubon Society
USEPA

University of North Carolina

Rennselaer Polytechnic Institute
Petro-Canada
University of California
USEPA

USEPA
 Shahbeg  Sandhu
 Adel  Sarofim
 William  Sayers
 Edward Schuck
 Mark  Seagal
 Hanwat Singh
 William  Spencer
 USEPA
 Massachusetts  Institute of Technology
 USEPA
 USEPA
 USEPA
 SRI  International
 USDA
                                       XVlll

-------
                         LIST OF PARTICIPANTS (Continued)
James Stemmle
Arthur M. Stern

Al Tiedman
Mason Tomson
Charles Trichilo
Henry Tsuchiya
Bruce Turner

Phillip Walsh
C. H. Ward
Walter J. Weber, Jr.
Claire Welty
Richard Wiegert

W. Zielinski
Gunter Zweig
USEPA
USEPA

State of Virginia
Rice University
USEPA
University of Minnesota
USEPA

Oak Ridge National Laboratory
Rice University
University of Michigan
USEPA
University of Georgia

National Bureau of Standards
USEPA
                                       xix

-------
              1.0  TRANSPORT AND FATE IN THE AQUATIC ENVIRONMENT
1.1  INTRODUCTION

     Recently, Congress has enacted legislation which requires the United
States Environmental Protection Agency (EPA) to regulate many toxic chemicals
in the environment.  This regulatory pressure has begun to highlight
fundamental inadequacies in the understanding of transport and transformation
processes affecting the fate of chemicals in the water environment and the
ability to properly assess exposure.

     In an effort to clarify present research goals, the Office of Research
and Development of the EPA invited prominent non-Agency scientists to discuss
research frontiers in the area of aqueous transport and fate of toxic
chemicals with EPA scientific/managerial officials.  The participants were
asked to identify the principal research obstacles to the prediction of
toxicant fate and exposure and to attach relative priorities to these research
needs.

     Recommendations for high, medium or low priority research needs refer to
a relative allocation of resources based on the adequacy of existing knowledge
and environmental importance.  Thus, some important but well understood
processes have been assigned low priority.  In such cases the group does not
wish to imply less priority than at present.

     The report reflects the strong consensus developed among the participants
that development of our abilities to predict fate and exposure of toxic
chemicals in the aquatic environment depends heavily on:  (1) future research
developments in the accumulation of reliable physicochemical property data;
(2) increased understanding of dominant transport routes to and from the water
column (especially for transport in the sorbed state and through the action of
aquatic biota);  and (3) continued development of our understanding of the
kinetics of transformation processes in the aquatic environment.  Important and
unique needs were also identified for the special aquatic environment
represented by ground and saline water.  In addition, the participants
concentrated on the characteristics of sources of toxic chemicals that most
heavily influence fate and persistence, such as methodology for accurate total
flux data, dissolved-particulate fractionation, particle size distribution, and
chemical speciation.

-------
1.2  PHYSICOCHEMICAL PROPERTIES

     While there have been advances in the state-of-the-art technology,  a need
exists for standardization of testing to reconcile wide differences in the
existing literature regarding physicochemical properties.  In all discussions,
it is assumed that the correct chemical structure of any organic material would
be established and available, though this was not initially the case with
Mirex.

1.2.1  Aqueous Solubility

     Aqueous solubility is regarded as an essential property in any assess-
ment.  For solid nonionic materials the National Bureau of Standards (NBS)
May-Wasik-Freeman "Generator Column" method (May et al. 1978a, 1978b) is
preferred, though some doubt exists about its applicability to ionic solids,
some halogenated hydrocarbons, and liquids.  The Haque-Schmedding method is
also excellent (Haque and Schmedding 1975).  Further development work by NBS
and others is strongly recommended to define procedures and scope, and to
standardize methods.

     Solubility is influenced by temperature, electrolytes, surface active
agents, and possibly by natural organic substances such as fulvic acids.  The
magnitude of these effects should be elucidated, with fulvic acids given the
highest priority.

     For sparingly soluble compounds, accuracy within a factor of two is
adequate but many published data are highly inaccurate.  A standard compila-
tion of critically reviewed data would be invaluable.  Work on structure -
solubility relationships should be pursued as a potential method of prediction
and as a means of exposing bad data.

1.2.2   Octanol/Water Partition Coefficient (Kow)

     This quantity is indicative of the tendency of a solute to partition into
lipid phases and onto organic carbon in sediments (Leo et al. 1971; Neely et
al. 1974; Kenega and Goring 1978; Karickhoff et al. 1979; and others).  Method
standardization is desirable because extremely inaccurate values may be
obtained by poor techniques.  The continued development work on the Hansch-Leo
approach to calculating Kow should be encouraged.

     Measurement techniques using HPLC are promising but not yet fully vali-
dated.  Further work is recommended, especially for screening water effluents,
to identify the presence of high Kow compounds.  The applicability of KQW to
ionic species such as carboxylic acids, phenols, and amines requires better
definition.  Unresolved questions relating to an apparent "upper limit" to Kow
should be clarified.  Methods development and further data acquistion are
straightforward and can be accomplished with relatively low expenditures.

     More expensive, and thus meriting higher priority, are continued efforts
to relate Kow to environmental behavior, notably bio-uptake, sorption and
toxicity.  Such work should not only attempt to elucidate these relationships
but should clearly define any limitations or areas of nonapplicability.  Since
the correlations are based on data for moderately hydrophobic compounds, their

-------
reliability when applied to low or high Kow compounds should be examined
carefully as a matter of high priority.  Due to their economy and usefulness,
a risk exists that these correlations may be over-applied.

1.2.3  Vapor Pressure

     Vapor pressure is an essential quantity for prediction of rate of loss to
the atmosphere.  Published data are often in disagreement by a factor of 10 or
more (Spencer and Farmer 1978); an extreme example is lindane, for which the
range was 3000.  Static methods are satisfactory for reasonably volatile
compounds, but gas saturation flow systems (Spencer and Cliath 1969) are
preferred for low volatility compounds.  A standard technique should be
defined.  Very little work correlating vapor pressure with chemical structure
has been attempted.  Such correlations would be useful and should be carried
out.  This topic merits medium priority.

1.2.4.  Henry's Constant (H)

     This quantity is the ratio of vapor pressure to solubility.  For some
sparingly soluble, low volatility, compounds direct determination by a
stripping technique is preferable (MacKay et al. 1979).  A simple test to
determine the possible significance of volatilization would be very useful and
should be devised.  Determinations of H for actual waters containing other
contaminants is advantageous since the effective H (based on total liquid
concentration) may differ substantially from the distilled water value.  Medium
priority was assigned.

1.2.5  Ionic Properties

     For substances (primarily inorganic) forming dissolved ionic species, work
should be devoted to acid-base and solubility equilibrium measurement, complex
formation, redox equilibria, and sorption.  Complex formation and sorption
should be given the highest research priority of the four ionic properties.

1.2.6  Complex Formation

     Data on the potential of organic substances to complex, chelate (or to be
chelated), or speciate in some unusual manner are desirable to elucidate any
mechanism of solubilization or transport.  Low priority was assigned.

1.2.7  Sorption

     A knowledge of the potential for sorption of organics on sediments, biota,
and suspended solids is necessary to study transport and fate of toxic chemi-
cals in the aquatic environment.  In particular, information on the fractions
of the substance in sorbed and dissolved states is critically important for
predicting or correlating behavior in the water column.

     Work relating the sorption partition coefficient to sorbent properties
such as organic carbon content is useful as a mechanism of generalization.
Further correlation with Kow appears possible (Karickhoff et al. 1979), but the
scope and limitations require definition.  A body of information is available

-------
for sorption on activated carbon which may be useful in discovering the rela-
tive behavior of various sorbates.  The entire area of sorption equilibrium is
regarded as high priority.

1.2.8  Accuracy and Physical Property Data

     Considerable discrepancies exist in the literature for physical property
data, and inaccurate values are still being reported.  The effort required to
obtain accurate and precise data is usually underestimated.  The NBS and EPA
should play a more active role in this complex and demanding area by supporting
activities such as methods development, standardization, interlaboratory com-
parisons, use of standard reference materials, and the compilation and revision
of available data.  The acquisition of physicochemical data for toxic sub-
stances is demanding and expensive.  Not all data should necessarily be of
the same degree of accuracy.  For example, a substance with properties similar
to ethylene glycol will not volatilize from water significantly because of its
high solubility and low vapor pressure (i.e., very low H), so an accurate vapor
pressure is not necessary.  Similarly, accurate sorption data may be
unnecessary for freons.  It is not immediately obvious what accurate data are
required until some form of "sensitivity analysis" can be conducted.  Such
analysis can only be done if estimates for all data are available.

     The implication is that it may be more economical to proceed through
several levels of sophistication initially using estimated values (from cor-
relations, for example) with experimental determinations made to an accuracy
dictated by the sensitivity of the exposure to the physicochemical property.
A great contribution in this area can be made by "evaluative models" (Baughman
and Lassiter 1978).  Since physicochemical data are inherently reproducible in
contrast to other environmental data, there is no excuse for the present,
less-than-satisfactory situation.
1.3   TRANSPORT PROPERTIES

1.3.1  Bulk Flow or Hydraulics

     A detailed knowledge of the prevailing hydraulic regime is critical for
any site-specific work.  These sites include river flow, lake stratification,
estuarine mixing, tidal flushing, storm behavior, oceanic currents, and ocean
stratification.  The time scales and dimensions of these mixing or diffusion
processes, horizontally and vertically, will often require quantification.
Such work should be on an ad hoc basis because little optimism exists that any
significant quantitative generalizations will be possible.  The qualitative
site-specific generalizations are already established.

1.3.2  Transport in Sorbed State (Vertical and Horizontal)

     The water column dynamics of solutes which are appreciably sorbed are
believed to be strongly influenced or even dominated by processes such as
sinking in association with such solids as mineral or organic matter and fecal
pellets.  Examples include hydrocarbon transport in copepod fecal pellets from
surface waters to ocean sediments (a process which is orders of magnitude more
rapid than simple particle settling) and PCB transport to lake bottoms

-------
following wet or dry deposition from the atmosphere.  In estuarine waters the
salinity change may induce flocculation or desorption, and density differences
may cause stratification.

     This entire subject area, including organic and inorganic solutes, is one
of high priority and requires better quantification, especially regarding
acquisition of data on pollutant fluxes in various water columns.

1.3.3  Release Rates from Sediments

     Since sediments are often the repository of significant quantities of
toxic solutes, a knowledge of their release rates and mechanisms is necessary
for effective regulation.  This information is particularly needed if sediment
alteration processes are naturally possible as a result of severe storms, or
are man-induced by dredging or stream-flow control.  The need for parallel
bio-and chemical-transformation work is also apparent.  Medium priority was
assigned.

1.3.4  Air-Water Exchange

     Diffusive volatilization or absorption processes appear to be capable of
reliable mathematical description, but validation is required in actual
environmental water bodies.  The role of sorption in volatilization rates
should be examined further as should the effects of surface organic
microlayers and the rates of spray transfer.  Reliable data on transfer
coefficients for lakes, oceans, and rivers should be gathered and correlated.
Relationships to oxygen reaeration constants should be exploited.  This area
deserves medium priority for organic solutes and low priority for spray
transfer processes.

1.3.5  Biotic Transport

     The dynamics and equilibrium of the processes of uptake, degradation,
metabolism, and accumulation of toxic substances by various levels of biota is
an area of growing scientific concern.  Cometabolism by mixed, continuous
microbial cultures should be studied and the distinction between food web
magnification (predator-prey) and equilibrium lipid partitioning deserves more
attention.  The potential for bioaccumulation should be the topic of more
study, especially for solutes such as hydrocarbons where existing data may be
inconclusive.  High priority should be assigned to work with organics and
medium priority to inorganics.
1.4  TRANSFORMATION PROCESSES

     Procedures for prediction of environmental fate in aquatic systems which
are based on measurements of specific transport and transformation processes
have a reasonable conceptual foundation.  Moreover, no good alternatives to
this approach appear to be presently available.  This conclusion offers a
point of departure for examining some of the detailed assumptions made con-
cerning important transformation processes for organic and inorganic chemi-
cals.  Current assessment procedures are concerned primarily with the fate of
organic chemicals, and interest has focused on the specific reactions that

-------
will account for their transformations in freshwater (Smith et al.  1978;  Mabey
and Mill 1978).  This section attempts to summarize the processes that control
transformations of both organics and inorganics and to develop some
generalizations for inorganic chemical processes to assist in developing fate
assessment methods similar to those used for organic chemicals.

1.4.1  Transformations of Organic Chemicals

     The following processes are believed to be major routes for transforma-
tions of organics in water:

     •  Hydrolysis by water, H3O+ and OH"

     •  Photolysis by direct absorption of light in the solar region (>290nm)

     •  Oxidation by free radicals and singlet oxygen

     •  Microbially mediated transformations

     •  Reduction in anoxic systems.

     In one kinetic treatment each process is expressed as a rate (equation 1)
which is a function of the concentration of the chemical, an environmental
parameter such as pH, biomass or oxidant level, and a rate constant charac-
teristic of the chemical, the process, and the temperature.

                           -d(C)/dt = -Rn = kn (C)                   (1)

For the case described by equation 1, C is the concentration of chemical and
kn is the rate constant.  kn usually depends on temperature and composition of
the medium, while the total rate of loss of C is simply the sum of individual
rates:

                                   %. = lRn                          (2)


     An additional simplifying assumption, that kn is essentially constant
during the measurement interval, converts equation 1 into a pseudo first-
order expression:

                              -d(C)/dt = k '(C)                      (3)
     There are questions concerning the validity of the foregoing assumptions
such as the importance of including in the rate equations environmental
parameters which can vary widely in aquatic systems and may not be considered
as constants.

1.4.2  Metal Ion Catalysis of Oxidation and Hydrolysis

     Square planar  iron and cobalt complexes such as phthalocyanines are
powerful catalysts  for oxidation of sulfur dioxide and sulfide to sulfate in
water, being effective even at  10~9 to 10~11 M  (Hoffman and Lim 1979).  These

-------
observations raise the question of whether iron and manganese complexed with
natural organics might also prove to be oxidation catalysts for selected or-
ganic structures in both dark and light natural waters.  Reliable information
on the role of trace metals - especially iron, copper, and manganese - in
oxidation needs to be developed both for fresh and marine waters.

     Also needed are some careful measurements of hydrolysis rate constants
for organic structures which are known to be susceptible to metal ion cata-
lysts (such as copper ions) under metal ion-free conditions and in fresh and
saline waters.

1.4.3  Microbially Mediated Transformation

     There is a dearth of reliable data concerning the behavior of mixed
culture organisms in the biotransformation of organic molecules in natural
waters.  The need to understand synergistic effects of mixed cultures
such as cooxidation requires more emphasis, as does the need for more investi-
gations at very low concentrations of chemicals in natural waters where
natural organics are present in large excess.

     Several other problems in biological transformation include the possible
role of photosynthetic algae in transformations, the need for greater under-
standing of anaerobic transformations, and the need to recognize that fate and
effects are bound together when considering interactions of microorganisms
and chemicals.  Of major concern in evaluating microbial transformation in
streams and rivers is whether organisms in these water bodies can acclimate to
trace levels of organics and, if so, by what mechanisms and over what periods
of time.

1.4.4  Sediments

     The role of aquatic sediment in transformations is largely ignored in
current assessment procedures.  Sediments serve only as reservoirs of sorbed
organics, and models assume that no significant transformations occur during
a sorbed state.  Sorption is not generally considered a transformation
process, although it does contribute to the total loss of material.  There are
suspicions that this assumption is at least partly incorrect, but specific
data on the question are lacking.  One example is the enhanced rate of
microbial transformations observed when clay particles are added to dilute
solutions of organics.  Other evidence points to inhibition of microbial
action in sediments where sorption into deep pores removed organics from the
access of microorganisms.

1.4.5  Transformations of Inorganic Chemicals

     Transformations of organic molecules in the environment almost always
have zero or negative free energies and for practical purposes are irreversi-
ble.  Kinetic parameters are needed to evaluate these processes quantita-
tively.  However, transformations of metal ions, complexes, and anions are
often very rapid and reversible.  As a result, thermochemical estimates and
models are valuable tools for understanding the speciation of inorganic
chemicals among different complexation and oxidation states.

-------
     Despite many apparent dissimilarities between organic and inorganic
chemicals, some parallels exist in transformation/transport processes,  and the
following processes must be considered in developing models for fate of in-
organic chemicals:

     •  Oxidation and reduction

     •  Complexation

     •  Precipitation and dissolution

     •  Sorption to particulate matter

     •  Metal-organic identification, synthesis and cleavage

     •  Photochemistry

     •  Bioaccumulation of metals.

1.4.5.1  Oxidation/Reduction

     Species such as Fe, Mn, Cr, Cu and As may be oxidized at rates that de-
pend on their complexation states; Fe2+ oxidizes more slowly as the chloride
complex, and the rate is therefore slower in marine environments.  Some rates
of oxidation are pH dependent (Stumm and Lee 1961).  For the oxidation


                                 Fe 2+ -^  Fe3+
the rate law is

                       -d[Fe2+]/dt = k[Fe2+][02][OH~]2

Manganese is oxidized from +2 to +3 or +4 by 02 under abiotic conditions
(Morgan 1967), and the process is possibly mediated by microorganisms
(Sorokin 1970).  Cu+ is so rapidly oxidized that it is unstable unless
strongly complexed (Baes and Mesmer 1976).  Some evidence exists that Fe2"1"
also has a role in oxidation of organics (Theis and Singer 1974).  More
remarkable is the possibility that Cr3"1" is reoxidized to Cr6+ in coastal water
(Jan and Young 1978).  None of these oxidations are well understood, and some
are slow enough that kinetic data are needed to properly describe the systems.
The oxidation of HS*" to S6+ catalyzed by Fe and Co complexes has been studied
in detail (Hoffmann and Lira 1979).

1.4.5.2  Complexation

     The complexation of Pb, Cu, and Cd by naturally occurring organics needs
additional study as does the hydrolysis of Cu.  Among common divalent cations,
Cu2+ exhibits the largest stability constants for natural organics.  Rate
constants for many complexation reactions are known, but there are significant
gaps for trivalent metal species.

-------
1.4.5.3  Precipitation and Dissolution

     Knowledge of the state of inorganics as precipitates or in solution is
essential for understanding their other reactions.  Much more information on
these processes is needed including pH dependence, anoxic/oxic dependence, and
the influence of other ions (CO^~, PO^") on their rates.  The state of
precipitates (i.e., amorphous versus crystalline) will control their apparent
equilibrium solubilities, and rates of dissolution may confound thermochemical
estimates.  For a number of metal ions the solubility product constant for
CO32~, S2~, and PO^" complexes are not well known and are badly needed.  A
detailed review of physicochemical properties for Cd has been prepared (Baes
1973).

1.4.5.4  Bioconcentration of Metals

     This process usually does not significantly affect the concentrations of
metal ions available in the water column; however, complexation of metal ions
into cellular materials can concentrate metals by 102 to 104 times their water
concentrations and affect either the organism itself or other organisms higher
in the food chain.  Concentration is selective, pH dependent, and appears to
compete with complexation by Cl~, CO^~, SO^~ and organics in water (Sunda
and Guillard 1976).

1.4.5.5  Generalizations on Characterizing Inorganics and Their Fate

     As a minimum, the following information is needed to characterize in-
organic species in water:  pH, temperature, oxygen level, ionic composition and
concentrations (Cl~, C0j2~, S0^~, P043~, S2~), sediment composition load, and
size distribution.  In some cases thermochemistry alone will suffice to calcu-
late equilibrium species distribution attained in one to several weeks; but
when speciation is required in shorter times, existing data on kinetics of
precipitation are inadequate.

1.4.5.6  Organometal Synthesis

     Synthesis of carbon-metal or metalloid bonds occurs in sediments via
cobalamin complexes derived from biological sources (probably microorganisms).
For example, (CH3)2Hg is formed in both oxic and anoxic systems.  Kinetic
parameters for all metal-carbon forming reactions and their reoxidation are
poorly characterized at present and require additional investigation to
facilitate their use in fate models.

1.4.5.7  Photochemistry

     Insufficient data are available on the photochemistry of aquatic inorganic
species to warrant significant generalizations.  However, some metal organic
complexes photolyze with transformation of the organic structure, and limited
data suggest that organic multidentate ligands, which bind strongly to Cu or Fe,
may have photolysis rates which are greatly enhanced by the presence of metal
ions (Lockhardt and Blakeley 1975).  This process could also occur for metal
ions bound to biomass.

-------
1.4.5.8  Adsorption to Particulate Matter

     The dominant factors governing the adsorption of trace metals and trace
anions to particulates in water are pH, temperature, nature of the particulate
phases, amount of surface available, and the overall ionic composition of the
water.  In general, particulates such as metal oxides, biological particles,
and clays exhibit strongly pH dependent adsorption.  The charge characteristics
of the particles and their surface functional groups (-OH, -NH2, -COOH) in-
fluence the intensity and extent of species binding.  Competition for adsorp-
tion sites between minor and major metal species (e.g., Pfo2+, Cu2+ an<$ c
-------
     •  Effects of changing from fresh to marine waters on kinetics

     •  Equilibrium processes.

     Test methods to evaluate kinetic constants need to be improved, and in-
creased efforts should be made to use estimation methods for structure-
reactivity and kinetic relations to place upper and lower bounds on rate
processes.  These two preliminary steps would minimize unnecessary testing
in the laboratory.


1.5  TRANSPORT AND PERSISTENCE IN MARINE AND GROUNDWATERS

     This section discusses those environments and aspects of transport and
transformation which are not easily described by linear dynamic models.  In
particular, the emphasis is on dispersion of the pollutant and the adaptation
of the environmental system to the impact of the pollutant.  The only special
methodological consideration in this section is sampling.  Because of compli-
cated ocean circulation, marine samples and particulate matter or living
organisms have a wide variability.  Therefore, in choosing analytical methods,
the accuracy of the method need not greatly exceed the reproducibility of
samples.

1.5.1  Marine Environment

     There is no single "marine environment" because of the widely differing
transport situations which prevail in different locations.  However, the marine
environment is not totally site-specific, and it is possible to identify five
types of environments which may be ranked in order of decreasing energy of
mixing.

     Dispersion of wave action and mesoscale circulation dominate the open
ocean environment.  Environmental impact is minimal provided the amount of
pollutant is not too large or chronic.  This has been documented in the case of
the Ekofisk Bravo blowout (Mackie et al. 1978; Johnson et al. 1978) and the
Argo Merchant oil spill (Grose and Mattson 1977; Center for Ocean Management
Studies 1978).

     A high energy coastline with a rocky or sandy shore subjected to strong
surf action and frequent storms will normally not retain much waterborne
pollution.  Pollution dispersion will be rapid unless there is a long calm
period where floating material can be deposited at high-tide level (Owens
1978).

     In a low energy coastline, sedimentation will be more prevalent and buried
sediment will be occasionally uncovered and dispersed by storms.  More
biota (especially mollusks used for food) make this an area of greater concern
as pollutants will linger in the environment (Hann 1977; Hess 1978; Gilfillan
and Vandermeulen 1978; Sanders 1978).

     Estuarine waters, a complex regimen of fresh water flowing into salt water
compounded by tidal mixing, can result in unexpectedly long retention times
provided there is no stormy weather or strong offshore current.  Sedimentation

                                      11

-------
is very important, and accumulation of toxic materials in sediments can proceed
unchecked and undispersed for many years (Mann and Clark 1978;  Teal et al.
1978).  Storms or dredging can suddenly reintroduce these materials to the
water.  The high level of productivity of unpolluted estuaries  and the high
levels of human activity in urbanized areas and harbors make this an important
region.

     Some salt marsh and some freshwater wetland areas act as important puri-
fiers of the marine environment, either as filters for material from rivers
and terrestrial runoff, or as active centers of chemical and biological
transformation (Keefe 1972; Nixon and Oviatt 1973; Valiela et al. 1973).
These areas should be of high concern because little if any knowledge is
available of the capacity of wetlands to absorb toxic materials or of when
their ability to absorb toxic materials has been exceeded.

     Additional topics on dispersion include bioturbation and riverine input.
Bioturbation occurs when organisms live in the sediment and mix it by burrowing
(Schink and Guinasso 1978).  An outstanding example is the burrowing shrimp,
Callanassia major, which lives in subtropical subtidal carbonate environments.
A single animal can dig a burrow 1 meter deep and turn over 50,000 cm^ of
sediment a year (Morris et al. 1977).  The result is a blurring of the
sedimentary record and a rapid burial of recent material.  This process may be
diminished if the burrowing animals are vulnerable to pollutant damage.

     The input of most materials from rivers to the marine environment is
poorly known.  The National Academy of Sciences study of petroleum in the
marine environment noted the lack of data, but little has been done to
remedy the situation (Van Vliet and Quinn 1977, 1978).  Little if any informa-
tion is available on the input of other toxic substances.  Unknown is the
residence time of such materials in surface waters before they reach the
ocean and the flux of organic materials of different kinds to the sea, as well
as the rate of degradation and dispersion.  The answers to these questions will
be site-specific, not only with respect to terrain and inputs but also with
respect to transport processes involving sedimentation.

     When one goes from a freshwater to a marine environment, simple chemical
factors such as salinity, dissolved oxygen, and pH may undergo sharp changes.
A change in salinity will modify the speciation of dissolved metals, especially
important toxic metals (Pb, Hg, Cd) which form strong chloride complexes.  Clay
minerals which are cation exchangers modify the composition of their adsorbed
material, exchanging cations when confronted by high concentrations of sodium
ion.  Flocculation and sedimentation may be rapid because of changes in surface
charge.  Dissolved oxygen is normally plentiful (except in relatively unusual
anoxic basins such as deep trenches and the Baltic Sea) and represents a
simplification over the easily depleted stream and lake systems characteristic
of terrestrial waters.  Almost all marine waters are within a few tenths of pH
8, which also makes the modeling process (Stumm and Morgan 1970) somewhat
simpler than in fresh waters where pH may vary from 2 to 10.

     Transport phenomena in the oceans are only beginning to be elucidated.
Except for Gulf Stream eddies, which can be monitored by satellite, and the
rare, large-scale international study such as the MODE program (Richman et al.
1977; Robinson 1976; Wunsch 1976), little is known about the circulation of the

                                      12

-------
ocean on a scale of 100 m to 10 km, yet this is the scale that is most impor-
tant for dispersion of pollutants.  The motion of particulate material in the
open oceans was assumed to be simple settling in a quiescent tank; now it has
been established by the use of deep sea sediment traps (Honjo 1978) that the
primary flux of material from the surface of the ocean to the surficial sedi-
ments is rapid (46 mg/m^ day at 5367 meters in the Sargasso Sea), mediated by
copepod grazing and excretion of fecal pellets which are large (hundreds of
microns) smooth packages that sink at a rate of about a hundred meters per day.
Thus, the sediments of the deep ocean are chemically more like the surface
waters than they are like the nearby deeper waters.  This discovery points out
that scientists constantly must be on the lookout for unknown or unexpected
pathways in describing or predicting the environmental fate of pollutants.

1.5.2  Groundwater

     Presently, 20 percent of the nation's water for industry and domestic use
is supplied from groundwater, but 50 percent of drinking water is of ground-
water origin. Subsurface waters are currently rather free from pollution com-
pared to surface waters, but if they were to become contaminated they would be
exceptionally difficult to purify and probably remain contaminated for long
periods.  The recent landfill chemical leakage in Niagara Falls points to the
potential long-term danger of subsurface water contamination.  Most
information and regulations concerning water pollution relate to surface
waters.  Laboratory or field studies seldom examine the special conditions
which characterize groundwater such as methods of entry, unusual physical,
chemical, and biological conditions, abnormal persistence, and possible
chromatographic versus quasi-plug flow of water movement (American Water Works
Association et al. 1973).

     Pollutants can gain access to the groundwater via injection wells, land-
fills, surface mining, and rapid infiltration.  Injection wells are used to
prevent saltwater intrusion or subsidence and often use tertiary treated waste-
water.  These wells add water directly to the underground aquifer.  Landfills
are typically located near cities, which are normally near rivers and shallow
aquifers.  Landfills often penetrate into the water table and are possible
sources of pollution leaking into the groundwater for years.  Plastics in
landfills have been shown to release phthalate esters into groundwaters and
other refractory organics are probably released (Dunlap 1976).  The Niagara
Falls industrial landfill incident forced people to vacate their homes because
of the slow leakage of toxic chemicals into an underground aquifer.

     Surface mining is an increasingly important potential source of shallow
aquifer contamination, because mining operations are often below the shallow
water table.  Shallow groundwaters can have a humic acid content from decaying
surface plant matter, and these acids can have strong heavy metal chelating and
organic leaching capacities which could lead to contamination of shallow
aquifers near surface mines (Steelink 1977).

     Land application of wastewater for treatment by rapid infiltration is
encouraged by EPA.  A biologically active mat is built up at the surface of
the infiltration basin, and it is expected that this active layer will adsorb
as well as decompose organics by bacterial action.  Many aspects of this process
should be examined, such as the effectiveness of the living mat to retain and

                                      13

-------
decompose refractory organics, optional application/aeration periods as well as
soil and climatic variations, and the relative kinetics of sorption versus
biological degradation in the top few inches.  In addition, little information
is available on the adaptation time required for new sites to establish
themselves biologically.  If rapid infiltration for water treatment is to be
encouraged, an important need exists for the evaluation of the associated
chemical, engineering, and biological factors.  This evaluation is critical
because the infiltrated water often enters shallow flowing groundwater with
downstream human uses and, because it is well water, it is assumed to be
potable.

     Insufficient attention has been given to the special physical and chemical
conditions which exist in groundwater and which may affect transport, degra-
dation, and biological activity.  There is no UV light, so photolysis, important
in surface waters, is not possible.  Typically, the organic content of the sub-
surface soil is low, and thus absorptive mechanisms for refractory organics are
probably not as important as in surface soils and sediment.  Since groundwaters
are normally anaerobic, oxidation is not likely.  Also, if bacteria are present
they must be anaerobes or facultative aerobes, and both generally degrade
organic matter more slowly and less completely than aerobes (Metcalf and Eddy
1972).  Because of the low potential BOD of most groundwater even anaerobic
microbiological activity is probably limited.  Most studies of microbiological
activity have focused on aerobic organisms, while little is known about the
microbiology and metabolism in anaerobic groundwaters.  The importance of
mineral redox couples, e.g., Fe^+/Fe^+ or Mn^+/Mn02, in the decomposition of
organics should be examined.

     Sampling and analytical techniques for very low level materials are
difficult, especially in deeper aquifers where suction type pumping is no
longer feasible.  There is a need for improved sampling methodology, particu-
larly for trace level hydrocarbons (Dunlap 1977).  Once developed, a coordi-
nated sampling regime in space and time should be realized to establish the
presence and movement of trace level contaminants about land application sites
in various soil and climatic conditions.

1.5.3  Microbial Interactions with Toxic Substances in Marine Environments

     Responses of microbial populations to continuous exposure from low levels
of pollutants may differ from responses to "one time" introduction of these
pollutants.  Two distinct research approaches are needed to investigate the
influence of microorganisms on the rates of degradation of the pollutant.  In
the first case, the resident microbial assemblage may have become adapted to
the pollutant through selection for resistant strains and induction of enzymes
capable of attacking the pollutant.  For such situations relatively long term
degradation studies using microcosms would be appropriate.  Data are needed,
however, regarding the length of exposure time necessary for population adap-
tation and the threshold concentrations (if any) of pollutants at which selec-
tion for resistant types and enhanced degradation rates will occur.  Conversely,
studies intended to give information on the response of pre-existing microbial
populations to the presence of a toxic substance need to be designed to follow
the initial rates of degradation and must therefore involve shorter time
periods.  The problems inherent in following short-term (minutes or hours)
turnover of a pollutant in marine systems might be resolved by the development

                                      14

-------
of radio-tracer techniques for those pollutants which are available in
radiolabeled form.

     Recent studies have shown conclusively that, with regard to microbial
uptake and mineralization of organic compounds by marine microbial populations,
the fate of the compound depends largely on the concentration at which it is
present in water (Wright 1973).  At very low concentrations, free-living
bacteria appear to out-compete other components of the microbial assemblage for
organic compounds.  At much higher concentrations, organisms such as algae and
bacteria attached to detrital particles compete more successfully (Wright and
Hobbie 1965; Azam and Hodson 1977).  Therefore, any experimental design for
following the microbial fate of a compound must consider the concentration at
which the substance is present in the environment.  Degradation studies should
be conducted over a wide range of pollutant concentrations if the substance
concentration can vary.

     Models for predicting the fate of a toxic substance in a marine system
characteristically contain terms describing the contribution of microbial
activity to the net loss of compound from the environment.  These terms contain
a factor which estimates the biomass of the microbial population.  There are a
number of techniques presently in use for determining microbial biomass includ-
ing plate counts (Carlucci and Pramer 1957), particulate adenosine triphos-
phate (ATP) determination (Holm-Hansen 1973; Hodson et al. 1976), and total
adenylate determination (Karl and Holm-Hansen 1977).  All of these methods
have severe limitations.  For instance, plate counting of bacteria from sea-
water often under-estimates population size by a factor of 1(H or more.
Studies are needed which will identify the most reliable methods for estimating
microbial biomass in seawater and sediments so models can be predicated on some
standardized estimate.

     Bacteria and other marine microorganisms accumulate substances from water
both by adsorption and by "active transport" (Wright and Hobbie 1965; Hodson
and Azam 1977).  In the case of compounds which are merely adsorbed onto cell
surfaces, the microorganisms are functioning merely as particles, transporting
but not transforming the pollutant.  Conversely, active transport into the cell
often leads to metabolism of the compound.  Studies assessing the importance of
microbes to the fate of a toxic substance often do not distinguish between
these two fundamentally different processes, and further work is needed to make
such a distinction.  It may be possible to classify toxic substances according
to whether they are adsorbed onto or transported into microbial cells (or
both).

     Presently, there are no data on the residence times of microbial cells in
marine systems.  If marine bacteria which accumulate toxic substances rapidly
grow and are preyed upon by higher organisms, a potential exists for the
bacteria to serve as "links" in passing the substances to organisms at higher
trophic levels, including man.  If marine bacteria grow slowly or not at all in
seawater and are insignificant as links in marine food webs, they may serve as
terminal "sinks" for toxic substances retaining them until they are deposited
in sediments, or metabolized to CC>2 and organic metabolites.

     To date, nearly all studies of microbial degradation of organic substances
in marine environments have been conducted under well oxygenated conditions.

                                      15

-------
Studies are needed which will give information on rates of degradation in
oxygen depleted environments such as coastal sediments and salt marshes where
the anaerobic microbial processes (fermentation, denitrification, sulfate
reduction, and methanogenesis) predominate.

     Studies of pollutant degradation in aqueous environments typically involve
an initial concentration measurement, incubation period, and filtration fol-
lowed by subsequent concentration determination.  There has not been much uni-
formity with regard to the types of filter membranes used, and recent studies
indicate that frequently utilized glass fiber filters retain only a few percent
of marine bacteria (Hodson and Azam, unpublished data).  Thus, subsequent
concentration measurements include the sum of dissolved and "microbially
associated" pollutant.  Such methodology needs improvement and standardization.

     A number of recent studies indicate that the concentrations of many
organic compounds in surface films on seawater can be 3 to 6 orders of
magnitude higher than in bulk seawater.  Similarly, the density of bacteria in
the surface microlayer is many times higher than that of bulk seawater (Odham
et al. 1978). In such an organically enriched environment, microbial activity
might also be expected to be many times higher than in bulk seawater.  Perhaps
a significant percentage of the degradation of certain toxic organics in
seawater occurs in this microlayer.  Existing data are conflicting; i.e.,
inhibition of microbial activity in this layer by UV radiation may negate the
effect.  More studies seem appropriate, especially considering the hydrophobic
nature of the many important toxic organic compounds.

     In any given marine environment, microbial processes play a significant
role in determining the overall chemical condition of pH, EH, oxygen availabil-
ity, etc.  If the presence of a toxic substance alters the rates of microbially
mediated geochemical processes, then the physicochemical environment and
fate of the substance will also be altered.  It is essential to develop
techniques for assessing effects of toxic substances on naturally occurring
microbial biogeochemical processes.  In the past, modelers have assumed that
concentrations would always be below the levels at which microbial processes
could be affected.  The implications of the possible biogeochemical effects are
too important to be dismissed with this assumption.

1.5.4  Recommendations for Further Study of Transport and Persistence in Marine
       Groundwaters

High Priority:

 1.  Present scientific understanding does not permit accurate prediction of
     the fate of any toxic substance injected into groundwater systems.  This
     inability is particularly acute for substances needing photolysis,
     microbial action, or dispersion to render them innocuous.  Research on the
     fate of such materials in ground water systems is badly needed.

 2.  Monitor groundwater and study mechanism of decay of toxic chemicals.

 3.  Study changes in distribution of pollutants between particulate and
     dissolved phases, as well as speciation of components, on transition from
     fresh to salt water.

                                       16

-------
 4.  Give considerations to and study the effects of ocean disposal of waste
     in high energy coastal regions.

 5.  Study adaptation of natural microbial populations to pollutants, and
     determine threshold concentrations at which selection for resistant types
     and enhanced degradation rates will occur.  Radiotracer methods may be
     helpful in these studies of short-term processes.

Medium Priority:

 6.  Determine rates of microbial degradation in oxygen-depleted environments
     such as coastal sediments and salt marshes, and study such sites as
     possible areas for natural processing of less-toxic pollutants.

 7.  Degradation studies should be made over a wide range of pollutant
     concentrations since microbial processes are expected to be nonlinearly
     dependent on concentration.

 8.  Methods are needed for estimating microbial biomass in seawater and
     sediments so that models can be predicated on some standard estimate.

Long-term research topics related to the above:

 9.  Distinguish between uptake of pollutants by adsorption on microbial cell
     material and active transport into the cell which can lead to metabolism
     o f the compound.

10.  Investigate microbial activity in the sea surface microlayer and the
     influence of photolysis on this activity.

11.  Determine residence times of microbial cells in marine systems.

12.  Investigate the alteration of natural microbially mediated geochemical
     processes by the addition of pollutants,  improve methodology for
     pollutant studies involving filtration to separate microbial fractions
     from dissolved fractions.
1.6  SOURCES OF TOXIC CHEMICALS IN AQUATIC ENVIRONMENTS

1.6.1  Introduction

1.6.1.1  Flow and Mass Balance Concepts

     The concepts of a mass flow of pollutants to and through different envi-
ronmental compartments (water, air, sediments, soils, biota) and of a chemical
mass balance for elements and compounds are essential features of any effort to
accurately relate pollutant emissions to environmental exposure levels.  An
assessment of exposure level for a toxic chemical in an aquatic environment
therefore demands identification of all important sources for the chemical and
physical-chemical characterization of each source.  The pertinent information
for each source includes:  total emission rate of flux of the chemical to the
                                      17

-------
aquatic environment, the physical state of the chemical (dissolved,  particu-
late, or gaseous in the source),  and specific molecular forms or speciation.

1.6.1.2  Sources and Exposure Assessment

     A rationale for predictive exposure assessment is summarized by the
following scheme for a chemical in an aquatic environment:

          Fluxes

   SOURCES - - ->• TRANSPORT - - >  TRANSFORMATION - - -»• FATE

               (Physical Forms)        (Species)

                   Environmental  Exposures


The chemical mass balance concept may be expressed as:

               Accumulation = Inputs - Outputs ± Transformations

As discussed in Section 1.2 the physical-chemical state of inorganic and organ-
ic chemicals is important in determining environmental behavior such as trans-
port rates, biological availability, reactivity, and toxicity.  The resultant
accumulation and exposure for an element, compound, or derived chemical species
will generally depend upon input fluxes, transport out of the system, and
decomposition and formation through reactions.  Thus, it is impossible to
reliably treat transport, fate, and exposure without a thorough knowledge of
source identity, mass emission rates, and physical-chemical forms.

1.6.2  Source Identification

     From a broad perspective, sources of toxic chemical flux to the
environment can arise at one or more points in the material stream of
chemicals including extraction, manufacture, transportation, use, accidental
losses, discharge, and disposal (National Academy of Sciences 1977a).  It may
be necessary to identify a variety of different chemical sources for a
specific aquatic environment.  Distinctions between "point sources" and
disperse or "diffuse sources" are standard.  Significant examples of diffuse
sources for the aquatic environment are atmospheric deposition of metals
(Huntzicker et al.  1975), gas absorption, release of pollutants from sediments
(Neely 1977), and chemical infiltration of ground waters from polluted rivers.

     General methodology for treating pollutant flows in a region has been dis-
cussed by Davidson and Friedlander (1978) with respect to metals and by White
and Friedlander (1978) with respect to the suspected carcinogens, benzo[a]
pyrene and chloroform.  Their work emphasizes that an inaccurate emission
inventory for a region can sometimes result in failure to identify an important
environmental pathway.

     The example of lead (Pb) in Southern California is instructive (Patterson
and Settle  1975).  The mass balance for Pb in the coastal waters required
input flux data for aerosol deposition, rain, wastewaters, storm runoff, and

                                      18

-------
dry weather flows.  The diffuse atmospheric sources accounted for ~35 percent
of total inputs, while coastal wastewater discharges (point sources) accounted
for ~45 percent, with river flows adding ~20 percent.

     Not all chemical species are introduced directly to the water environment
by primary emission from a source or transported via another environmental
compartment.  Transformation reactions among species represent secondary
sources of hazardous materials in the aquatic environment.  Important examples
currently recognized include trihalomethane formation by chlorination of
various organics (Barcelona and Morgan 1978), synthesis of methylmercury and
other metalalkyl compounds (National Academy of Sciences 1977b),  and reaction
of nitrite with amines forming nitrosamines.

     In the case of chloroform in the Los Angeles Basin, Barcelona and Morgan
(1978) estimated from an analysis of mass flows that about 14 kg/day of chloro-
form are emitted to the atmosphere from primary sources, whereas  65 kg/ day are
emitted from the secondary sources involving chlorination of water.  Daily
dose/exposures to chloroform via water and air were calculated at 21 yg,
and 14 pg, respectively.

     Via reactions with chlorine, the precursor organics in water are the
sources of chloroform and other trihalomethanes.  Thus, humic acids and other
naturally-occurring organics have been identified as sources of a toxic
chemical in water.  In general, sources of reactive materials which may yield
toxic substances need to be systematically identified.

1.6.3  Source Characterization

1.6.3.1  Fluxes

     For each identified primary or seconday source of a toxic chemical to an
aquatic environment one needs to know the total flux (g/hour, kg/day, etc.) of
the actual chemical or its precursors.  This information is necessary for a
total element or compound mass balance in the system.  However,  it may not be
sufficient for assessing transport, persistence, exposure and fate for any but
purely conservative substances whose physical-chemical form is unchanged in the
environment.  Thus, for many chemicals attention must be directed to the
physical states of an element, e.g., dissolved or gaseous versus  particulate
fractions, and to species present, e.g., oxidized versus reduced  forms,
inorganic versus organic or metallorganic compounds, free versus  complex ions,
and monomers versus polymers.

1.6.3.2.  Dissolved Versus Particulate Fractions

     Toxic chemicals may exist in particulate matter in an adsorbed, precipi-
tated, solid-solution, or emulsified state.  The size range may be wide or
narrow.  Information on the dissolved and particulate fluxes of chemicals from
sources to aquatic environments is important in predicting transport, persis-
tence and fate behavior.  The need for such information is greater the longer
the time scale for transfer between these fractions in comparison to the
residence time of the chemical in the aquatic environment.  Chemicals which are
predominantly particulate at the source, and which tend to remain so with time,
have a "robust" association with particulate matter.  These chemicals will be

                                      19

-------
deposited to sediments rapidly, have less availability to some biota,  and
will show a reduced tendency to volatilize.  On the other hand, chemicals with
a rapid particulate-solution exchange rate ("labile") will show little
influence of their initial source characteristics once in the environment.
Thus, the initial distribution in a source and the kinetics of particulate-
solution interchange are important characteristics of any given chemical.
There is considerable evidence for many organic and inorganic chemicals that a
substantial particulate fraction exists for a variety of sources, especially
effluents and sludges, as well as for rivers or estuaries with high suspended
loads.  Almost all available information on particulate chemicals in water is
based upon filtration or sedimentation procedures.  The distinction between
"particulate" and "dissolved" has been operationally defined by means of
membrane or other filters with sizes of separation on the order of 0.5 ym to
0.1 ym.  Fundamental considerations, and experience with atmospheric particu-
late pollutants, suggest that information on the size distribution of particu-
late fractions of chemicals will allow more accurate predictions of transport
and exposure in aquatic environments.

1.6.3.3  Implications for Chemical Analysis

      Trace organic analysis and trace element analysis for source and ambient
environmental samples must take cognizance of the existance of particulate and
soluble fractions.  Certain isolation procedures such as purging of volatile
organics may be incomplete because of strong adsorption and slow release from
particulate matter.  It appears that present analytical methodologies for
organic chemicals need to be critically examined with respect to the effective-
ness of isolation procedures for chemicals in the particulate state.

1.6.3.4  Particle-Size Distributions and Toxic Chemicals

      The use of 0.1 to 0.5 nm membrane filters to distinguish between
dissolved and particulate fractions of a chemical is reasonable when
essentially all of the chemical resides in particles of a size greater than the
filter pore size.  If the particle size distribution in a source extends well
down in the colloidal range (particle diameters much less than 0.1 pm) and if
adsorption holds a chemical in the particulate state, then it is predicted that
the chemical will have a significant mass that will escape conventional
filtration separation procedures.  The joint particle-size and distribution for
a chemical is an important piece of information in guiding the choice of
analytical methods and prediting transport and fate in the wter environment, as
it has proven to be in the atmospheric environment (Davidson and Friedlander,
1978; Abrott et al., 1978).

     There is a need for a new research effort to identify both the form of
particle-size distributions and the toxic chemical content of different
particle-size fractions in sources and aquatic environments.  It will be neces-
sary to develop new sampling and separation procedures for sub-micron particu-
late fractions in order to obtain results of wide applicability.  The findings
of Hoffmann and Lim (1979) on the size fractions of copper, lead, cadmium, and
carbon in surface waters obtained by ultrafiltration represent one point of
                                      20

-------
departure.  Their methodology might be extended to characterizations of other
chemicals in sources.  Ultracentrifugation is another method which should be
explored more fully.

1.6.3.5  The Need for Chemical Species Information

     For both organic and inorganic chemicals, the species present in the
system can strongly influence transport, transformation, distribution, and
biological effects. The number of organic chemicals of concern is far greater
than the number of elements, and each organic chemical is a species in the
present sense.  Precursors of organic transformation and products of reactions
represent additional organic chemical species with which to contend.  Modern
trace organic analysis procedures such as chromatography, mass spectroscopy,
and other spectroscopic methods detect and quantitate on a species basis.
While it is increasingly recognized that a number of distinct chemical forms
of each element can exist in water, exposure assessment for aquatic systems
has been based almost entirely on total elements for trace inorganic pollutants
in water (Hoover 1978).  The reason lies in present limitations in analytical
capabilities to differentiate among species of most elements.

     A species matrix for the elements arsenic, chromium, cadmium, lead,
copper, nickel, zinc, mercury, nitrogen, and selenium would contain at least 40
distinct chemical species comprising oxidized and reduced forms, hydrolysis
products, complex ions, polymers, precipitates, and organic derivatives.
Solution species for elements are strongly dependent upon pH, redox conditions,
and major anion and cation constituents.  To an unknown degree, species are
also dependent upon naturally-occurring and pollutant organic chemical consti-
tuents some of which are chelating agents.  Solution speciation is linked to
particulate-dissolved fractionations for many elements (Hoover 1978; Vuceta
and Morgan 1978) .

     The need for species characterization in sources exists because physical-
chemical forms provide initial-concentration conditions for
transport-transformation fate models.  For example, chromium can exist in
wastewater effluents in two oxidation states, Cr(III) and Cr(VI).  The Cr(III)
may exist in solution as hydrolysis products and complexes or in a particulate
state, either adsorbed or precipitated; Cr(VI) may be in solution or adsorbed
(Morel 1975; Jan and Young 1978).  The total flux of Cr to the water environ-
ment, FCr, may consist of FCr(III) part, FCr(III) diss' FCr(VI)part' and
FCr(VI) diss*  'I^le individual fluxes are products of the total element flux and
the different speciation fractions for the element in the source.  If redox
reactions and particulate-solution interchanges such as desorption and
dissolution are slow with respect to transport processes, the initial
speciation at the source will be preserved into the water environment.

     Similar considerations, different in detail for each element and source,
apply to essentially all elements and compounds.  Available analytical capabi-
lities for chemical species and a lack of kinetic information on many transfor-
mations presently limit detailed speciation-exposure predictions to very few
elements.
                                      21

-------
1.6.4  Recommendations for Improved Source Characterization Methods

     Significant gaps exist in availability of proven procedures of sampling,
isolation, resolution, identification, and quantitation for both organic
chemicals and inorganic species in aquatic sources.  Adequate measurement
techniques do not exist for many of the more than 100 organic chemicals which
EPA must regulate.  In principle, isolation procedures (purging, adsorption,
ion exchange, liquid-liquid extraction, freeze drying, reverse osmosis,
ultrafiltration), resolution techniques (gas and liquid chromatography in
various modes), and identification-quantitation devices (flame ionization,
electron capture, conductivity, microcoulometry, mass spectrometry,
spectroscopy) can be applied in suitable combinations to yield methods for
every organic chemical in water (Trussell and Umphres 1978).  In fact, current
analytical methods are often not as accurate as required nor are they
standardized.  An enormous effort of development and demonstration is now
required to bring forward analytical methods which are acceptable in terms  of
detection, accuracy, and appliability to wastes, effluents, sludges, and
natural waters.

     Areas requiring critical attention are representative sampling of water
containing volatile species, separation and resolution of particulate-
associated chemicals, and the approximate methodological sensitivities to be
used for chemical flux and material balance calculations.  The National
Organics Monitoring Survey quantitated two dozen organic chemicals, partly
because analytical methodology existed for them.  Heesen and Young (1977) noted
that the normal collection and composition procedures lead to significant
losses of halogenated hydrocarbons from wastewater samples.  They estimated
that current methods probably acount for but a small percentage of halogenated
hydrocarbons in effluents.  A more intensive program of analytical development
is clearly needed.

     Perhaps the highest priority need in sources research is the development
and demonstration of methods which will yield accurate total flux data for
chemicals of interest.  Research should focus on sampling, isolation, and
quantitation improvements.  Accurate total fluxes are essential for predictive
models.  Research on measurements and models for dissolved-particulate fractio-
nation and particle size distributions of toxic chemicals will be of signi-
ficant value in developing improved transport-fate-exposure models.  Chemical
speciation of inorganic and organic chemicals in aquatic sources is needed for
predicting transport, availability, and toxicity in the water environment.
Analytical research is needed on complexation, oxidized versus reduced chemical
forms, and organic derivatives of elements.
 1.7  MODELING IN AQUATIC SYSTEMS

     This  section  includes an  identification of the transport and reaction
 mechanisms that can operate  in aquatic environments, an assessment of the
 information needed to describe these mechanisms quantitatively, and an evalua-
 tion of  research needs  for the effective modeling of toxic substances in
 aquatic  systems.   The fate of  the  substances introduced into an environmental
                                      22

-------
system depends on the rate of input, output, and reaction of the material
within the system.

1.7.1  Inputs and Outputs

     Inputs and outputs include transport by advection and dispersion, trans-
port across land-water boundaries (point and nonpoint land-based sources),
transport across air-water boundaries (atmospheric deposition, volatilization
and dissolution, aerosol formation), and transport across sediment-water boun-
daries (deposition, resuspension, diffusive exchange).  Characterization of
these input and output rates involved the following:

     •  Advective transport - requiring description of large-scale fluid
        motions such as river and tidal flows;

     •  Dispersive transport - requiring description of smaller-scale fluid
        motion using dispersion coefficients determined from established
        empirical relationships or field measurements with tracers;

     •  Point sources - requiring description of source flows and total
        concentrations of pollutants.  Knowledge of dissolved and particulate
        fractions is usually required and descriptions of temporal variations
        can be needed;

     •  Nonpoint sources - total pollutant loadings or fluxes, the parti-
        tioning between suspended and dissolved fractions, and temporal changes
        can require determination;

     •  Volatization and dissolution - analysis can use Henry"s solubility
        coefficient, mass transfer coefficients at the air-water interface,
        and such factors as wind speed, water velocity, water depth, and
        temperature;

     «  Atmospheric deposition - this factor may require experimental deter-
        mination in field situations or it can be an output from an atmospheric
        model of the transport and fate of the substance;

     •  Aerosol formation - the transport of toxic materials from the water
        to the atmosphere by the formation of aerosols appears to be uncate-
        gorized at this time;

     *  Deposition to sediments - this can be described experimentally with
        considerable effort.  Deposition depends upon hydraulic flow regime
        and the physical properties of particulates, including their size,
        distribution, and density.  Biological effects from such processes as
        zooplankton feeding may also be important.  Sediment deposition is the
        principal removal flux of most nonconservative substances in aquatic
        systems;

     •  Suspension of sediments - this can be caused by man-made disturbances
        (ships, dredges) and by fluid flow such as storms.  For anoxic
        sediments, pore waters may contain high soluble concentrations of
                                      23

-------
        pollutants which then contaminate overlying waters.   Iron and manganese
        in resuspended sediments can be important in transporting both organic
        and inorganic pollutants.  These inputs are difficult to quantify at
        present;

     •  Diffusional exchange - oxic sediments can support organisms that mix
        the surface sediments (bioturbation), thereby enhancing vertical trans-
        port across the sediment water interface.  Anoxic sediments may permit
        vertical transport by molecular diffusion only.   These fluxes are not
        quantified at present.

1.7.2  Reactions

     Important reactions of toxic substances within aquatic  systems include
hydrolysis, photolysis, sorption and desorption, oxidation and reduction,
association and disassociation, precipitation, biological degradation and bio-
logical uptake or bioaccumulation.  These reactions involve the following:

     •  Hydrolysis - description requires determination of rate constants
        and evaluation of effect of pH, temperature, and possible catalysts?

     •  Photolysis - evaluation includes determination of the absorption
        spectrum of the compound of interest at wavelengths greater than
        290nm, quantum yield, and radiation intensity.  Other factors
        affecting this rate are surface area, depth, competing absorbers such
        as particulates and humic substances, sensitization, and certain
        metals;

     •  Sorption and desorption - description utilizes sorption partition
        coefficients, the concentration of adsorbing solids, kinetic constants
        for adsorption and desorption, the charge of the adsorbing species,
        chemical characteristics of the solids, and the concentrations of
        other substances that can compete for adsorbing sites.  The particle
        size distribution of the adsorbing solids can affect the concentration
        of surface area available for absorption and the rate at which these
        solids are removed from the system deposition.  The rate of removal of
        the pollutant from the water column is related to the rates of supply
        and deposition of the adsorbing surfaces;

     •  Oxidation and reduction - evaluation uses kinetic constants, the
        concentrations of dissolved oxygen and other oxidants, consideration of
        catalysts including metals, acids and bases, and ultraviolet light, and
        determination of pH and temperature effects;

     •  Association and dissociation - description requires the thermodynamic
        constants for both ligand and metal interactions, knowledge of the
        concentrations of reacting and competing metals and ligands,
        evaluation of the effects of pH, temperature, ionic strength, and
        perhaps kinetic constants when polymerization may occur;

     •  Precipitation - description requires thermodynamic constants, the
        concentrations of species forming precipitates, evaluation of the
                                      24

-------
        effects of pH, temperature, and  ionic strength,  and consideration of
        the effects of particulates and of surface active agents such as humic
        substances;

     •  Biological degradation - characterization can use kinetic constants for
        the degradation of the substance of interest by acclimated organisms,
        determination of organism populations, and can require description of
        the acclimation period.  Effects depend upon the electron acceptor
        available and will vary considerably with oxic and anoxic conditions.
        Temperature and pH effects can be significant.  As in the cases of
        removal by deposition and by absorption, the environmental substance
        (in this case biomass) reacting with the toxic substance is controlled
        by the inputs of other materials to the system;

     •  Biological uptake - this may be assessed using a food web model with a
        resultant need for many additional coefficients.   Tt may be possible to
        approximate biological uptake by laboratory measurements such as
        octanol-water partition coefficients.

1.7.3  Modeling Recommendations

     It is not feasible or necessary to develop a single, comprehensive,
general model for the transport and fate of every toxic substance that may have
been or will be introduced into all of the diverse physical, chemical, and
biological conditions occuring in aquatic systems.  The main driving force for
any model is the total input of sources to the system.  The sources will be
site-specific.  Model predictions depend predominantly on source characteriza-
tion.  The second important class of factors is the output of the material of
interest from the system.  Outputs include deposition and volatilization in
addition to advective and dispersive flows.  All are site-specific and
deposition is usually the most difficult to evaluate.  Regarding reactions in
the system, it is not likely that a single pollutant will require evaluation of
all pathways.  Some reactions, such as adsorption and biological degradation,
are difficult to predict and are affected by other inputs into the system.  It
is also plausible that other reactions not described here can be significant.

     The number and diversity of the described inputs, outputs, and reactions
should not be a mathematical or computational problem in most cases.  However,
the measurement requirements for obtaining the fluxes and coefficients are
large and probably overwhelming.  It is useful to evaluate some reactions in
laboratory experiments in order to determine their potential need for
consideration in actual environmental models.

     Table 1-1 lists the processes which have been judged to require further
study to permit effective modeling of transport and fate of toxic substances in
the environment.
                                      25

-------
                   TABLE  1-1.   MODELING  RESEARCH  PRIORITIES
                              Highest Priority

                       Outputs by  depositional processes

                Inputs  and outputs by resuspension  of  sediments

                         Sorption-desorption reactions

                       Biological  degradation reactions

                         Inputs  from nonpoint sources

             Inputs and outputs  by volatilization and  dissolution
                             Significant  Priority

                           Inputs from point sources

  Inputs and outputs by diffusional exchange at the  sediment-water  interface

                       Inputs by atmospheric deposition

                             Hydrolysis reactions

                             Photolysis reactions
1.8  SUMMARY OF HIGH PRIORITY RECOMMENDATIONS FOR AQUATIC RESEARCH

     A.   Physicochemical Properties

         1.   Further development of methods for measuring solubility of solid
             non-ionic materials.

         2.   Influence of natural organic substances (especially fulvic acids)
             on aqueous solubility or apparent solubility.

         3.   Standardized compilation of critically reviewed data.

         4.   Relationship of Kow to environmental behavior.

         5.   Effect of complex formation and sorption on the measurement
             of solubility equilibria for substances forming dissolved
             ionic species.
                                      26

-------
B.  Transport Properties

    1.  Increased quantitative data on the influence of sorption on
        pollutant flux in water columns.

    2.  More information on the processes of uptake, degradation,
        metabolism and accumulation of toxic organic solutes by
        various biotic levels.

C.  Transformation Processes

    1.  Microbial transformation in dilute solutions with mixed cultures.

    2.  Sediment-promoted reactions for sorbed organics.

    3.  Metal-ion catalyzed oxidations, photolyses and hydrolyses.

    4.  Transformations in marine or estuarine waters.

    5.  Oxidation by oxygen.

    6.  Formation and cleavage of metal-organic species.

    7.  Competitive sorption of cations and anions to particulate matter.

    8.  Effects of changing from fresh to marine waters on kinetics and
        equilibrium processes.

D.  Marine and Ground Waters

    1.  Prohibit injection (or other types of disposal affecting
        groundwater) of any toxic substance needing photolysis, microbial
        action, or dispersion to render it innocuous.

    2.  Monitor groundwater and study mechanisms of decay.

    3.  Study changes in distribution of pollutants between particulate
        and dissolved phases, as well as speciation of components, on
        transition from fresh to salt water.

    4.  Give considerations to, and study the effects of, ocean disposal
        of waste in high energy coastal regions.

    5.  Study adaptation of natural microbial populations to pollutants,
        and determine threshold concentrations at which selection for
        resistant types and enhanced degradation rates will occur.
        Radiotracer methods may be helpful in these studies of short-term
        processes.
                                 27

-------
E.  Sources of Toxic Chemicals in Aquatic Systems

    1.   Perhaps the highest priority need in sources research is the
        development and demonstration of methods which will yield accurate
        total flux data for chemicals of interest.   Research should focus
        on improvements in sampling, isolation, and quantitation.
        Accurate total fluxes are essential for predictive models.

    2.   Measurements and models for dissolved-particulate ffractionation
        and particle size distributions of toxic chemicals will be of
        significant value in developing improved transport-fate-exposure
        models.

    3.   Chemical speciation of inorganic and organic chemicals in
        aquatic sources is needed for predicting transport, availability,
        and toxicity in the water environment.  Analytical research is
        needed on complexation, oxidized versus reduced forms, and organic
        derivatives of elements.

F.  Modeling in Aquatic Systems

    1.   Outputs by depositional processes.

    2.   Inputs and outputs by resuspension of sediments.

    3.   Sorption-desorption reactions.

    4.   Biological degradation reactions.

    5.   Inputs from nonpoint sources.

    6.   Inputs and outputs by volatilization and dissolution.


The Aquatic Environment group participants are listed in Table 1-2.
                                  28

-------
                TABLE 1-2.  AQUATIC ENVIRONMENT GROUP PARTICIPANTS
Non-Agency Participants

James N. Butler
Harvard University
Cambridge, MA

Russell F. Christman (Chairman)
Department of Environmental Science
  and Engineering
University of North Carolina
Chapel Hill, NC

Robert E. Hodson
Department of Microbiology
University of Georgia
Athens, GA

Michael Hoffmann
Department of Civil and Mineral
  Engineering
University of Minnesota
Minneapolis, MN

Donald Mackay
Department of Chemical Engineering
  and Applied Chemistry
University of Toronto
Toronto, Ontario
Canada

Theodore Mill
SRI International
Menlo Park, CA

James J. Morgan
Dean of Students
California Institute of Technology
Pasadena, CA

Charles O'Melia
Department of Environmental Sciences
University of North Carolina
Chapel Hill, NC
EPA Representatives

George Baughman (Lead)
Environmental Research Laboratory/
Office of Research and Development
Athens,  GA

Michael Callahan
Office of Water Planning and Standards/
Office of Water and Waste Management
Washington, D.C.

Jess Cohen
Municipal Environmental Research
  Laboratory/
Office of Research and Development
Cincinnati, OH

Asa Leifer
Office of Toxic Evaluation/
Office of Toxic Substances
Washington, D.C.

Ronald Ney
Office of Pesticide Programs/
Office of Toxic Substances
Washington, D.C.

William Sayers
Office of Air, Land and Water Use/
Office of Research and Development
Washington, D.C.

James Steramle
Office of Energy Mineral Industries/
Office of Research and Development
Washington, D.C.

Charles Trichilo
Office of Drinking Water/
Office of Water and Waste Management
Washington, D.C.
                                        29

-------
                              TABLE 1-2 (Continued)
Non-Agency Participants

Mason Tomson
Department of Environmental Science
  and Engineering
Rice University
Houston, TX

Henry Tsuchiya
Department of Chemical Engineering
University of Minnesota
Minneapolis, MN

C. H. Ward
Department of Environmental Science
  and Engineering
Rice University
Houston, TX
                                        30

-------
                                  REFERENCES
Abrott, T. J.,  W. H. White, A. H. Miguel and S. K.  Friedlander.  Human Source
     Emission Relationships for BaP.  In:  Human Dosage/Emission Source
     Relation-ships for Benzo[a]Pyrene and Chloroform in the Los Angeles Basin.
     Report to USEPA. Keck Laboratories, Caltech.  1978.

American Water Works Association, Department of Civil Engineering at Univ. of
     Illinois at Champaigne-Urbana, and Illinois Environmental Protection
     Agency.  Organic Matter in Water Supplies.  Proceedings of the 15th Water
     Quality Conference, 1973.

Azam, F. and R. E. Hodson.  Size Distribution and Activity of Marine Micro-
     Heterotrophs.  Limnol. Oceanog., 22:492, 1977.

Baes, C. F. Jr.  The Properties of Cadmium.  In: Cadmium, the Dissipated
     Element, W. Fulkerson and H. E. Goeller, eds.   Oak Ridge National
     Laboratory, Oak Ridge, Tenn., 1973.

Baes, C. F. Jr. and R. E. Mesmer.  The Hydrolysis of Cations.  Wiley, New York,
     1976.

Barcelona, M. J. and J. J. Morgan.  Human Dosage/Emission Source Relationships
     for Chloroform.  In: Human Dosage/Emissions Source Relationships for
     Benzo[a]pyrene and Chloroform in the Los Angeles Basin.  Report to
     USEPA, Keck Laboratories, Caltech,  1978.

Baughman, G. L. and R. R. Lassiter.  Prediction of Environmental Pollutant
     Concentration.  A.S.T.M. S.T.P., 657.  Philadelphia, 1978.

Carlucci, A. F. and D. Pramer.  Factors  Influencing the Plate Method for
     Determining Abundance of Bacteria in Sea Water.  Proc. Soc. Exp. Biol.
     Med., 96:392,  1957.

Center  for Ocean Management Studies.  Symposium: In the Wake of the Argo
     Merchant.  Univ. Rhode Island.  Narragansett, R. I., 1978.

Davidson, C. I. and S. K. Friedlander.   A Filtration Model for Aerosol Dry
     Deposition: Application to Trace Metal Deposition  from the Atmosphere.
     J. Geophys. Res., 83:2343,  1978.

Dunlap, W. J.   Isolation and Identification of Organic  Pollutants in Ground
     Water.  In:  Identification and Analysis of Organic Pollutants in Water,
     pp. 447-453, Lawrence H. Kieth, ed.  Ann Arbor Sci. Pub., Inc., Ann
     Arbor, Michigan,  1976.

                                      31

-------
Dunlap, W. J. Sampling for Organic Chemicals and Microorganisms in the Sub-
     surface.  Technology Series of USEPA, EPA 600/12-77-176,  Ada, Oklahoma,
     1977.

Gilfillan, E. J. and J. H. Vandermuelen.  Alterations in Growth and Physiology
     of Soft Shell Clams Mya arenaria, Chmromically Oiled with Bunker C from
     Chedebucto Bay, Nova Scotia 1970-76.  J. Fish Res. Board Canada,
     35:630, 1978.

Grose, P. and J. Mattson.  The Argo Merchant Oil Spill: A Preliminary Report.
     U. S. Dept of Commerce (NOAA).  Boulder, Colorado, 1977.

Hann, R. G.  Fate of Oil from the Supertanker Metula.  Proceedings of the
     1977 oil Spill Conference of America.  Petroleum Institute, Washington,
     D. C., pp. 465-468, 1977.

Haque, R. and D. Schmedding.  A Method of Measuring the Water Solubility of
     Very Hydrophobic Chemicals.  Bull. Environ. Contam. and Toxicol.,
     14: 13, 1975.

Heesen, T. C. and D. R. Young.  Halogenated Hydrocarbons in Wastewaters: Knowns
     and Unknowns.  S.C.C.W.R.P. Annual Report, pp. 33-38, 1977.

Hess, W. N.  (ed.).  The Amoco Cadiz Oil Spill: A Preliminary Scientific Report.
     U. S. Department of Commerce (NOAA) and U. S. Environmental Protection
     Agency.  Washington, D.C., 1978.

Hodson, R. E. and F. Azam.  Determination and Biological Significance of
     Dissolved ATP in Seawater.  In: Proceedings of the Second Bi-Annual
     Methodology Symposium, G. A. Borun, ed., pp. 127-140. SAI Tech. Coll.,
     San Diego, CA, 1977.

Hodson, R. E., O. Holm-Hansen and F. Azam.   Improved Methodology for ATP
     Determination in Marine Environments.  Marine Biology, 34:143, 1976.

Hoffman, M.  R. and B. C. H. Lira.  Kinetics and Mechanisms of the Oxydation of
     Sulfide by Oxygen: Catalysis of Homogenous Metal-Phthalocyanine Complexes.
     Env. Sci. Tech., 11:1406, 1979.

[Holm-Hansen, O.  Determination of Total Microbial Mass by Measurement of
     Adenosine Triphosphate.  In: Estuarine Microbial Ecology, pp. 73-89,
     Univ. South Carolina Press, Columbia, S. C., 1973.]

Honjo, S.  Sedimentation of Materials in the Sargasso Sea at a 5367 Meter
     Deep Station.  J. Marine Research, 36:469, 1978.

Hoover, T. B.   Inorganic Species in Water: Ecological Significance and
     Analytical Needs.  EPA-600/3-78-064, 1978.

Huntzicker,  J. J., S. K. Friedlander and C.  E. Davidson.  A Material Balance
     for Automobile Emitted Lead in the Los  Angeles Basin.  In:  Air-Water-Land
     Relationships for Selected Pollutants in Southern California.  Keck
     Laboratories, Caltech.,  1975.

                                      32

-------
Jan, T. K. and D. R. Young.  Chromium Speciation in Municipal Wastewater and
     Seawater.  J. Water Pollution Control Federation, 50:2327, 1978.

Johnson, J. J., P. W. Brooks, A. K. Aldriger and S. J. Rowland.  Presence and
     Source of Oil in the Sediment and Benthic Community Surrounding the
     Ekofisk Field After the Blowout at Bravo.  Proceedings of the Conference
     on Assessment of Ecological Impacts of Oil Spills.  (Keystone Co.),
     American Institute of Biological Science, Washington, D. C.,  1978.

Karickhoff, S. W., D. S. Brown and T. A. Scott.  Sorption of Hydrophobic
     Pollutants on Natural Sediments.  Water Research 13:241, 1979.

Karl, D. M. and O. Holm-Hansen.  Adenylate Energy Charge Measurements in
     Natural Sea Water and Sediment Samples.  In: Proceedings of the Second
     Bi-Annual ATP Methology Symposium, G. A. Borun, ed., pp. 141-169.  SAI
     Tech. Coll., San Diego, CA, 1977.

Keefe, C. W.  Marsh Production: A Summary of the Literature.  Contrib.  Mar.
     Sci., 16:163, 1972.

Kenega, S. E. and C. A. I. Goring.  Relationship Between Water Solubility
     Soil-Sorption Octanol Water Partitioning and Bioconcentration of
     Chemicals in Biota.  Paper Presented at the ASTM Third Aquatic Toxicology
     Symposium, New Orleans, LA., 1978.

Leo, A., C Hansch and D. Elkins.  Partition Coefficients and Their Uses.
     Chem. Rev., 71:525, 1971.

Lockhardt, H. B. and R. V. Blakely.  Aerobic Degradation of Fe (III)
     Ethylenedinitrilo Tetracetate.  Env. Sci. Tech., 9:1035, 1975.

Mabey, W. R. and T. Mill.  Critical Review of Hydrolysis of Organic Compounds
     in Water Under Environmental Conditions.  J. Phys. Chem. Ref. Data,
     7:383, 1978.

Mackay, D., R. Sutherland and W. Y. Shiu.  Determination of Air-Water Henry's
     Law Constants for Hydrophobic Pollutants.  Environ. Sci. Technol., 13:333,
     1979.

Mackie, P. R., R. Hardy and K. J. Whittle.  Preliminary Assessment of the
     Presence of Oil in the Ecosystem at Ekofisk After the Blowout of April
     22-30, 1977.  J. Fish Res. Board Canada, 35, 1978.

Mann, K. H. and R. B. Clark.  Long-Term Effects of Oil Spills on Marine
     Intertidal Communities.  J. Fish Res. Board Canada, 35:791, 1978.

May, W. E., S. P. Wasik and D. H. Freeman.  Determination of the Aqueous
     Solubility of Polynuclear Aromatic Hydrocarbons by a Coupled Column
     Liquid Chromatographic Technique.  Anal. Chem., 50:175, 1978a.

May, W. E., S. P. Wasik and D. H. Freeman.  Determination of the Solubility
     Behavior of Some Polycyclic Aromatic Hydrocarbons in Water.  Anal. Chem.,
     50: 997, 1978b.

                                      33

-------
Metcalf and Eddy, Inc.  Wastewater Engineering.  McGraw-Hill, New York,
     pp. 415 ff., 1972.

Morel, F. M. M.  Fate of Trace Metals in L. A. County Wastewater Discharge.
     Env. Sci. Technol., 8:756, 1975.

Morgan, J. J.  Chemical Equilibria and Kinetic Properties of Manganese in
     Natural Waters.  In: Principles and Applicators of Water Chemistry, S.  D.
     Faust and J. V. Hunter, eds., p. 56.  Wiley, New York, 1967.

Morris, B. F., J. Barnes, F. Brown and J. Markham.  The Bermuda Marine
     Environment.  Bermuda Biological Station Special Publication No. 15,
     1977.

Neely, W. B.  A Material Balance Study of Polychlorinated Biphenyls in Lake
     Michigan.  Science of the Total Environment, 1:117, 1977.

Neely, W. B., D. R. Bronson and G. E. Blau.  Partition Coefficients to Measure
     Bioconcentration Potential of Organic Chemicals in Fish.  Env. Sci.
     Technol., 8:1113.  1974.

National Academy of Sciences.  Fates of Pollutants: Research and Development
     Needs.  Analytical Studies for USEPA, 1977a.

National Academy of Sciences.  Sources of Residuals and Techniques for Their
     Control.  Analytical Studies for USEPA, 1977b.

Nixon, S. W. and C. A. Oviatt.  Analysis of Local Variations in the Standing
     Crop of spartina Alterniflora.  Bot. Mar.,  16:103, 1973.

Odham, G. , B. Noren, B. Norkrans, A. Sodergren and H. Losgren.  Biological and
     Chemical aspects of the Aquatic Lipid Surface Microlayer.  Prog. Chem.
     Fats and Other Lipids,  16:31, 1978.

Owens, E. H.  Mechanical Dispersal of Oil Stranded in the Littoral Zone.  J.
     Fish Res. Board Canada, 35:563, 1978.

Patterson, C. C. and D. Settle.  Contributions of Lead via Aerosol Deposition
     to  the Southern California Bight.   In: Air-Water-Land Relationships for
     Selected Pollutants in  Southern California.  Keck Laboratories, Caltech.,
      1975.

Richman, J. G., C. Wunsch and  N. G. Hogg.  Space and Time Scales of Mesoscale
     Motion in the Western North Atlantic.  Space Physics, 15:385, 1977.

Robinson, A. R.  Eddies and  Ocean Circulation.   Oceanus, 19:2,  1976.

Sanders,  H. L.  Florida Oil  Spill Impact on the  Buzzards Bay Benthic Fauna,
     West Falmount.  J. Fish Res. Board  Canada,  35:717, 1978.

Schink,  D. R. and L. Guinasso, Jr.  Redistribution of Dissolved and Adsorbed
     Materials in Abyssal Marine Sediments Undergoing Biological Stirring.
     Amer. J.  Sci., 278:687, 1978.

                                      34

-------
Smith, J. H., W. R. Mabey, N. Bohonos, B. R. Holt, S. S. Lee, T. W. Chou,
     D. C. Bomberger and T. Mill.  Environmental Pathways of Selected Chemicals
     in Freshwater Systems, Part I.  EPA 600/7-78-074, 1978.

Sorokin, Y  U.  Interrelations Between Sulfur and Carbon Turnover in Meromictic
     Lakes.  Arch. Hdyrobiol., 66:391, 1970.

Spencer, W. F. and M. M. Cliath.  Vapor Density of Dieldrin.  Env. Sci.
     Technol., 3:670, 1969.

Spencer, W. F. and W. J. Farmer.  Assessment of the Vapor Behavior of Toxic
     Organic Chemicals.  Personal Communication, 1978.

Steelink, C.  Humates and Other Natural Organic Substances in the Aquatic
     Environment.  J. Chein. Ed., 54:599, 1977.

Stumm, W. and G. F. Lee.  Oxygenation of Ferrous Iron.  Ind. Eng. Chem.,
     53:143, 1961.

Stumm, W. and J. J. Morgan.  Aquatic Chemistry, chapters 8, 9, 10.  Wiley,
     New York, 1970.

Sunda, W. and R. R. L. Guillard.  The Relationship Between Cupric Iron Activity
     and the Toxicity of Copper to Phytoplankton.  J. Mar. res., 34:511,  1976.

Teal, J. M., K. Burns and J. Farrington.  Analyses of Aromatic Hydrocarbons
     in Intertidal Sediments Resulting from Two Spills of No. 2 Fuel Oil in
     Buzzards Bay, Mass.  J. Fish Res. Board Canada, 35:510, 1978.

Theis, T. L. and P. C. Singer.  Complexation of Iron(II) by Organic Matter and
     its Effect on Iron(II) Oxygenation.  Env. Sci. Technol., 8:569, 1974.

Trussell, A. R. and M. D. Umphries.  An Overview of the Analysis of Trace
     Organics in Water-  J. American Waste Water Association, 70:595, 1978.

Valiela, I., J. Teal and W. Sass.  Nutrient Retention in Salt Marsh Plots
     Experimentally Fertilized with Sewage Sludge.  Estuarine Coastal Mar.
     Sci., 1:261, 1973.

Van Vliet, E. S. and J. G. Quinn.  Input and Fate of Petroleum Hydrocarbons
     Entering the Providence River and Upper Narragansett Bay from Wastewater
     Effluents.  Env. Sci. Technol., 11:1086, 1977.

Van Vliet, E. S. and J. G. Quinn.  Contribution of Chronic Petroleum Inputs to
     Naragansett Bay and Rhode Island Sound Sediments.  J. Fish Res. Board
     Canada, 35:536, 1978.

Vuceta, J. and J. J. Morgan.  Chemical Modeling of Trace Metals in Fresh
     Waters: Role of Complexation and Adsorption.  Env. Sci. Technol., 12:1302,
     1978.

White, W. H. and S. K. Friedlander.  Human Dosage/Emission Source Relationships
     for Benzofa]pyrene and Chloroform in the Los Angeles Basin.  Report to
     USEPA.  Keck Laboratories, Caltech, 1978.
                                      35

-------
Wright, R. T.  Some Difficulties in Using CV4-Organic Solutes to Measure
     Heterotrophic Bacterial Activity.  In:  Estuarine Microbial Ecology,
     L. H. Stevenson and R. R. Colwell, eds., pp. 199-217.  University of
     South Carolina Press, Columbia, S. C., 1973.

Wright, R. T. and J. E. Hobbie.  The Uptake of Organic Solutes in Water.
     Limnol. Oceangr., 10:22, 1965.

Wright, R. T. and J. E. Hobbie.  Use of Glucose and Acetate by Bacteria and
     Algae in Aquatic Ecosystems.  Ecology, 47:447, 1977.

Wunsch, C.  The Mid-Ocean Dynamics Experiment (MODE).  Oceanus, 19:45, 1976.
                                       36

-------
            2.0 TRANSPORT AND FATE IN THE ATMOSPHERIC ENVIRONMENT
2.1  INTRODUCTION

     Research on transport and fate of toxic chemicals in the atmosphere
concerns (1) the relationship between environmental quality and the various
sources currently emitting toxic compounds, and (2) the effects upon
environmental quality of a new technology or industry which will emit toxic
chemicals to the atmosphere.  Transport and fate parameters of toxic agents
depend upon the agents' chemical and physicochemical properties; however,
priorities for studying their flow through the environment should be based on
their toxicity and emission rates.  Much of the discussion in this section is
devoted to recommendations for a research program designed to meet EPA needs
in establishing regulatory guidelines for emission and control of toxic
chemicals in the atmosphere.

     Figure 2-1 illustrates how transport and fate problems were approached in
this workshop and lists the general area of Section 2.0 which discusses the
subject.  The diagram traces the flow of pollutants from the source through the
atmosphere where transformations may take place, to the point(s) where the air
quality parameters are measured.

     An emission source data base (Section 2.2) of high quality is essential to
achieve a satisfactory understanding of the transport and fate of atmospheric
pollutants.  This element is sometimes overlooked and when programs on pollu-
tant transport and fate are compartmentalized the atmospheric compartment is
frequently treated (incorrectly) as if it were uncoupled from the source.

     Rates of atmospheric transformation processes (Section 2.3) are needed to
determine the lifetimes of organic compounds in the atmosphere.  Such informa-
tion can be used to guide source resolution and mass balance studies.  In many
cases it will be sufficient to establish that transformation rates are either
very high or very low compared with the transport time of interest so that
detailed information may not be necessary.  The products of chemical conversion
may themselves pose toxicity problems and methods are needed for predicting the
nature of such products.

     An air quality data base (Section 2.4) of research quality (as distin-
guished from routine air monitoring information) should be assembled.  These
data should include detailed information on chemical speciation for organic
gases, aerosols, particle size distribution, and the distribution of chemical
species with respect to particle size.  The development of instrumentation
necessary to accomplish this goal should continue.
                                       37

-------
                      O
                                           CO
                                           LU
                                         COO
u.2
oi-
                     Qo
XDC;
Q-oL  -
C051OCNI
                                           CO
                  CO
                  LU
                  O
                  CO
                        38

-------
     Research quality data are required for source resolution studies (Section
2.5) and environmental mass balances (Section 2.6) as well as for toxicity
assessment.  It may be possible to carry out source resolution studies for
organic compounds using a chemical species balance method analogous to the
chemical element balance (CEB) method developed for aerosol source resolution.
A program of research should be initiated in this field using data available
in the literature.

     Mass balance studies are particularly useful for establishing the flow of
a pollutant among various environmental compartments.  Such studies should be
initiated for organic compounds such as polynuclear aromatic compounds and
halogenated organics.

     While dealing with current air pollution poses many problems, an even
greater task involves predicting the effects on environmental quality caused by
introduction of new technologies or industries which release organic chemicals
to the atmosphere.  The ability to make such predictions will improve with the
experience developed in treating existing sources.  A good way to approach new
sources will be by analogy with the methods discussed for treating existing
sources.

     These recommendations for research and development have attempted to
strike a balance between basic research needs and the practical needs of the
United States Environmental Protection Agency with associated time constraints.
Recommendations for research have been limited to the direct needs of an
integrated program on transport and fate and should be considered as minimum
goals.
2.2  SOURCE CHARACTERIZATION

     Recently a detailed picture of the characteristics of automobile emissions
and emissions from pulverized coal-fired power plants has emerged with data on
both the aerosol and gas phases available.  Aerosols, particle size
distributions, and chemical composition as a function of particle size have
been obtained and a wide variety of organic compounds have been identified in
automobile exhaust gases.  These measurements have been made with advanced
technology methods, sometimes especially developed for the purpose.  Such data
can be defined as research quality and are necessary for predictive modeling,
source resolution, and mass balances.  They are also needed for asssessing the
health and ecological effects of air pollutants.

     Few data of research quality are available for emissions from the chemical
industry.  There are many more types of chemical industry sources than auto-
mobile or power plant sources.  As a result, it may be more difficult to
generalize concerning the nature of such emissions.  With proper measurements,
it may be possible to generally characterize the emissions of a major class of
sources, such as petrochemical plants, by doing detailed studies of just a few
examples.  Studies of this type have already been done in the case of particu-
late trace element emissions from several types of major sources.  Coal-fired
power plants are an example since there are great variations in the nature of
                                       39

-------
emissions of a given element from one plant to another,  depending on the type
of coal used, the design of the boiler, the efficiency of pollution control
devices, and stack temperature.  However, detailed studies of a few examples
of coal-fired plants (usually including measurements of the composition of the
coal, various ash fractions, and suspended particles in the stack)  have led to
an understanding of the principles of fractionation of elements within plants
such that with knowledge of the type of coal used and the plant design
parameters one can now make reasonable estimates of emissions from a particular
plant (see Davison et al. 1974; Klein et al. 1975; Kaakinen et al.  1975;
Gladney et al. 1976).

     Furthermore, it has been found that some types of sources that were a
priori thought to be quite variable over time and from one plant to another
were actually not so variable.  For example, Greenberg,  Zoller and Gordon
(1978) and Greenberg et al. (1978) found that the composition of particles
released from various municipal incinerators is quite similar and almost
constant with time.  The ability to treat most sources of a given class in a
similar way allows formulation of reasonable chemical element balances by
lumping sources of a given class (Section 2.4).  Similar approaches may be
successful in characterizing emissions of organic compounds from certain
classes of sources without having to measure emissions on a plant-by-plant
basis.

     EPA sponsored field studies enlisting academic and industrial groups with
different research capabilities have made important contributions to the
understanding of the transport and fate of atmospheric pollutants.  An example
is the VISTTA (Visibility in Sulfur Transport and Transformation Areas) Project
currently sponsored by the Agency.  However, source characterization is
frequently not carried out in conjunction with the field measurements; the
laboratories which specialize in source measurements and field studies tend to
work independently.  When field studies of toxic organic compounds are
sponsored by EPA, a strong effort should be made to include groups capable of
careful source measurements (not routine air monitoring).

     A detailed emissions inventory for atmospheric organics should be prepared
based on available information and include the type of source and classes of
organic emissions when more detailed information on chemical species is not
available.  The industry groups would be evaluated in the following descending
order:

     •  Petrochemical based industry

     •  Metallurgical and inorganic chemical industry

     •  Coal based chemical industry

     •  Wood and cellulose based chemical industry

     •  Fossil fuel combustion

     •  Transportation
                                       40

-------
The emissions inventory should include not only industrial sources but also
chemical waste dumps, pesticides, and other sources.

     A major program should be initiated on detailed source characterization
for the most important sources identified in the inventory.  Priorities should
be set based on toxicity and emission rates (Geomet 1977). Examples of
possible sources are coke ovens and catalytic converters.  Data should be
obtained on chemical speciation for both the gas and aerosol phases.  Aerosol
data should include particle size distributions and distributions of chemical
species with respect to particle size.  Data should be of research quality
using advanced measurement techniques and not of a routine air monitoring
nature.  The data should be of a quality sufficient for source resolution and
mass balance studies.  The source characteristics should be determined over
time using a range of production conditions at a number of facilities to
develop statistical information on the variability of the sources.

     Measurements of source characteristics should be carried out in conjunc-
tion with field studies designed to measure atmospheric transport and trans-
formation.  Such integrated studies will enhance efforts to identify the
complex pathways leading from pollution sources to uptake by humans.

     An intra-Agency task force or committee should be established to provide
direction on the priorities of chemicals and industries to be evaluated by the
EPA.  The counsel of the Science Advisory Board should be sought at an early
stage.  The Interagency Testing Committee's priority list of chemicals may
offer direction though there are other chemicals of special interest to the EPA
regulatory programs.
2.3  ATMOSPHERIC TRANSFORMATION PROCESSES

2.3.1  Atmospheric Lifetimes

     The lifetime of a compound in the atmosphere depends on the compound's
chemical reactivity in both the gas and aerosol phases, interaction with solid
surfaces including vegetation, and solubility in aqueous phases taking chemical
reaction into account.  Section 2.3 discusses in some detail chemical reacti-
vity in the air which also determines the generation of secondary products.
The lifetimes of various organic compounds in air, based on chemical
reactivity, are given in Table 2-1 along with comments on the feasibility of
mass balance and smog chamber studies.  A very long lifetime will lead to
stratospheric sinks with dry deposition and washout becoming important.  A
short residence time such as a primary pollutant will increase the importance
of studying the nature of the reaction products which may be present at high
concentration in an urban and industrial region.

     The toxicity, transformations, and sinks of any secondary products must
also be considered.  The transformation of a nontoxic chemical to a potentially
toxic one is also a possibility.  This transformation could be the case for
polynuclear aromatic compounds, some of which are not initially active.
                                       41

-------
0>
a
6 in
* 
e*

£t| O1 JJ
0 is"
W !
B
H Ij
<
D )
5 <
£tf CO i
cj w
S r^ !
w < |
K tJ
fL <
CO PQ
O i'
S3 W
EH CO 1 „,
y ! *
_ ** u
& en
O

>,
3 3 0) 3

0) (l) -H 0)
tn tO rH tO rH rH rH
33333 3
4J UH U-i U-i
4J 4J tO 4->  > > > fl ft »



QJ

JJ
C
•H a> x,
V4 C C C C
tn tn SH
>iCOOU U O >i Ifl
0 JJ QJ d) 3J CD 0
0 >H 0 rH
X-POO O Ofi -H
fttnEX X XDn 2





N
iH rH U
rH rH C C >
IB flJ 0 0 r-f
0 0 tT> tT1 U rH
rH rH HJ O O -H



\^ 1
H
Pj
B
CM

B
1-1
§
EH








H
pi







(A

1
a
e
0
CJ
(U
0
IO
* u
CM JJ
fa 11
fM
rH •"
CJ CM
U fa
- u
fa U
m fa
r-H CM
U rH

4 *
rH 1C
U fa
U TO
IB
(U
>i





X
CJ

^
m
rH
8
m

U
1
o





m
U
X
u

CJ tj
m CM
X X
U U


ro
rH
U
X
CM
CJ


•CJ
m

(N
U



*.
CO
X

CJ

»,

X

CJ


ix


tn
0)


f^
d>


fd
iH
0
tn
o
3
0
cu

>1

nj
•O
C
O
u
01

0)

rH
J-l
0
JC



rH


H
fi
CO
X
u

U)
CM
X



C

4J
3

rH
0
O^
1
CU

J;


CO
0>
c
0>


a;
JJ
tn
0)

>A
o>
T3

<


CM
0
2

O

42

-------
2.3.2  Species and Processes Important in the Chemical Conversion of Organics

     A. useful approach to classifying compounds in terms of atmospheric
chemical transformations is to consider their likelihood of participating in
the following processes:  (1) photolysis; (2) attack by hydroxyl radicals;
(3) attack by ozone; (4) reaction with other radical intermediates, e.g.,
hydroperoxyl radical (HC>2)/ 0(3p) atoms, etc.; (5) reactions with molecular
species such as NOX, SOX, O2(1Ag), ground state oxygen, etc.; (6) hydrolysis;
(7) thermal decomposition; and (8) aerosol formation and reactions.

     If a compound undergoes photolysis in the actinic UV (290 < A < 410 nm) or
visible region it is important to characterize the resulting products and
intermediates and to establish their individual fates in the atmosphere.  Some
of the key properties which need to be known are the UV absorption spectrum,
the photolysis quantum yields in air, the products formed, and their rates  of
formation and decay (Pitts 1978).

     It has been well established in the last five years that the OH radical is
a key reactive intermediate in both the lower and upper atmospheres.  For
example, the dominant loss process for alkanes and alkenes as well as for many
other organics in the polluted troposphere is attack by OH radicals (Demerjian
et al. 1974; Falls and Seinfeld 1978; Atkinson et al. 1979).  The rate of
reaction of the chemical substances with OH radicals is critical for any study
of atmospheric reactions.  In a number of cases this reaction rate can be
determined either experimentally or by comparison to similar compounds or both.
Furthermore, it has been shown that reaction with OH can, in some cases, be
indicative of the reactivity with respect to oxidant formation of a given
compound and can be used as a means of classifying the reactivity of chemicals
(Darnall et al. 1976; Pitts et al. 1976).

     Reactions with ozone will be of major importance for unsaturated com-
pounds.  Unsaturated compounds with elevated levels of ozone can be competitive
with or even exceed in importance reactions with OH (National Academy of
Sciences 1977a, 1977b).  Elucidation of the products of ozone-alkene reactions
remains an important research area.

     In general, organic reactions with HO2 and with O(3p) at the ambient
concentrations of these species will not be important relative to reactions
with OH or ozone.   Although present in photochemical smog at atmospheric
concentrations of approximately 10  radicals cm"3, HO2 appears to react slowly
with most organics  (Graham et al. 1979).  Conversely, while rates of reaction
with O( P) are fast (Hampson and Garvin  1977), its atmospheric concentrations
are so low as to diminish its importance.

     The radicals formed by attack by OH and other species are known to react
subsequently with oxides of nitrogen and sulfur as well as with ground state
molecular oxygen in many cases to form the ultimate products of photooxidation
in the atmosphere.  Thermal reactions with molecular species can also occur.

     For chemicals  and intermediates which may undergo hydrolysis it is
necessary to determine rates and products at various humidities, both with and
                                       43

-------
without aerosols, which may catalyze the reactions.  In certain cases,  thermal
decomposition may be a significant factor in determining the stability  and
lifetime of a chemical in the atmosphere.

2.3.3  Predictive Modeling for Reactive Species

     A methodology has been developed for the prediction of air quality from
emission source data which accounts for transport (the wind field)  and  atmos-
pheric conversion processes.  The models have been developed with particular
reference to photochemical smog and the prediction of ozone concentrations
resulting from emissions of organics and nitrogen oxides under given meteorolo-
gical conditions.  Extensive reviews and comparisons of such models have been
given in Ozone and Other Photochemical Oxidants (National Academy of Sciences
1977a) and Air Quality Criteria for Ozone and Other Photochemical Oxidants
(U. S. Environmental Protection Agency 1978) and will not be repeated here.
The results of such models should be of direct applicability to predicting
reaction rates for organic compounds of interest.  This applicability will be
true for models which accurately predict the concentrations of reactive species
such as OH radicals and ozone.

2.3.4  Hierarchy of Evaluation

     The levels of evaluation of the fate of chemical substances in the atmo-
sphere should be a function of usage, presumed toxicity, and emission factors.
At least three degrees of complexity, involving the processes previously
mentioned, can be suggested.

     The first level consists of an initial screening procedure which would be
based on a comprehensive assessment of the existing literature concerning the
physical, chemical, and photolytic properties of a given compound.  Available
information such as the vapor pressure, structural groups, and absorption
spectrum can provide a basis for estimating the likely paths and rates  of
transformation processes to determine the fate of an organic substance.  Where
such information is lacking it may be necessary to conduct fundamental
laboratory studies to derive these basic properties.

     A second level of evaluation would involve relatively straightforward and
inexpensive experimental studies.  An example would be the photolysis of a
compound (including photooxidation), thermal reactions in the presence  of
oxygen, and (if any) hydrolysis reactions.  At this level, simple irradiation
experiments in small Teflon bags using typical ambient air pollutant concentra-
tions could be conducted to obtain initial data concerning the nitrogen oxides
(NOX) photooxidation reactions of the chemical species.  In such experiments,
preliminary data can be obtained concerning both gas phase and aerosol  phase
products formed in such reactions.  The rates of reaction and lifetimes may be
determined by following reactant decay and mass balance.  If the rate is very
slow further chamber experiments would not be useful.  Typical analytical
tools which would be employed in such studies include:  gas chromatography; wet
chemical analysis; continuous gas phase analyzers  for species such as ozone,
NO, and NOX/ aerosol analyzers such as condensation nuclei counters; perhaps
combined gas chromatography-mass spectrometry; infrared spectroscopy; thin
                                       44

-------
layer chromatography; high pressure liquid chromatography;  and analysis of the
filtered reaction mixture.  From these experiments basic reactivity and pro-
ducts estimates from the initial screening process may be verified and the need
for further analysis determined.

     If, on the basis of high usage or indications of substantial environmental
impact, it is necessary to carry out complete and detailed assessments of the
atmospheric transformations and fates of given chemical species having a signi-
ficantly short lifetimes, an extensive program employing more sophisticated
environmental chambers and other techniques such as flow reactors should be
used.  Analytical techniques such as long path FT-IR spectroscopy for the
detection of labile compounds may be necessary.

     When experimental evaluations of atmospheric transformations are required,
it may be cost-effective to develop collaborative efforts between industry,
university laboratories, and research institutes as well as governmental
agencies to avoid unnecessary duplication of expensive experimental facili-
ties.

2.3.5  Atmospheric Transformation Processes Recommendations

     Based on hierarchy of evaluation, specific recommendations for the trans-
formation of substances in air includes the following:

     •  Research should continue into tabulation and evaluation of rate
        constants for OH and 03 reactions, evaluation of photolysis rates, and
        development of estimation schemes for atmospheric lifetimes of toxic
        substances.  Additional experimental studies may be needed for key
        reactions or substances to validate estimation schemes.

     •  The elementary chemical kinetics of toxic substances or toxic substance
        precursors should be studied in detail.  For substances which react
        rapidly in the atmosphere, survey type chamber studies should be
        undertaken and products identified.

     •  Protocols should be developed for testing potentially toxic substances
        in environmental chambers varying parameters such as HC/MOX ratios,
        hydrocarbon composition, and presence of aerosols.  Attention should be
        paid to mass balances in such experiments.

     •  Using literature and experimental data, predictive schemes for evaluat-
        ing toxic products of atmospheric reactions based on a general
        atmospheric model should be developed.  Existing models which predict
        gas and aerosol phase products of hydrocarbons with photochemical
        oxidants should be extended to new agents of interest.

     •  There is a need for kinetic measurements and chamber studies for the
        chemistry of aromatic compounds in the atmosphere because of the
        increasing aromatic content of unleaded fuel and the potential toxicity
        of the products.  Ozone-olefin reactions to form toxic epoxides should
        also be studied.
                                       45

-------
        Substances which react slowly in the atmosphere or in some cases those
        which have fast homogeneous loss pathways may also be removed from the
        atmosphere through interaction with aerosol particles, vegetation
        surfaces, rain-out, washout, and dry deposition.  Properties such as
        vapor presssure, solubility, and gas-surface interaction rates should
        be tabulated.
2.4  SOURCE RESOLUTION:  CHEMICAL SPECIES BALANCES

     Much progress has been made on the determination of the sources of sus-
pended particulate matter in urban air with the chemical element balance (CEB)
method.  To use the CEB method one must know the elemental composition (for 20
or more elements) of particles released from all important sources in the area
under study as well as the composition of ambient aerosols.  The basic assump-
tion of CEB is that the concentration of element ± in a particulate sample is
given by:

                                  Ci = EmXi                      (1 )
where nij is the mass of material in the sample originating from source
j and 3Cji is the concentration of element i_ in material released by source
j (Friedlander 1973).  One usually performs a least-squares fit to the observed
concentrations of a subset of the elements equal to or greater than the number
of sources used in the fit to calculate the source strengths, mj from a set of
concentrations, C_^.

     For example, Kowalczyk et al. (1978) did a fit to eight elements (Na, Pb,
V, Zn, Al, Fe, Mn and As) to determine the strengths of six sources (sea salt,
motor vehicles, coal- and oil-fired plants, soil and municipal incinerators in
the Washington, D.C., area.  Ideally, elements used in the fit should
originate mainly from one dominant source (Na, Pb, V and Zn from sea salt,
motor vehicles, oil, and refuse combustion, respectively).  However, soil and
coal fly ash have such similar compositions that Al and Fe indicate the sum of
the two, Mn is sensitive to soil, but depleted in coal emissions, and As is
enriched in coal emissions.

     The predicted concentrations of the measured elements not used in the fit-
ting are compared with the observed values to test the reliability of the fit.
Concentrations of many test elements were severely underpredicted in several
early attempts to apply CEB (Friedlander 1973; Miller et al. 1972; Gatz
1975; Gartrell and Friedlander 1975) mainly because of limited information
on the compositions of particles from important sources.  However, in a recent
CEB study of Washington, D.C., Kowalczyk et al. (1978) predicted 15 elements
on the average to within a factor of two with most of the errors being under-
predictions.  Although this represents a rather good fit, work continues in
various laboratories to eliminate some remaining weaknesses of the method.
This additional work includes compositions of particles from additional sources
that are important in some areas being determined (e.g., cement plants, steel
                                       46

-------
mills, non-ferrous smelters), and separate CEB's being performed on small
and large particles to allow for different residence times of different sized
particles from a given source.

     The CEB method works best for the primary aerosol components.   For the
secondary components such as NH4+, SO^2~ and NO3~,  it is necessary to have a
gas-to-particle conversion model.  Secondary components can also be treated
using the statistical approaches of multivariate analysis, including factor and
cluster analysis (Hopke et al. 1976; Henry 1977; Gatz 1978; Hopke 1977).
For example, factor analysis makes no a priori assumptions about the number or
composition of sources, but finds a minimum number  of common factors that will
explain most of the variation of concentrations from one sample to another in a
large data set.

     Although not totally complete, with the improvements and testing now in
progress CEB and multivariate methods should soon be able to determine the
magnitude of contributions of various sources to TSP and even the sources of
most non-volatile elements on aerosols.  There is less optimism about the
ability of these methods to handle elements such as Hg, the halogens and, to a
lesser extent, Se, which are so volatile that they may exchange with the gas
phase depending on temperature, concentration, and the presence of reactive
chemical species.

     In view of the success of chemical element balances, an investigation of
the applicability of a more general approach to chemical species balances (CSB)
is recommended in which all chemical species (molecular and ionic species,
elements, and both gas- and particulate-phase species) are used to establish
relationships between sources and atmospheric concentrations.  It is not a
priori obvious that this approach is universally applicable to molecular
species.  The CEB approach has the advantage that elements are consevative;
even though they may be transformed from one chemical form to another or from
the gas phase to the particulate phase, one element does not transform into a
different element.  By contrast, if one attempts to apply CSB to a series of
organic molecules released by various sources, one may find that the original
concentration pattern is not conserved because some of the species may be
transformed more readily than others into new species (Section 2.3).
Nevertheless, the potential usefulness of CSB warrants an investigation of the
possibilities.

     The most obvious application of CSB to organic species would be to a
series of closely related compounds.  To be of value, the concentration pattern
of species emitted should be strongly characteristic of each important source
with large differences between types of sources.  In order that the pattern be
preserved between the time of release and the subsequent collection of samples,
the various members should have comparable rates for major sink reactions, such
as photolysis or attack by OH radicals.  It would also be desirable for the
volatilities to be such that all members of the series are mainly either the
gas phase or the particulate phase.

     Little has been done on the application of CSB to organic species but
some classes of compounds appear promising, including the series of polynuclear
                                       47

-------
aromatic hydrocarbons (PAHs).  Although the PAH compounds differ considerably
in vapor pressure, under ambient conditions most of them are probably in the
particulate phase (though this may not be true at high stack temperatures).
Futhermore, the pattern of PAH emissions from different types of sources is
quite different, offering the possibility of source identification from the
observed patterns (Gether and Seip 1979; Kites et al.  1977).  Preliminary
studies have, in fact, indicated that PAH patterns can be used for source
identifications.

     Other classes of compounds might be used in CSB.   Mayrsohn and Crabtree
(1976) have shown that the series of G-J to €5 hydrocarbons can be used to
identify sources in the Los Angeles basin.  In this case the compounds are in
the gas phase.  Other series which might be useful in  CSB applications include
organic halogen compounds, organic acids, and terpenes from natural sources,
although the latter group may be too reactive over long distances from sources
(Finlayson and Pitts 1976).

     Another identification technique that may be helpful in identifying
sources of carbonaceous species is the ratio of   C/  C (or /total C).  Carbon
derived from fossil fuels is devoid of   C, whereas that from live or recently
dead biological sources has a  ^C/12C ratio in equilibrium with the atmosphere.
The recently developed accelerator methods of 14C/  C  measurements are much
more sensitive than traditional   C 8 counting, making possible sensitive
measurements with small sample sizes (Currie and Klouda 1979).

     To test CSB for classes of organic compounds it will be necessary to make
accurate measurements of the patterns in emissions from major sources of each
class (Section 2.2).  Care should be taken, especially in the case of sources
with high temperature stacks, to insure complete collection of each species,
including both gas and particulate phases.  Furthermore, in testing the method
in ambient air, it will be useful to measure not only the class of compounds of
interest, but also trace elements characteristic of the sources (especially
those found on small particles with long residence times).  Since the major
sources of many elements have been established, it will be possible to check
some aspects of the CSB.

2.4.1  Sources Resolution Recommendations

     CEB can be used to make reliable estimates of the sources of many toxic
elements if they are not too volatile.  Further refinements and application of
the approach should be made.  Improved methods for collection and measurement
of volatile elements such as Hg, halogens, and Se are needed.  When ambient
samples are analyzed with a view towards CEB analysis, measurements should in-
clude Al and/or Ni, Mn and As as a minimum, or elements that are established as
surrogates for these elements.  Desirable additional elements for measurement
include Fe, Ca, and Cd.

     Some success has been achieved in the application of CSB to the identifi-
cation of sources of organic species, but the method should be further tested
on various classes of closely related compounds.  When using CSB it is
essential that measurements of hot sources include both gas and particulate
                                       48

-------
phase emissions.  As an additional test of this method it is recommended that
simultaneous measurements of the elements noted using CEB be made since the
sources of those elements are well established.  Investigation should continue
into the use of  '^C/'^C (or total carbon) ratios of carbonaceous materials to
determine the ultimate source of carbon (i.e., fossil fuel or recent living
material).

2.5  AIR QUALITY CHARACTERIZATION

     Air quality characterization for atmospheric organic compounds requires
the identification of such chemicals in the atmosphere and the separate
measurement of their concentration in the gas and aerosol phases.  The volati-
lity and reactivity of certain organic compounds can make this task difficult.
For example, pure benzo[a]pyrene (BaP), a compound of relatively high vapor
pressure (Pupp et al. 1974) would be expected to be present in the gas phase
at concentrations observed in the Los Angeles atmosphere.  However, measure-
ments confirm that BaP is present in the aerosol phase (Miguel and Friedlander
1978) probably as a result of adsorption on primary aerosol components.
Compounds such as BaP when collected on a filter may also react with species in
the gas phase as they are drawn through the filter.

     Aerosol characterization requires information not only on the total con-
centration of the chemical species of interest but also its distribution with
respect to particle size.

     A careful characterization of air quality is necessary in studies of
transport and fate for the following purposes:

     •  The estimation of atmospheric reaction rates (Section 2.3);

     •  The estimation of atmospheric scavenging by incloud processes
        and rain;

     •  The estimation of deposition rates on ground and vegetation;

     •  The completion of mass balances for various chemical
        species (Section 2.6);

     The development of methods of measuring low concentrations of organic
compounds in the gas and aerosol phases is an important part of a program for
tracking the flow of toxic substances through the environment.  Decisions on
instrument development by EPA should be based on priorities set for specific
classes of pollutants.  Once specific agents have been chosen in the gas and
aerosol phases, it will be easier to select the most appropriate types of
measurement systems.

2.5.1  Air Quality Characterization Recommendations

     Although many organic compounds have been detected and measured in the
atmosphere, there is no central repository for the data.  An inventory of data
on atmospheric organic, organometallic, and volatile inorganic compounds should
                                       49

-------
be established and include the name and chemical formula of the compound,
concentration ranges, measurement methods, physical states (gas and/or aerosol),
and measurement sites.  Literature references should be cited and some measure
of the quality of the data should be given.  Because of the large number of can-
didate compounds and measurements, priorities should be set based on toxicity
and emission rates (Geomet 1977).  Where useful, compounds should be classed
in family groups such as polynuclear aromatic compounds and halogenated
organics.  A limited inventory should initially be set up to gain experience.

     Data for the inventory should be assembled from the literature and by
surveying laboratories engaged in atmospheric chemical research.  The informa-
tion should be stored in an appropriate computer data bank from which the data
could be retrieved and displayed in a number of ways (e.g., compound class,
physical state, geographic location, temporal trends).  This data bank would be
maintained by EPA but would be available to the general scientific community.
As new compounds are discovered and toxicity data developed, they would be added
to the inventory.

     Based on the inventory and toxicity information, a group of compounds
should be selected for an atmospheric measurement program.  Both gases and
aerosols (in "respirable" and "non-respirable" fractions) would be sampled and
analyzed.  Measurement sites would be established at 5 to 10 locations around
the country.  Each site should be completely equipped with both sampling and
analytical instrumentation (GC/MS, liquid chromatography, etc.,) and all
chemical analyses would be done on-site.  Sites would be chosen to represent
different types of source impacts.  If the sites are well chosen, the
information obtained at this small number of sites will be applicable to other
regions.

     It would be desirable to have such sites located at academic institutions
to maximize the research capability and the application of the results to air
quality modeling (Section 2.3), source resolution (Section 2.4), and mass
balances (Section 2.6).  Because it is likely that the analytical system will
contain a mass spectrometer with a computer data system, information related to
compounds not on the survey list (or unidentified mass peaks) could be stored
for retrospective study should the need arise.  These data would be useful for a
variety of purposes including epidemiological studies and chemical species
balance research (Section 2.6).

     United States Environmental Protection Agency should continue to support
research on the development of new, highly sensitive methods for the detection
and measurement of organic gases in the atmosphere.  An essential part of such a
program should be field applications of the new methods.  The application of
high resolution mass spectrometry and gas chromatography/mass spectrometry to
atmospheric analysis has provided ample evidence for the merits of such
exploratory studies.

     While many methods of chemical sensing are available, priority in the
development of instrumentation will be made by  selecting the chemical agents of
most concern to the toxic chemicals program.
                                        50

-------
     There is evidence that the organic component of the aerosol contains car-
cinogenic constituents.  Methods for the sampling and analysis of aerosol
organics are primitive - even more so than for the inorganic components.
Filters are usually used for sampling and the filtration process may result in
changes in the chemical nature of the deposited material.  Research should also
be supported on the development of continuous, real-time methods of aerosol
analysis for organics to avoid sample storage and delays in analysis.

     More data are needed on the distribution of organic compounds with respect
to particle size, particularly in the submicron size range.  Such data are
required for estimations of lung deposition and atmospheric loss rates.

     Further studies of the biological effects of collected atmospheric samples
should be made.  Protocols and procedures should be developed for sample
collection to maintain the activity of the sample of its atmospheric level.
2.6  MASS BALANCES

     Mass balances between inputs to the atmosphere and outputs to other envi-
ronmental compartments can be used to quantify the flows of pollutants through
the environment.  Such balances can be developed for geographic scales ranging
from the urban to the global.  The basic requirement is that the emissions
to the atmosphere be balanced by flows out of the atmosphere.  If a balance
cannot be completed, the possibility of either an unknown source or an unknown
environmental pathway must be considered.

     Mathematically, the mass balance can be expressed as a series of simul-
taneous equations

                               oc    a    a      a
                              C=E+D+ZT                      (1)
                               i    i    i   k  ki

Where:  C^ is the mass rate of accumulation of chemical a in environmental
        compartment i;

         a      a
        E  = Z E   is the sum of emission of a from all
         i   j  iJ
        sources j into compartment i;

         a       (prod)    (dest)
        D  = Z (D       - D      )a is the sum over all
         i   r   ir        ir
        chemical reactions r which either produce or
        destroy a in compartment i;

         a     a      a
        T   = T    - T    is the rate of mass transfer of a between
         ki    k+i    i>k
        compartments i and k.
                                       51

-------
     The condition of mass balance requires that:
                             a       a    a      a
                          Z C  = Z (E  + D  + E T  )                   (2)
                             i   i   i    i   k  ki

In the absence of chemical reactions this reduces to:

                                     a       a
                                  £ C  = £  E                         (3)
                                  i  i   i  i

     The types of information which are required for the construction of a mass
balance are given in Table 2-2.  Most mass balances appearing in the literature
have used a combination of measured parameters or estimates based on such
theoretical considerations as the relationship between particle size and depo-
sition rate.
  TABLE 2-2.  TYPES OF INFORMATION NECESSARY FOR CONSTRUCTING A MASS BALANCE
     Vapor pressure

     Aqueous solubility

     Identity of sources

     Mass emission rate for each source

     If aerosol, source particle size distribution

     Ambient atmospheric concentration

     If aerosol, ambient particle size distribution

     Dry deposition rate

     Rain-washout rate

     Chemical reactivity in the atmosphere, soil, and water system

     Sorption by particulate matter

     Atmospheric transport rate from the region of interest
     Once an acceptable mass balance is determined for a given chemical, it can
be used as a model  for predicting the environmental flows of chemicals with
similar emission characteristics and chemical and physical properties.  Such an
                                       52

-------
analogy was used to predict the environmental flows of Mn if it were sub-
stituted for Pb in gasoline (Huntzicker 1975).

     The geographic scale over which an atmospheric mass balance is feasible
depends on the atmospheric lifetime of the pollutant (Table 2-1).   For a highly
reactive or easily removed pollutant, a mass balance would be possible only in
a region localized around the source.  For a substance with a lifetime of the
order of days to weeks urban and regional scales can be used, and  for
pollutants with very long lifetimes global scales can be considered.  In
general, the atmospheric lifetime specifies the maximum extent of  the geogra-
phic scale with all smaller scales also being feasible.

     A partial list of mass balances which have been carried out for chemicals
in the atmosphere is given in Table 2-3.  The list can be divided  into trace
elements and organic compounds.  Because the inherent identities of the trace
elements are unaffected by chemical reactions, mass balances for these species
are somewhat easier to construct than for organic compounds which  can undergo a
complex series of chemical reactions in all environmental compartments.
           TABLE 2-3.  MASS BALANCES FOR CHEMICALS IN THE ATMOSPHERE
SUBSTANCE                          SCALE                         REFERENCE

   DDT                             global             Woodwell et al. 1971

   DDT                             global             Cramer 1973

   Hg                              global             Xothny 1973

   PCB                             regional           McClure 1976

   Pb                              regional           Getz et al. 1977

   Pb, Zn, Cd                      regional           Van Hook et al. 1977

   Pb                              urban              Huntzicker et al. 1975

   Zn, Cd                          urban              Huntzicker and Davidson
                                                      1975

   Benzo(a)pyrene                  urban              Abrott et al. 1978
   An example of a trace element mass balance is given in Figure 2-2 in which
the flow of automobile emitted Pb is traced through the Los Angeles basin
(Huntzicker et al. 1975).  The input data were obtained from literature values
for gasoline sales in California, the concentration of Pb in Southern
California, the concentration of Pb in Southern California gasoline, and the
                                       53

-------
             INPUT
            23.7 ± 2.4
                                                     REMOVED BY
                                                        WIND
EVAPORATION
    ~0.3
                          VAPOR
                            0.9
                         AEROSOL
                           16.7
              ATMOSPHERE
                17.9± 2.6
  RETAINED IN CAR
      5.8 ± 1.1
    (1.4)
     9.5 ± 2.2
                   NEAR DEPOSITION
                   STREET
  REMOVED BY
    STREET
   CLEANING
                (RAINY)
                 (DRY)
                 0.64
  SEWAGE
     LAND
      (7.7)
                    2.0 ± 1.0
                    (DRY)
    FAR
DEPOSITION
  (LAND)
RUNOFF
  (0.4)
  0.03
                                                       VAPOR
                                                         0.3
                                                      AEROSOL
                                                       5.3 ± 3.0
                   0.3
                   (DRY
               DEPOSITION)
     COASTAL WATERS
                                                          0.09
                                                         RAIN
FIGURE 2-2.   THE  FLOW OF AUTOMOBILE EMITTED LEAD  THROUGH THE LOS
     ANGELES BASIN.   (With the exception of far deposition and wind
     removal fluxes, which are for dry weather only,  the values are
     daily averages calculated by dividing the yearly totals by 365.
     All fluxes are in metric tons/day.  Numbers  in parentheses are
     model-dependent and in need of a more extensive  data base.
     From Huntzicker et al. 1975a).
                               54

-------
percent of consumed Pb exhausted to the atmosphere.   Deposition near roadways
was estimated from consideration of the known particle size distribution of
auto exhaust Pb, and "far" deposition (away from sources)  was measured.   The
amount of of Pb blown out of the basin was estimated by assuming the basin to
be a continuously stirred flow reactor and using the relationship:
                               qPb = qco [pb/CO]                         (4)
where qj_ is the mass flow rate of species i out of the basin and [Pb/CO]  is the
average value of the ratio of Pb to CO at receptor sites in the basin.   Because
                                                       in    out
CO is essentially unreactive on the urban time scale, q   = q   ,  and q  /[CO]
                                                       CO    CO        CO
is a measure of the flow of air out of the basin.  With this procedure it was
estimated that 18±3 metric tons/day of Pb were emitted into the atmosphere and
that the environmental pathways of near source deposition, far deposition, and
removal from the basin by wind accounted for 17±4 tons/day.  Because of the
close agreement between input and output it is likely that all major sources
and environmental pathways were included.  The analysis also showed that the
contribution of Pb from auto exhaust to the coastal waters was comparable to
that from sewage.

     A similar procedure was applied to Zn and Cd in the Los Angeles atmosphere
(Huntzicker and Davidson 1975).  More Cd was found in roadway deposition, far
deposition, and removal by wind than could be accounted for by known sources.
This suggested an unknown Cd source.

     The situation with respect to toxic organic chemicals is considerably more
complex than for most trace metals for the following reasons:

   •  Both photochemical and thermal reactions can transform the parent com-
      pound into more or less toxic daughter compounds (Pitts et al. 1978;
      Cliath and Spencer 1972).

   •  Significant reintroduction into the atmosphere of deposited compounds
      with low vapor pressure and low aqueous solubility can occur (Mackay
      and Wolkoff 1973; SpenCer and Cliath 1975; Glotfelty and Caro 1975).

   «  Many organic compounds will exist in both the gaseous and particulate
      phases in the atmosphere.

   •  In many cases adequate analytical methodologies have not been developed.

   •  Adequate source information is often not available.

     Because of these difficulties mass balances are available for only a few
organic compounds (Table 2-2).  Woodwell et al. (1971) and Cramer (1973)
considered the global circulation of DDT, but because of the large number of
assumptions involved in their models, some disagreement as to the importance of
various environmental pathways occurred.  One point which did emerge, however,
                                       55

-------
is that the biological importance of a pathway can be much greater than
suggested simply by the fraction of the total mass which flows through that
pathway.  Pathways carrying a small fraction of the mass can be very important
from a biological standpoint and should not be overlooked.

     Relatively little work has been done on the regional and urban flows of
organic compounds.  In a study of BaP in Los Angeles, good agreement was found
between estimated BaP emissions and the amount of airborne BaP calculated from
measured BaP/CO ratios during the winter and spring (Abrott et al. 1978).
During the summer smog season much less than the expected amount of ambient BaP
was found suggesting a chemical transformation of BaP with a half-life of
about a day.  McClure (1976) has studied the flow of atmospheric PCB's in
Southern California by combining atmospheric concentrations and deposition
measurements with a Gaussian puff dispersion model.  .An actual mass balance,
however, was not possible because the source strength was not a priori known
but was treated as a variable parameter.

2.6.1  Mass Balances Recommendations

     The environmental flows of model compounds in both the gas and aerosol
phases should be measured.  Where appropriate, the environmental flows of other
toxic compounds may be determined by analogy.  Emphasis should be on organic
compounds which comprise the bulk of known toxic substances.

     Basic research in the following "interfacial processes" is needed,
including:  aerosol deposition processes; deposition of organic vapors;
volatilization of organic compounds from realistic soil and water systems; and
atmospheric detection of the volatilized compounds.

     Studies should be conducted of the partitioning of organic and other vola-
tile compounds between the aerosol and gas phases with specific attention to
the role of sorption.

     A complete mass balance for an organic compound for which substantial in-
formation is already available (BaP) should be worked up.  The mass balance
should take into account the effects of chemical reactions and the environ-
mental flows of the reaction products.

     Prototype mass balance studies of other selected organic compounds should
be conducted.  These would include compounds primarily in the vapor phase,
compounds primarily in the aerosol phase, and compounds existing  in both
phases.  Such mass balances should involve the determination of an emission
inventory in the region of interest and measured mass flows of the compounds
through the various environmental pathways.  Such  studies would follow from a
mass balance for an organic compound with substantial available information and
would provide a test of the applicability of the mass balance method to organic
compounds.

             The Air Group participants are  listed in Table 2-4.
                                        56

-------
            TABLE 2-4.  ATMOSPHERIC ENVIRONMENT GROUP PARTICIPANTS
Non-Agency Participants

Sheldon K. Friedlander (Chairman)
Department of Chemical Engineering
California Institute of Technology
Pasadena, CA

Glen F. Gordon
Department of Chemistry
University of Maryland
College Park, MD

John Herron
National Bureau of Standards
Gaithersburg, MD

J. Huntzicker
Oregon Graduate Center
Beaverton, OR

Casimer Jackimonski
NASA Headquarters
Washington, D.C.

Ben Mason
Geomet Corporation
Gaithersburg, MD

James N. Pitts, Jr.
Department of Chemistry
University of California
Riverside, CA

Hanwat Singh
SRI International
Menlo Park, CA

W. Zielinski
Office of Environmental
  Measurements
National Bureau of Standards
Washington, D.C.
 EPA Representatives

 Joseph Breen
 Office of Program Integration and
   Information/
 Office of Toxic Substances
 Washington, D.C.

 Marcia Dodge (Lead)
 Environmental Sciences Research
   Laboratory
 Research Triangle Park, NC

 D. Golomb
 Office of Energy, Minerals and
 Industry/
 Office of Research and Development
 Washington, D.C.

 Richard Johnson
 Office of Air Quality and Planning
   Standards/
 Office of Air, Noise, and Radiation
 Research Triangle Park, NC

Robert Kellam
 Office of Air Quality and Planning
   Standards/
 Office of Air, Noise, and Radiation
 Research Triangle Park, NC

 Stanley Kopcznski
 Environmental Sciences Research
   Laboratory/
 Office of Research and Development
 Research Triangle Park, NC

 Edward Schuck
 Office of Research and Development
 Washington, D.C.

 Bruce Turner
 Environmental Sciences Research
   Laboratory/
 Office of Research and Development
 Research Triangle Park, NC
                                      57

-------
                                  REFERENCES
Abrott, T. J., M. J. Barcelona, w. H. White, S. K.  Friedlander and
     J. J. Morgan.  Human Dosage/Emission Source Relationships for
     3enzo[a]pyrene and Chloroform in the Los Angeles Basin.  Special
     Report to the USEPA.  Keck Laboratories, Caltech., 1978.

Atkinson, R., K. R. Darnall, A. C. Lloyd, A. M. Winer and J. N.  Pitts, Jr.
     Kinetics and Mechanisms of the Reaction of the Hydroxyl Radical with
     Organic Compounds in the Gas Phase.  Adv. Photochem., 11:374, 1979.

Cliath, M. M. and M. M. Spencer.  Dissipation of Pesticides by Volatilization
     of Degradation Products, I. Lindane and DDT.  Env. Sci. Technol., 6:910,
     1972.

Cramer, J.  Model of the Circulation of DDT on Earth.  Atmos.  Environ.,
     7:214, 1973.

Crosby, D. G.  The Environmental Photochemistry of Toxic Substances.  Paper
     Presented at the 176th National Meeting of the American Chemistry
     Society, Miami Beach, Fla., 1978.

Currie, L. A. and G. A. Klouda.  Advances in the Discrimination of Natural
     From Anthropogenic Carbonaceous Pollutants via Isotopes of Carbon.  Paper
     Presented at the American Chemical Society National Meeting, Hawaii,
     1979.

Darnall, K. R., A. C. Lloyd, A. M. Winer and J. N. Pitts, Jr.   Reactivity Scale
     for Atmospheric Hydrocarbons Based on Reaction with Hydroxyl Radical.
     Environ. Sci. Technol., 10:696,  1976.

Davison, R. L., D. F. S. Natusch, J. R. Wallace and C. A. Evans, Jr.  Trace
     Elements in Fly Ash — Dependance of Concentration on Particle Size.
     Advan. Environ. Sci. Technol., 8:1107,  1974.

Demerjian, K. L., J. A. Kerr and J. G. Calvert.  The Mechanisms of Photo-
     chemical Smog Formation.  Adv. Environ. Sci. Technol., 4:1-262,  1974.

Falls, A. H. and J. H. Seinfield.  Continued Development of a Kinetic Mechanism
     for Photochemical Smog.  Environ. Sci. Technol.,  12:1398, 1978.

Finlayson, B. J. and J. N. Pitts, Jr.  Photochemistry  of the Polluted
     Troposphere.  Science,  192:111,  1976.
                                      58

-------
Friedlander, S. K.  Chemical Element Balances and Identification of Air
     Pollution Sources.  Environ. Sci. Technol., 7:235, 1973.

Gartrell, G., Jr. and S. K. Friedlander.  Relating Particulate Pollution to
     Sources: The 1972 California Aerosol Characterization Study.  Atmos.
     Environ., 9:279, 1975.

Gatz, D. F.  Relative Contributions of Different Sources of Urban Aerosols:
     Application of a New Estimation Method to Multiple Sites in Chicago.
     Atmos. Environ., 9:1, 1975.

Gatz, D. F.  Identification of Aerosol Sources in St. Louis Area Using
     Factor Analysis.  In: Proceedings of the Conference on Environmental
     Sensing, New Orleans, LA., November, 1977, PP. 443-445.  American Chemical
     Society, Washington, D. C., 1978.

Geomet.  Assessment of the Contributions of Environmental Carcinogens to Cancer
     Incidence in the General Population.  Report to the USEPA under Contract
     No. 68-03-2504, 1977.

Gether, T. and H. M. Seip.  Analysis of Air Pollution Data by the Combined
     Interactive Graphic Presentation and a Clustering Technique.  Atmos.
     Environ., 13:87, 1979.

Getz, L. L., A. W. Haney, R. W. Larimore, J. W. McNurney, H. V. Leland,
     P. W. Price, G. L. Rolfe, R. L. Wortman, J. L. Hudson, R. L. Solomon
     and K. A. Reinbold.  Transport and Distribution in a Watershed Ecosystem.
     In: Lead in the Environment, W. R. Boggess, ed.  NSF/RA - 770214, 1977.

Gladney, E. S., J. A. Small, G. E. Gordon and W. H. Zoller.  Composition and
     Size Distribution of In-Stack Particulate Material at a Coal-Fired Power
     Plant.  Atmos. Environ., 10:1071, 1976.

Glotfelty, D. E. and J. H. Caro.  Introduction, Transport and Fate of
     Persistent Pesticides in the Atmosphere.  In:  Removal of Trace
     Contaminants from the Air, V. R. Dietz, ed., pp. 42-62.  American Chemical
     Society Symposium Series, 17.  American Chemical Society, Washington,
     D.C., 1975.

Graham, R. A., A. M. Winer, R. Atkinson and J. N. Pitts, Jr.  Rate Constants
     for the Reaction of HC>2 with HC>2 SC>2/ CO, N2O, Trans-2-Butene and
     2,3-Dimethyl-2-Butene at 300K.  J. Phys. Chem., 83:1563, 1979.

Greenberg, R. R., G. E. Gordon, W. H. Zoller, R. B. Jacko, D. W. Neuendorf
     and K. J. Yost.  Composition of Particles Emitted from the Nicosia
     Municipal Incinerator.  Environ.  Sci. Technol., 12:1329, 1978.

Greenberg, R. R., W. H. Zoller and G. E. Gordon.  Composition and Size
     Distributions of Particles Released in Refuse Incineration.  Environ. Sci.
     Technol., 12:566, 1978.
                                      59

-------
Hampson, R. F. and D. Garvin.  Reaction Rate and Photochemical Data for
     Atmospheric Chemistry.  NBS Special Publication 513,  1977.

Henry, R. C.  The Application of Factor Analysis to Urban  Aerosol Source
     Identification.  Paper Presented at the 5th Conference on Probability
     and Statistics, Las Vegas, Nevada, 1977.

Kites, R. A., R. E. LaFlamme and J. W. Farrington.  Sedimentary Polycyclic
     Aromatic Hydrocarbons:  The Historical Record.  Science,  198:829,  1977.

Hopke, P. K.  The Use of Multivariate Statistical Techniques to Interpret
     Multielement Environmental Analysis.  In:  Proceedings of the Third
     International Conference on Nuclear Methods in Environment and Energy
     Research, J. R. Vogt, ed., pp. 49-57. TID Document No. CONF-771072, 1977.

Hopke, P. K., E. S. Gladney, G. E. Gordon, w. H. Zoller and A. G. Jones.
     The Use of Multivariate Analysis to Identify Sources  of Selected Elements
     in the Boston Urban Aerosol.  Atmos.  Environ., 10:1015,  1976.

Huntzicker, J. J.  Environmental Forecasting: The Projected Flow of Automobile
     Emitted Manganese Through the Los Angeles Basin.  In: Air-Water-Land
     Relationships for Selected Pollutants in Southern California,
     J. J. Huntzicker, S. K. Friedlander and C. I. Davidson, eds.  Final
     Report to the Rockefeller Foundation (Grant No. RF-72049), 1975.

Huntzicker, J. J. and C. I. Davidson.  Atmospheric Trace Metal Flows in the
     Los Angeles Basin: Zn, Cd, Ni.  In: Air-Water-Land Relationships for
     Selected Pollutants in Southern California, J. J. Huntzicker, S. K.
     Friedlander and C. I. Davidson, eds.  Final Report to the Rockefeller
     Foundation  (Grant No. RF-72049), 1975.

Huntzicker, J. J., S. K. Friedlander and C. I. Davidson.  Material Balance for
     Automobile-Emitted Lead in the Los Angeles Basin.  Environ. Sci. Technol.,
     9:448, 1975.

Kaakinen, J. W., R. M. Jorden, M.  H. Lawasani and R. E. West.  Trace Element
     Behavior in a Coal-Fired Power Plant.  Environ. Sci.  Technol., 9:  862,
     1975.

Klein, D. H., A. W. Andren, J. A.  Carter, J. F. Emery, C.  Feldman,
     W. Fulkerson, w. S. Lyon, J.  C. Ogle, Y. Talmi, R. I. Van Hook and
     N. Bolton.  Pathways of Thirty-Seven Trace Elements Through Coal-Fired
     Power Plant.  Environ. Sci. Technol., 9:973,  1975.

Kothny, E.  The  Three Phase Equilibrium of Mercury in Nature.  In: Trace
     Elements in the Environment,  E. Kothny, ed., pp. 48-80.  Advances in
     Chemistry Series No.  123, .American Chemical Society, New York,  1973.

Kowalczyk, G. S., C. E. Choquette  and G. E. Gordon.  Chemical Element Balances
     and Identification of Air Pollution Sources in Washington, D. C.
     Atmos. Environ.,  12:1143, 1978.
                                      60

-------
Mackay, D. and A. W. Wolkoff.  Rate of Evaporation of Low Solubility Contami-
     nants from Water Bodies to Atmosphere.  Environ. <3ci. Technol.,
     7:611, 1973.

Mayrsohn, H. and T. H. Crabtree.  Source Reconciliation of Atmospheric
     Hydrocarbons.  Atmos. Environ., 10:137,  1976.

McClure, V. E.  Transport of Heavy Chlorinated Hydrocarbons in the
     Atmosphere.  Environ. Sci. Technol., 10:1223, 1976.

Miguel, A. H. and S. K. Friedlander.  Distribution of Benzofa]pyrene and
     Coronene with Respect to Particle Size in Pasadena Aerosols in the
     Submicron Range.  Atmos. Environ., 12:2407, 1978.

Miller, M. S., S. K. Priedlander and G. M. Hidy.  A Chemical Element Balance
     for the Pasadena Aerosol.  J. Colloid Interface Sci., 39:165, 1972.

National Academy of Sciences, Committee on Medical and Biological Effects of
     Environmental Pollutants.  Ozone and Other Photochemical Oxidants, Chapter
     5.  National Academy of Sciences, Washington, D. C., 1977a.

National Academy of Sciences, Committee on Medical and Biological Effects of
     Environmental Pollutants.  Vapor-Phase Organic Pollutants.  National
     Academy of Sciences, Washington, D. C.,  1977b.

Pitts, J. N., Jr.  Mechanisms, Models and Myths: Fiction and Fact in
     Tropospheric Chemistry.  NASA Reference Publication 1022,  p. 1, 1978.

Pitts, J. N., Jr., C. K. A. VanCauwenbergh, 0. Grosjean, J. P.  Schmid,
     D. R. Fitz, W. L. Belser, Jr., G. B. Knudsen and P. M. Hynds.  Atmospheric
     Reactions of Polycylic Aromatic Hydrocarbons: Facile Formation of
     Mutagenic Nitro Derivatives.  Scinece, 202:515, 1978.

Pitts, J. N., Jr., A. M. Winer, K. R. Darnall, A. C. Lloyd and G. L. Doyle.
     Hydrocarbon Reactivity and the Role of Hydrocarbons, Oxides of Nitrogen
     and Aged Smog in the Production of Photochemical Oxidants.  ^aper No.
     14-2 presented at the International Conference of Photochemical Oxidant
     Pollution and Its Control.  Raleigh, N.  C., September 13-17.
     EPA-600/3-77-0016, 1976.

Pupp, C., R. C. Lao, J. J. Murray and R. F. Pottie.  Equilibrium Vapor
     Concentrations of Some Polycyclic Aromatic Hydrocarbons AS4O6, and
     SeO2 and the Collection Efficiencies of These Pollutants.  Atmos.
     Environ, 8:915, 1974.

Spencer, W. F. and M. M. Claith.  Vaporization of Chemicals.  In: Environmental
     Dynamics of Pesticides, R. Hague and V.  H. Freed, eds., pp. 61-78.
     Plenum Press, New York,  1975.

U. S. Environmental Protection Agency, Office of Research and Development.  Air
     Quality Criteria for Ozone and Other Photochemical Oxidants, Chapter 6.
     EPA-600/8-78-004, 1978.
                                      61

-------
Van Hook, R. I., W. F. Harris and G. S. Henderson.  Cadmium, Lead and Zinc
     Distributions and Cycling in a Mixed Deciduous Forest.  Ambio,
     6:281, 1977.

Woodwell, G. M., P. P. Craig and H. A. Johnson.  DDT in the Biosphere: Where
     Does it Go?  Science, 1974:1101, 1971.
                                      62

-------
             3.0.  TRANSPORT AND FATE IN SOILS AND SEDIMENTS
3.1  INTRODUCTION

     Comprehensive assessment of the potential hazards of a contaminant within
the environment requires detailed information on the distribution of dominant
forms and related concentrations of the contaminant within each of the major
environmental compartments: air, water, biota, soils and sediments.

     Many contaminants exhibiting toxic, carcinogenic, mutagenie and/or
teratogenic properties are known to associate strongly with soil and sediment
materials via sorption phenomena.  Soils and sediments thus function as
significant sinks and sources in the transport and distribution of persistent
contaminants.  Further, such associations can have pronounced effects on
chemical, physical, and biochemical transformations affecting the fate of
contaminants, and on the bioavailability of parent compounds and transformation
products to aquatic and terrestrial organisms, including man.

     Association of a contaminant with soils and/or sediments frequently
reduces the environmental hazard of the chemical.  For example, it has been
established that the percentage of total PCB content of water which tends to
accumulate in fish is much higher in water bodies in low suspended solids than
it is in waters having higher suspended solids content.  Conversely, there are
other chemical and/or biochemical reactions of adverse environmental impact
which are promoted by such soil/sediment associations.  For example, bottom-
feeding organisms are exposed to high concentrations of contaminants in water
bodies in which such associations have resulted in enriched levels of
pollutants on solids at the sediment-water interface.

     There is little doubt that the association reactions of contaminants with
soils and sediments play dominant roles in determining the environmental
behavior and significance of many of those contaminants of greatest potential
concern.  Any meaningful environmental hazard assessment program for a chemical
contaminant must include an estimate of the extent to which that contaminant
becomes associated with soils and/or sediments.  This, in turn, is determined
and controlled by a variety of physicochemical and biochemical phenomena, many
of which are complex and interactive.

     It is difficult to prescribe straightforward methodologies for the
screening and testing of potentially hazardous compounds that will ensure
precise prediction of environmental transport and fate.  Nonetheless, it is
necessary that methodologies be developed which afford at least a reasonable
                                      63

-------
estimate of the major environmental characteristics of such compounds.  With
respect to the soil/sediment compartment of the environment, such methodolo-
gies must take account of at least the following important phenomenologic
factors:  a) the thermodynamics and kinetics associated with sorption/desorp-
tion processes, b) transport phemonena associated with leaching and exchange
processes, c) photodegradation and volatilization, and d)  physicochemical and
biochemical transformations.  The properties and characteristics of soils and
sediments are important determinants of the relative extent and significance
of these phenomena.

     This chapter reviews state-of-the-art understanding of each of the
phenomenologic factors identified as significant to the role of soils and
sediments in the transport and fate of toxic chemicals in the environment, and
considers major information deficiencies and selected research needs.  In
addition, these concerns are addressed in relation to meaningful characteri-
tion of soils and sediments and their influences on the phenomenologic
factors.  Finally, land disposal of chemicals - one of the most direct routes
of transport of toxic chemicals to this compartment of the environment - is
considered.

     It should be noted that the factors considered in this chapter are
complex and interactive, that the body of contributors - although representing
a reasonable breadth and depth of related knowledge and experience - does not
necessarily constitute one of comprehensive expertise on all aspects of these
factors, and that the time alloted for development of the dicussion was short.
This report is not advanced as a totally exhaustive and definitive document
and should not be taken as such.  Rather, it is an attempt, constrained by the
factors mentioned above, to provide one contribution to better definition of
research needs relating to the role of soils and sediments in the transport
and fate of toxic chemicals in the environment.
3.2  SORPTION-DESORPTION PROCESSES

     Extensive work on the sorption of synthetic organic chemicals from water
has been done in the past decade, and much has been learned about factors
which affect and control the sorption of such compounds.  Much of this effort
has focused on the use of activated carbon for purification of drinking
and waste waters, but many of the findings with respect to concepts and sorp-
tion dynamics can be extrapolated to naturally occurring soils and sediments.
Detailed descriptions and discussions of sorption concepts and dynamics have
been given by Weber and colleagues (1972, 1977, 1978, 1979a), are readily
available in the literature, and will not be repeated here.

     Much of the sorption work regarding the effects of potentially toxic
compounds on soils and sediments has related to pesticides (Bailey and White
1964; Bailey et al. 1968,' Lotse et al. 1968; Shin et al. 1970; Adams and Li
1971; Helling 1971; Goring and Hamaker 1972; Weber 1972).  These studies have
included characterizations of the effects of pesticide properties (pKa, pKjj,
water solubility, polarity) and soil properties (class, size distribution,
organic matter content).  With respect to chemicals other than pesticides,
                                      64

-------
detailed studies have been performed by Haque et al. (1974) and Weber et al.
(1979b) on the kinetics and equilibrium aspects of sorption-desorption
reactions of polychlorinated biphenyls (PCB's) on different types and sizes of
sediments, and on suspended soils under different solution conditions.

     Sorption equilibria have been most commonly described in terms of the
Freundlich sorption isotherm equation:

                                qe = % Ce

where qe is the amount of contaminant sorbed per unit amount of sorbent, Ce is
the equilibrium solution concentration of the contaminant, and Kp and 1/n are
characteristic constants relating to the extent and intensity of sorption,
respectively (Weber 1972).

     While a variety of methods are available to determine distribution co-
efficients from sorption isotherms, laboratory batch tests of the type
recommended by the American Institute of Biological .Sciences (A.I.B.S. 1974)
and cited by the U. S. Environmental Protection Agency (1975) are generally
satisfactory.

     For natural aquatic systems in which the ratio of quantity of contaminant
to soil and/or sediment is low, the value of the constant 1/n in Equation 1
approaches unity, and the relationship between qe and Ce approaches linearity.
Sorption in these systems is frequently described in terms of the empirical
equation:

                                                                         (2)

where S is the sorbed concentration (in yg/g), C the solution concentration
(in ug/g~yg/ml), and K
-------
     2.   When normalized by the carbon content of  a  sediment  or  soil,
         the sorption of neutral organic  compounds on  sediments  or
         soils from different areas varies by a factor of  only 2 to 3.
         This conclusion,  if substantiated by additional studies, could
         greatly simplify prediction of the sorption behavior of toxic
         organic compounds since the results of tests  conducted  on one,
         or a few,  sediments or soils would in general be  applicable  to
         natural water systems throughout the world.

     3.   While the  sorption of most toxic organic  compounds on natural
         particulate matter is generally  rapid (approaching near-maximum
         levels within a few hours) and reversible,  there  is  an  indication
         of continuing slow sorption.  Further, there  are  some compounds
         which apparently do not desorb,  or do so  only very slowly.

     4.   In general, sorption on natural  particulate matter has  a low
         temperature coefficient (low activation energy).

     5.   One of the most promising developments is the correlation
         between the distribution coefficient for  the  uptake  of  neutral
         organic molecules by natural water solids and the octanol/water
         partition coefficient.  A relationship of this type  is  to be
         expected for a large group of organic chemicals,  especially
         nonionic compounds.

3.2.1  Sorption-Desorption Recommendations

     Sorption studies to date have yielded some important  tentative conclu-
sions which may greatly facilitate prediction of the environmental behavior
of contaminants which tend to associate with soils and sediments.  A  case in
point is the observation that sorption frequently  correlates  well with  the
organic matter content of soil or sediment.  The limitations  of  these
conclusions should be determined for various classes of organic  compounds in
association with various types of sediments and soils.  Where deviations  are
found for certain groups of organic compounds, studies should be conducted  to
develop generalizations applicable to that group relative  to  their uptake and
release from natural fresh and marine water sediments, soils, and particulate
matter.

     For specific situations involving evaluation  of the behavior of  a  parti-
cular contaminant at a certain location,  soils or  sediments  from the  area of
interest should be used.  However, in situations where attempts  are being made
to predict the environmental behavior of a chemical  which is  expected to  have
broad distribution in the environment, soils and sediments typical of those
found throughout the United States should be used.  Reference soils and/or
sediments should be  incorporated in testing schemes  to develop a broad data
base which can be used to estimate the environmental behavior of new
chemicals.

     Most existing information concerning the  sorption behavior  of  toxic
chemicals on soils has been obtained in experiments  of relatively short
                                      66

-------
duration, and for systems with water/soil ratios much higher than those
encountered in field situations.  Further, under natural conditions the
surface of a soil is commmonly subjected to substantial variations in
temperature, moisture content, and relative humidity.  The use of short time
periods for equilibration is dictated by the need to minimize complications
arising from hydrolysis, microbial decomposition, volatilization, and other
factors.  Although published data suggest that later stages of sorption
reactions continue over weeks or months, there is a lack of quantitative data
regarding these long-term sorption phenomena.

     From the standpoint of the long-term behavior of toxic chemicals in soils
and sediments, desorption studies are also of critical importance.  Desorption
generally appears to differ from sorption, particularly with respect to
reaction rate.  Desorption reactions may be responsible for the fact that it
is generally more difficult to extract an "aged" soil residue than it is a
freshly incorporated soil.  Therefore, a better understanding is needed of the
effects of time, cyclic variations in temperature and moisture content over
time on the sorption and desorption of toxic chemicals on soils and soil
constituents.

     The liquid/solid ratios employed during sorption tests is a related area
of investigation requiring further pursuit with respect to testing
methodologies and the validity of obtained data.  Because of the markedly
different character of these systems, especially with respect to the
concentrations of contaminants, oxidation-reduction conditions must be
considered in designing sorption and leaching tests for materials in landfills
and concentrated sediment systems such as those associated with dredged
sediment disposal (Weber et al. 1976; Lee 1978).  These systems are
generally anoxic, which tends to promote the release of many contaminants
(Weber et al. 1976).  However, with few exceptions, the concentration of
contaminants in the anoxic leachate is a poor measure of the potential impact
of the leachate on environmental quality.  Eventually these leachates will
come in contact with dissolved oxygen, which will result in removal of many
contaminants through sorption on iron hydroxide floe.  Accurate assessments of
the release of contaminants from landfills and from sediments dispersed in
natural waters require better understanding of the role of oxidation-reduction
reactions in influencing sorption/desorption reactions.

     Conventional sorption studies generally involve analysis of the
equilibrium concentration of a compound remaining in solution.  A better
understanding of sorption processes on soils and sediments may also result
from examination of that portion of the compound that has reacted with the
soil or sediment surface.  Information on any changes in compound speciation
and/or character resulting from interaction with the surface of a soil or
sediment is required to provide a better basis for estimation of hydrolysis,
degradation, desorption, and other transformation phenomena.
3.3  LEACHING AND EXCHANGE PROCESSES

     Consideration of environmental impacts would be relatively
straightforward if potentially toxic chemicals were retained in soils and/or
                                      67

-------
sediments at the site of application or deposition.  However,  the transport of
toxic chemicals from points of release, and their subsequent dispersal
throughout the various compartments of the environment,  result in exposure to
both human and other target species.

     Exchange processes between soils and sediments and  the atmospheric and
aquatic compartments of the environment are in large measure determined by
such factors as volatilization and sorption/desorption.   Additionally, such
exchange processes are markedly affected by aerodynamic  and hydrodynamic
forces operative at the soil-air and sediment-water interfaces.  Uptake
and release of contaminants by natural water particulate matter is frequently
controlled by the hydrodynamic characteristics of both the water and the
sediments.  Mixing processes within the sediments and between the sediments
and overlying waters therefore often govern the environmental significance of
sediment-associated contaminants.  These mixing processes arise from both
physical forces and biological activities (Lee 1970).  Further, hydrodynamic
factors control the transport of particulate solids and associated pollutants
within the water column and into and out of the sediments resulting in
differential accumulations of potentially toxic compounds in the environment
(Weber et al. 1979b).  It is now becoming clear that meaningful modeling of
environmental behavior and fate of solids-associated contaminants will require
a substantial research effort devoted to characterizing and quantifying mixing
and dispersion processes.

     Of principal concern are leaching and exchange phenomena affecting
migration and movement of potentially toxic chemicals within soils and
sediments.  Leaching and diffusion of toxic chemicals in soils and sediments
represent two of the basic transport mechanisms for dispersal of chemicals
within this compartment.  In order to evaluate the potential transport of
chemicals via these mechanisms, it is necessary to understand the factors
which determine the extent of leaching and diffusion.  This information is
required in making decisions regarding exposure from chemical use and
disposal.

     The extent to which compounds are leached from soils and sediments is
strongly influenced by rates of degradation, sorption, and precipitation
reactions.  For purposes of this discussion, degradation will be considered
negligible.  Because leaching is a relatively slow process, only persistent
compounds are environmentally significant.  While it is recognized that in
many situations factors of degradation must be considered, sorption and
precipitation reactions mediate the transport of most persistent toxic
chemicals and will therefore be considered in greatest detail.

     Leaching and diffusion processes are treated separately.  In most trans-
port situations one or the other predominates.  Leaching is usually the more
important transport process in soils when considering how much or how far a
compound is moved.  One situation in which diffusion becomes significant in
soils is in the transport of volatile  compounds, such as the diffusion of
fumigants through soil or the loss of volatile compounds from soil surfaces.
                                      68

-------
     The relative contributions of leaching and vapor transport can be esti-
mated from solubility and vapor pressure data.  Goring (1962)  and Helling
et al. (1971) have suggested that compounds with water-air ratios (weight/
volume basis) under 10^ will diffuse primarily in air, while those over
3x10  will diffuse mainly in water.  Using these estimates as a rough
approximation, it could be assumed that those compounds predicted to move
primarily in the vapor phase would have the potential for rapid transport by
a diffusion controlled process.  Considerable data are available in the
literature on the aqueous solubility of compounds.  However, reliable vapor
pressure data are obviously lacking, and considerable effort needs to be
expended to correct this.  These vapor pressure data are similar to those
required for evaluating the volatilization potential of compounds.

     Much of the literature on predicting transport of toxic chemicals
in soils and sediments deals with the transport of organic pesticides in soils
(Letey and Farmer 1974; Leistra 1973; and Hamaker 1975), although several
reviews discussing some aspects of the transport of heavy metals in soils and
sediments are also available (Hutchinson 1975; Fuller 1977).

3.3.1  Leaching

     A simple approximate equation for describing the flux, Js, of a non-
interacting or conservative compound through soil or sediment as a function
of water flux, Jw, is:

                                  JS = Jw C                              (3)
where C is the weight-ratio concentration of the compound in the soil water.
This equation describes the mass flow of a conservative species, such as the
chloride ion, for situations where diffusion and dispersion within the liquid
phase are negligible relative to convective transport.  A non-interacting
species will move with the water front though most chemical species will
interact with the soil or sediment and lag behind the water front, depending
on the extent of such reactions as sorption or ion exchange.  Therefore,
characterization of sorption and related processes is most important for
prediction of leaching.

     Leaching of chemicals through a soil profile is dependent upon the
direction and rate of water flow as well as the sorption characteristics of
the chemical with the soil.  Water flow through soil depends on soil proper-
ties and can be quite complex.  A number of investigators have examined this
subject and, based on their work, several conclusions can be made concerning
research needs, particularly in the area of data collection in laboratory
column experiments and field studies.

     Most leaching models assume a linear isotherm (equation 2) to describe
the distribution of a chemical between the solution and the sorbed phases.
The linear isotherm is often satisfactory for a relatively narrow concentra-
tion range.  When dealing with large concentration ranges, as might be the
                                      69

-------
case for many toxic chemicals in the environment, the Freundlich isotherm
(equation 1) is more applicable.  The sorption of most organics by soils will
follow the nonlinear Freundlich isotherm at higher concentration, and serious
errors in predicting the extent of leaching can result in such cases if the
linear isotherm is assumed.  When sorption isotherms are used to improve the
predictability of models, it is imperative that sorption phenomena be
characterized as to the type of isotherm which best describes a particular
system.  Although in the modeling process it may be possible to treat most
sorptions in terms of a linear isotherm over limited ranges of conditions, it
is necessary to know the actual nonlinear relationship in order to properly
choose a linear relationship and associated limits of application.

     Well-defined isotherm relationships and information relative to the
factors and mechanisms influencing sorption reactions are essential for
description of the basic processes controlling leaching.  Quantitative
description of these basic processes must be encouraged if there is any
expectation of improving the ability to predict transport.  However, immediate
data needs for predicting pollutant transport in the natural environment are
of necessity somewhat different from those associated with basic research.
This difference exists because predictions under natural conditions often
require methodologies that have wide application by many people under
conditions where variability is uncontrolled.  Large errors can be expected
under these conditions.  Therefore, the predictive tool adopted need only
provide rough estimates to be useful in application to the environment.

     State-of-the-art methodology does not permit prediction of the precise
distribution of a toxic chemical within a soil profile with any success in
the natural environment.  Long-term estimates of when a material may be
expected to appear in the ground-water appear possible, especially if rough
estimates can be made for the sorption reactions.  Such estimates
would describe the arrival time for the bulk of material from a surface
event and would assume uniform flow across a field.  Certain parts of
the drainage would arrive sooner or later than the average due to spatial
variation in percolation rates.

     The correlations between sorption of un-ionized organic compounds and the
organic matter content of soils and sediments is reasonably well established.
This class of compounds includes most pesticides and most of the trace
organics found in drinking water.  From the soil/water distribution, K
-------
The fraction of chemical in the soil solution can thus be assessed from K^, P,
or Q.  The determination of P is a simple laboratory procedure.  From a
determination of P and the use of equation 5 or a similar equation, a rough
estimate of the mobility of neutral molecules in soils is possible.  Further
investigation into the applicability of such relationships seems warranted
since there is already considerable interest in the use of octanol/water
ratios for estimating the bioaccumulation potential of organic chemicals.

     Laboratory leaching columns and soil thin-layer chromatography (TLC) have
both been used successfully to correlate sorption characteristics with the
relative mobility of compounds in soils and sediments.  The use of soil
columns is often preferred because water flow in such systems more closely
simulates that in an undisturbed soil or sediment profile.  Soil TLC has the
advantage of being a more rapid means for determining the relative mobility of
a number of compounds in a number of soils and/or sediments.  Soil TLC may be
extremely useful as a screening tool to determine which compounds might pre-
sent a leaching hazard and warrant further study with soil leaching columns.
Adequate methods are available for the application of either technique.
However, there is a need for uniformity of techniques between laboratories
using leaching columns as a means to improve predictive capability.  Many of
the data available in the literature on the leaching of organics in column
tests are not comparable because of differences in techniques used by
different labs.  In addition, investigators often fail to collect all the
data, especially water flow characteristics, necessary to evaluate and compare
leaching experiments.  In a review of 49 commonly used pesticides, no leaching
data were found for 27 of the compounds (Farmer 1976).  For those compounds
for which data were available, the information was difficult to use in
predictive evaluations due to lack of uniformity of leaching techniques and
soils employed by the different investigators.

     The modeling of pollutant transport due to leaching in laboratory columns
has been fairly successful.  Using sorption and water flow characteristics
the distribution of chemicals within a soil profile in the laboratory can be
simulated relatively well.  Problems can be expected when these models are
applied to the field.  Spatial variability is likely to be a serious problem
in simulation of field events.

     Laboratory modeling has looked at sorption and the exclusion volume
(dead-end pores) concept to simulate laboratory columns.  These usually assume
a diffusion process mediating the compound in solution in the stagnant phase
entering the convection stream.  In these models sorption processes are
usually linearized to give

                                  BS = R 0 C                        (6)

where R is the sorption coefficient, K
-------
     One problem that is only beginning to be approached by investigators is
the influence of wetting-drying cycles on the upward mobility of chemicals
after they have entered the soil profile.  The ability to predict upward
mobility is essential for predicting the runoff and volatilization of
soil-incorporated water-soluble compounds from the soil surface.  This is a
complex problem involving water flow and water evaporation to and from the
soil surface.  Several models describing chemical runoff from non-point
sources for humid regions of the U.S. are available (Hutchinson 1975;
Donigian and Crawford 1976; Stewart 1976; and Wauchope 1978).  There is a
need for field testing and verification of existing chemical runoff models,
and for expansion of such models to include more arid regions of the U.S.

     The methodology for field investigations of leaching in soils has not
progressed sufficiently to permit recommendation of a definitive procedure.
As with laboratory columns there is a lack of uniformity in procedures and
often a deficiency of data collection on water flow characteristics.  The
problem is exaggerated in field experiments because of the natural variability
involved when working with soils in their undisturbed (native) state.

     It is important that field investigations be continued.  They are
necessary to verify the models developed with laboratory leaching columns.
Though unlikely there will be much success in predicting the absolute distri-
bution of a chemical from a soil profile in the field, there is the
possibility of estimating the depth of maximum penetration or finding a range
of expected concentrations at expected depths.

     Current leaching studies with soils are characteristically performed with
distilled water or water of low ionic strength to simulate irrigation water.
The application of current leaching information to the problem of landfill
leaching is probably not justified.  Leachate from land disposal practices can
be expected to contain high concentrations of soluble organics and inorganics
which may interfere and/or compete with each other for sorption sites.  The
effects of concentration parameters on leaching are largely unknown and
require further investigation.  A likely result of competitive forces on
sorption is increased leaching and analytical procedures are expected to be
more complicated with land disposal leachates.

     Studies on metals over the past decades have provided sufficient data to
predict short-term consequences of metal contaminations of agricultural soils.
Most knowledgeable researchers in the field are of the opinion that
phytotoxicities and quantities of metals sorbed can be assessed accurately
where metals are applied to soils in the form of wastes for periods up to a
decade.  The longer-term consequences of metals added to soils, as well as
predictions of what occurs when the metal additions cease, are more uncertain.
A basic understanding of the chemistry of the metals in soils is required for
prediction of long-term consequences as well as of the ultimate fate of metals
added to soils in the form of waste.  Factors controlling the solubilities and
the nature and rate of chemical and microbial transformations need to be
delineated.  Factors which control the availability of metals to plants are
also not well established.  These are dependent upon chemical forms present,
                                       72

-------
soils and sediments, and upon competitive interactions.  A thorough knowledge
of the chemistry of metals in soils and sediments should facilitate the
required long-term predictions.

     Release or mobilization of metals from soils and sediments by the action
of added complexing agents is a substantial concern.  Similarly, the role of
metal speciation on the removal of metals from solution onto soils and
sediments is of importance in regulation of metal transport.  At present,
little information other than total metal uptake and release is available.
Different types of sites in soils and sediments have different affinities and
strengths of binding for heavy metals.  Lack of consideration of speciation
with respect to extent and strength of uptake, or release of metals from the
solids as a consequence of the addition of complexing agents, precludes use of
such information except for the given conditions of testing.

     Research should be conducted to establish which phases of sediments and
soils are active in metal uptake, and whether binding strength is similar
among various soils and sediments.  Information regarding the ability of
complexing agents to release metals from metal-amended and unamended soils and
sediments is required.  This research should consider release from each of the
several metal phases present.  The degree to which metal release capability is
constant between the same phase of various soils and sediments should be
ascertained.

3.3.2  Diffusion

     Non-vapor diffusion in soil is not likely to be a predominant transport
mechanism so this discussion will be limited to vapor diffusion.  If reliable
vapor density-temperature data are available for a compound, it should be
possible to develop models for successfully predicting gaseous flux.

     A gas flow model has been applied to the landfill disposal of wastes
containing volatile components by Farmer et al. (1978).  A possible avenue
for the escape or migration of a pollutant from a landfill is via
volatilization through the soil cover.  The assumption can be made that the
compound is nearly insoluble in water so that diffusion in the vapor phase in
the soil pores is the only mechanism by which it can move through the soil
cover.  Such an assumption is valid for certain compounds, such as certain
chlorinated hydrocarbons for which diffusion is controlled by the air-filled
porosity of the soil.  The limiting step for volatilization is the rate at
which the compound diffuses through the soil cover to the surface.  Farmer et
al. (1978) suggest that for any particular land disposal situation, the
volatilization flux will be determined by the soil air-filled porosity and
depth of the soil cover, both of which can be managed by the landfill designer
to meet existing regulations for vapor emissions.

     If cracks or other small openings develop in the soil cover an
appreciable increase in flux through the cover will result.  The practice of
placing toxic waste with materials such as municipal solid waste, which
is subject to settlement, could cause such cracking and increase flux.
                                      73

-------
     Another consideration to be given to the use of land disposal for
volatile toxic chemicals is the need for long-term arrangements for ensuring
the integrity of soil covers.  Some compounds that pose a volatilization
potential from land disposal will also be resistant to degradation.  Assuming
no degradation, calculations for hexachlorobenzene indicate that it can
continue to volatilize at maximum rate for several centuries when placed on
land.  The integrity of the soil cover must be maintained for this period by
preventing such disruptions as erosion or digging.

     In the course of investigations into the land disposal of volatile wastes
Farmer et al. (1978) found that some wastes contained a dense organic liquid
in which hexachlorobenzene is highly soluble (23,000 yg/ml).  Thus, organic
solvents as well as water must be considered as the carrier in mass flow.

     A variety of forces have been identified as significant factors operative
in the exchange of toxic substances between sediments and water.  A clear need
exits for research directed toward quantification of the relative influences
of physical disruption or mixing (turbulence, gas movement), biological mixing
(bioturbation) and diffusive transport factors in different sediment and water
environments.
3.4  PHOTODEGRADATION AND VOLATILIZATION

3.4.1  Photodegradation

     Absorption of light by a chemical is absolutely required for direct
photochemical reaction.  However, the molecule of interest does not itself
have to absorb light energy if other molecules, so-called photosensitizers,
are available to participate in the primary photochemical event.  Once a
photosensitizer molecule has been promoted to an excited state it can transfer
its energy either directly or indirectly to the molecule of interest which,
once excited, can dissipate its energy by several mechanisms including the
breaking of formation of chemical bonds.

     Environmental photodegradation occurs exclusively through the interaction
of sunlight with molecules.  Photodegradation experiments performed in the
laboratory have frequently relied on artificial light due to the variable and
often unpredictable quality and quantity of sunlight available over any given
period of time.  The high energy end of the sunlight spectrum that reaches
Earth (290 to 400 nm) can be duplicated in the laboratory through use of such
sources as mercury arc lamps (Zepp et al. 1975), conventional fluorescent
lamps with borosilicate glass filters (Crosby and Moilanen 1973) and xenon
arc lamps (Lockhardt and Blakely 1975).

     Knowledge concerning the photochemical fate of synthetic chemicals
purposefully applied to land (eg., pesticides, waste effluent resulting from
more stringent water quality standards, municipal and industrial solid wastes,
etc.,) will increase in importance in the next few years.  Therefore, an
important transition must be made between the laboratory and the real
environment to identify and understand those factors in the environment which
modify the results of laboratory photodegradation experiments.
                                       74

-------
     Photochemical reaction of chemicals in soils and on sediments does not
occur throughout the various soil and sediment layers but only at the surface
(Crosby 1972).  Some researchers have tried to duplicate the soil environment
in the laboratory.  Hautala (1976) has examined the photochemical decomposition
of 2,4-D, sevin, and parathion on three types of arable soils.  His experiments
determined that moisture content of the soils was a critical factor as was the
type of soil used and the presence or absence of a detergent.  For example,
anionic surfactant decreased the extent of photolysis of 2,4-D on dry or
saturated soils.  In the presence of excess water ("wet" soils), detergent
dramatically increased the rate of photolysis.

     Carey et al. (1976), in an effort to simulate the interactions between
PCB's and suspended sediments, studied their photodechlorination in the
presence of TiO2-  Based on their results, PCB photolysis in the presence of
suspended solids does not proceed along the same pathway as in the absence of
such sediments.

     Spencer et al. (1978) have used a thin layer of soil dust in the labora-
tory to simulate the upper soil layer for studies of parathion photodegrada-
tion.  Smith et al. (1978) have also performed laboratory photodegradation
studies using a thin-layer soil technique.

     Based on these works and on reviews of the subject by Crosby (1972) and
Plimmer (1972), it can be concluded that the phototransformation of certain
synthetic chemicals may be an important way in which these chemicals are
altered in the environment.  Only very limited field work on chemical
photodegradation in the environment has been performed so a number of
important questions remain unanswered.  Since a large group of new compounds
may be either directly placed on the land or discharged to the land after use,
there remain many undocumented areas regarding their photodegradability.

     The circumstances under which photoreaction of materials on soils and
sediments may be important should be researched, and the physicochemical
properties or structure-activity relationships which would predict
phototransformation of one compound on various types of soils should be
explored.  More information is needed about which properties to consider when
predicting photodegradability of land-applied materials not in direct contact
with soils and sediments, as well as the soil and sediment characteristics and
parameters which determine routes and rates for photochemical reactions of
specific compounds.  Laboratory procedures for photodegradation should be
identified and field tests made to verify any laboratory models.  The
importance of phototransformation of chemicals on airborne particulate matter
or in the vapor phase following volatilization from soil or sediment should be
considered.  Finally, a check should be made to see if there are laboratory
screening tests based on results of experiments designed to answer questions in
the aforementioned areas to allow relatively rapid and predictive
photodegradation experiments on a large number of compounds.

3.4.2  Volatilization

     Organic pollutants principally enter the atmosphere by volatilization, by
direct injection such as pesticide sprays or stacks from industrial plants,
                                      75

-------
and by movement of wind-blown dust particles (Spencer et al.  1973).   The soil
becomes both sink and source for a large portion of the chemicals that enter
the environment by intentional application to the land, inadvertent  spillage,
or release into the air.  One of the management tools required for the
utilization of land as a waste disposal medium is the ability to control
volatilization of the chemicals in the waste, whether in landfill or mixed
into the soil after application.

     Post-application volatilization of pesticides under field conditions was
thoroughly reviewed by Taylor (1978) and will not be repeated here.   The
reader is also referred to other reviews on pesticide volatilization by
Hamaker and Thompson (1972), Spencer et al. (1973), Wheatley (1973), Guenzi
and Beard (1974), and Plimmer (1976).

     Potential volatility of a chemical is related to its inherent vapor
pressure; however, actual vaporization rates will depend on environmental
conditions and other factors that modify or attenuate the effective  vapor
pressure or behavior of the chemical at a solid-air or liquid-air interface.
Vaporization from surface deposits depends only on the vapor pressure of the
chemical and its rate of movement away from the evaporating surface.
Vaporization from aqueous systems depends on the vapor pressure of the
chemical and its water solubility.  Vaporization from soil is controlled by
solubility and sorption as well as vapor pressure.  Consequently, no single
physicochemical property is adequate to describe and predict the probable
vapor behavior and fate of a chemical in the environment or its likely method
of transport in the atmosphere.  However, estimates of relative vaporization
rates useful for environmental indices can be calculated from basic  physical
properties of vapor pressure, water solubility, sorption, and persistence if
reliable values for each are known at various temperatures.

     The vapor pressures of many organic chemicals of environmental interest
increase aproximately 3 to 4 times for each 10°C increase in temperature.
Consequently, reliable values for vapor pressures at various temperatures are
necessary to estimate vapor losses of the chemical from surface deposits, to
predict partitioning between water and air, and soil, water, and air relative
to volatility from water solutions and from wet soils, and to calculate
residence times in the atmosphere of chemicals in droplets and aerosols.

     A search of available literature on the vapor pressure of 49 commonly
used pesticides revealed both a lack of vapor pressure data for many of the
pesticides and a wide variability in vapor pressure of the same compound
reported by different authors (Spencer 1976).  The inconsistency in reported
vapor pressure values attests to the need for accurately determining vapor
pressures of pesticides and other toxic organic chemicals by a standard
procedure, preferably one which relies analytically upon a specific method for
determining the chemical of interest.

     The gas saturation method has proven to be a reliable method of measuring
vapor pressures of pesticides (Spencer and Cliath 1969; Spencer et al.  1978;
U. S. Environmental Protection Agency  1975) and should be equally reliable for
determining vapor pressures of other toxic organic chemicals with vapor
                                      76

-------
pressures in the range of those exhibited by most pesticides.  When this
method was recently used to measure the vapor pressure of technical grade and
purified ethyl and methyl parathion, no significant differences in vapor
pressure were observed between technical grade or purified materials of either
chemical (Spencer et al. 1978).  The observed vapor pressures, which
indicated methyl parathion is approximately twice as volatile as ethyl
parathion, are consistent with their relative persistence on foliage as
reported by Ware et al. (1972, 1974).  In contrast, the most frequently
quoted literature value for vapor pressure of parathion (Bright et al. 1950)
is approximately four times higher than the reported vapor pressure of methyl
parathion (Guckel et al. 1973).

     The relationship between vapor loss rate and vapor pressure by measure-
ments of vapor loss rates from various surfaces types under the same condi-
tions is helpful in comparing relative volatility of potentially toxic
chemicals.  This measurement can be accomplished by a procedure such as that
described by Spencer et al. (1978) or by Guckel et al. (1973).

     Incorporation of an organic chemical into the soil decreases the
concentration at the evaporating surface thereby greatly reducing the
volatilization rate.  When a chemical is mixed into the soil loss by
volatilization involves desorption of the chemical from the soil, upward
movement to the soil surface, and vaporization into the atmosphere.
Consequently, the volatilization loss rate of a chemical in soil will be
related to the vapor pressure of the chemical within the soil and its rate of
movement to the evaporating surface.  The rate of loss initially will be a
function of the vapor pressure of the chemical as modified by sorptive
interactions with the soil surface.  Volatilization rate decreases rapidly as
the concentration at the surface of the soil is depleted, and the rate soon
becomes dependent upon rate of movement of chemical to the soil surface
(Spencer, Farmer and Cliath 1973; Farmer et al. 1972, 1973).  The two general
mechanisms whereby pesticides move to the evaporating surface are diffusion
and mass flow in evaporating water.  Usually both mechanisms operate together
in the field where water and the chemical vaporize at the same time unless the
chemical is essentially insoluble in water or has a sufficiently high vapor
pressure to result primarily in vapor phase movement through the soil pores
rather than movement within the soil solution.

     In the absence of evaporating water volatilization rate depends upon
rate of movement of the chemical to the soil surface by diffusion.  If
diffusion coefficients for the chemicals in the soil are known, diffusion
equations can be used to predict changes in concentration of the chemical
within the soil and its loss rate at the soil surface.  Mayer et al. (1974)
and Farmer and Letey (1974) have obtained good agreement between volatiliza-
tion rates predicted from mathematical models using diffusion coefficients and
volatilization rates observed in the absence of evaporating water.  Diffusion
processes in the soil will also control volatilization losses of chemicals in
the presence of evaporating water if the chemical does not move significantly
by mass transfer with the water due to its insolubility or to its much greater
mobility in the vapor rather than in the liquid phase of the soil system
(Farmer et al. 1978).
                                      77

-------
     In addition to factors directly affecting vapor behavior,  the proportion
of the total amount of the chemical in the soil lost to the atmosphere will
depend upon the resistance of the chemical to degradation and leaching in the
soil.  Any models for predicting vapor behavior must take into account the
relative degradation rate of the chemical compared with its movement into the
atmosphere.  With methyl parathion (Spencer et al. 1978) and parathion
(Yang 1974) biological degradation in the soil, rather than volatilization,
is the major pathway for dissipation from soil.  The total methyl parathion
volatilized over a 33-day period when added at 10 ppm to Flanagan silt loam
averaged only 0.25 percent of the amounts incorporated into the soil, but
after 33 days only 0.5 percent of the added methyl parathion remained in the
soil.  This is in direct contrast with lindane for which much higher
volatilization rates were observed even though the vapor pressure of lindane
is somewhat lower than that of methyl parathion; its greater persistence
results in much greater losses by volatilization.

     Many industrial wastes are disposed of in landfills.  These wastes may or
may not be subsequently covered with a layer of soil to prevent or reduce
vapor loss of any potentially toxic organic chemicals.  Volatilization of the
chemicals from uncovered wastes will follow the principles of vapor loss from
inert surfaces, and the actual rate of loss will depend upon weather variables
affecting vapor pressure and air turbulence at the waste site.   Vapor loss of
chemicals from a landfill in which the waste is covered with a layer of soil
will be at a greatly reduced rate in accordance with the above principles on
vapor loss from soil.  Volatilization losses will be related to the rate of
movement of the chemical through the soil cover into the atmosphere.

     Soil cover provides an efficient means of decreasing the loss of
volatile chemicals placed in a landfill (Farmer et al. 1976).  One of the
management tools required for utilization of landfill as a waste disposal
site is an ability to control the volatilization of hazardous chemicals from
the wastes.  Farmer et al. (1976, 1978) recently completed a study designed
to gather data useful to the landfill planner in controlling vapor movement of
hexachlorobenzene by utilizing a soil cover to limit vapor loss to the
surrounding atmosphere to an acceptable level.

     Hexachlorobenzene (HCB) is a persistent, water-insoluble, fat-soluble
organic compound.  Farmer et al. (1976, 1978) used a simulated landfill to
determine the parameters necessary to predict volatilization of this compound
from wastes placed under a soil cover.  Covering the wastes with only 1.0 cm
of soil reduces vapor flux of HCB from 363 to 5 ng/cm /hr, or by 98.6 percent,
indicating that soil is a very effective covering material.  They found that
soil depth and soil air-filled porosity were the two prime factors controlling
movement of the vapor in soil.

     The relationship between vapor flux, soil depth, and air-filled porosity
is predictable on the basis of known factors controlling vapor behavior in
soil.   Since HCB is essentially insoluble in water, mass flow in the evapora-
ting water will not be important in moving this compound to the soil surface;
consequently, diffusion in soil pores will be the only mechanism available.
Because of the water-air partition ratio of compounds such as HCB, and the
                                      78

-------
higher rate of diffusion in air than in water, diffusion of these compounds
will be primarily in the vapor phase through the soil pores.  Thus, vapor loss
can be considered as a diffusion controlled process involving only vapor phase
diffusion through air-filled soil pores.

     Since soil porosity is controlled mainly by soil compaction and soil
water content, vapor flux can be decreased by increasing compaction and
increasing soil-water content as well as by increasing soil depth.  Vapor flux
of HCB through a soil cover of 30 cm or greater would be less than 1 ng/cm /hr
(1 kg/ha/year) compared with a vapor flux of 363 ng/cm /hr from the uncovered
waste under the same conditions.  For compounds such as HCB, increasing
soil-water content greatly decreases vapor loss rate by decreasing vapor phase
diffusion through the soil due to its effect on air-filled porosity.  For
compounds which are more water soluble, soil-water content may have varying
effects and may increase or decrease vapor loss depending upon the potential
for downward or upward movement of the chemical in the water phase.

     There is a definite need to develop and test models for predicting
volatilization of various organic chemicals from the soil.  These models
should include simultaneous movement of the chemical toward the surface by
diffusion and mass flow in the evaporating water, while also considering
changes in concentration from degradation of the chemical in the soil.

       A second serious gap in the knowledge of vapor behavior concerns the
extrapolation of laboratory data on relative volatilization rates to field
situations.  Volatilization models developed from physicochemical parameters
and controlled vapor loss studies in the laboratory should be calibrated with
field measurements of volatilization rates with the chemical under study
applied to the soil surface and incorporated within the soil to test
predictive capabilities.

     Vaporization from water-containing sediments or water-sediment systems
should be governed by the same factors affecting vapor losses from water after
considering the reduction in concentration of the chemical due to sorption on
the sediment.

     Success in predicting vapor loss rates with any model depends upon the
availability of reliable values for vapor pressures, water solubility, sorption
characteristic, and persistence or degradability of the chemical in the
environmental system whether it be soil, water, or sediment.  A need exists
for collection of basic information on chemicals to be studied utilizing
standard procedures so that data gathered by different investigators can be
utilized with confidence.

     It is impossible for all new chemicals to initially be examined in detail
for screening purposes.  Applying the baseline data on the dynamics of repre-
sentative compounds will be more practical in providing comparative data on the
photodecomposition and volatilization of the new chemical.  These comparative
data could become much more useful if a set of reference, or so-called
"benchmark," chemicals were established with representative properties.
                                      79

-------
Reference compounds can be chosen from the most widely used compounds for which
models could be validated against results obtained in field studies and model
ecosystems.

     Selection of benchmark compounds should be based on classes of chemicals
and their widespread uses.  Subdivision of the classes may become necessary
where widely differing octanol/water partition coefficients or other sorption
parameters are found within one class.  The physical and chemical constants for
reference compounds should be examined including vapor pressure, hydrolysis
rate as a function of pH, water/octanol partition coefficient, quantum yield
for photolysis, solubility, etc.  These criteria can be used to identify the
situation most likely to promote photodegradation or volatilization.  The
chemical behavior of the benchmark compounds should be determined in a set of
well defined natural systems chosen to represent as wide a range of environ-
ments as possible.  Benchmark data should be incorporated into simulations of
the expected dynamic behavior of the chemicals and validated against suitable
field data.  Other chemicals relative to the benchmark compounds should be
assessed by lifetime/concentration measurements for free and sorbed chemicals
and active metabolites in the critical test systems.  Associated chemical and
physical data can be assembled and used to predict behavior to provide simula-
tions of photodecomposition or volatilization patterns which can be examined in
the light of results obtained in test systems and to delineate the limitations
of these simulations.

     As the library of chemodynamic information grows and the models are
refined, deviations of new compounds from predicted behavior will be readily
apparent.  Deviations of new compounds from predicted behavior should be
examined closely since they may indicate where further research is needed in
assessing the behavior of new chemicals.
3.5  PHYSICOCHEMICAL AND BIOCHEMICAL TRANSFORMATIONS

     Synthetic chemicals introduced, deliberately or inadertently, into soils
or sediments may be subjected to one or more kinds of transformation.  Many
toxicants are volatilized and lost from the site.  A large number of
organic compounds are subjected to nonbiologically effected chemical changes;
such nonbiological alterations in structure are characteristic of many classes
of simple molecules.  Other organic substances that are present at the very
surface of soils undergo relatively common photochemical changes.  In addition,
organic and inorganic chemicals can be acted upon biologically with biological
or enzymatic reactions bringing about changes in the structure of synthetic
molecules.  These biotransformations have been noted in diverse soils and
sediments.

     The agents responsible for major biologically induced changes on soils
and sediments are usually bacteria or fungi microorganisms.  The microflora
stand out in the biological realm because of their role in the degradation of
synthetic chemicals, including most of the major changes that are deemed to be
of ecological and human health significance in these environments.  Although
abiotic processes are extremely significant in transforming synthetic toxi-
cants, the chief (and often the sole) means by which these compounds are
                                      80

-------
transformed to inorganic products is a result of raicrobial activity in aquatic
and terrestrial environments.  There are few documented instances where an
organic chemical, particularly one of some degree of complexity, is completely
converted abiotically to an inorganic product.  Microbial inhabitants appear
to be the chief agents for complete destruction of such chemicals though
fires may also lead to complete destruction.  Mineralization is the term used
to designate the total destruction of an organic compound and its conversion
to carbon dioxide and other inorganic products.  Mineralization is a
characteristic of microorganisms and is rarely noted in abiotic transforma-
tions.  Conversely, some inorganic pollutants are converted into persistent
organic compounds by microbiotic action in soils and sediments.

     Many synthetic organic compounds which are not mineralized or transformed
accumulate in the environment.  These persistent molecules may be hazardous to
humans, domestic animals, or wildlife (DDT and PCB's), create nuisance
problems because of the odor or taste they impart to waters (chlorophenols),
cause serious injury to crops grown in rotation (triazine herbicides),
increase the cost of municipal waste disposal (the synthetic polymers making
up a significant portion of the solid waste of cities), or their influence may
largely be associated with aesthetic deteriorization (plastic containers in
the countryside or some surfactants in waters).  The susceptibility to
biological and chemical transformation of each organic compound entering
soils and sediments must be determined in order to discover which chemicals
accumulate in soil and sediments creating potentially hazardous environmental
situations.

     Incomplete degradations, caused by biological or abiotic means, are often
of great practical concern.  The products of these incomplete reactions may
have greater toxicity than the parent molecules, exhibit toxicity to organisms
different from those affected by the original substance, show greater
persistence in soils and sediments than the parent chemical, or may be subject
to bioaccumulation greater than that of the parent molecule.  In addition, the
toxicological art is constantly growing and a product that today is deemed to
have no toxicological significance may subsequently be revealed to harm one
or more species.  As a result, the incomplete breakdown of chemicals that is
characteristic of abiotic as well as many microbial processes is of great
practical concern.  The mineralization of organic compounds must be establish-
ed.  Simple and inexpensive test procedures are adequate for assessment of
mineralization though the existing techniques must be validated.

     Classically, three general sources of biomass have been used for labora-
tory biotransformation studies.  One of these classes, a microbiologically
pure culture, is used in the study of biotransformations of an individual
compound by one species of microorganism.  Pure culture studies are attractive
from a microbiological standpoint, but the relevance of the results of such
studies is not always evident.  Enrichment cultures of microorganisms derived
from soils, waters, or sludges have also been used.  Studies performed with
enrichment cultures are often difficult to reproduce because of microbiologi-
cal variability but are attractive because of the environmental relevance of
the original biomass source.  The use of articifically mixed cultures (i.e.,
                                      81

-------
combinations of several pure cultures)  is less attractive than the other
biomass sources because data from studies performed are neither easily
explained microbiologically nor are they environmentally relevant.

     Some chemicals considered nonhazardous at environmental concentrations
may in fact be converted to highly toxic chemicals.  For example, dialkyla-
mines and nitrate at environmental concentrations rarely are problems, but
microbial and/or chemical reactions in waters, and soils lead to the formation
of carcinogenic nitrosamines from these precursors (Ayanaba et al. 1973;
Ayanaba and Alexander 1974).  Further,  an innocuous precursor may disappear
readily but convert to a persistent and possibly toxic product.  These
considerations suggest that major research is needed on the products of
biological and abiotic transformation of synthetic chemicals in soils and
sediments.  The need for identification should be tempered by the available
toxicological information.

     The transformation of products from nontoxic to toxic forms is evident in
the conversion of the fungicide thiram to a carcinogenic nitrosamine (Ayanaba
et al. 1973) and the conversion of the antifungal pentachlorobenzyl alcohol to
phytotoxic chlorinated benzonic acids (Ishida 1972).  The need for such
research is reinforced by observations of the conversion of a chemical
retained on sediments or soil surfaces into a product which is taken up in
biological food chains, such as the methylation of mercuric ions so that the
mercury retained by cation exchange on colloidal surfaces is available for
uptake into animal tissues (Wood 1971).  Few of the products formed from the
chemicals that are released into waters and soils have been identified, and
the few established products are typically those derived from the microbial or
abiotic transformation of pesticides.  Chemicals that originate in industrial
or domestic operations have received essentially no attention.

     It is important to determine whether particular synthetic organic
compounds are acted upon by biotic or abiotic mechanisms to understand the
kinetics of the process, possible products, and information on the likelihood
of formation of polymeric, complexed or "bound" residues of the chemical.
Microorganisms may use an organic molecule as a nutrient and convert much of
the carbon in the chemical to carbon dioxide, provided oxygen is available.
Simultaneously, the organism will assimilate  some of the carbon in the
molecule into microbial cells.  With information that microorganisms are
responsible for a process, the carbon not accounted for as carbon dioxide or
known products can frequently be ascribed to  the formation of biomass which
is of no toxicological significance.  If the  mechanism is abiotic, no cellular
constituents are generated and additional chemical information is required.
Further, knowledge that microorganisms are responsible for transformation
will aid in kinetic investigations in so far  as the rates of bacterial
processes parallel the logarithmic growth of  these microorganisms, at least
under ideal circumstances.

     Relatively little is known about the role of  soil- and sediment-dwelling
macroinvertebrates in directly causing transformations of synthetic chemicals
or indirectly affecting these processes as they influence the microbiota that
                                      82

-------
carries out the transformation.  In addition to whatever role these inverte-
brates might play in mechanically transporting materials on or within
soils and sediments, research is required to determine the role of these
invertebrates in transforming synthetic chemicals,  either directly or
indirectly.  Attention might be given to their role in the transfer and
transformation of litter and detritus, and in reduction of particle sizes to
facilitate microbial action.

     Knowledge that a chemical is transformed in soils and sediments is not
sufficient for regulatory action.  Research is needed to establish the types
of kinetics and kinetic expressions of biological and abiotic transformations
to allow meaningful predictions of the environmental behavior of chemicals.

     Field evidence shows that chemicals which are destroyed rapidly in one
environment are largely or wholly resistant in other environments.  This is
evident in the longevity of polysaccharides and other natural polymers in peat
as long as the peat remains under water; once the area is drained the
polymers are readily converted to carbon dioxide and other simple products.
Similarly, hydrocarbons endure in nature in anaerobic regions but they are
often quickly destroyed in aerobic environments (Alexander 1973).  The
effects of environmental factors on biological and abiotic transformations in
natural environments should be explored.  Although considerable attention has
been given to temperature and pH, the significance of surfaces in sorbing or
complexing toxic chemicals, the importance of anaerobiosis, the effects of
chemical mixtures, and the role of plants require further characterization.

     Sorption of synthetic chemicals to colloids in soils and sediments as
protection against biodegradation requires attention.  Such sorption
frequently increased the resistance of organic compounds to attack.  For
example, proteins bound to clays are frequently depolymerized by enzymes far
more slowly than the same protein found in solution (Esterman et al. 1959).
Similarly, the complexing of a normally degradable substance may render it
refractory to microbial attack.  Polysaccarides are less available to
enzymatic degradation by microorganisms when they are bound to polyaromatics
of environmental importance than when they are free of the binding agent.  In
addition to surface adsorption, clay minerals may entrap organic substances
within their lattice structures and account for the occasional resistance of
chemicals that are usually quickly destroyed.

     The role of one chemical or a mixture of chemicals in affecting the
transformation of specific compounds of interest requires definition.  Few
studies have been concerned with the interaction of chemicals as they affect
transformation reactions in nature.  Thus, one chemical may be used by the
active population, and these microorganisms may then not transform the primary
chemical of concern, although they would have done so in the absence of the
second compound.  Conversely, if co-metabolism is important, refractory
organic compounds might be co-metabolized more readily if the environment is
rich in readily biodegradable compounds that support microbial proliferation.

     The significance and means of enhancing co-metabolism should be esta-
blished.  Co-metabolism refers to the metabolism by a microbial population of
                                      83

-------
a chemical it cannot use as a nutrient source.  A compound acted upon in this
manner, though transformed biologically, will not be converted to microbial
cell constituents and therefore the kinetics of the transformation will not
parallel microbial growth inasmuch as the population is not increasing in
response to addition of the chemical to water or soil.  Furthermore,  chemicals
undergoing co-metabolism are not mineralized (Alexander 1979).  Hence,
products derived from the parent molecule will be likely to occur.  The small
amount of existing knowledge of co-metabolism suggests that the parent
toxicant is not significantly modified enzymatically and it is likely that the
products are themselves toxic.  Surprisingly few data are available on the
precise role of co-metabolism in natural environments, but the foregoing
considerations suggest that additional attention ought to be paid to this
phenomenon.

     Should the chemical be acted upon largely or solely by biological
mechanisms, the lack of increase in microbial numbers or biomass, as is
typical of co-metabolism, will be reflected in a prolonged persistance of the
toxicant provided that the responsible microbial population is initially
small.  It is possible that co-metabolism can be enhanced by stimulating the
growth of the responsible population.  Such stimulation would require the
addition of organic nutrients other than the toxicant, and the possibility of
this type of environmental manipulation to enhance the destruction of
toxicants should be explored.  In addition, the kinetics of co-metabolic
processes in nature have not received attention.  Since this phenomenon
appears to be wide and significant, the kinetics need to be delineated.

     When measuring the biodegradability of particular compounds for
regulatory, developmental, or research purposes, a simple model of a natural
ecosystem is employed in order to maximize convenience and minimize expense.
Typically, one or very few concentrations of the test chemical are used and
characteristically the level of the chemical in the laboratory model  is much
higher than found in nature.  However, Michaelis-Menten kinetics suggest that
the rate in nature would be lower than in the laboratory.  The reduction in
rate may be directly proportional to the lowered concentration of the chemical
(as suggested by the linear portion of the Michaelis-Menten plot), but the
concentrations equivalent to the linear portion of the plot are unknown for
nearly all synthetic chemicals entering soils and sediments.  Moreover, some
evidence now exists that there is a threshold for biodegradation below which
there is no microbial destruction (Boethling and Alexander 1979).  Alterna-
tively, the concentration in the laboratory tests may be high enough  to
inhibit microbial transformation that would otherwise occur at the low
concentrations in nature.  Thus, a chemical in the laboratory model either
would be decomposed at a greater (or sometimes lesser) rate than at the lower
concentration in nature, or it may be decomposed in the laboratory model but
not at all in soils and sediments.  Hence, the influence of chemical  concentra-
tion on the rate of its biological transformation in natural environments needs
to be determined.

     Alternate enzyme substrates in waters or soils or the presence of surfaces
upon which a synthetic chemical can be concentrated may alter the apparent
concentration dependence of rates of biological transformation so the effect of
                                      84

-------
additional chemicals and surfaces in regulating the rates in nature must be
evaluated.  Plausible microbiological explanations can be advanced to explain
the concentration dependence of biodegradation and the presence of a threshold
concentration below which microbial attack does not take place, but these
hypotheses have not been tested under laboratory conditions simulating natural
environments.  The effect of inoculum size on rates of microbial transforma-
tions in laboratory tests must also be evaluated.

     Organic compounds as well as potentially toxic inorganic chemicals are
assimilated from soil by plants.  These chemicals may affect animals feeding  on
the plants and thus be important to herbivores or grazers as well as to other
organisms in food chains.  If the toxicants enter food crops, they may affect
humans.  Except for pesticides and heavy metals little attention has been
given to the uptake of toxic molecules by higher plants.  Research on those
toxics relative to food chains is of great importance and should consider
factors (including soil type) affecting assimilation by plants and animals or
classes of molecules involved.

     Many toxicants as well as other organic compounds are resistant to
destruction or appreciable modification in natural environments.  The list of
resistant molecules is long but as yet incomplete.  These chemicals are
employed for many purposes and the significance to man, wildlife, and natural
ecosystems may sometimes be appreciable.  The reasons that certain organic
compounds are resistant in all environments (that is, are intrinsically
resistant to degradation) and the reasons others persist in one environment
but are destroyed in another must be established.  Some speculation on the
basis for resistance of recalcitrant molecules to microbial attack exists,
but few of the hypotheses have been tested.

     An extremely limited number of correlations between chemical structure
and microbial transformations in soils and sediments have been made (Alexander
1973).  It is essential to determine what structural characteristics are
correlated with, or account for, resistance to microbial mineralization or
co-metabolism.  This type of information should permit regulatory agencies to
determine the chemicals that are more likely to be environmental problems
because of their persistence and availability for biomagnification.  The
information will also be of use to industry as it seeks new chemicals that will
not endure in waters and soils.  The few generalizations that exist in this
area relate to a limited number of surfactants, insecticides, herbicides, model
aromatic compounds, and aliphatic acids.  From the very few investigations in
this field, it appears that slight changes in the structure of many molecules
appreciably alter the rate of their destruction in natural environments.  These
changes allow a molecule which previously was not able to support microbial
growth to be co-metabolized into one which is readily decomposed.

     With regard to the recommendations above, generalizations must be provided
on classes of molecules which are susceptible to mineralization or co-metabo-
lism and the effect of concentration, aeration, or surfaces as well as other
environmental factors on biodegradation of classes of chemicals.  In the
absence of such generalizations each one of the vast number of compounds of
concern to EPA (and regulations for TSCA) would have to be evaluated
                                      85

-------
separately.  These generalizations should ease the task of regulatory agencies
and private industries in determining whether particular compounds or classes
of compounds pose environmental problems, and the information should facilitate
identification of those organic molecules which may be serious environmental
pollutants.
3.6  CHARACTERIZATION OF SOILS AND SEDIMENTS

     As discussed in preceding paragraphs, the fate of potentially toxic
substances in soil systems is dependent on a variety of factors,  including but
not limited to:  sorption-desorption phenomena, leaching and diffusion
phenomena, chemical and biochemical transformation, photodegradation,  and
volatilization.  The magnitude and nature of the effects of the last four
factors are largely influenced by the nature and extent of sorption-desorption
reactions.  These reactions are in turn regulated by the physical and chemical
nature of the sorbent.  Section 3.6 identifies major properties of soils and
sediments which are of importance relative to the sorption-desorption of
chemicals and characterization of the leaching, diffusion, and environmental
fate of these chemicals.

     Aquatic sediments are similar in some respects to terrestrial soils.
Erosion of soils is a major source of sediments.  It should follow that
sediments would be similar in many aspects to the soils from which they
derive and this is the case in some instances.  Erosion is a size-selective
process; smaller particles are most vulnerable to erosion.  Frink (1969) found
that the clay content of a lake sediment was fivefold greater than that of the
upland soils, while the sand content was substantially reduced.  The most
obvious difference between soils and sediments is perhaps the most important:
soils are only periodically exposed to a mobile water phase; the extent of
hydration of the soil particles diminishes over the prolonged periods during
which they are essentially dry.  Conversely, sediments are in continuous
contact with the overlying water.

     Soils and sediments are comprised of organic and inorganic materials
whose physical and chemical properties affect sorption and desorption of
dissolved constituents.  The three major components of soil which have been
reported to be of significance to sorption are clay minerals, organic matter,
and hydrated metal oxides (Hamaker and Thompson 1972).

     In addition, soil texture or particle diameter distribution is of great
importance.  Soils and sediments are generally described by their sand (2 mm
to 50 pm diameter), silt (50 ym to 2 ym), and clay (<2 pm) content.  There are
12 named textural classes (Ahlrichs 1972).  Texture is important in the
evaluation of potential fate of materials because specific surface areas of
soils of various textural classes vary markedly.  Typical values of specific
surface areas for different textural classes have been give by Kohnke (1968).
The surface area of coarser materials can be greatly increased as a
consequence of a high humus content.  However, for sediments, sand or larger
size material, surface area is relatively unimportant with respect to sorption
                                      86

-------
of compounds and may be physically excluded from sorption studies.  Later,
mathematical corrections can be applied to relate the experimental results  to
those conditions existing in the natural system.

     Clay minerals present in soils and sediments may also differ in a number
of important properties (Weber 1972).  Both clay mineral composition and
texture are of importance with regard to hydraulic conductivity though the
physical test conditions are of controlling importance.

     The organic matter present in soils and sediments may greatly modify the
sorption properties of clay minerals.  Hamaker and Thompson (1972) cite the
results of a number of studies which report increases in the sorption coeffi-
cient, K
-------
     No single soil or sediment parameter can account for the observed
sorption of various compounds.  There appear to have been no systematic sorp-
tion studies in which both a range of sediments and soils and of sorbates have
been employed.  Helling (1971) has reported one related leaching study in
which a matrix of 13 pesticides and 14 soils was investigated and multiple
regression analysis employed to elucidate the relative importance of selected
soil parameters affecting pesticide movement.  The results of similar
investigations with respect to sorption reactions would be of substantial
value in defining the limits of predictability of sorption-desorption
phenomena for soils and sediments, as well as in providing information on the
sorption behavior of a standard series of compounds.

     Such information is necessary to provide a data base with which to compare
the results of other studies, including those for screening new materials.
This information is of particular importance in the case of sediments for which
it may be impossible to provide a "bank" of materials to be used as sorbents.
Soils may be air-dried prior to the sieving and homogenization steps necessary
for preparation of a characterized materials suitable as a reference sorbent.
Conversely, if their significant properties with respect to sorption reactions
are to be maintained, sediments cannot be dried.  Size separation is achieved
by sieving and settling of slurries, and prolonged storage of sediments can
result in major alterations as a result of oxygen depletion, biochemical acti-
vity and coagulation.  It is recommended that the time lapse between sampling
and testing be limited, and the conditions of sampling and testing be
rigorously controlled.  To assure validity of the data, sorption of selected
reference compounds of known behavior should be simultaneously measured.  Such
an approach will likely provide data which could be extrapolated to a wider
variety of conditions.  It is also recommended that the effects of drying of
sediments on the sorption of various classes or organic compounds be
established.
3.7  LAND DISPOSAL OF WASTES AND CHEMICALS

     Land disposal is not a basic process affecting transport and transforma-
tion in the same sense as sorption, diffusion or volatilization, but the land
disposal facility is a source of potential movement of materials into soil,
air, and water.  The physical and chemical characteristics of concentrated
wastes disposed directly on land have such a significant effect on the proce-
dures for studying basic processes that land disposal is treated as a separate
topic in this section.  Additionally, research on site monitoring and site
selection (not discussed elsewhere) are needed both to improve disposal
practice and to check the effectiveness of new practices.

     Land disposal involves a number of practices including landfilling, land
cultivation and spreading of microbially degradeable wastes, and overland
runoff treatment of dilute wastes.  There is a great range in the physical and
chemical characteristics of wastes, and in the characteristics of aqueous
extracts of wastes and the non-aqueous liquids that drain more or less freely
from the waste.  Procedures for measuring sorption-desorption properties of
substances have been developed for aqueous solutions containing low levels of
the materials of interest as well as low levels of other dissolved substances.
                                      88

-------
The work by Griffin and Shimp (1978) is an example of a sorption study that has
been conducted with high levels of other substances in solution.  Leachates and
other waste liquors are known to have higher levels of dissolved materials
(Houle et al. 1978), are more viscous, and may have one or more non-aqueous
phases.  Measurement of sorption-desorption parameters for single substances
from such liquids is expected to require substantial modifications of current
procedures.  Aeration status in the soil at the disposal facility must be
considered, and if wastes will be subject to leaching under anoxic conditions
the waste must be similarly leached or extracted.  If waste leachates/fluids
will pass through anoxic and oxygenic soil, adsorption must be measured under
both conditions.

     A related problem is development of an appropriate background solution
for sorption studies.  Background solutions for municipal-waste-landfill-
destined substances would be different from those of substances destined for
land-spreading or industrial waste landfills.  Some work has been done with
municipal solid waste landfill simulants (.Mitre Corporation 1978), but none for
simulating industrial waste landfill leachates which result from the mixing of
large numbers of different wastes.

     The permeability of undisturbed and compacted soils at disposal facili-
ties is measured using water as the liquid even though such facilities
commonly hold mixtures of water and organic liquids.  Lee (1978) has shown
that mixtures of water and organic solvents flow through clay mineral samples
at different rates than does water alone.  This effect must be considered when
designing landfills and lagoons, and when attempting to predict leaching rates
at such facilities.

     Another route for loss of materials from disposal facilities is
volatilization and vapor phase transport.  The amounts and rates of loss by
this route are not known; some studies of vapor concentrations have been made
(Oberacker 1978; Arozanena 1978) but available information is not sufficient
to determine whether amounts lost by this route are substantial enough to
warrant concern about specific types of disposal facilities.  Farmer et al.
(1978) reported a procedure for estimating flux from low-to-moderate vapor
pressure, low-solubility organic wastes, using data on soil air-filled pore
space and on the vapor pressure of the waste constituents.  The procedure has
not been tested in the field and it is not known how well the method will
predict vapor flux through soil covers over organic wastes with high vapor
pressure.  Until this tool is further developed and field tested, evaluation
of transport from land disposal facilities into the atmosphere will be
difficult.

     Problems relative to collecting and interpreting data on chemical
composition of waters in saturated and unsaturated zones of the soil will
limit evaluation of aqueous transport from disposal facilities and collection
of information for developing methods for evaluating disposal facilities.
Devices for extracting water samples from unsaturated soils have been
developed for agricultural use but have not been extensively tested on
leachates from disposal facilities under field conditions.  Cartwright et
al. (1978) have installed pressure/vacuum lysimeters under and around
                                      89

-------
landfills in central Illinois and report success with the technique.  A
closely related problem is taking a groundwater sample representative of the
chemical composition of water at specific points in the formation.   Regarding
the numbers of well volumes to be removed from a monitoring well before the
sample is collected, there are recommendations ranging from 3 to 10 well
volumes, or pumping until electrical conductivity is constant.  In  one
instance (Gibb et al. 1978), the concentration of zinc in solution  in the
eighth well volume pumped was 20 percent of the original value and  appeared
not to have stabilized.  Until this uncertainty is resolved there will be
doubt about the accuracy of monitoring well data which will inhibit
development and verification of predictive methods for pollutant movement.

     Even when accurate groundwater monitoring well data can be collected at
land diposal facilities, the problem of predicting concentrations at a point
of water withdrawal distant from the facility still remains.  Much  work
related to this topic has been conducted, but the information has not been put
into the form of a procedure useful under different hydrologic conditions and
different types of disposal facilities.

     Finally, tested procedures for predicting the impact of a disposal faci-
lity are not available.  Although soil is known to have significant capacity
for removing constituents from waste leachates, the rates and amounts of
removal cannot be quantitatively predicted, nor can the long-term leaching
behavior of the waste (the input term) be simulated.  The lack of this
capability promotes a very conservative approach to design of disposal facili-
ties and makes it impossible to determine the maximum amount of waste loading
that may be placed in a facility with a given amount and type of soil between
the facility and water bodies that may be impacted.  A long time (~10 years)
will be required for development of conceptual models for this purpose though
empirical procedures can be developed in a shorter time (U. S. Environmental
Protection Agency 1978).
3.8  THE USE OF MICROCOSMS IN TRANSPORT AND TRANSFORMATION RESEARCH

     Controlled model ecosystems, or microcosms, may have the potential for
revealing certain information regarding the transport and fate of chemicals in
the environment.  The degree or extent to which this information can be
extrapolated to the macroenvironment is markedly dependent upon how well the
microcosm simulates actual transport and transformation processes operative
in natural systems.

     It is impossible to incorporate a full range of environmental dynamics in
a bench-scale microcosm system.  Larger systems which might provide a
reasonable simulation of a full range of conditions are necessarily more
complex and costly, but promise better reliability with respect to at least
qualitative prediction of the fate, transport, and potential impacts of
chemicals in the environment.  Smaller and less complex microcosms may provide
insight into the fate and movement of a compound within one compartment of the
environment, e.g., through a particular soil column.
                                      90

-------
     A significant disadvantage of small, single-compartment microcosm systems
is the lack of a method for coupling with other environmental compartments
even though the dynamics of chemicals in the environment are remarkably
influenced by intercomponent phenomena.  Examination of only intracomponent
factors in simple bench-scale microcosms may lead to conclusions which prove
erroneous in natural systems because of intercomponent reactions.  The
limits of extrapolation from small and simple microcosms must be fully
recognized.

     Given the previously discussed constraints, appropriate microcosm systems
may prove useful as research and screening methods for developing a better
understanding of the transport and transformation of chemicals in the
environment.  If properly designed, such microcosms might:

     «  Provide preliminary information on anticipated losses via various
        routes, the rates of those losses, and the ultimate consequences in
        terms of concentrations in soil, air, water and biota;

     •  Demonstrate the utility of a given monitoring procedure or instrument
        in a practical setting;

     •  Provide opportunity for determining precision of predictions
        of environmental transport and transformation from elementary
        laboratory data on physicochemical and biochemical properties;

     •  Demonstrate the effects of different environmental conditions, altered
        components, or alternative routes of entry of chemicals into potential
        target and non-target environments; and

     •  Provide for testing of interactions between components and processes
        in a more complex system approaching respresentation of actual
        environments.

     Microcosm systems can avoid the risk of irreversible damage to natural
ecosystems or human beings since they are "contained" or "bounded" systems.
Structure, composition, operations, and environmental conditions are under the
control of the investigator and may be arranged for specific experiments.
This control of experimental factors allows attention to be directed to the
processes involved in transport and transformation and to the inherent vari-
ability derived from natural biologic systems, rather than to uncontrolled
environmental forces.  Hundreds or even thousands of chemicals might be tested
in a microcosm, whereas only dozens of chemicals can be so thoroughly inves-
tigated in the field.  The results of transport and transformation studies in
a rigorously designed and controlled microcosm can be expected to reflect the
outcome of the properties of the chemicals and interactive processes involved,
particularly when employed in a comparative manner.  Systems can be made
site-specific (e.g., soil core microcosms) and permit certain biologic effects
to be assessed directly with exposure to a chemical (Van Voris et al. 1978).
However, microcosms are not self-sustaining, cannot contain all processes
represented in the environment, and may not operate validly for long time
periods.  There are also practical limits to compexity and scale.
                                      91

-------
     To the extent that microcosms validly represent real environments,  they
can be used to test mathematical models developed from basic properties  and
potentially reveal discrepancies in the model or disclose unknown processes
for transport and transformation.  In particular, this approach requires
rigorous process and systems analysis thus strengthening comprehension of the
general principles involved in transport and transformation.

     Caution should be exercised because "verification" of mathematical  models
in microcosms does not ensure their validity for predicting  conditions  in
the real environment; they must be modified and specifically tailored to
account for differences between the micro and macro systems.  This is particu-
larly true if such models are both calibrated and verified with microcosm
data.  There is a definite need for research on the extrapolation from small
microcosms to more complex microcosms and to actual environmental systems.

     Preliminary studies reviewed in Gillett and Witt (1979), and evaluated by
a peer panel in great detail, have demonstrated that there are several terres-
trial microcosm systems capable of being employed for testing environmental
movement of chemicals (leaching or soil percolation, volatilization)  and
transformation (loss of parent material, appearance of metabolites and "bound
residues"), including bioaccumulation.  The soil core microcosm (Draggan 1976;
Ausumus et al. 1979) is the simplest of these systems and shows the greatest
promise for use in screening for transformation and soil movement.  It is in
the stage of interlaboratory testing for both fate and effects assays of
pesticides, toxic substances, and industrial effluents.

     The Cole/Metcalf systems (Cole et al. 1976; Cole and Metcalf 1979)  have
been tested with about 20 chemicals, the CERL Terrestrial Microcosm Chamber
(Gile and Gillett 1979) with about a dozen, and the microagroecosystem (Nash
and Beall 1979) with about ten.  These are much larger and more expensive
systems and include more extensive sets of tests.  These systems currently
lack adequate research bases for immediate application to screening systems.
The degree of representation of transport processes (scaling, boundary
conditions) has not been thoroughly investigated.  Transformation processes
have been thoroughly studied in the large microcosms and only preliminarily in
soil core microcosms.  Multi-organism and multi-media tests were judged
more representative of actual environmental disposition.  Achievement of
environmental conditions (media flux, stochastic dispersion of energy and
media flux) approximating reality has not been systematically sought, although
some experimentation is in progress.  Thus far, the systems lack demonstrable
photochemical processes.  The problems due to added "large" animals (and
their attendant effects on structure and abiotic processes) have not been
investigated.

     Many of these difficulties can be studied and possibly overcome;
alterations have already been made in some systems as a result of this
review.  Full endorsement of terrestrial microcosms cannot be given without
further improvement of the present state-of-the-art though many observers
believe that adequate laboratory systems can be achieved.
                                      92

-------
     The Gillett and Witt (1979) state-of-the-art review revealed specific
areas where information is needed.  Continued development is required to
assess the degree of environmental control necessary to achieve outcomes of
transport and transformation representative of field results.  Criteria for
operation need to be obtained that will permit evaluation of microcosms of
different structure, composition, or operational variables on a commensurate
basis.  The precision in any given microcosm system needs to be validated by
interlaboratory testing and compared against field studies specifically
designed for that purpose.  Current verification is achieved by comparison to
results of monitoring or fate studies under broadly differing circumstances
than the microcosm test.  Precision and accuracy in microcosm studies need
only be similar to field results; because of expectations of greater accuracy,
microcosms have often been viewed as unsuccessful.

     Mathematical models should be constructed that would predict the outcome
of transport and transformation studies from simple laboratory measurements of
basic properties of the chemical and microcosm studies compared to the model.
Innovative means of data reduction need to be introduced to handle the
relative richness of microcosm outputs.  Cost-effectiveness studies of
microcosms should be conducted to determine if various sets of tests can be
replaced with the simple soil core microcosm.

     The foregoing discussion indicates several thrusts of terrestrial micro-
cosm research.  The most appropriate screening systems need to be validated by
interlaboratory testing on the same chemicals from broadly differing classes.
Subsequently, validated microcosm systems need to be verified in specifically
designed field studies.  The variation between laboratory and field systems
needs to be carefully determined, particularly with respect to the impact of
variables such as temperature, wind and water flux, soil type, and mode of
application of the chemical.  The range of results for transport and transfor-
mation need to be more thoroughly explored within specific classes of chemi-cals
(e.g., an extensive homologous series) and between widely differing chemical
classes for which some environmental data in the field are available.  Data
management models and mathematical models of transport and transformation need
to be set forth.  Operational simplification should be investigated, particu-
larly aimed toward developing the necessary and sufficient information to
predict likely environmental fates.

     One of the chief objectives of microcosm technology is the development of
systems which can be used to carry out research on interactive processes and
thereby predict the results of competing properties expressed as elementary
characteristics of a chemical (e.g., sorption, solubility, volatility, sites
of chemical transformation).  Of primary concern is whether or not realistic
air and water flows could be achieved such that transport processes would be
adequately simulated in the laboratory.  Moreover, the lack of modeling is a
major concern.

     Clearly, the more complicated systems appear to have greater utility in
this phase of microcosm application, since temporal and spatial studies could
                                      93

-------
be performed.  Hence, one might divide terresterial microcosms according to
whether they would be applied to screening (highly standardized,  limited
focus) or to research on processes such as vapor loss,  leaching,  bioaccumula-
tion, and biotransformation (more complete as to processes and levels of com-
plexity included, broadly focused).  The former might be tested against many
chemicals and involve variable substrates and environmental conditions.  The
latter would be employed with only a few chemicals and carefully constructed
and operated for comparison to a specific field situation.  It is through this
latter approach that Gillett and Witt (1979) believed that mathematical
models might be extended to the field via microcosm studies and complete the
predictive sequence from basic laboratory bench data on physical and chemical
characteristics.  In this iterative process, the reviewers felt that the
microcosm would thereby be an intermediate technology permitting extension
ultimately from the basic data with relatively few confirmatory tests in the
microcosm or field.  Such a predictive capability would greatly reduce the
cost of testing for screening purposes, provide better understanding of the
complexity of the interacting environmental processes,  and permit attention to
be directed at application of mathematical and laboratory model systems to
other aspects of chemical control.

     Terrestrial microcosms for predictive techniques suffer the same problems
in terms of data deficiencies as for screening and testing techniques.  Due to
the predictive focus of this report, the absence of adequate modeling efforts
is particularly acute.  Moreover, the validation and verification efforts
would necessarily be more extensive, and more expensive, for this class of
investigation.  The problems of necessary and sufficient studies of scale
(size), scope (relative complexity), and data "richness" are amplified in
larger and more complete systems.  Criteria need to be established for
determining just how accurately a microcosm must represent field experience
given the problems of spatial and temporal heterogeneity and instability of
environmental variables in the field.  These criteria must then be applied to
determine the need for accuracy of environmental controls in microcosm
operations, the detail of chemical tests to be applied, and the extent of data
resolution.  It would not be cost effective to demand of microcosm systems a
degree of accuracy unachievable in natural systems.

3.9  SUMMARY OF RESEARCH NEEDS

     The soil and sediment group did not prioritize their research needs
because it was felt priorities would be impossible without some type of
budget framework.  As a result, the following recommendations are listed in
the order in which they appeared in Section 3.

      1.  Relate the structure of chemicals to sorption over a wide range of
soils and sediments, and the effects of soil/sediment surface and structural
properties to sorption/desorption reactions. Study the prediction and
measurement of linear free energy relationships between chemicals and soil/
sediment surfaces.  These studies should include evaluation of other
particulate solids - such as fly ash, plant residues, street dirt, etc., -
which function as sorbents and transport media for toxic chemicals in the
environment.
                                      94

-------
     2.  Develop an extensive reference library of sorption/desorption data on
different chemicals and soil/sediment systems.   This should include reference
soils and sediments and reference classes of chemicals.

     3.  Quantify and evaluate the significance of long-term sorption and
desorption kinetics.  These studies should include characterization of the
chemical state of the sorbed species to provide a better basis for predicting
and understanding sorption and transformation reactions  in the environment.

     4.  Determine the effects of various types of soil/sediment preparation,
sterilization, and storage techniques on the sorption/desorption, organic, and
biochemical character of these sorbents.

     5.  Determine the effects of cyclical variations in soil temperature and
moisture content.

     6.  Verify the applicability of normalizing sorption on sediments and
soils to carbon content, Koc, and/or octanol/water partition coefficient.
Determine with which types of chemicals and/or soils these predictive methods
are inappropriate.

     7.  Determine and quantify the importance of mixing processes between
sediments and overlying water, and between soils and overlying air, as they
affect uptake and release of contaminants.  These processes include
hydro-dynamic dispersion, physicochemical disruption, and bioturbation.

     8.  Develop mathematical models for predicting transport of soil
and sediment materials - and associated chemicals - through the atmosphere and
water column, and the possible differential accumulation of these materials in
various compartments of the environment.

     9.  Study and establish methodologies for measuring chemical sorption by
ferric oxides in sediments, as affected by redox potential changes.  Study how
widespread the occurrence of sorption onto ferric oxides is in nonsediment
environments, such as soils and drainage systems.

    10.  Improve reliabilty and availability of vapor pressure data on
specific chemicals to allow assessment of volatilization losses from soil,
water, and plant surfaces.

    11.  Improve data collection and uniformity of laboratory and field
leaching simulations.  To predict leaching in landfills determine how sorption
reactions in soils are influenced in the presence of the leachate.  Improve
capability for predicting leaching in the natural environment by accounting for
spatial variability.

    12.  Develop thorough knowledge of the chemistry of metals in soils and
sediments to enable long term predictions of the transport of heavy metals,
including the effects of added complexing agents in regard to release of metals
from amended and unamended soils and sediments.
                                      95

-------
    13.  Develop predictive models for leaching and diffusion which include
appropriate degradation terras.

    14.  Develop, refine, field test, and validate models for prediction of
transport of chemicals by run-off from non-point sources.

    15.  Determine the particular circumstances under which photodegradation on
soils and sediments is important, and the physicochemical properties and/ or
structure-activity relationships that are needed to predict photochemical
transformations.

    16.  Determine conditions that are appropriate to simulated laboratory
tests, and which must be measured in field experiments, in order to generate
photodegradation processes of environmental significance and identify their
pathways and products.

    17.  Characterize the photodegradation of chemicals on airborne particu-
laltes or in the vapor phase following volatilization from soils and/or sedi-
ments.

    18.  Develop and test models for predicting vapor loss of various organic
chemicals from soils.  These models include simultaneous movement of the
chemical toward the surface by diffusion and mass flow in evaporating water,
while at the same time considering changes in concentration from degradation
of the chemical in the soil.

    19.  Determine an approprite set of reference or benchmark compounds that
can be exhaustively tested in field studies and model ecosystems to permit
testing, calibration, and verification of predictive mathematical models.

    20.  It is important to determine the relative degree to which particular
synthetic organic compounds are acted upon by biotic or abiotic mechanisms,
what structural characteristics are correlated with or account for resistance
to abiotic transformations and to microbial mineralization or co-metabolism,
and how mineralization and/or co-metabolism reactions can be enhanced.

    21.  There is a need to determine the role of invertebrates in trans-
forming synthetic chemicals, either directly or indirectly.  Research is
also needed on how plants in soils, and macrophytes in waters, transport and
transform synthetic chemicals.  Uptake of toxic chemicals by plants, and
thereby the significance of soils as a reservoir or source of chemicals to
animals, should be established.

    22.  The types of kinetics and kinetic expressions appropriate to biotic
and abiotic transformations must be determined to allow for meaningful pre-
dictions of environmental behavior of chemicals.

    23.  The effects of environmental factors on biological and abiotic
transformations in natural environments should be explored.  The sorption of
synthetic chemicals to colloids in soils and sediments as protection against
biodegradation requires attention, and the role of oxygen and anaerobiosis in
                                      96

-------
biodegradation must be clarified.  The role of one chemical or a mixture of
chemicals in affecting the transformations of specific compounds of interest
requires definition, and the influence of concentration on the rate of
abiotic and biological transformation of a chemical in natural environments
needs to be determined.

    24.  Test the suitability of existing sorption/desorption procedures for
land disposal of concentrated chemicals using liquid wastes and leachates/
extracts of wastes to identify problems due to the unique physical/chemical
nature of the wastes and to identify needs for changes in the procedures.
Develop a simulant for municipal landfill leachate so the sorption/desorption
characteristics may be predicted for single components that may go to
landfill.

    25.  Develop information or test procedures for predicting changes in
conductivity of landfill soils/clay liners due to organic and inorganic
constituents of concentrated wastes.  Evaluate reactions and transport of
non-aqueous liquids in soils.  Develop methods for predicting the rates of
movement through soil of constituents of wastes and combinations of wastes
placed in specific disposal facilities.

    26.  Monitor the air around land disposal facilities to determine whether
volatilization is producing appreciable air pollution.  Develop methods for
predicting rates and amounts of release of vapors from disposal facilities.

    27.  Evaluate devices for taking liquid samples in unsaturated soils at
disposal facilities.  Develop methods for taking groundwater samples that are
representative of the chemical composition of water at a specific location in
the water-bearing formation.  Using chemical composition of groundwater at
specific locations around the disposal facility as inputs, develop methods for
predicting the composition of waters at locations distant from the disposal
facility and the composition of waters withdrawn from wells (of various design
and yield) at such locations.

    28.  Through interlaboratory testing, determine the precision with which
microcosm experiments can be replicated.

    29.  Determine the accuracy with which microcosms can be used to predict
transport and fate of chemicals in field situations, taking into account the
variability of replicated field tests.  These data are required to evaluate
the ultimate utility of microcosms.  Determine the applicability of microcosm
systems for predicting the transport and fate of several chemicals within each
of a wide variety of chemical classes.  Ranges of results both within chemical
classes and between classes need to be determined.

    30.  Ultimately develop a mathematical model which is capable of reliable
prediction of the transport and fate of a chemical in the environment based
upon measurable properties of the chemical and of the major components of the
environment, and a reasonable knowledge of the dynamics and interactions of
transport and transformation reactions.

    The Soils and Sediments Group participants are listed in Table 3-1.
                                      97

-------
               TABLE 3-1.  SOIL AND SEDIMENT GROUP PARTICIPANTS
Non-Agency Participants

Martin Alexander
Department of Agronomy
Cornell University
Ithica, N.Y.

Herbert E. Allen
Department of Environmental
  Engineering
Illinois Institute of Technology
Chicago, IL

Walter J. Farmer
Department of Soil Science
University of California
Riverside, CA

Farley Fischer
National Science Foundation
Applied Service and Research
  Applications
Washington, D.C.

Charles Helling
U.S. Department of Agriculture
Beltsville, MD

Shahamat Khan
Institute of Chemistry and
  Biology
Agriculture Canada
Ottawa, Ontario
Canada

G. Fred Lee
Department of
Colorado State University
Fort Collins, CO

Haines Lockhart
Eastman Kodak Company
Rochester, N.Y.
EPA Representatives

Allan Crocket
Environmental Monitoring Systems
  Laboratory/
Office of Research and Development
Las Vegas, NV

James Dragun (Lead)
Office of Toxic Evaluation/
  Office of Toxic Substances
Washington, D.C.

James Gillett
Corvallis Environmental Research
  Laboratory/
Office of Research and Development
Corvallis, OR

Samuel Karickhoff
Environmental Research Laboratory/
Office of Research and Development
Athens, GA

Martin Kovacs
Office of Pesticide Programs/
Office of Toxic Substances
Washington, D.C.

Michael Roulier
Municipal Environmental Research
  Laboratory/
Office of Research and Development
Cincinnati, OH

Claire Welty
Office of Water Management/
Office of Water and Waste Management
Washington, D.C.
                                     98

-------
                            TABLE 3-1   (Continued)
Non-Agency Participants

William Spencer
U.S. Department of Agriculture
Riverside, CA

Walter J. Weber, Jr. (Chairman)
Department of Civil Engineering
University of Michigan
Ann Arbor, MI
                                     99

-------
                                  REFERENCES
Adams, R. S.,  Jr. and P. Li.  Soils Properties Influencing Sorption and
     Desorption of Lindane.  Soil Science Society of America Proceedings,
     35:78, 1971.

Alrichs, J. L.  The Soil Environment.  In:  Organic Chemicals in the Soil
     Environment, C.A.I. Goring and W. Hamaker, eds.  Vol. 1.  Marcel Dekker,
     New York, 1972.

A.I.B.S.  Analysis of Specialized Pesticide Problems, Volume II.  American
     Institute of Biological Sciences Report No. 68-01-2457, 1974.

Alexander, M.   Nonbiodegradable and Other Recalcitrant Molecules.  Biotechnol.
     Eng., 15:611, 1973.

Alexander, M.   Role of Cometabolism. In:  Microbial Degradation of Pollutants
     in Marine Environments, A. W. Bourquin and P. H. Pritchard, eds., pp.
     67-75.  U.S.E.P.A. Publication 60019-79-012, Gulf Breeze, Florida, 1979.
Arozanena, L.  Draft Proceedings of a Pesticide Disposal Research and
     Development Symposium, Reston, Virginia.  Solid and Hazardous Waste
     Research Division, EPA-MERL, Cincinnati, Ohio, 1978.

Ausumus, B. S., D. R. Jackson and P. Van Voris.  The Accuracy of Screening
     Techniques.  In:  Terrestrial Microcosms and Environmental Chemistry, J.
     W. Gillett and J. M. Witt, eds.  Proceedings of the Colloquia at Oregon
     State University, Corvallis, Ore., June 13-14, 1977.  National Science
     Foundation, Washington, D. C., 1979.

Ayanaba, A. and M. Alexander.  Transformation of Methylamines and Formation of
     a Hazardous Product in Samples of Treated Sewage and wastewater.  J.
     Environmental Quality, 3:83, 1974.

Ayanaba, A., W. Verstraete and M. Alexander.  Formation of Dimethylnitrosamine,
     a Carcinogen and Mutagen, in Soils Treated with Nitrogen Compounds.  Soil
     Science Society of America Proceedings, 37:565, 1973.

Bailey, G. W. and J. W. White.  Review of Adsorption and Desorption of
     Organic Pesticides by Soil Colloids, with Implications Concerning
     Pesticide Bioactivity.  Agr. Food Chem., 12:324, 1964.
                                       100

-------
Bailey, G. W., J. L. White and T. Rothberg.  Adsorption of Organic Herbicides
     by Montmorillonite:  Role of pH and Chemical Character of Adsorbate.
     Soil Science Society of America Proceedings, 32:222,  1968.

Boethling, R. S. and M. Alexander.  Microbial Degradation of Organic Compounds
     at Trace Levels.  Environ. Sci. Technol., 13:989, 1979.

[Bright, N. F. H., J. C. Cuthill and N. H. Woodbury.  J.  Sci. Food Agr.,
     1:344, 1950.]

Carey, J. H., J. Lawrence and H. M. Tosine.  Photodechlorination of PCBs in
     the Presence of Titanium Dioxide in Aqueous Suspension.  Bull. Environ.
     Contam. Toxicol., 16:697, 1976.

Cole, L. K. and R. L. Metcalf.  Terrestrial/Aquatic Laboratory Model Ecosystem
     for Pesticide Studies.  In: Terrestrial Microcosms and Environmental
     Chemistry, J. W  Gillett and J. M. Witt, eds.  Proceedings of the
     Colloquia at Oregon State University, Corvallis, Oregon, June 13-14, 1977.
     National Science Foundation, Washington, D. C., 1979.

Cole, L. K.,  R. L. Metcalf and J. R. Sanborn.  Environmental Fate of
     Insecticides in a Terrestrial Model Ecosystem.  International J. Environ.
     Studies, 10:7, 1976.

Crosby, D. G.  Degradation of Synthetic Organic Molecules in the Bioshpere.
     National Academy of Sciences, Washington, D. C., p.  289, 1972.

Crosby, D. F. and K. W. Moilanen.  Photodecomposition of Chlorinated Biphenyls
     and Dibenzofurans.  Bull Environ. Contam. Toxicol.,  10:373, 1973.

Draggan  S.  The Microcosm as a Tool for Estimation of Environmental Transport
     of Toxic Materials.  International Journal of Environmental Studies,
     10:65, 1976.

Donigian, A. S. and N. H. Crawford.  Modeling Pesticides and Nutrients on
     Agricultural Lands.  EPA-600/2-76-043.  USEPA, Washington, D. C., 1976.

Esterman, E. F., G. H. Peterson and A. D. McLaren.  Digestion of Clay-Protein,
     Lignin-Protein, and Silica-Protein Complexes by Enzymes and Bacteria.
     Soil Science Society of America Proceedings, 23:31,  1959.

Farmer, W. J.  Leaching, Diffusion and Sorption. In:  A Literature Survey of
     Benchmark Pesticides.  George Washington University Medical Center,
     Washington, D.C., pp. 185-245, 1976.

Farmer, W. J., K. Igue and W. F. Spencer.  Effect of Bulkensity on the
     Diffusion and Volatilization of Dieldrin from Soil.   Journal of Environ-
     Quality, 2: 107, 1973.

Farmer, W. J., K. Igue, W. F. Spencer and J. P. Martin.  Volatility of
     Organochlorine Insecticides from Soil, I:  Effects of Concentration,
     Temperature, Air Flow Rate, and Vapor Pressure.  Soil Science Society
     of America Proceedings, 36:433, 1972.

                                      101

-------
Farmer, W. J. and J. Letey.  Volatilization Losses of Pesticides from Soils.
     Environmental Protection Technical Series, EPA-660/2-74-054.  U. S.
     Environmental Protection Agency, Washington, D.C.,  1974.

Farmer, W. J., M. Yang, J. Letey and W. F.  Spencer.  Residual  Management  by
     Land Disposal.  EPA-600/9-76-015, pp.  117-185, 1976.

Farmer, W. J., M. Yang, J. Letey, W. F. Spencer and M. H. Roulier.   Land
     Disposal of Hexachlorobenzene Wastes:   Controlling Vapor  Movement in
     Soils.  In:  Proceedings of the Fourth Annual Research Symposium, Land
     Disposal of Hazardous Wastes.  EPA-600/9-78-016,  pp. 182-190,  1978.

Frink, C. R.  Chemical and Mineralogical Characteristics of Eutrophic Lake
     Sediments.  Soil Science Society of America Proceedings,   33:369, 1969.

Fuller, W. H.  Movement of Selected Metals, Asbestos,  and Cyanide in Soils:
     Applications to Waste Disposal Problems.  EPA-600/2-77-020.  USEPA,
     Washington, D. C., 1977.

[Gibb, et al.  Reference to Illinois Institute of Environmental Quality
     report, 1978]

Gile, J. D. and J. W. Gillett.  Fate of C'4-Dieldrin in a Simulated Terrestrial
     Ecosystem.  Arch. Environ. Contam. Toxicol., in press 1979.

Gillett, J. W. and J. M. Witt, eds.  Terrestrial"Microcosms.  Proceedings of
     the Workshop on Terrestrial Microcosms, Otter Rock, Oregon, June 15-17,
     1977.  National Science Foundation, Washington, D.  C., 1979.

Goring, C.A.I.  Theory and Principles of Soil Fumigation.  In: Advances in Pest
     Control Research, R. L. Metcalf, ed.,  Vol. 5, pp.  47-84.  Interscience
     Publishers, New York, 1962.

Griffin, R. and N. Shimp.  Attenuation of Pollutants in Municipal Landfill
     Leachate by Clay Minerals.  EPA-600/2/78/157.  Solid and Hazardous
     Waste Research Division, EPA-MERL, Cincinnati, Ohio, 1978.

Guckel, W., G. Synnatschke and R. Rittig.  A Method for Determining the
     Volatility of Active Ingredients Used in Plant Protection.  Pesticide
     Science, 4:137, 1973.

Guenzi, W. D. and W. E. Beard.  Volatilization of Pesticides.   In:  Pesticides
     in Soil and Water, W. D. Guenzi, ed.,  pp. 107-122. Soil Science Society of
     America Proceedings, Madison, Wisconsin, 1974.

Hamaker, J. W.  The Interpretation of Soil Leaching Experiments.
     In: Experimental Dynamics of Pesticides, R. Haque and V.  H. Freed, eds.,
     pp.  115-123.  Plenum Press, New York, 1975.

Hamaker, J. W. and J. M. Thompson.  Adsorption.  In: Organic Chemicals in the
     Soil Environment, C. A. I. Goring and J. W. Hamaker, eds., Volume I,
     pp.  341-397.  Marcel Dekker, New York,  1972.
                                       102

-------
Haque, R., D. W. Schmelding and V. H. Freed.  Aqueous Solubility Adsorption
     and Vapor Phase Behavior of A-1254.  Environ.  Sci. Technol., 8:139,  1974.

[Hautala, R. R.  Paper presented at the Symposium on  Nonbiological Transport
     and Transformation of Pollutants on Land and Water, Gaithersberg,
     Maryland, May 11, 1976.]

Helling, C. S.  Pesticide Mobility in Soils,  Til:   Influence of Soil
     Properties.  Soil Science Society of America Proceedings,
     35:743, 1971.

Helling, C. S., P. C. Kearney and M. Alexander.  Behavior of Pesticides in
     Soils.  Advan. Agron., 23:147, 1971.

Houle, M., et al.  Migration of Hazardous Substances Through Soil.  Draft Final
     Report from Dugway Proving Ground to Solid and Hazardous Waste Research
     Division.  IAG-D4-0043.  EPA-MERL, Cincinnati, Ohio, 1978.

Hutchinson, T. C., ed., International Conference on Heavy Metals in the
     Environment.  Symposium Proceedings, Volume II: Pathways and Cycling
     (Parts 1 and 2).  Toronto, Ontario, Canada, 1975.

Ishida, M.  Phytotoxic Metabolites of Pentachlorobenzyl Alcohol.  In:
     Environmental Toxicology of Pesticides, F. Matsumura, C. M. Bousph
     and T. Misato, eds., pp. 281-306.  Academic Press, New York, 1972.

Karickhoff, S. W., D. S. Brown and T. A. Scott.  Sorption of Hydrophobic
     Pollutants on Natural Sediments.  Water Research,  13:241, 1979.

Kohnke, H.  Soil Physics.  McGraw-Hill, New York.   1968.

Laskowski, D.  Determination of Adsorption/Desorption Constants.  A. S. T. M.
     Committee E 35.21.02 Draft No. 2,  1978.

Lee, G. F.  Factors Affecting the Transfer of Materials Between Water and
     Sediments.  University of Wisconsin, Eutrophication Information Program,
     Literature Review No. 1, 1970.

Lee, G. F.  Personal Communication regarding project being conducted for the
     USEPA.  R. S. Kerr Environmental Research Laboratory, Ada, Oklahoma, 1978.

Leistra, M.  Computation Models for the Transport of Pesticides in Soils.
     Residue Reviews, 49:87, 1973.

Letey, J. and W. J. Farmer.  Movement of Pesticides in  Soils.  In: Pesticides
     in Soil and Water, W. D. Guenzi, ed., pp. 67-97.   Soil Science Society
     of America, Inc., Madison, Wisconsin, 1974.

[Lockhart, H. B. and R. V. Blakely.  In:  Environ.  Sci. Technol., 9:1035,
     1975.]
                                      103

-------
Lotse, E. C., D. A. Graetz, G. Chesters, G. F. Lee and L.  W.  Newland.   Lindane
     Adsorption by Lake Sediments.  Environ.  Sci. Technol.,  2:353,  1968.

Mayer, R., J. Letey and W. J. Farmer.  Models for Predicting Volatilization of
     Soil-Incorporated Pesticides.  Soil Science Society of America
     Proceedings, 38:563, 1974.

Mitre Corporation.  Comparison of Three Waste Leaching Tests — Executive
     Summary.  EPA-600/8-79-001.  Solid and Hazardous Waste Research
     Division, EPA-MERL, Cincinnati, Ohio, 1978.

Nash, R. G. and M. L. Beall, Jr. A Microagroecosystem to Monitor the
     Environmental Fate of Pesticides.  In: Terrestrial Microcosms and
     Environmental Chemistry, J. W. Gillett and J. M. Witt, eds..
     Proceedings of the Colloquia at Oregon State University, Corvallis,
     Oregon, June 13-14, 1977.  National Science Foundation,  Washington,
     D. C., 1979.

Oberacker, D.  Personal communication regarding measurements of vinyl
     chloride monomer around landfills.  Solid and Hazardous Waste
     Research Division, EPA-MERL.  Cincinnati, Ohio, 1978.

Plimmer, J. R.  Degradation of Synthetic Organic Molecules in the Biosphere,
     p. 179.  National Academy of Sciences, Washington, D. C., 1972.

Plimmer, J. R.  Herbicides:  Chemistry, Degradation and Mode of Action, pp.
     891-934.  Marcel Dekker, N. Y., 1976.

Shin, Y., J. J. Chodan and A. R. Walcott.  Adsorption of DDT by Soils, Soil
     Fractions, and Biological Materials.  J. Agr. Food Chem., 26:1129, 1970.

Smith, C., T. Iwata and F. Gunther.  Conversion and Disappearance of Methida-
     thion on Thin Layers of Dry Soil.  J. Agr. Food Chem., 26:959,  1978.

Spencer, W. F.  A Literature Survey of Benchmark Pesticides.  George
     Washington University Medical Center, Washington, D. C., pp. 72-165,
     1976.

Spencer, W. F., J. Adams and T. Shoup.  Oxidation of Parathion to Paraoxon in
     Soil Surface Dust as Affected by Atmospheric Conditions.  Presented at the
     4th International Congress of Pesticide Chemistry, Zurich, Switzerland,
     July, 1978.

Spencer, W. F. and M. M. Cliath.  Vapor Density of Dieldrin.  Environ.
     Sci. Technol., 3:670,  1969.

Spencer, W. F., M. M. Cliath and W. J. Farmer.  Vapor Density of Soil-Applied
     Dieldrin as Related to Soil-Water Content, Temperature, and Dieldrin
     Concentration.  Soil Science Society  of America Proceedings, 33:509,  1969.

Spencer, W. F., W. J. Farmer and M. M. Cliath.  Pesticide Voalatilization.
     In: Residue Reviews, 49:1,  1973.
                                       104

-------
Taylor, A. W.  Post-Application Volatilization of Pesticides Under Field
     Conditions.  J. Air Pollution Control Association, 28:922, 1978.

U. S. Environmental Protection Agency. Guidelines for Registering Pesticides in
     the U. S.  Federal Register, 40 (123), 26829.  June 25, 1975.

U. S. Environmental Protection Agency.  Pollution Prediction Techniques for
     Waste Disposal Siting.  USEPA Office of Solid Waste, SW-162c.  Washington,
     D.C.

Van Voris, P., R. V. O'Neil, H. H. Shugart and W. R. Emanual.  Functional
     Complexity and Ecosystem Stability: ?Vn Environmental Approach.
     Environmental Sciences Division Publication No. 1123, ORNL/TM-6199,
     Oak Ridge National laboratory, Oak Ridge, Tenessee, 1978.

Ware, G. W., B. J. Estesen and W. P. Cahill.  Organophosphate Residues on
     Cotton in Arizona.  Bull. Environ. Contain. Toxicol., 8:361, 1972.

Ware, G. W., D. P. Morgan, B. J. Estesen and W. P. Cahill.  Establishment of
     Reentry Levels for Organophosphate Treated Cotton Fields based on Human
     Data.  Arch. Environ. Contam. Toxicol., 2:117, 1974.

Wauchope, R. D.  The Pesticide Content of Surface Water Draining from
     Agricultural Fields — A Review.  Journal of Environmental Quality,
     7:459, 1978.

Weber, J. B.  Interaction of Organic Pesticides with Particulate Matter in
     Aquatic and Soil Systems.  In: Fate of Organic Pesticides in the Aquatic
     Environment, S. D. Faust, ed.,  Advances in Chemistry Series Til,
     American Chemical Society, Washington, D. C., 1972.

Weber, W. J., Jr.  Physicochemical Processes for Water Quality Control.
     Wiley Interscience, New York, 1972.

Weber, W. J., Jr., M. Pirbazari and M. D. Herbert.  Removal of Halogenated
     Organic and THM Precursor Compounds from Water by Activated Carbon.
     Proceedings of the 98th Conference, American Water Works Association,
     Denver, Colorado, 1978.

Weber, W. J., Jr., M. Pirbazari, M. D. Herbert and R. Thompson.  Effectiveness
     of Activated Carbon for Removal of Volatile Halogenated Hydrocarbons from
     Drinking Water.  In:  Viruses and Trace Contaminants in Water and
     Waste-water, J. A. Borchardt et al., eds.  Ann Arbor Science Publishers,
     Inc., Ann Arbor, Michigan, 1976.
Weber, W. J., Jr., J. D. Sherrill, M. Pirbazari, C. G. Uchrin and T. Y. Lo.
     Transport and Differential Accumulation of Toxic Substances in River-
     Harbor-Lake Systems.  In: Dynamics, Exposure and Hazard Assessment of
     Toxic Chemicals in the Environment, R. Haque, ed.  Ann Arbor Science
     Publishers, Inc., Ann Arbor, Michigan, in press 1979.
                                      105

-------
Weber, W. J., Jr. and B. M. Van Vliet.  Fundamental Concepts for Application of
     Activated Carbon in Water and Wastewater Treatment.  In:  Activated
     Carbon Adsorption of Organics from the Aqueous Phase, I.  H. Suffet and
     M. J. McGuire, eds. Ann Arbor Science Publishers, Inc., ton Arbor,
     Michigan, in press 1979.

Wheatley, G. A.  Pesticides in the Atmosphere.  In:  Environmental Pollution
     by Pesticides, C. A. Edwards, ed., pp. 365 - 408, Plenum Press, London,
     1973.

Wood, J. M.  Environmental Pollution by Mercury.  Environ. Sci. Technol.,  2:39,
     1971.

Yang, M-S.  Processes of Adsorption, Desorption, Degradation,  Volatilization,
      and Movement of Parathion in Soils. Ph. D. Dissertation, University of
      California -Davis, 1974.

Zepp, R. G., N. L. Wolfe, J. A. Gordon and F. L. Baughman.  Dynamics of
     2,4-D Esters in Surface Waters.  Environ. Sci. Technol.,  9:1144, 1975.
                                       106

-------
                  4.0 EFFECTS OF BIOTA ON TOXIC SUBSTANCES
4.1  INTRODUCTION

     This section elucidates the state-of-the-science and research priorities
necessary to allow risk assessments of toxic substances through biological
components of ecosystems, and focuses on the biological aspects of transport,
transformation and storage of toxic substances.  The research priorities are
oriented towards and constrained by the need to obtain anticipatory assessments
based upon generic properties of new substances.

     In general, predictive techniques should be based upon mechanistic pro-
cesses. Many of these processes have been discovered and analyzed in simpli-
fied, laboratory conditions.  It is essential to evaluate the utility and
limitations of incorporating laboratory studies into ecosystem models.  Many of
the recommendations involve efforts to identify and quantify factors of
complexity and scale as one goes from prototype to full scale systems.

     Further information is needed on the biotransformation of substances in
particular environments.  Microbial activity, particularly in anaerobic
environments, needs considerable articulation.  The role of higher organisms
is unclear and this issue should be resolved at an early date.

     The transport of toxic substances by organisms has seldom been quanti-
tatively analyzed.  A few classic case studies exist which are natural
opportunities for analysis of regional transport.  The kepone episode in the
Chesapeake Bay is probably the best example since the input took place at a
single point in space and time.

     This chapter is the consideration of opinions of some sixteen individuals
(Table 4-1) who are currently involved with research and regulatory activities
associated with toxic substances in the environment.  Their discussions focused
on the biological data required to formulate and initiate predictive models
dealing with biotransformation and biotransport.  Furthermore, they concen-
trated on specific research recommendations that could be realistically
initiated and completed.  The recommendations reflect the collective concerns
at the time of the conference (December, 1978).  The issues are arranged into
seven sections each accompanied by a short explanatory statement.  The
recommendations are incorporated into the appropriate text.

4.2  BIOTRANSFORMATION

     A vast amount of literature exists on the biochemical pathways through
which various toxic substances might pass but the primary areas of interest are
microbes, higher plants, and higher animals.  Ecological analyses have
                                      107

-------
repeatedly demonstrated the critical role of microbes in energy flow and
biotransformations of materials.  Very few data exist on rate functions
determined under field conditions.  The interest in higher plants resulted
from the increasing use of terrestrial ecosystems for the recycling of munici-
pal and industrial waste waters and sludges while the emphasis on higher ani-
mals resulted from questions as to whether or not these organisms have any
major direct influence on biotransformation.

     Predictive models must be based upon mathematical representations of
functional mechanisms if one wishes to characterize unobserved events.  Know-
ledge of the mechanisms of biotransformation at the biochemical level will be
needed for certain types of processes.  The emphasis on standardized in vitro
analogs for activation and detoxification processes is motivated by the need to
express causal mechanisms in the desired models.

4.2.1  Microbial Transformations

     Microbial degradation is important to the fate and transport of a chemical
in the environment.  Little is known concerning biodegradation in nature and
most of our current information on biodegradation results from laboratory
studies in which selected pure or mixed microbial cultures were used.  A major
question remains concerning the validity of extrapolating these laboratory
results to the natural environment.  Microbial degradation, persistence, and
environmental effects on microbial transformations have been discussed
previously (see Microbial Degradation of Pollutants in Marine Environments:
Consensus to Methodology, Interpretation, and Future Research, 10-14 April
1978, Pensacola Beach, Florida).

     Currently, rate constants for microbial degradation or transformation are
obtained in the laboratory from pure cultures utilizing a given chemical as a
sole carbon source. These methods may generate environmentally unrealistic
information because co-metabolic phenomena could have a significant effect on
degradation yet involve minimal changes in microbial biomass or physiological
activities.  Verification of in situ rate constants should be undertaken and
the influences of environmental parameters on rate constants should be defined.
It remains to be determined whether a measurement of the physiological activity
of microorganisms (ATP, lipids, heterotrophic uptake) may provide a useful
basis for comparison between laboratory and environmental systems.

     Much of the natural environment is anaerobic.  Unfortunately, little is
known about the microbial degradation processes and the transformation products
of chemicals under in situ anaerobic conditions.  Such information on the
anaerobic microbial process is crucial for predicting the fate of chemicals in
the environment.

     In addition, anaerobic microbial degradation should be investigated to
determine the effects of anaerobic treatment plant effluents, anaerobic
microflora of the human and animal gut, and the microbial reactions that occur
in sediments.  More suitable methodologies are needed for easily studying the
anaerobic process in the laboratory.
                                       108

-------
     Natural and synthetic organic nutrients often occur in very low concentra-
tions and bacterial populations in aquatic environments are specifically
adapted to growth at these low nutrient levels.  Thus, common naturally occur-
ring organics such as urea, glycollic acid, and amino acids may be biodegraded
in preference to synthetic organic pollutants.  Consequently, threshold concen-
trations of pollutants may exist in nature below which degradation will not
occur.  Similarly, due to concentration effects and preferences for naturally
occurring organic nutrients, microbial populations in natural environments may
never become adapted for specific degradative capacities.  The effect of con-
centration must be experimentally verified using laboratory and field degrada-
tion studies which specifically maintain low pollutant concentrations.

     Most available information concerns microbial biodegradation and little is
known concerning biodegradation by fungi, algae, and protozoa.  Efforts should
be undertaken to define the mechanisms and pathways of biodegradation for these
other organisms.

4.2.2  The Fate of Toxic Materials in Higher Organisms

4.2.2.1  Plants

     Although the chemical and physical properties of toxic substances deter-
mine their transport and fate, at present neither the action of new chemicals
in plants nor the effects of vegetation on the fate of these substances can be
predicted.  A great deal is known about the fate of toxicants in or on plants
(U. S. Environmental Protection Agency 1975).

     Plants can translocate substances downward from leaves to roots and other
organs or in the reverse direction depending on the point of entry, the physi-
cal state, and chemical properties of the materials.  Substances can preferen-
tially accumulate and be stored in certain organs in their original form.  They
can add to and be indistinguishable from existing amounts of these same
substances accumulated from natural sources.  They may undergo transformation
resulting in storage of degradation or reaction products which can be more or
less toxic than the original substances.  Materials can move from the interior
of plants by exudation to leaf or root surfaces where they can be lost by the
action of rain or dew (leaching), wind, abscission, or as a result of consump-
tion by herbivores or omnivores.  Loss of organs to soil can lead to further
conversion or degradation by microorganisms, storage in soil, or transfer to
water supplies.  Some substances can be volatilized from plants directly to the
atmosphere.

     As a consequence of these manifold and complex processes taking place in
vegetation, research is needed to derive both general principles and quantita-
tive descriptions for the construction of predictive models.  The question is,
which processes deserve highest priority?

        The most important factors concerning the transport and fate of toxic
substances in plants are degradation and conversion of toxic substances to more
or less active forms (metabolism), the extent to which plants transfer sub-
stances to other media (soil, water or air), or to other organisms, and the
                                      109

-------
degree of accumulation of toxic substances in plants and the conditions under
which this accumulation is fostered or inhibited.

4.2.2.2  Animals

     Animals affect the fate of exogenous chemicals either directly through
assimilation and metabolic transformation, or indirectly through their control
in the structure and microbial activity of ecosystems.

     Direct uptake by feeding assimilation and metabolic transformation of
chemicals by individual animals can be studied in the laboratory (greenhouse
and metabolic cage studies).  It is not apparent that this work has been
systematically used to estimate the magnitude of importance of these biotrans-
formations on the environmental fate of chemical pollutants.  For example,
benthic aquatic insects and terrestrial invertebrates consume and process
great quantities of detritus and soil but there is little knowledge of the im-
portance of these groups in directly biotransforming chemicals in relation to
the microbial populations present. Fish, aquatic and terrestrial crustaceans,
and bivalves should be similarly examined.

     Animals are important in influencing the active microbial biomass present
in aquatic and terrestrial systems.  They would, therefore, be important
factors in indirectly determining the capacity of a system to biotransform
chemicals microbially.  The microbial stimulation from the bioturbation and
detritus-processing activities of invertebrates in sediment and soils provides
examples of the influence that animals have on microbial activities.  The
alteration of microbial biotransformation rates occurs as a result of organisms
incorporating chemicals into microbially available forms (zooplankton, fecal
pellets, animal tissue).

     Higher animals may also indirectly affect microbial biotransformation of
chemicals by controlling the movement and concentration of these materials in
systems.  Uptake and excretion of a chemical in a more concentrated waste
product may affect biotransformation rates in microbial populations utilizing
those wastes as substrates.  Nesting, food storage, and waste disposal acti-
vities may also affect local concentrations of chemicals.

     Experimental manipulations of animal species have been shown to have
significant impacts on the structure and function of ecological communities
(Paine 1966; Hall et al. 1970; Neill 1975; Dayton 1971).  Changes in
foraging patterns, prey selectivity, competitive interactions, and habitat
modifications associated with alterations in species arrays will affect the
composition and abundance of the microbial fauna and flora.  This feedback of
effects will directly influence the biotransformation capability of a given
ecological community.  The primary role of higher organisms might not be
apparent from studies oriented solely towards activation and deactivation
potentialities.

     Work with artificial outdoor streams has shown that removal of crayfish
from these systems resulted in notable changes in the stream species composi-
tion.  Mechanical exclusion of crayfish, aquatic insects, or other important
                                      110

-------
detritus-processing groups from similar streams could demonstrate the effects
such exclusions have on microbial activity and biotransformation of chemical
contaminants.

4.2.3  In Vitro Activation/Detoxification Models for Whole Animal
       Biotransformations

4.2.3.1  Biochemical Investigations

     Standard methods to use microsomal mixed-function oxidases should be
developed and utilized.  Representatives of various animal groups, including
those which are commonly used in laboratory studies (rat, mouse, chicken, blue-
gill or other fish, insect, etc.), or those which are important in regard to
transport and fate of toxic chemicals, should be utilized.  The investigator
studying the metabolism of the toxic chemical and the mixed-function oxidase
should provide information about the activity of the enzyme system towards some
standard assays such as aldrin epoxidation (Stanton and Khan 1975), aniline or
benzopyrene hyroxylation, and 0- and N-demethylations of, respectively, para-
nitroanisole and d-benzphetamine, and cytochrome P-450 contents (Omura and
Sato 1964; Mazel 1972; Burke and Meyer 1975). Microsomal oxidases are an
important detoxification system and perform aliphatic and aromatic
hydroxylations, O- and N-dealkylations, desulfurations, and hydrolysis.
The levels of activity of this system in various animals should be known and
included in the ecological models.

4.2.3.2  Other Detoxification Systems

     Microsomal enzyme systems such as epoxide hydratase attack epoxides, many
of which are known carcinogens.  The effects of these enzymes on epoxide or
oxides of toxic chemicals should be investigated to understand whether the
epoxide intermediates produced in animals in the ecosystem will accumulate or
be further metabolized (Oesh et al. 1971; Bentley et al. 1978).  Other
microsomal enzyme systems such as glucuronyl-transferases and esterases should
be investigated as their activity will indicate whether certain ester type
toxic chemicals can be hydrolysed.  Th,e investigation should also show
whether the animal is able to further handle, by conjugation, the metabolites
of the toxic chemical  (Doroug 1979; Yang 1976).

     Cytosolic detoxification systems include esterases, conjugases of various
kinds, glutathione-dependent transferases, and dechlorinases (Yang 1976;
Doroug 1979; Dauterman 1976; Hollingworth 1976).  These enzymes should be
used as models to understand the ability of the test animal(s) to metabolize
toxic chemicals. These detoxification systems are standardized in certain
laboratory-reared mammals, fish, and birds and it will be easy to use them as
models for the study of the fate of the toxic chemical in vitro.

     The selection of test animals should include animals that are most
commonly used in the laboratory, easily raised in the laboratory, easily ob-
tained from the field, and play an important role as target or non-target
organisms in the ecosystem regarding the fate and transport of the toxic
                                      111

-------
chemical.  It is also recommended that radioactive tracers should be used and
sufficient analyses be carried out to chemically characterize the nature of
these biotransformations.

4.2.3.3  The Use of Laboratory Obtained Parameters in Estimating the
         Distribution of Organic Compounds in Environmental Compartments

     It has been shown in hundreds of examples that log P(P = octanol/water
partition coefficient) can be used to correlate the binding of organic com-
pounds to proteins, fat phases, tissue, soils, and fabrics as well as the
movement of organic compounds through various membranes (natural or synthetic).
Equations of the following type are found to hold:

                       log k = a log P + c                                (1)

                       log k = a log P + b (log P)2 + c                   (2)

     k may be a rate or equilibrium constant.  Neely and co-workers at Dow have
shown that for a very small stable set of pesticides equation (1) held for the
uptake by trout.  However, this study plus some work by Metcalf's group is so
limited that one can be quite sure it will not hold for a large set of organic
compounds in a real ecosystem.  Such a large set of compounds would have
varying degrees of chemical and biochemical reactivity through which they would
be modified and hence their distribution affected.  One might expect the
following general equation to hold:

              Cone, biophasei = f (log p, Special Reactivities.^)          (3)

     There are a large number of special reactivities and it appears that
microsomal and photochemical processes are two of the most important.  In
addition, hydrolysis is important but might be lumped in with photchemical in a
single parameter.  With these parameters:

                            VP = Vapor pressure

                            MO = Microsomal oxidation

                            PC = Photochemical Change

                           Hyd = Hydrolysis

which would be relatively easily determined under a set of standard laboratory
conditions, a correlation equation could be formulated:

log Cone. Biophasei =
          a log P + b (log P)2+dVP+eMO+fPC+g Hyd + c           (4)

Once a-f are established for a representative set of chemicals found in the
various phases of a microecosystem, the fate of new chemicals could be
estimated.
                                       112

-------
     This stochastic model assumes that octanol/water partition coefficients
will be a good approximate measure of the thousands of lipophilie/aqueous
partitioning ratios organic compounds will experience from source to sink.
This has been established to be true for a drug passing through animals where
partitioning is occurring with thousands of different macromolecules, fat
depots, and membranes.

     In an ecosystem there are thousands of biochemical degradation systems.
The rat-liver microsomes (s~"' could represent a prototype in the same way
octanol/water partitioning approximates movement through thousands of different
lipophilic phases in an animal.  One cannot consider all possible routes of
biodegradation.  Hence, the stochastic approach appears better than a
deterministic model.

4.2.3.4  The Utility of Biochemical Processes in the Macroscopic Models

     Incorporation of biochemical processes into macroscopic population
models requires several levels of system integration.  This activity is
extremely complex and requires considerable scientific documentation.  Proto-
type systems should be selected for which population, community, and ecosystem
data currently exist.

     The following series of analyses should then be initiated:

     1.  Organisms from various phyla important in the ecological concentration
of the toxic chemical or those that can be raised in the laboratory should be
used, e.g., Daphnia, mussel, lobster, fish (bluegill, carp, goldfish), bird
(chicken, duck), mammals (rat, rabbit).

     2.  The physiological state of the organisms should then be described.
This includes age, size, sex, nutritional status, season of the year and pre-
vious exposure of the animal to other xenobiotics.

     3.  The routes of exposure of the test organisms should be determined.
For example, absorption of the chemical can occur from the aquatic environment,
ingestion of suspended particles in water, or consumption of detritus, animal
or plant tissues.

     4.  Standard methods used by other federal agencies (FDA, NIH, USDA and
USEPA) should be reviewed to determine the protocol for such studies.

     5.  Bioaccumulation studies of chemicals should use standard procedures
for such assays.  Existing procedures should be further improved to make them
adaptable for these studies.  Most bioaccumulation values are derived from the
concentration of the toxic chemical in the whole animal.  It is recommended
that, in addition to the concentration within the animal, the concentration of
the chemical in certain specific tissues (such as muscle, liver, kidney, brain,
gonads, fat) should also be described.  This may eventually lead to physiolo-
gical models of the toxico-dynamics of the chemical in organisms.  This is a
necessary step in the integration process.
                                      113

-------
     6.  The half-life of the chemical in the whole organism as well as in
various organs of the animal should be determined.  Standard methods to study
this should be outlined to be used by investigators to provide data for the
regulatory procedures.

     The previous information on bioaccumulation, storage, metabolism,  and
elimination by various animals in the laboratory should be integrated with
knowledge of the population dynamics of these species or similar species in
prototype ecosystems to determine the magnitude of importance of biochemical
transformations by higher organisms.

4.2.4  Biotransformation Recommendations

     Rate constants for biodegradation of toxic organics regulated by EPA
should be defined.  The influences of environmental parameters, nutrient
concentration, chemical concentration and microbial populations on biodegrada-
tion rates should be tested in laboratory and field systems.  A determination
of the feasibility of using various physiological parameters to relate the
result of laboratory and natural biodegradation processes should be included.

     Methodologies should be developed for easily studying and screening
anaerobic microbial transformation processes in the laboratory.  Much more
information is needed concerning the magnitude of anaerobic microbial transfor-
mation of toxic organics in prototype environments (marine sediments, flood
plains, saturated soils).

     The role that other microbial forms (fungi, algae, and protozoa and their
communities) play in degradation or transformation should be determined.  Path-
ways for degradation or transformation by these other microbial forms should be
studied for prototype ecosystems of particular concern.

     Prototype plant species should be selected based upon their use in human
food chains or their dominant role in critical ecosystems.  The role of higher
plants in biotransformation and microdistribution should be systematically in-
vestigated utilizing organic and organo-metallic complexes as test chemicals.
Emphasis should be given to those residuals commonly found in municipal
sludges.

     Field plots should be examined for the effects of various populations of
higher plants (monoculture and mixed communities) on the rate of microbial bio-
transformation of chemical contaminants.

     Utilizing a well studied, experimental ecosystem, the relative magnitude
of direct and indirect effects of higher animals on biotransformation of
chemicals in the environment should be more systematically assessed and
compared with microbial transformation rates in  similar systems.

     The environmental importance of direct biotransformations described in
metabolism cage studies with individual organisms must be evaluated and the
factors affecting the initial uptake and assimilation of environmental
                                       114

-------
chemicals by populations of higher organisms must be determined.  A knowledge
of community-level interactions (predation, competition)  will be important in
making the extrapolation from single organism metabolism studies to
environmental biotransformations by populations.

     A few key ecosystems frequently exposed to exogenous chemicals should be
examined in the above manner to determine the indirect effects of higher
animals on microbial biotransformation.

     A  program for development of a series of in vitro systems utilizing
microsomal and specific enzyme activation and detoxification processes should
be initiated.  Prototypes should be selected that are feasible and available to
general laboratory operations.  Furthermore, the prototypes should reflect
critical ecological components influencing the transport and fate of toxic
substances.  Utilization of these physical prototypes should be a standard
aspect of the assessment protocol.

     A reasonable start for the stochastic approach might be to analyze five
sets of twelve congeners each.  Families such as chlorinated hydrocarbons,
phenols, aromatic amines and phosphate esters with good variation in
physicochemical properties in each should be carefully selected.  Important
industrial chemicals must be included with other molecules.  Crucial to the
project is a reasonably good microecosystem on which to validate parameters.
Such a project would best be done as a cooperative effort by several
laboratories with expertise in each area including a photochemist to get
photochemical constants, a laboratory experienced in working with microsomal
oxidation, and an environmental lab with experience with microecosystems.

     Prototype ecosystems should be selected and utilized to document the pro-
cedures of integration across levels of structure in the ecological hierarchy
of complexity.  This integration includes the mechanisms of transformations and
the magnitudes of the activity given population and community characteristics.
4.3  BIOLOGIC TRANSPORT

4.3.1  Existing Case Study

     Since many species of animals and plants move from one location to another
due to migration or translocation by air or water currents, the biological mode
of toxicant transport should be incorporated into mathematical models.  The
particular life cycle or feeding characteristics of an animal may influence the
distribution of a substance in the system.  Therefore, the biological segment
of the ecosystem should be recognized not only from a toxicological or bio-
transformation standpoint but also as an active transport mechanism.

     Many harmful substances, such as PCB's and pesticides, have entered the
ecosystem from varied sources which confuses the interpretation of field data
relative to biological transport.  An exception is Kepone which has contami-
nated the estaurine portion of the James River.  This compound has been trans-
ported from the Chesapeake Bay to New England waters by bluefish and striped
                                      115

-------
bass which were resident in the Bay before migrating north (Huggett et al.
1979).  Blue crabs in North and South Carolina have been shown to contain
Kepone, presumably due to consumption of contaminated fish migrating south  from
the Chesapeake (Huggett et al. 1979).  Even though the amounts transported  by
these fish are but fractions of a percent of the total amount present in the
James River, it is sufficient to contaminate ecosystems far removed from the
"polluted source."

     Microdistribution can be determined by biotic activity that mixes sub-
stances and sediments and concentrates by filter feeding.  Kepone pene-
trates the sediments of the James River far in excess of expected diffusion or
sedimentation rates due to the fact that burrowing organisms are constantly
mixing the sediments (Huggett et al. 1979).  Other examples of the biota in-
fluencing the distribution of toxicants exist.  The biotic component of
transport must be integrated into models designed to predict the transport  of
toxicants in the environment.

     At the present time there are few mathematical models capable of modeling
bioaccumulation of toxic substances in complex ecosystems.  Currently, none of
these are being used to predict biological transport from one part of the
ecosystem to another, with the exception of the obvious water column, suspended-
sediment coupling.  Furthermore, most of the models are based on simplyfying
assumptions that are incompatible with the detail required for modeling biotic
migration.

     An example of a model that can be used to represent this type of transport
is PEST (Park et al. 1977; Leung 1978), which considers twenty carrier com-
partments including four age classes in two groups of fish.  With such detail
toxic substance loads and migratory patterns can be distinguished for each age
class.  These can be verified in the field where migratory populations are
periodically exposed to point-sources of toxicants.  A good example is
anadromous fish that come into contact with Kepone as they migrate into the
James River and leave with a relatively high Kepone load.

4.3.2  Information System

     There is a vast amount of literature on transfer rates between organisms
and their environments.  Consumption rates, defecation rates, migration
patterns, sorption and desorption coefficients, and mortality rates are just a
few of the measurements obtained from ecological studies.

     Predictions of chemical transfers within ecosystems for regulatory deci-
sions should be based upon measured environmental transfer rates whenever
possible.  To aid in these predictions, the information concerning chemical
transfer coefficients among ecosystem components which currently exists should
be abstracted, indexed, and placed on a computer retrieval system.  The index
system should employ a large array of prioritized key words which will provide
maximized cross-referencing.

     The abstracts should be worded in a manner which is understandable,
pertinent to the hazard evaluator, and readily usable in current hazard evalua-
tion calculations.  Tranfer rate data for higher trophic levels should receive
                                       116

-------
priority attention as hazards to macroorganisms are most likely to influence
regulatory decisions.

4.3.3  Anthropogenic Influences

     The PBB episode in Michigan provides a classic example of the potential for
a toxic material to be rapidly transported throughout a region.  The dominant
process of transport involved the spatially oriented activity of the agro-
business industry.  The accidental addition of the toxicant into an animal feed
component of the dairy industry took place at a single location.  The materials
were then distributed, remixed, redistributed, and eventually sold to production
units broadly distributed throughout the midwest.

     Attempts to reconstruct the patterns and processes of distribution demon-
strated a serious lack of data on the intra- and inter-state transport
mechanisms within this industry.  A majority of transactions involved
independent truckers and farm businesses for which there is no monitoring or
reporting system available.  In fact, the existing reporting system indicated
that the toxic substance failed to cross the political boundary of Michigan.

     Similar issues are now surfacing with the residual management of hazardous
substances.  One should develop the predictive ability to simulate probability
isopleths of distributions of a toxic material given a point-source origin, a
time period from the initial event, and a pattern of commercial transport.  This
capability will be of particular importance in the allocation of human and mone-
tary resources to post-episode monitoring.

4.3.4  Transport Recommendations

     A model such as that developed by Park et al. (1977) and Leung (1978)
should be modified to be applicable to the Kepone incident in the James River.
Taking advantage of the existing monitoring program (Huggett et al. 1979), the
model should be utilized to predict the distributional pattern of Kepone
throughout the Chesapeake Bay and associated waters resulting from the new expo-
sure of migratory fish in the spring.  This appears to be a unique opportunity
to explore and validate a regional transport model.

     An information retrieval system should be developed that would allow those
involved in environmental assessment to have rapid access to the diffuse litera-
ture involving measurements of transport and transfer rates among ecosystem
components.  The emphasis should be on coefficients, thus complementing the
Toxline system.

     Utilizing a specific case study like the PBB in dairy products or the mer-
cury in tunafish, EPA should develop a case history analysis that characterizes
the anthropogenic distribution of a toxicant.  The analysis should include the
man-made transport, storage, utilization, and residual management components.
From case history analyses, a determination should be made as to whether or not
it is feasible to develop a generic system that predicts distributional
characteristics for any given environmental episode.
                                      117

-------
4.4  MODELS:  MATHEMATICAL AND PHYSICAL ANALOGS

4.4.1  Further Studies in Model Development

     Most existing biological fate models are point models that ignore the
spatial heterogeneity that is so important in a natural system.  Such models are
adequate for evaluative purposes but are inadequate for assessing impacts at
particular sites.

     Only a few horizontal and vertical segments are required for many site-
specific applications.  The connectivity of these segments can be specified by
the user on an ad hoc basis.  Depending on the model, this may require signi-
ficant reprogramming.  However, at some sites the spatial resolution will be
critical and a fine grid encompassing many segments should be employed.  Capa-
bility for representing this resolution is not present in existing bioaccumula-
tion models.  Furthermore, specification of the biologic and physical connecti-
vity of the segments is a monumental task if left to the individual user.
Comparable problems have been encountered in modeling thermal plumes in a
variety of aquatic systems and algorithms have been developed that can be
adapted to bioaccumulation models.

     The concept of mass balance is sufficiently accepted as an attribute of
systems dynamics so as to need little further elaboration here.  Instead, a
detailed consideration of where and why the measurement of mass balance has been
impractical and/or incomplete should prove valuable.  In particular, such a
review should examine the way in which mass balance studies are linked to the
problems of transport and fate of environmental pollutants.  Elemental mass
balance measurements are usually the easiest and cheapest but may have little use
when toxic compounds and daughter products are the substance of concern.  There
are exceptions to the ease and cost of mass balance.  For example, the measure-
ment of N2 released from soil and sediments as a result of denitrification is a
difficult measurement without the use of isotopes.

     Clearly the establishment of a mass balance table in any ecosystem requires
clear and unambiguous definition of the boundaries defining the system of
interest, knowledge of the transport pathways and rates of movement into and out
of the system, the capability of analyzing for the compound(s) of interest, and
knowledge of the important physical sinks and sites of bioaccumulation within
the system so that an adequate sampling program can be designed.  Thus, a mass
balance study is directly related to the details of transport and fate of the
compound(s).

     By varying enough parameter values in a model it is possible to fit almost
any observed data; this negates the usefulness of the model for any purpose
other than empirical description.  Therefore, it is important that a philosophy
of model formulation and parameterization be established in order to guarantee
that models will be applicable and defensible in assessing the impacts of toxic
chemicals.

     All actions of living organisms are influenced to some degree by probabilis-
tic considerations.  Furthermore, the distribution of substances in nature,
                                       118

-------
whether in a physical sink or in the tissue of living organisms,  is not uni-
form.  It varies both temporally and spatially, within a range of extremes,
often with a distribution that is far more normal in the Gaussian sense.  This
distribution is a central problem for systems ecology.

     The question of stochasticity in the behavioral action of individual
organisms can be handled reasonably well by averaging, provided a large number
of individuals within the scope of the system modeled and the probability dis-
tributions are monotonic.  For example, questions of the timing of feeding,
reproducing, defecating, etc., by individuals in pond A as opposed to those
same species in pond B are unimportant when the model summarizes the net result
of thousands of such individual decisions.  The realized average specific rate
of growth and ingestion is an adequate predictor.

     Different analytical techniques must be utilized when the behavior of
small numbers of individuals is to be simulated or the spatial and temporal
patterns produce non-monotonic probability distributions.  One important
variable is the range.  The toxic effect can be expected to be related to the
extremes as well as to the chronic mean value, often in a complicated non-
linear manner.  The model must have built into it the capability of simulating
such different temporal and spatial variations in concentration in order that
"best" and "worst" case simulation estimates can be obtained.  Considerable
research effort should be devoted to ecological modeling with this general best
and worst case scheme or goal.

     This inherent variability in biological systems affects the loads of toxic
chemicals that organisms may contain.  For example, in the James River many
adult bluefish have high concentrations of Kepone, while some have very low
concentrations because they have just recently migrated into the estuary. This
bimodality has to be considered in monitoring body loads.  It also should be
considered in predicting the body loads using mathematical models.

     The variability of bioaccumulation can be incorporated into predictive
models in at least two ways at the present time.  Mean parameter values can  be
used to obtain mean values of toxicants and distributional statistics can be
superimposed on these (Schofield and Krutchkoff 1974).  Different values of
individual parameters can be used with a Monte Carlo technique to obtain repe-
titive simulations and the output can be summarized in terms of best, worst,
mean and standard deviation estimates.  In either case, it is necessary to have
at least minimal statistics for the important parameters.

4.4.2  Validity and Reliability of Ecosystem Models

     Biological mediation of the transport and fate of toxicants can only be
modeled through the use of mathematical descriptions that connect chemical and
physical properties of the toxicant to observable properties of ecosystems.
The resulting models will inevitably contain a mix of probabilistic correla-
tions, process relationships, and system-level interactions.  Experimental
studies involving toxicants must include a parallel characterization of the
environmental processes that control the expression of the phenomenon in real
systems, including those factors that govern the dynamics of biotic components.
                                      119

-------
     The major area where research is imperative includes research that asso-
ciates ecosystem properties with toxicant dynamics.  Properties that indicate
the ability of systems to degrade toxicants cannot be generalized until the
environmental determinants of microbial activity are discovered.  At present
there is little rationale for predicting gross population sizes, the propor-
tional representation within the population of those organisms that will
actively degrade a toxicant, the time required for the microbiota to adapt to
the toxicant, or the net degradation rate that will develop over time in an
ecosystem setting.  In many cases toxicant processes may be intimately asso-
ciated with normal energy processing within the community.  This suggests that
research relating gross primary and secondary productivity, respiratory meta-
bolic activity, and energy flow through the community to toxicant dynamics
would be important.

     In a more general sense, the ways in which a toxicant can modify the "fate
processes" through its toxic properties per se are still largely unknown.
Microbiological studies using a toxicant as a sole carbon source carry the
hidden implication that the toxicant is itself a significant energy source for
the system.  Alternative plausible assumptions include the possibility that
degradation of the toxicant constitutes a simple drain on the metabolic capabi-
lity of the community or that it is an energy source whose use pattern will
depend on the availability of more readily degradable substrates.  Research is
needed to determine, as a function of the physical-chemical properties of the
toxicant, which of these assumptions is most likely to hold for important
classes of real ecosystems.  More importantly, both true "toxic effects" and
the introduction of exotic energy sources may distort normal ecosystem
functioning by differential effects within component subsystems.

     For example, the microbial community that normally functions in recycling
of inorganic nutrients could be modified either by toxic effects or by compe-
tition with toxicant degraders.  In an aquatic system is it easy to imagine a
chain of causation leading to a decrease in productivity, a consequent modifi-
cation of the pH regime, and a diminution of the rate of base-catalyzed
hydrolytic decomposition of the toxicant.  This causal chain would give a
positive feedback loop that could systematically destroy the system.  Fortu-
nately, real ecosystems are buffered against these processes by their struc-
tural complexity and by high diversities and niche overlaps that provide re-
dundancies in functional components.  However, research into the tolerable
levels of functional simplication, as determined by the magnitude of loading or
the properties of toxicants, is at present virtually non-existent.

     Research that associates ecosystem properties with toxicant transport is
also required.  There are several classes of biologically-mediated transport of
toxicants that have a major impact on the fate of the material.  In aquatic and
marine systems, filter-feeding organisms can capture large quantities of toxi-
cants sorbed to particulate materials and vastly accelerate the rate of incor-
poration of toxicants into the sediment phase of the system over the rate
suggested by simply physical principles.  Furthermore, bioturbation of aquatic
sediments and soil turnover by terrestrial invertebrates are the prime deter-
minants of the magnitude of total capture of a toxicant within the soil/sedi-
ment phase of many real systems.  Although the degree to which toxicants will
                                       120

-------
be entrained in these processes can perhaps be predicted from the sorptive
properties of the toxicant, the magnitude of entraining flows and the bulk of
materials involved (i.e., depth of mixed layers)  need further research at both
the descriptive and functional levels of ecosystem ecology.

     Some types of biologic transport may have only minor impact on the "fate"
of the compound, but nevertheless are important in that they amplify the toxic
impact of the material and shunt the toxicant into regions or components that
have not been directly contaminated by physical transport processes.  Transport
of toxicants across phase boundaries, through geographic barriers, and among
subsystems within a single ecosystem are important types of food chain trans-
port that can be accomodated only in the context of ecosystem analysis.

     Unfortunately, the general significance of the phenomena epitomized by the
consumption of DDT-contaminated fishes by bird life, the direct transport of
Kepone to crab populations in non-contaminated areas by bluefish migration.
and the advection of toxicants into detrital food chains via uptake by higher
plants, cannot be determined from anecdotal information alone.  Research is
needed to determine the minimum spatial scale required to accurately portray
ecosystem effects and the ecosystem properties that would impart a plausible
descriptive capacity to models that include toxicant movement through food
chains.  Although some pertinent ecological information is certainly available
such as home ranges and migratory ranges for animal species, toxicant body
burdens as a function of sexual differences and life-history stages and other
large-scale integrative effects have not yet been systematically incorporated
in models that are accessible to small-scale testing and validation.

4.4.3  Validation of Ecosystem Models

     A model that is theoretically sound and assembled in a way that allows for
extrapolation does not utilize arbitrary "calibration factors" or correlations
extrapolated beyond their range of observation.  Questions of reliability and
validity can be reduced to a series of technical assessments of the assumptions
and process designations used in constructing the model.  These assumptions
usually include several levels of integration ranging from molecular chemistry
to ecosystems.  Validation requires testing at the several levels of integra-
tion.  The objective of validation is to attempt to test the plausibility and
robustness of the model's assumptions.  Although all assumptions can be tested
in concert at the highest level of resolution represented in the model, a
simple test at this level will fail to identify those assumptions that are
causing discrepancies.

     The assumptions at the lowest level of integration are usually the most
accessible to laboratory experimentation, while ecosystem effects can only be
validated through critical monitoring and case studies.  However, it is impor-
tant to initially validate at lower levels of integration and proceed toward
field validation in a logical series of investigations.  In this way, the high-
level assumptions can be reduced to constant values, and the particular set of
assumptions to be tested in each investigation can be clearly specified.
                                      121

-------
     Time and space are probably interchangeable entities in ecological models
in that they both can be expressed as probability distributions for given eco-
logical interactions.  Because of this, it is not possible to restrict the
spatial dimension of an experimental arena without altering the frequency of
interactive events.  This is particularly true with macroscopic plants and
animals.  Many of the stability characteristics of ecological communities
result from the co-evolution and convergence of these same probability
distributions.  Abundant research with predation and inter-specific competition
has demonstrated the validity of these observations (Huffaker 1958;
Luckinbill 1973).

     It is essential, therefore, to develop a hierarchical experimental design
to determine the validity of predictive models that range from simple labora-
tory estimation of parameters to full scale hypothesis testing at the eco-
system level.  Parameters whose values are determined primarily by physical and
chemical processes should maintain their integrity through the hierarchy.
Parameters whose values are products of ecological interactions should demon-
strate convergence as the influence of scale is diminished.

     Experiments designed to determine the "calibration for scale" must involve
a degree of environmental regulation in order to buffer the experimental system
from uncontrolled exogenous influences.  These larger scale experiments will
involve online monitoring, pulsation experiments in series, and rapid assess-
ments of errors for model adaptation.

     The hierarchy involves estimates of rate coefficients from reactions
involving isolated processes.  First generation models then serve as complex
hypotheses which then can be challenged through controlled microcosm experimen-
tation.  Differences between expected and observed behaviors are then utilized
to modify the models.  Outdoor mesocosms (farm ponds, experimental streams,
isolated salt marshes, and field plots) can then be utilized to test the
validity and importance of ecological interactions.  This step will require
careful selection of the experimental.unit and/or partial environmental
regulation.  Finally, large scale episodic events like the Kepone incident can
be utilized to test hypotheses (models) at the regional level.  Environmental
regulation at this scale will not be feasible.  Continuous environmental
monitoring programs will be necessary to prevent "drift" of the environment
over time.

     Although particular tests can only be specified given a particular model,
a general scheme for organizing research programs can at least be outlined.
Program offices and granting agencies should consider the propriety of funding
particular projects as a function of the degree to which necessary antecedent
projects have been initiated.  Furthermore, a process of integration of the
efforts at various levels should be developed.

     1.  Laboratory studies:  given a sound relationship between environmental
variables and toxicant properties, the statistical noise and parameter values
of the relationships can be determined through direct laboratory studies.
                                       122

-------
     2.  Microcosm studies:  given a set of validated unit process relation-
ships, the assumptions used to assemble the units can be tested in the context
of a controlled, reduced set of environmental variables.  This test series also
examines one kind of completeness of the unit processes.

     3.  Mesocosm studies:  in larger scale tests (e.g. experimental ponds and
channels), the assumptions used to account for stochastic variation in environ-
mental functions can be assessed.  At this level, ecological ecosystem vari-
ables can be described (e.g., bioturbation of bottom sediments, population
densities), but their dynamics cannot be tested.

     4.  Case studies:  actual case studies serve to test model assumptions
concerning ecosystem-level effects of biological forcing functions on transport
and tranformation.

     Note that the spatial scale of the models and the level of integration of
the assumptions under investigation continually expand in the test series.  It
is important that the series be executed in the order specified because each
level of assumptions must be tested in a context that allows assumptions at
higher levels to be reduced to constant values.  It is important that the same
model be tested throughout the validation series, i.e., the assumptions used in
assembling "sub-models" must also be subjected to critical examination.

4.4.4 Microcosms

     In a broad sense, a microcosm is a laboratory approach to the study of
environmental processes as they occur in natural systems.  Microcosms are
capable of simulating only specific situations or portions of the environment.
To be useful a microcosm should incorporate the important components and pro-
cesses of natural systems and should be studied under controlled laboratory
conditions to yield reproducible results.

     Because of variations in species composition commonly experienced in
microcosm studies, process oriented measurements appear to be the best methods
for characterizing and defining parameters of microcosms.  Production to res-
piration ratios, nitrogen fixation, and heterotrophic activity can show good
consistency from one microcosm study to another.  As a result, microcosms can
be useful tools for assessing some fate and transport processes.  The rela-
tively small size of microcosms permits excellent versatility in determining
the effects of common environmental parameters.  The ability to test for these
effects will lead to definition of parameter limits and validation of the pro-
cess analyzed.

     Microcosms, as laboratory systems, do possess some inherent limitations.
Their small size is desirable for reproducibility and versatility, but makes
microcosms poor for simulations of total ecosystems.  Also, microcosms are not
capable of simulating fate and transport processes for larger animals whose
biomasses are generally greater than zooplankton or soil/sediment inverte-
brates.  They are, however, useful in performing fate studies with higher
plants.
                                      123

-------
     Synthetic or gnotobiotic microcosms are limited since complex biological
interactions are minimized and only species readily cultivated in the labora-
tory are used.  Naturally derived microcosms are affected by scaling factors,
lack of consistency in the inocula, and previous history.

4.4.5  Mathematical and Physical Model Recommendations

     Biological fate models that are to be used in assessing site-specific
impacts should be capable of representing separate segments simultaneously so
that biological heterogeneity can be taken into consideration.  Techniques for
representing final spatial resolution and/or automatically connecting adjacent
spatial segments should be adapted and refined for use in biological fate
models.

     Because a complete mass balance study is a major undertaking, criteria
should be established for deciding when such a study is necessary and when
measurement of only selected transport pathways and sinks may be substituted.

     Models should be formulated so that all parameters have physical signifi-
cance and are subject to estimation using laboratory or field techniques.  So-
called "free parameters" that do not have physical significance and that are
adjusted during the calibration of the model in order to obtain a better fit to
data should not be used.  To do so would result in the validity of the model
being subject to challenge in an adversary proceeding. All parameter values
should be documented in such a way that their validity can be established.
Where applicable, ranges of values should be given so that constraints can be
placed on the limits of parameters.

     Predictive models should provide information on variability of toxicant
loads in the biota appropriate for assessment purposes.  The variability should
be utilized in establishing probabilistic confidence intervals for any quantity
utilized in a regulatory fashion.

     For many classes of toxic substances, the transport and fate dynamics are
directly influenced by ecological processes.  Existing ecosystem models devel-
oped around energy and nutrient dynamics should be modified to "enslave" the
associated dynamics of particular toxic substances.  Existing models should be
utilized for which there exist physical analogs amenable to direct, experimen-
tal verification.

     Case studies should be selected and the hierarchical experimental design
implemented to calibrate the quantitative influences of "scale" and to iden-
tify the qualitative distortions due to typical ecological interactions.  The
case studies chosen should have a regional monitoring program already in opera-
tion.  The Kepone incident in the Chesapeake Bay or the Mirex incident in Lake
Ontario are good candidates.

     Microcosms, like any simulation of natural processes, should be validated.
Validation methods need to be established and tested.  Presently available
validation techniques which should be further tested and applied are:
                                       124

-------
     1.   Comparative studies relating microcosm and ecosystem experi-
         ments in which the environment is perturbed by a  chemical.

     2.   Calibration with standard reference chemicals for which
         information is known concerning fate and transport.

     3.   Development of correlations between the activities of
         various biological processes of natural systems and
         microcosms.

     4.   Mathematical modeling of the microcosms should be under-
         taken.  Efforts must be directed toward obtaining verified,
         quantitative measures of the various physical, chemical
         and biological parameters (e.g., chemical hydrolysis,
         photolysis, biodegradation and sorption-desorption)  from
         microcosm studies.

     5.   Design features of microcosms should be carefully examined
         and optimized prior to use for fate assessment.  Scaling
         factors, substratum to air or water rations, turnover of
         air or water, sensitivity to sampling variation in the
         inocula, nutrient cycling rates, and long-term stability
         should be investigated experimentally.
4.5  MONITORING

4.5.1  Use of Biological Monitors as Transport and Fate Indicators

     There are a number of chemicals which have been found to be long lived in
the environment and are scarcely affected by chemical or biological trans-
formation.  These chemicals may be more easily handled than those compounds
which are greatly transformed as they pass through the environment.  The
analysis of key organisms could provide positive indications for a large number
of chemicals which enter the environment.

     Surveys can be run using indigenous species,  such as has been done in
Mussel Watch (Golberg et al. 1978), looking for specific chemicals.  If the
chemical is found, one can work backwards and identify the pathways of these
chemicals through the environment.  If pathways are hypothesized, they could be
studied using introduced biological monitors as concentrators of the chemical.
This method has been shown to be viable by placing mussels in baskets along a
pollution gradient and retrieving them after a specified period of time for
tissue analysis.

4.5.2  Monitoring for Unknowns

     Utilizing state-of-the-art extraction and analysis technologies, Huggett
et al. (1979) observed over 400 individual species of halogenated hydrocarbons
in oysters from the Chesapeake Bay.  Only about a  dozen of these compounds  have
currently been identified and characterized.  The  majority of organic toxicants
                                      125

-------
moving through ecological pathways are unidentified and uncharacterized.  The
information is not currently available to structure and initiate productive
models for the vast majority of toxic substances.

     Early indication of the transport of a new toxic substance can be obtained
from time series analyses of organic "finger prints" taken from key organisms
associated with transport mechanisms.  Further research and regulatory options
are retained if similar materials are stored in a specimen bank such that the
chemical integrity of the samples are maintained.

     The temporal and spatial pattern of a new emerging peak is sufficient
evidence to mobilize human and monetary resources.  Furthermore, these patterns
give direction to the research and sampling efforts required to identify and
characterize the unknown.  If the indicator organisms are chosen correctly, the
major transport mechanism associated with specific ecosystems will be monitored
for the passage of new toxicants.  This approach will not preempt the release
of toxic substances, but will reduce the response time of the regulatory
agency.  In addition, the limited available human and monetary resources can be
allocated in a most effective fashion.

4.5.3  The Use of Chemical Surrogates in Transport and Fate

     The vast majority of toxic chemicals are proving to be organic in nature.
Analysis for these chemicals is typically very expensive, especially when
dealing with a large scale monitoring or survey program designed to define the
transport and fate of these chemicals in the environment.  One alternative is
to use a more easily analyzed chemical surrogate which has been shown to
function as a tracer for the chemical of interest.

     There are several possibilities of tags on the actual compound of interest
or an environmental tag useful on the process involved.  Radioactive tracers
are an obvious example of the first type.  A similar but less common example is
the possibility of following the path of halogenated hydrocarbons by analyzing
for the halogen, rather then the compound itself.

     The use of an easily analyzed chemical species, which can be shown to
follow the chemical of interest in a particular environment, is also a possi-
bility.  It has been shown, for example, in Narragansett Bay, R. I., that body
burdens of nickel can be directly correlated with the level of total hydro-
carbon in the tissue of Mytilus edulis.  Since the analysis of an organism for
metals can be one or two orders of magnitude lower in cost than for hydrocar-
bons, the use of such a surrogate may be a necessity as well as an attractive
alternative.  In a large mapping program, a surrogate could be determined and
used as the major tag.  A subset of samples could also be analyzed periodi-
cally for the chemical of interest to reconfirm validity of the technique.

4.5.4 Monitoring Recommendations

     EPA should support further research in the use of biological organisms
as concentrators for the study of transport and fate of various chemicals in
the environment.  For example, many organics are dumped into upper Narragansett
                                       126

-------
Bay and their movement through the bay system could he followed through the
analysis of mussels placed in specific patterns throughout the bay.  Software
should be developed to allow sufficient analysis of existing mussel watch data
to evaluate the application of such a monitoring program to the transport and
fate of chemicals in the environment.

     EPA should continue to support the development of a monitoring program for
"unknowns" and an associated specimen bank.  Specific attention should be
directed toward indicator organisms that reflect the major pathways of biologic
transport.  Particular emphasis should be given to the development of analyti-
cal techniques for pattern identification with time series data.

     A research effort should be initiated to determine the viability of using
comparatively easily analyzed materials as surrogates for the organic toxic
chemicals of interest.  For example, the data resulting from the first three
years of Mussel Watch could be run through statistical analysis to see if any
of the five metals analyzed could be used as surrogates for any of the organics
also analyzed in the same samples.  Large scale survey studies of the transport
and fate of chemicals which are prohibitively expensive due to the high cost of
analysis can become feasible if appropriate surrogates are available.
4.6  METHODOLOGIES

4.6.1  Software for Analyzing Monitoring Data

     A number of multivariate techniques exist for discerning patterns of
occurrence and for identification of anomalous occurences (Park 1975).  These
include cluster analysis, factor analysis, ordination, multiple regression, and
discriminant analysis.  Such techniques would be quite useful in analyzing
Mussel Watch data on "unknown" compounds.  These analyses would provide insight
into efficient allocations of human and monetary resources devoted to the
monitoring activity.  In addition, they would assist in the identification of
surrogate compounds.

     Multivariate analytical strategies have been utilized in geological and
ecological investigations for years (Parks 1966; McCammon 1978; Park 1974;
Allen and Koonce 1974), and are appropriate for objectively and systemati-
cally processing large quantities of data on body loads.  They reduce an almost
incomprehensible array of data matrices to indices and graphic patterns that
can be readily interpreted.

4.6.2  Speciation of Metals

     Research on trace metals in the aquatic environment often centers around
laboratory experiments utilizing the metal in the inorganic form.  Results from
these experiments are then extrapolated to the real world via mathematical
models.  A major pitfall in this logic is that the metals may not be in the in-
organic form in the environment due to chelation or complexation by organic
substances.  These organic forms can in turn be more or less biologically
available than the inorganic ones.
                                      127

-------
     One reason the organo-metallic species have been slighted is because of
the analytical difficulty in identifying and quantifying the complexes.  Com-
monly used methods for analyzing metals such as atomic absorption spectropho-
tometry or activation analysis do not differentiate between chemical species
unless special extraction techniques are employed, and even then the state-of-
the-art is marginal.

     New analytical methods to identify speciation are being developed.  An
example is the coupling of high pressure liquid chromatography with Graphite
Furnace Atomic Absorption.  Research should be funded that develops new
techniques for identifying and experimenting with the transport of
metallo-organic compounds.  Without this experimentation work will continue
with inorganic metals as surrogates of real world complexes which could result
in errors in estimating biological effects, biological transformations, and
biological transport of trace metals.

4.6.3  Analtyical Techniques

     Toxic compounds which enter the environment from anthropogenic sources are
supposedly controlled via the permit system.  When a permit is issued for dis-
charge, there is an implied assumption that the discharge will be monitored to
assure compliance.  For many of the substances entering the environment, ana-
lytical methods to determine the compound at environmental levels (i.e., yg/g,
yg/g) in water, tissue, and sediments do not exist.  For many substances which
are permitted, the ability to monitor is hampered by inadequate manpower and
instrumentation at both state and federal levels when analytical methodologies
exist.  Therefore, the role of organisms in transport and transformation of
these substances is inadequately known for natural field situations.

     The modeler must incorporate into the mathematical expressions, which
approximate the real world, those pathways which will affect the distribution
of the substance.  If better monitoring systems were in effect for anthropoge-
nic compounds in the environment, including the biotic compartment, then these
data and the resulting conclusions could be incorporated into future models.

4.6.4  Exposure Determination

     Prediction of risk requires knowledge of:  the chemical and physical form
and the biological availability of the toxic substance, the total amount in-
corporated by the receptor, the rate and pathway of incorporation, the fre-
quency and duration of exposure, the time intervals between the periods of
exposure, and an understanding of the life history and population characteris-
tics of the organism involved.  In the past, aggregate statistics on the con-
centration of substances in soil, water or air medium have been used to
evaluate hazards and the above mentioned details have been lacking.

     The aggregate statistic approach is inadequate because the process of bio-
accumulation and biotransformation can markedly alter the experiences or
effective concentration of the substance.  Another factor is that normal
population processes produce variations in uptake and accumulation rates as a
function of seasonality, diurnal rhythms, genetic state of the population, and
                                      128

-------
stage of life history development.  Also, the repair and recovery processes
operate during intervals between periods of incorporation and the adaptive,
avoidance or exclusion responses can function as behavioral alternatives during
periods of exposure to toxic substances.

4.6.5  Methodology Recommendations

     Generally accepted multivariate analytical strategies should be employed
in the analysis of monitoring data (like Mussel Watch)  so that patterns,
trends, and anomalous occurrences can be readily perceived.

     Work should commence immediately to examine existing analytical techni-
ques for analyzing metallo-organics and their applicability for utilization in
microsm-mesocosm-megacosm research.  These metallic species must then be incor-
porated into existing or new mathematical models on the fate and transport of
trace metals.

     Analytical techniques must be developed to allow better monitoring of
existing systems into which anthropogenic substances are discharged.  The dis-
charges chosen should be those which contain compounds with a wide range of
organic structures.  Techniques should be developed which allow extraction and
analyses of multiple substrates - i.e. animal tissues,  plant tissues, sedi-
ments.

     The concept of exposure must be expanded to include active ecological and
behavioral responses of organisms to the presence of toxic substances.  Factors
like those listed above should be evaluated and incorporated into risk assess-
ment models.
4.7  EXISTING DATA REQUIREMENTS

4.7.1  A General Overview of the Types of Information Currently Reviewed by
       EPA in Considering the Registration of Pesticides in the Area of Fate
       and Transport

     EPA reviews a number of physical and chemical parameters including melting
point, boiling point, water solubility, solubility in organic solvents, vapor
and quantity of impurities, octanol/water partition coefficient, explosive-
ness, miscibility, corrosive characteristics, specific gravity, and pH.

     Other data reviewed in the area of environmental chemistry include:

     A.  Physicochemical Degradation

         1.  Hydrolysis (at different pH values)

         2.  Photolysis (soil surfaces and aqueous)
             Sufficient data are required for these studies so that rate
             constants can be calculated.
                                      129

-------
B.  Metabolism

    1.  Plant (rotational crop uptake)
        These studies require the identification of the nature
        of plant residues in mature crops.

    2.  Soil (aerobic and anaerobic)
        These tests are qualitative in nature and emphasize the nature
        of the degradation products rather than their amounts.
        However, studies outlined under Section D require
        field persistence data of parent compound and signi-
        ficant degradation products.

    3.  Aquatic (aerobic and anaerobic)

    4.  Bird (poultry)

    5.  Microbial

        a.  Effects of pesticides on microbes

            These studies can involve either the effects on microbial
            function or microbial populations.  Some effects cannot
            be measured directly and population studies may be the
            only recourse.  Data obtained include 02 consumption,
            CO2 evolution, nitrogen cycle reactions, and measure-
            ment of enzyme activity for dehydrogenase and phos-
            phatase.  When the functional approach is used, data
            on the effects of nitrogen fixation, nitrification
            and degradation of cellulose, starch and proteins
            may be utilized.

        b.  Effects of microbes on pesticides •

            Effects are determined on pure or mixed culture popu-
            lations of representative microorganisms from soil
            or water  including free-living nitrogen fixing
            bacteria  and blue-green algae such as Czatobocler,
            Clostridium, Nostoc, and nitrofiers such as Nitro-
            somanos and Nitrobacter.  For cellulose, starch,
            pectin and similar substrates, selected soil
            bacteria, actinomyses, and molds should be utilized.

    6.  Activated Sludge

C.  Mobility

    Column  leaching studies and/or soil TLC and/or adsorption/desorp-
    tion  studies are  done on  solid and sediment.
                                  130

-------
     D.  Persistence Under Field Use Condition

         1.  Soil

         2.  Sediment

         3.  Water

         4.  Specialized Ecosystems

     E.  Accumulation

         1.  Irrigated Crops

         2.  Fish

             a.  Flow thru systems

             b.  Static laboratory systems containing treated soil
                 (sediment)

         3.  Field Study

     Frequently the conduct of the above-mentioned studies is facilitated by
the use of radio-labeled compounds with routine mass balance for laboratory
type studies.  Identification of major degradation products is routinely
performed either by co-chromatography with model compounds and/or mass spectro-
metric confirmation.  The most recent statement of EPA's need for data can be
found in the Federal Register of July 10, 1978, Part III,  Registration of
Pesticides in the United States:  Proposed Guideline for Registering Pesticides
in the United States.

     Additional information required of pesticides for which tolerances on food
are established include the decline of residues versus time between treatment
and harvest and metabolism in plants and rats (or other animals).  These data
requirements in support of a pesticide tolerance are provided for by certain
parts of the Federal Food, Drug and Cosmetic Act.

     The exact type and quantity of data required of a pesticide registrant is
dependent on the proposed pattern(s) of use.  More data are required for pesti-
cides to be used on or near water than for certain limited terrestrial uses.

4.7.2  A General Overview of the Types of Information to be Reviewed for Toxic
       Substances

     The chemicals that will have to be regulated by the Office of Toxic
Substances (OTS) under TSCA fall into the categories of new chemicals (Section
5) and existing chemicals (Section 4).  In general, the premanufacturing
notification form for new chemicals will include provisions for the informa-
tional elements listed for pesticides (except for those marked as "not
applicable" (n/a)).  Since Section 5 is not a set of requirements but merely
                                      131

-------
serves as a guideline, it is problematical as to how much of these data will
actually be provided by the potential manufacturer.  Additional information can
be obtained through court action under Section 5e only if EPA can make a case
that it is essential to make a risk assessment.

     Test protocols for existing chemicals are going to be published as re-
quirements.  However, of the 70,000 chemicals already in commerce, only a few
will be selected annually for inspection.  This process begins with the pre-
paration of a dossier on each of the selected chemicals or chemical classes
after which data deficiencies are noted and manufacturers notified that certain
critical testing has to be performed to correct these deficiencies.  It is
estimated that EPA will consider 50-150 such chemical/classes per year.  So
far, a review of about 20 dossiers indicates that most contained essential
physical/chemical data but had little persistence data (none photochemical),
and none contained much in the way of ecological effects data except for short
term (acute) toxicity data on relatively few species.

     Thus, ball park figures indicate that the score is:

     •  Physical/chemical data - 70 percent

     •  Persistence data - 5 percent

     •  Ecological data - 15 percent

     •  Photochemical - 0 percent

     Since EPA Section 5 test guidelines are only in draft form and Section 4
test rules are just being formulated, it is not yet possible to spell out the
complete spectrum of testing guidelines and requirements that the Office of
Toxic Substances will eventually develop.  TSCA mandates that all test
protocols be reviewed annually so they will always be amenable to changes as
required.

4.7.3  Parameters Required by PEST Model (Park et al. 1978)

Solution

     SOLS        -  solubility of the solid
     EK          -  dissolution constant

Adsorption-Desorption

     K1          -  rate constant for adsorption
     K2          -  rate constant for desorption
     MO          -  limiting number of moles adsorbed per grain

Volatilization
     H*          -  Henry"s law constant
     P^          -  Partial pressure of solute
                                       132

-------
      AG
   liquid mass transfer coefficient
   gas mass transfer coefficient
-  partial pressure of A at film-gas interface
-  partial pressure of A. at film-liquid interface
Chemical Degradation
     KO
     KH
     KHO
     KHA
     KHB
     HA
     HB
     NA
     NB
     OPTPH
     SIGPHA
     SIGPHB
     KSI
     EPT

Excretion

     KEXFAT
     KEXPRT

Defecation
   uncatalyzed rate constant
   hydrogen ion constant
   hydroxide ion constant
   rate constant for hydrolysis by the
   rate constant for hydrolysis by the i
                                    Lth
                                     th
acid
base
                        ith
                            acid
                            base
concentration of the
concentration of the
number of acids
number of bases
optimum value of pH
parameters used to "shape" reduction
factors for non-optimum pH conditions
quantum yield of reaction considered
molar absorptivity of TOM at particular wave length
-  proportion of TOM excreted under starvation conditions
   proportion of TOM excreted under normal maintenance
     E           -  proportion of TOM not assimilated

Microbial Metabolism
     AA
     MUO
   four regression coefficients of second-order fit of
   growth and mortality to TOM concentrations

   growth rate coefficient in absence of TOM
     KDO         -  mortality coefficient in absence of TOM

4.7.3.1  Data/Model Recommendation

     The information required to initiate and utilize an ecosystem model like
PEST should be compared with the existing requirements for data under EPA rules
and regulations.  A strategy for addressing the discrepencies should be
immediately addressed.  Otherwise, even existing models will not be available
for routine risk assessment.


          The Biota Group participants are listed in Table 4-1.
                                      133

-------
                     TABLE 4-1.  BIOTA GROUP PARTICIPANTS
Non-Agency Participants

Robert T. Belly
Eastman Kodak Company
Rochester, NY

William Cooper (Chairman)
Department of Ecology
Michigan State University
East Lansing, MI

Corwin Hansch
Department of Chemistry
Pomona College
Claremont, CA

Robert J. Huggett
Virginia Institute of Marine
  Science
Gloucester Point, VA

Jay Jacobson
Boyce Thompson Institute
Cornell University
Ithaca, NY

John Matheson
Bureau of Veterinary Medicine
U.S. Food and Drug Administration
Rockville, MD

Richard Park
Center for Ecological Modeling
Rennselaer Polytechnic Institute
Troy, NY

Richard G. Wiegert
Department of Zoology
University of Georgia
Athens, GA

M. A. Q. Khan
Department of Biological Sciences
University of Illinois (Chicago
  Circle)
Chicago, IL
 EPA Representatives

 Lawrence Burns
 Environmental Research Laboratory
 Athens, GA

 Caroll Collier
 Office of Pesticide Programs/
 Office of Toxic Substances
 Washington, D.C.

 Walter Galloway
 Environmental Research Laboratory/
 Office of Research and Development
 Narragansett, RI

 Robert K. Hitch
 Office of Pesticide Programs/
 Office of Toxic Substances
Washington, D.C.

 Melvin Nolan
 Office of Health and Ecological
   Effects/
 Office of Toxic Substances
 Washington, D.C.

 Hap Pritchard (Lead)
 Environmental Research Laboratory
 Gulf Breeze, FL

 Shahbeg Sandhu
 Health Effects Research Laboratory/
 Office of Research and Development
 Research Triangle Park, NC

 Arthur M. Stern
 Office of Toxic Substances/
 Office of Research and Development
 Washington, D.C.
                                      134

-------
                                  REFERENCES
Allen, T. F. H. and J. F. Koonce.  Multivariate Approaches to Algal Strategems
     and Tactics in Systems Analysis of Phytoplankton.  Ecology/  54:1234,
     1973.

[Bentley, P., H. Schmassman, P. Sims and F. Oesch.  Epoxide Derived from
     Various Polycyclic Hydrocarbons as Substrates of Homogenous  and Microsome
     Bound Epoxide Hydratase.  Eur. J. Biochem., 1978].

Burke, M. D. and R. T. Mayer.  Inherent Specificities of Purified Cytochromes
     D-450 and P-448 Toward Bipheny Hydroxylation and Ethoxyresorufin
     Hydroxylation.  Drug Metabol. Dispos., 3:245, 1975.

Dauterman, w. C.  Extramicrosomal Metabolism of Insecticides.  In:  Insecticide
     Biochemistry and Physiology, C. F. Wilkinson, ed., pp. 149-176.  Plenum
     Press, New York, 1976.

Dayton, P. K.  Competition, Disturbance, and Community Organization:  The
     Provision and Subsequent Utilization of Space in a Rocky Intertidal
     Community.  Ecol. Monogr., 41:351, 1971.

[Doroug, W.  In:  Differential Toxicities of Insecticides, F. Matsumara, ed.
     Pharmacology and Therap., Pergamnon Press, 1979.]

Goldberg, E. D.,  et al. The Mussel Watch.  Environmental Conservation, 5:101,
     1978.

Hall, D. J., W. E. Cooper and E. Werner.  An Experimental Approach to the
     Production Dynamics and Structure of Freshwater Animal Communities.
     Limnology and Oceanography, 15:839, 1970.

[Hollingworth, R. M.  In:  Reactions to Environmental Agents.  Handbook of
     Physiology, Section 9, D. K. H. Lee, H. L. Falk, S. D. Murphy and S.  R.
     Geigler, eds.  Amer. Physiol. Soc., 1976.]

Huffaker, C. B.  Experimental Studies on Predation:  Dispersion Factors and
     Predator-Prey Oscillations.  Hilgardia, 27:343, 1958.

Huggett, R. J., M. Nichols and M. E. Bender.  Kepone Contamination of the James
    River.  Manuscript in preparation, 1979.
                                       135

-------
Leung, D.  Modeling in the Bioaccumulation of Pesticides in Fish.  Center for
    Ecological Modeling, Rensselaer Polytechnic Institute, Report, 5,  1978.

Luckinbill, L. S.  Coexistence in Laboratory Populations of Paramecium aurelia
     and its Predator Didinium nasutum.  Ecology, 54:1320, 1973.

Mazel, P.  General Principles and Procedures for Drug Metabolism in vitro.  In:
     Fundamentals of Drug Metabolism and Disposition, B. N. Ladu, H. G.  Mandel
     and E. L. Way, eds., Williams and Wilkins, 1972.

McCammon, R. B.  Multiple Component Analysis and its Application in Classifica-
     tion of Environments.  American Association of Petroleum Geologists
     Bulletin, 52:2178, 1968.

Neill, W. E.  Experimental Studies of Microcrustacean Competition, Community
     Composition and Efficiency of Resource Utilization.  Ecology, 56:809,
     1975.
Oesch, F., D. M. Jerina and J. W. Daley.  A Radiometric Assay for Hepatic
     Epoxide Hydratase Act
     Acta., 277:685, 1971.
Epoxide Hydratase Activity with 7  H-Styrene Oxide.  Biochem. Biophys.
Omura, T. and R. Sato.  The Carbon Monoxide Binding Pigment of Liver
     Microsomes, II:  Solubilization, Purification, and Properties.  J. Biol.
     Chem., 239:2379, 1964.

Paine, R. T.  Food Web Complexity and Species Diversity.  American Naturalist,
     100:65, 1966.

Park, R. A.  A Multivariate Analytical Strategy for Classifying Paleoenviron-
     ments.  Math. Geology, 6:333, 1974.

[Park, R. A.  Ecological Modeling and Estimation of Stress.  Second Joint
     U.S./U.S.S.R. Symposium on the Comprehensive Analysis of the Environment.
     USEPA. (Translated and published simultaneously in U.S.S.R.), 1975.]

Park, R. A., L. S. Clescari, P. DeCaprariis, R. Haimes, H. F. Herbrandson, W.
     Reeves, S. Ross and W. W. Shuster.  Modeling Transport and Behavior of
     Pesticides in Aquatic Environments.  Center for Ecological Modeling,
     Rensselaer Polytechnic Institute, Interim Report for USEPA Grant No.
     R804820010.  Troy, N.Y.,  1977.

Parks, J. M.  Cluster Analysis Applied to Multivariate Data.  J. Geology, 74:
     703, part 2, 1966.

Schofield, W. R. and R. G. Krutchkoff.  Stochastic Model for a Dynamic
     Ecosystem.  Virginia Water Resources Research Center, Virginia Polytechnic
     Institute Bulletin 60, 1974.

Stanton, R. H. and M. A. Q. Khan.  Components of the Mixed-Function Oxidase of
     Hepatic Microsomes of Freshwater Fishes.  Genes Pharmacol., 6:289,
     1975.
                                        136

-------
U. S. Environmental Protection Agency.  Fate of Toxicants in or on Plants.
     In:  Test Methods for Assessing the Effects of Chemicals on Plants.
     EPA-68-01-2249.  Final Report to Office of Toxic Substances.  Franklin
     Institute Research Laboratories, 1975.

Yang, R. S. H.  Enzymatic Conjungation and Insecticide Metabolism.  In:
     Insecticide Biochemistry and Physiology,  C. F. Wilkinson,  ed.,  pp.
     177-277.  Plenum Press, New York, 1976.
                                       137

-------
     5.0  EXPOSURE ASSESSMENT AND MODELING IN TRANSPORT AND FATE RESEARCH
5.1  INTRODUCTION

     The need for multimedia models of the transport and fate of toxic
chemicals in the environment has been widely recognized only in the past ten
years.  Prior to the late 1960's, comprehensive, detailed models of
environmental transport and fate had been developed primarily only for
radionuclides.  Models for restricted applications such as dispersion models
for air pollutants had also been developed prior to this date.  Other relevant
studies conducted at that period fall into the following categories:

     (1) Physical transport models for commonly recognized wastes;

     (2) Empirical studies of transfer processes such as soil
         erosion;

     (3) Specific studies of individual problems such as pesticide
         residue dynamics and occupational health exposures.

     The development of physical transport models for oxygen-demanding wastes
was begun in the 1950's and has continued to serve as the scientific base for
current efforts in this area.  These models were important precusors of
present-day exposure models, but until recently no attention has been given to
the application of these models to wastes other than domestic, municipal, and
industrial sewage.  During the same period, empirical methods for estimating
soil erosion rates such as the "Universal Soil Loss Equation" were developed as
numerical aids for field extension agents in attempts to improve erosion
control.  These methods were based on commonsense understanding of the major
mechanisms behind material loss and transport, but were largely restricted to
specific sites such as plowed agricultural plots and did not connect transport
to exposure in any real sense.  Finally, specific issues of pesticides and
occupational health exposures were considered as isolated problems that could
be analyzed without close analytical attention to multimedia transport.

     The discovery in the late 1960's that DDT is transformed into DDE and
causes important environmental effects away from the site of application was
perhaps the first stimulus for a systematic study of the transport and fate of
toxic substances.  The discovery of the methylation of inorganic mercury
focused concern on the linkage between biological and chemical transformations.
It was also recognized at that time that combinations of pollutants in one
medium could have total impact greater than the sum of the impacts of each
individual pollutant (e.g., photochemical smog).  These observations further
emphasized the need for complete assessments of pollutant transport and
transformation.

                                      138

-------
     Prior to the mid-1970's, most exposure models for toxic substances were
developed on an ad hoc basis for application to specific pollutants which were
known or suspected to be causing adverse effects.  In most such cases the
observed effects dominated discussion of the problem and thus guided the
formulation and development of the exposure models.  Only in very recent years
has the scientific community begun to develop a general methodology for
multimedia assessment of exposure to toxic substances which is independent of
immediate concerns with effects.  The concepts of "materials balance" (Caltech
1978) and "critical pathways" (Friedlander et al. 1970) have been recently
introduced into assessment of exposure to toxic substances, and are now being
employed as currently accepted techniques.
     Some of the earliest models of the environmental distribution of toxic
chemicals were based upon knowledge derived from models formulated for
radionuclides (for example, see Woodwell et al. 1971).  Throughout the period
of development of models for toxic substances, work has continued on models of
the transport, distribution, and fate of radionuclides, and these models are
still much better developed than those for other toxic substances (for a
summary of the available models see section 5.5.1 below).  We expect that the
development and improvement of comprehensive techniques for exposure modeling
of toxic chemicals will continue to benefit from experience gained in studies
of radionuclides.  In addition, multimedia studies will depend on the
development of improved methods for modeling distribution and transport of
chemicals in a single medium.  These methods are discussed in other chapters in
this report.

     The overall emphasis in this chapter is on the modeling of exposure levels
across more than one environmental medium.  Current multimedia studies are
described in detail in sections 5.5 through 5.11 of this chapter.  The small
number of available studies, combined with the discussion above, should be a
clear indication to the reader that the evolution of exposure modeling efforts
is at a very early stage and that historical experience provides only a broad
sense of the scope of effort required.
5.2  OBJECTIVES OF MULTIMEDIA MODELS

     The usual purpose of constructing an exposure model is to derive
predictions of the exposure  of target populations to a chemical resulting
from various types of use of release into the environment.  Most models are
developed in a regulatory context, to estimate whether or not specific uses of
a chemical will lead to exposures that will cause adverse effects on target
populations.  For this purpose, it is often necessary to construct models that
will establish cause-and-effeet relationships between specific releases of
chemicals into the environment and resulting exposures to sensitive
populations.
 1The term  "exposure" has been used in a number of different ways.  For the
 definition used in this chapter see Section 5.9 below.
                                      139

-------
     Multimedia models are necessary when a chemical either is released into
more than one environmental medium, or is transferred from one medium to
another after release, resulting in multimedia exposure.  Multimedia pollutants
are inherently more difficult to study than single-medium pollutants because
transfer coefficients are often difficult to measure and are very difficult to
calculate theoretically.  In cases where a pollutant has multiple sources,  it
is often very difficult to identify the critical pathways linking the most
important exposures with specific sources.

     Multimedia exposure models for toxic substances vary considerably in
complexity ranging from simple empirical relationships or arithmetic exercises
to complex computer models with many adjustable parameters.  To avoid wasting
resources on inappropriate levels of study or of regulatory control, it is
important to match the scale of the model to the importance of the problem  and
to the amount of data available.  In many cases it is desirable to use an
iterative approach, in which successively more complex models are used to
define the potential significance of a pollutant, to make rough estimates of
likely exposures, to design a monitoring program to measure actual
concentrations in the environment, and to make a detailed analysis of critical
pathways.  When a model predicts that hazardous exposure concentrations are
likely to occur, it is very desirable to conduct sensitivity studies to
determine whether the predicted exposure levels are strongly dependent on
assumptions made about source strengths or transfer coefficients.

      Because estimates of exposure derived from modeling exercises are often
uncertain, it is desirable to test the model at all stages of development
against field measurements.  The usual approach is to follow an iterative
pattern in relating the model exercises to field measurements.  Usually, the
first indication of the relative importance of the variables is apparent in
bodies of observational data.  The next step is to construct a model on the
basis of either intuition or a deterministic physical relationship that
reflects the trends seen in the data.  The model is then applied to the range
of conditions in the data base, and uncertainties as to the adequacy of the
model or of the completeness of the data base become evident.  Questions which
arise can usually be answered only through further measurement programs.  In
this way a model appropriate to the specific pollutant is developed.
5.3  CHARACTERISTICS OF MULTIMEDIA MODELS

     The basic structure of an exposure model is illustrated in Figure 5-1.
The system is initially characterized by the initial distribution of the
chemical and by the characteristics of the chemical and of each of the media
which control the behavior of the chemical within and between them.  The system
is driven by the source terms, which describe the rates of release of
the chemical as functions of space and time.  These input data are processed
according to the logic specified by the modeler and are used to derive the
output data, which are predicted concentrations, and the chemical as functions
of space and time.  In certain cases the model may be "run" backwards, using
the observed distribution of the chemical in the environment to derive
inferences about the relative importance of different sources.
                                       140

-------
 LUDCQ
   Ul
LU
DC
Z)
I-
   s.cc

     CO
                CO
                LU
                CC   H
                LLI   DC
                Q-   O
                O  -"Q-

                £-.32

                sS5&:
                253^3
                ,9Qot=m
                     CO
                     Z

                     <
                     QC
                                         D
                                         O
                                         S

                                         u
                                         H
                                         E-
                                         W
                                         H
                                         z
                                         w
                                         W
                                         D
                                         Cm
                                         O
K
D
E-i
L)
D
-o;
                                         a
                                         H
                                         u
                                         w
                                         D
                                         O
141

-------
     The specifications of the model and of its input and output data may vary
greatly depending on the characteristics of the chemical and the purposes for
which the output data are to be used.  For example, the spatial and temporal
scales of interest depend on spatial and temporal scales of release, on the
lifetime of the chemical, and upon transfer rates.  For a short-lived chemical
which is transformed rapidly into another chemical species, or transferred
rapidly into another medium, the primary interest may be in defining local
diffusion scales or short-term fluctuations in concentration.  For long-lived
or slowly-transferred chemicals the primary interest may be in long-range
transport or accumulation over long time-scales.  Depending on the
characteristics of the source terms and of the environmental transport and
transformation processes, interest may lie in predicting either long-term
average exposures or short-term fluctuations in exposure to specific sensitive
targets.  In each case the characteristics of the model should be adapted
approximately to the characteristics of the problem.
5.4  SOURCE TERMS

     One objective of modeling is to relate source emission data to ambient
environmental quality.  In many cases the ability to project steady state or
dynamic ambient loadings is limited by incomplete understanding of the source
terms.

     In addressing the transport and fate of hazardous substances in the
environment, data on physicochemical characteristics and on discharge rates
from specific sources are required.  These data are needed for all stages of
introduction of chemicals into the environment including losses during
production, distribution, usage, and ultimate disposal.  Any combination of
these environmental source modes may contribute significantly to the total
environmental burden of the chemical.  Consideration of specific examples of
loss into the environment will be used to illustrate the difficulties that may
be sometimes encountered in characterizing these source terms.

     During production, losses to the environment are determined usually by
monitoring of process effluents to air, water, and land.  These losses are
usually too small for determination by materials balance methods.  For chemical
processes the closure of material balances on feedstocks, intermediates, or
product streams to within a few percent would represent adequate closure for
purposes of monitoring the process, but would not quantify the residual or
effluent streams to the degree of accuracy required.

      Certain source terms - such as those for nitrogen oxides produced by
thermal fixation of atmospheric nitrogen - are strongly related to process
conditions.  Other source terms, such as those for particulate emissions, are
strongly dependent on the efficiency of the air pollution control devices.  For
these reasons, source emissions cannot be predicted in most cases and must be
determined by measurements on the effluent streams.  Often other factors such
as particle size may have an important impact on the transport of the
pollutants and characterizations other than the total quantity of effluents are
                                      142

-------
necessary.  For example, the distribution of lead emitted into the Los Angeles
Basin has been found to be governed by the distribution of the size of lead
particles emitted by automobiles .

     Many of the early episodes of environmental contamination could be related
to stationary and mobile combustion sources or chemical plants.  Systematic
measurements of effluents from these sources have provided statistical bases
for evaluating emissions (U. S. Environmental Protection Agency 1976).  These
data, although useful for establishing acceptable limits of emissions for
maintaining environmental quality, cannot substitute for the site-specific
monitoring information needed for purposes of validation of models.  The Toxic
Substances Control Act (TSCA) should be implemented to limit environmental
releases to levels which are estimated by exposure-modeling to constitute
"safe" ambient loadings.  Source monitoring will continue to be needed to
ensure that these emission levels are met or to provide data for validation of
models.

     Another significant factor in source description relates to the consumer
usage of the substance in question.  In cases where the principal use of the
chemical is dispersive (e.g., pesticides and aerosol propellants), the emission
rates can be readily estimated from statistics on production and sales;
additional information on geographical distribution of use may need to be
inferred from regional marketing data in order to identify regional source
terms.  Another category of use where widespread dispersal of chemicals into
the environment may occur, often in amounts equal to production, is that of
additives which may be vaporized or leached from a product during its useful
lifetime.  A well-publicized case of this type is the vaporization of
plasticizers such as phthalates out of consumer plastics.  In view of the wide
range of formulations of polymer products and the sensitivity of the
vaporization rate to both the physical and chemical properties, one can
anticipate major problems in obtaining reliable data on losses from all the
products that fall into this category.

     Models based on the physicochemical properties of the materials could be
helpful in providing a framework for correlating losses from products of widely
varying shape and composition.  Unscheduled use of hazardous substances
represents another path which can lead to significant exposures.  For example,
PCB's which had lost their functional dielectric properties are known to have
been used as anti-dusting agents in parking lots or as herbicides along
roadways; these are uses never envisaged by the manufacturer.  The above
examples of losses to the environment during usage constitute a class of
fugitive releases which will usually be very difficult to quantify by
integrating the spectrum of usages.  However, recognition that some fraction of
the total production of a chemical does not follow the normal track to eventual
disposal may allow conservative estimates of loss to be made.

     Ultimate disposal of hazardous compounds should not be confused with
sequestering of these compounds in storage or reservoir areas.  Understanding
the distinction between storage and disposal is a prerequisite to describing
the total environmental cycle and multimedia materials balance for the
substance.  During storage, in the absence of biodegradation, losses to the
                                      143

-------
environment proceed over time and ultimately may result in complete release.
In a steady-state economy in the longterm, the net effect of storage in "leaky"
dumps is to provide a delay time, of the order of magnitude of the residence
time in a dump, between time of disposal and time of release into the
environment.  The recent incident at Love Canal is an illustration of a delay
between the time of disposal of a chemical and its appearance in the
environment.  The cumulative effect of increasing numbers of hazardous chemical
storage areas can be expected to result in more frequent situations of this
type.  From the point of view of the modeler, temporary sequestering of
chemicals in dumps requires the inclusion in the model of compartments with
long retention times.

     An alternative means of disposal is by thermal degradation by pyrolysis  or
incineration.  In assessing these destruction processes for hazardous waste
materials, caution is necessary to insure that trade-offs of one toxic
substance for another do not occur.  High temperatures during incineration may
enhance the loss of volatile metals such as mercury and arsenic, and/or the
formation of hazardous organic compounds (e.g., polycyclic aromatic
hydrocarbons). These emission streams then become an additional source term
associated with the subject compound.

     Spills of hazardous substances constitute a generic class of intermittent
sources with potential for leading to locally acute exposures.  Although spills
may account for only a small fraction of the total materials balance, protocols
for integrating this source term into exposure modeling should be established.

     The goal of source assessment for hazardous substances is to set up
material flow diagrams for production through ultimate destruction which will
identify all entries into the biosphere and all reservoirs.  Research by Nisbet
and Sarofim (1972) may serve as an illustration of the value of simple material
balances.  In that work consideration of the fluxes of PCB's through the
environment resulted in identification of river and lake sediment as a
reservior for PCB's before field measurements identified the importance of this
temporary sink.  Materials balances are indeed proving to be a powerful tool  in
identifying and quantifying important environmental pathways in urban areas
(Abrott et al. 1978; Caltech 1978).

     Although it is desirable to use mass balances to provide a systematic
accounting of the flow of chemicals in the environment, there may be critical
pathways to man or other receptors that may involve a very minor use of a
chemical.  The use of solvents in confined quarters and of chemicals in
cosmetics are examples of major pathways to exposure to man which may represent
only minor uses of a chemical.  Although a number of such pathways are governed
by existing legislation, judegment will always be required in anticipating
problems associated with specified low volume uses of chemicals.  The
construction of scenarios (Friedlander et al. 1970) provides a possible
mechanism for identifying a few problems that would not be detected by
conventional modeling techniques because of the small amounts of chemicals
involved.  However, there will undoubtedly be incidents, such as the formation
of dioxins through the reactions of chlorinated phenols at high temperatures,
that will not be predicted by any model.
                                      144

-------
5.5  MATHEMATICAL MODELS

5.5.1  Available Models for Radiological Assessments
     The National Laboratories of the Department of Energy have accumulated
experience in the area of radiological assessment and have developed assessment
methodologies incorporating transport models and external and internal
dosimetry models.  The models for radionuclides are assessed in this section in
order to provide a starting point for the more complex considerations required
for modeling chemical pollutants.
                                              o
     Table 5-1 provides a compilation of codes  which may be implemented for
assessment of radiological consequences of discharges, routine or accidental,
from nuclear power facilities.  The table identifies the transport processes
and human uptake mechanisms covered by each code.  Blank areas in the table
occur if the column designation is not applicable to the code as determined
from the available documentation.  Supplementary information on some codes and
descriptions of additional codes for accidental release of radionuclides are
available (Winton 1969, 1971, 1974; Strenge et al. 1976).

     Figure 5-2 depicts the frequency of occurrence of the computer codes
compiled within various environmental transport and dosimetric categories.  The
figure indicates that the assessment of atmospheric dispersion, external
dosimetry and internal dosimetry via inhalation have predominated in the
development of radiological assessment computer codes.  This emphasis probably
reflects the need for such codes in safety evaluation work associated with
Preliminary Safety Analysis Reports (PSAR's) in which assessment of radiologi-
cal consequences of accidental releases is of primary importance.

     The incorporation of other pathways of exposure into computer codes for,
assessing radiological safety has been a fairly recent development and reflects
the need for satisfying the current requirements of environmental legislation
and the needs of regulatory agencies.  Previously, calculations of potential
exposures resulting from food chain transport and the subsequent ingestion of
food were performed for a few nuclides (e.g.,    I,   Sr,    Cs).  Computations
by hand were usually satisfactory for these cases.  This situation changes with
the inclusion of a large number of nuclides in the source term and the need to
estimate doses resulting from multiple exposure pathways.  Nevertheless, fewer
than 10 percent of the codes in the review have the capability of estimating
aquatic and terrestrial transport processes.
2 A "code" in the sense used in this section is the representation of a
mathematical model of transport and dispersal of a chemical in computer
language.  A "code" is equivalent to a "model" in the sense that both
incorporate the same information and logic.
                                      145

-------
w
W
H
< w
K W
O < — t
   K  '
Of J
H W
   z
   H
w
Q
OS
H
CM
S
O IB
U
o
2;
H D
EH H
W O
H CJ
J <

«!  II
 I
IT)



A
P"
W
^
«
r-
f^.
CT»
•
H
(C
4J
(U
«
II i
14-1
O
X
0
'd
0)
to
nt
«






RNAL
METRY
^« W
58
S
S
S
8
j
rt

0?
o>
a













ty
0
i
K
«
PS
«
a
a
K

PS





0^
K
cc;
j
soaaiv
*?
r^
o^
i-!
*
4-1
0>
C
.^
V4
>

«



IK








K
«
K
AIREM
Elaine and
Bramblett, 1964

< K



< K










a a.
\
II 3J.ISIW
in
r^
w
f.
o
&







t K


< K

< K




O
1
D
CX
*
r^
\D
Oi
iH
<0
4J
0)
So^
IS VD
22
















a
ARADS
Tveten, 1974














<

<
ARCON
Armstrong and Gloyna,
1968












< K




ARMSTRONG AND
GLOYNA
•J
r-
m
iH
TJ
^J
OJ
4J
10
T!
M
0
VI
K

£
K


«
K




K




ARRRG
Culkowski and
Patterson, 1976













«
«
o;
K
s
H
Watts, 1976
e





«
K




K




CARDOCC
Clarke, 1972
•* K
<: u















CEDRIC
Charak, 1967















rt
•t,
CHARAK
Bonzon and Rivard,
1970

<



<










<
CLOUD
                                                              146

-------
0)
3
C
O
u
 I
LO

w
J
m
<;
EH
M
11
i)
1
I
il
1!
|
Ij
C
" >
1' J CC
NTERN;
)OSIME1
ii
c
1' >H
j 1
f M
ii §
3
2
Ii g
1 *
\ K
j

11
|)
ii
!
l
i ?•
! g
1 !
' i
i "
- <
I *
! "
! £
§
1 H
; K


I!
i
REFERENCES

NOIJ.S30NI
NOIIVIVHNI
sunsodxa
aansodxa
3KnaaoHS
aansodxa
sansodxa HIV
saoiovd HOiAVHas
5 AHViaia Nvwnn
saooj oiivnov
saooj
NOIiVOIMMI
NOIXVJN3WIQ3S
H3XVM dNflOHS

NoisNsdsnsaa
NOIlISOdBd AHQ
NOIlISOdaO X3M
DIH3HdSOWIV
ftl
Q
O
U
H
TJ
4J
0)
CO
^H
(!)
•H 0
ia a*
D i-












< a,




ICOLHEAT
1
o m
r- r- -
*- r- M
V
•» - c
• • C
rH M 0
* (0 U
4J JJ 'O
0) 0) C
«
(A JJ
-H ,C ^
^-( o * m
•H D( JJ O\
S CO O *-





rt










-
COMRADEX
I, II, III
1
r-l
1C
4J
0>
T
i-H
(U
-H
*P m
01 0>
« '-

<



<










"
ICURIE/DOSE
ITHUNDERHEAD
•3
r-
<^
r-« m
itj r--
»t
ii
C OJ
o Cn
•P C
W (U
O -P
B tn

< a.














•t a.
bACRIN
|
ia and Church,
9 	
3 CTi
•J <-














rt B

«< 05
blFOUT
|
r-
T3 ^D
C O-
o) *-
c *
o e
en m
-H X
-H 3
U) O

< K



< K











CO
0
R
T3
C VO
ft) 1^
Oi
v< -
OJ 01
OT 4J
X 0
rt fc.












< «




a:
0,
p
'C
C ^D
nj (^
m
V* X
4J 0)
tn JJ
£ 0
«e ft,











< «





IDPRWGW
1
T1
•r-i t«
3
CTI »C vD
3 «-i r-
Q 3 CTi
Ch *-
U3 3
Q -
w en
01 .- 0)
> in >
QJ en 01
« «- pfi











< Pi





DUGUID &
REEVES
Donald, 1971
P





rf








<

....
<:
CCRIE
	
CN
r-
CTv
^
8,
0
CJ





a










OS
§
ft]
jelmann and Davis,
8
c m
ft] .-















< o:
<< a
ENGLEMANN AND
DAVIS
r*-
0)
a
o
u

















ERGAM
m
r--
N
U














< K
•U Oi
<«
ESDORA
idrickson, 1968
®
X














<

<
EXDOSE
E
c
5s


















EXGAM
co m
• a> o
rH Ql ?S £
4J *-
01 t8 X
01 0) 4) O
C C ,Q ^ 
-------
•O
0)
3
4-1
C
0
u

INTERNAL
DOSIMETRY
'
EXTERNAL DOSIMETRY
REFERENCES
NOIIS3ONI
NOI1VIVHNI
3HnSOdX3
H3J.VM
sansodxa
3Nn3HOHS
3HnSOdX3
annoHs
3HnSOdX3 HIV
J BSHO10Vd HOIAVH39
[ s AHviaia NvwnH
ENVIRONMENTAL TRANSPORT
saooj DiivnSv
saocw
IVIHlSSHHSi
NOIIVOIHMI
NOIJ,VJ,N3Wia3S
H3XVM QNnOHO
M3IVM SDVJHns
NOISN3dSnS3H
NOixisodaa AHa
NOIlISOdSd i3«
OIH3HdSOWIV
i §
!' °
i '
•s>
Ti
r-l
fl
*J
a)
-H
.G
w
•H
g












rt PS




g
s
Gu
^D
^
rH
(0
•U
01
ij

«3
4J
QJ
OJ
s

«c



«e
rt







rt
<
•<
bADOSE/DOSET
1
Thykier-Nielson,
1974, Hedemann, 1974
.





K










«
U!
8
u
Gogolak, 1973





K










K
boGOLAK
I
T
^
rH
•C
4J
0)
^J
(0
t3
r-l
0
to
K
K



«
K

K





Pi
K
CC
X
1
0
n
J>
tj
4J
0)
i-«

fu f
K
E
K
K
K
a
a
IX
C£
K
K
«
K

K
a
K
felERMES
[
vi
 !^ *
it 3 •
V O " rH
P l«! — 01
< «
rt «















INREM,
INREM II
jschmann & Nagel,
5
32





K










Q;
M
M
<
J
o
to
H
enge & Watson,
3
w »-





K










(£
TRONIC
^r
kD
CTi
to
,J

< (C












< K
< A
< a,
FANES
irt S Wendall,
4
tn *-
















4 «
FESODIF
fl
-P
0)
(0
^
!H
> *~














< K
< K
< a
METEO- 1
r
^*
ji
rH
«
JJ
OJ
to
E













< K
< a

rt «
j
<
s
01
J
>j
H
*;
Eckert, 1964

<



<










<
r0142
ID
CT»
03
4J

-------
•a
 a;
 3
 c
•H
+j


8
 I
in
r INTERNAL T
EXTERNAL DOSIMETRY DOSIMETRY
rc
w| I
lENTAL TRANSPORT
;; o
" M
i K
;;

;;
REFERENCES
NOIiSaOKI
in
m
0
's
S

K
NOI.L¥rIVHNI
annsodxa
H3X¥M
aynsodxa
annaaoHS
aansodxa
aynsodxa ai¥
30.L3V3 HOIAVH33
s Aaviaia N¥wnn

saooj onvnSv
SQOOJ

NOI1¥OIMMI
NoiiviNawiaas

H3I¥M 3D¥JHflS
in
2
S
C
rf





<







NOISN3dSnS3M
Noiiisodaa AHQ
K
Noiiisodaa jaw K
DIM3HdSOWi¥
O
0
L)


<

a <
^
IXOIMObS
6]
Couchman Rt al.,
1964

<


<
**

3
e"
r-l
X
< K
< K





j
l
l





^

<
JNURSE-1
CO
VD
r
,_,
a)
T3
Vi
0
C
•rH





<

Houston and Held,
1975

t-"
rr?
01
tn
c:
a) <-
4-> O\
CO <-

Q I
<£ [
i




|
1


|











t


1
1


i"
CD
z
[PLUME



O
<










<
Cooper, 1967
JJ
c
-U CTi
i

















t
r-1
-P
01
0)


0
T- r-i
cn
r~l
tP -
JJ ^-1
01 nt
•C 4-1
U 0)
U-j O
C C
CD <


Coates S Horton, 1966
Richarison, 1968


Wenzel, 1974


t i I
( t '
i
i











< tf.
1
1
1

f
1
i
i

i I





«
RACER
RADOS


S
(£

<
~1
SGVH|
ar
IREKVES ST AL
a-
(
rt

1
M
CT


< p;








< (t
< K

CO
ct




< «

< 05
CM
O
CO
PS
LI
r-
2
C
Ci
c
5
CD





CO
(1)
-p

-------
0)
3
C
•H
-P



I

I
; ><
M «
g g
|S E
H H
>g§
JH Q
1
^7
i "
s
1 H
j g
8
i g
! |
1 ^

(i
I'
Ii
j)
|i
|i
^
1 «
a
f
!: ^
i! ^
i' £
S
i
s
I
|
|

REFERENCES

NOIIS3ONI
NOIIVIVHNI
annsodxa
aaivM
^ansodxa
awnsHOHS
aansodxa

M01DVJ HOIAVH3S
9 AMV13IQ NVWflH
sooo, owvnsv
saooa
Noi,v9iaal
Noix™ia3S
H3XVM QNnOMO
H3XVM SOViJMflS
NOISN3dSnS3H
NOIilSOdSO AHQ
Nomsodaa X3«
OIH3HdSOW.IV
Ed
O
O
O
. & Houghton,
)
rH 10
0) ffl
CQ t-
















<
M
OH
S
to
in
r-I
fl
tu
c
.p
to





<










-!
ISUBDOSA
T 
nj r-
K

K
K
K
K
K
K
K
K
K

K

K

«
VADOSCA
.
§
r-J
L)

rt K



< K








< K

«*
r EERIE
VD
CTi
C
rt (S
X. «
rt















i
Oi
^}
CTi
M
n,
o
8





*










-
Q
S
3
It
•s •
2
C H
0)
M
1
"1
i
i
i


t
1
i
1

^
i
n



1
i
.
>j ]
                                                 150

-------
                                                  ORNL-DWG 76-17132
 LLJ
 o
 cc
 LU
cc
o
 QC
 I-
 LU
 CO
 O
 Q
  'ATMOSPHERIC
  WET DEPOSITION
  DRY DEPOSITION
  RESUSPENSION
  SURFACE WATER
  GROUND WATER
  SEDIMENTATION
  IRRIGATION
  TERRESTRIAL FOODS
  AQUATIC FOODS
  BEHAVIORAL FACTORS
  fAIR EXPOSURE
H GROUND EXPOSURE
uj SHORELINE EXPOSURE
  '-WATER EXPOSURE
HNHALATION
     ACCIDENTAL RELEASES
     ROUTINE RELEASES
                                 TOTAL NUMBER OF CODES = 83
                                 20      40      60      80      100

                                PERCENT TOTAL COMPUTER CODES
FIGURE 5-2.  FREQUENCY OF  OCCURRENCE OF VARIOUS FEATURES AMONG EIGHTY-THREE
    COMPUTER CODES FOR RADIOLOGICAL EXPOSURE  (Hoffman et al. 1977a).
                                  151

-------
     Table 5-1 shows that some codes are applicable to a variety of problems
whereas others are very specific in purpose and scope.  It should be noted that
neither the table nor the figure provide details of the structure of models
represented by each code.  Although 83 codes have been reviewed, many of these
codes share a similar mathematical approach for calculation of environmental
transport and dosimetry of radionuclide releases.

5.5.2  Atmospheric Transport

     Nearly all codes dealing with atmospheric transport are based on the
Gaussian plume dispersion model.  Most of the older codes use the Gaussian
model formulated by Sutton.  The newer ones are based on the formulation of
Pasquill (1961).  Some codes have modified these basic models to account for
ground deposition and depletion of the plume by one or more processes and to
account for the presence of an upper bound on the atmospheric diffusion layer.
However, most of these simple Gaussian models assume straight-line,
one-dimensional air flow and do not consider effects of spatial and temporal
meteorological variations.  A few codes use more complicated models for
calculating dispersion on a regional scale (10 to 100 miles) where these
effects become important.  MESODIF, for example, employs a two-dimensional puff
advection Gaussian dispersion model, and ADPIC is based on a three-dimensional
particle-in-cell trajectory model.

5.5.3  Aquatic Transport

     The few aquatic transport models which have been reviewed are of varying
degrees of sophistication.  For example, some codes such as ARRRG, CARDOCC,
HERMES, and VADOSCA calculate radionuclide concentration in water by assuming a
simple algebraic relationship between effluent discharge concentration, a
mixing factor, and average turnover rate of receiving water at the point of
interest.  More sophisticated treatments are represented by codes in which the
models are based on the solution to a transport equation.  Solution of a
one-dimensional transport equation for radionuclides introduced into aquatic
systems is computed by the code, TRNSPRT, and a code developed by Armstrong and
Gloyna (1968).  Two-dimensional transport equations are solved by the
finite-element Galerkin model used in GETRA and the finite-difference model
used in SERATRA.  These codes have been specifically developed for estimating
aquatic transport of radionuclides and sediments.  ADPIC, DPRWGW, and DPWRCR
are in the three-dimensional category.  The "particle-in-cell" approach
utilized in ADPIC was initially developed for atmospheric transport problems,
but the code can, with some limitations, also be adapted to surface water
transport.  The DPRW codes incorporate generalized transport models using the
"discrete-parcel-random-walk" approach.

5.5.4  Food-Chain Transfer

     The few codes developed for estimation of transfer and accumulation in
terrestrial and aquatic food chains employ a systems approach in which
empirically derived transfer coefficients are applied to calculate radionuclide
concentrations along various pathways.  Transfer from water to aquatic foods is
generally calculated through use of a single empirical transfer coefficient
called a "bioaccumulation  factor," whereas calculation of transfer from air to
terrestrial food employs a multiple series of transfer coefficients to account

                                      152

-------
for such phenomena as deposition, vegetation retention, animal grazing habits,
etc.  Most of these codes have been specifically intended for assessment of
routine releases in that the models assume steady-state or equilibrium
conditions.  The models incorporated in AQUAMOD and TERMOD,  however,  are
time-dependent and can be used potentially for both accidental and routine
releases.  Unlike other codes, AQUAMOD includes more detail  in the description
of radionuclide transfers in the aquatic ecosystem, but much of this  detail
cannot be used because of insufficient empirical data.  It is also of interest
to note that of the nine codes which can be used to calculate food chain
transfer, four of the codes (FOOD, GRONK, ARRRG, and CARDOCC) have been
developed from the initial models incorporated in HERMES.

5.5.5  External Dosimetry

     About one-half of the codes included in Table 5-1 permit calculation of
doses resulting from external modes of exposure.  The primary mode of external
exposure considered by these codes is exposure to contaminated air.  Only about
10 percent of the codes listed in Table 5-1 are concerned with the external
doses received from ground, shoreline, or water contamination.  The calculation
of air exposure is performed assuming a spatial distribution of radionuclides
determined by an infinite or semi-infinite uniform cloud model or a finite
Gaussian plume model (Slade 1968).  The semi-infinite or infinite cloud model
is assumed without exception for the calculations of exposure to beta
radiation, but nearly half of the codes employ the finite cloud model for
calculation of the exposure to gamma radiation.  The finite  cloud model is
used because the semi-infinite and infinite cloud models tend to underestimate
the dose received from gamma radiation at distances close to the point of
release from elevated sources.  Regardless of the type of radiation considered,
the codes which calculate external dose from exposure to contaminated water
assume an infinite or semi-infinite medium having a uniform  distribution of
radionuclides.  Codes which calculate ground and/or shoreline exposure usually
represent ground and shoreline as infinite planes of negligible thickness
having a uniform distribution of radionuclides (Hine and Brownell 1956).  Only
VADOSCA assumes a finite thickness for shoreline contamination.

     Although whole-body dose is calculated by all of the codes which consider
external dosimetry, only a few codes such as EXREM III, GRONK and SUBDOSA
consider the effect of dose attenuation with depth of tissue penetration by
radiation.  GRONK, however, is an example of a code which employs
dose-conversion factors to calculate external dose after concentrations of
radionuclides in an environmental medium have been determined.  These factors,
which consider the effective absorbed energy, the assumed spatial distribution,
and the tissue penetration of radiation emitted from radionuclides, are
calculated prior to input into the code.

5.5.6  Internal Dosimetry

     Most of the models used for calculation of internal doses are those
recommended by the International Commission on Radiological  Protection (ICRP
1959).  Some codes provide only a listing of values for a single parameter
                                      153

-------
which converts intake rate of a radionuclide into dose.  These dose conversion
factors must be calculated prior to input.  Other codes include the detailed
parameters of the ICRP models, such as mass of the organ and fractional uptake,
retention and effective absorbed energy of the radionuclides.   INREM is the
only code for which calculation of age-dependent internal doses is stated as a
specific objective.  The internal dosimetry models of ICRP Publications 10
(ICRP 1968) and 10 A (ICRP 1971) have been incorporated into INDOS I,  II, III,
and CEDRIC.  The ICRP Task Group Lung Model (ICRP 1966) represents the highest
degree of sophistication for calculating doses resulting from inhalation of
radionuclides.  The only codes which have attempted to implement this model to
date are AERIN and DACRIN.  Neither of these codes, however, considers the
additional contribution to total dose resulting from the complete production of
daughter nuclides in nuclide decay schemes, except insofar as this contribution
has been factored into effective absorbed energy parameters, which are supplied
as input.
5.6  AVAILABLE MODELS FOR EXPOSURE TO CHEMICAL POLLUTANTS

     The development of models of exposure for chemical pollutants has been
more recent and less concerted than that for radionuclides.  The literature on
the assessment of models for chemicals in the environment is sketchy and
fragmentary.  A number of bibliographies is available (Cophenhaver 1974a
through 1976b; Corrill et al. 1974; Cophenhaver and Wilkinson 1974a, 1974b;
Lehmann 1977, 1978a, 1978b; U. S. Environmental Protection Agency 1977;
Cavagnaro 1978a, 1978b; Harnden 1977; Ross 1967).

     Most of the models are still in the developmental state and have not been
systematically evaluated against concentration measurements in laboratory or
field experiments.  The available models are generally oriented toward either a
particular substance or a particular transport medium.  The chemical reactions
considered are most frequently lumped into an effective first order decay
process that may be valid over a limited range of environmental parameters.

     The most frequent formulations focus on the dispersion of a pollutant,
usually in a single medium, and in a well defined portion of the environment.
Few attempts have been made to formulate the general transport/dispersion
processes for the total environment.  A general model applicable for relatively
stable materials involved over large regions of the environment is sketched in
Figure 5-3.  Even this relatively complex model omits some potentially
important reservoirs such as lakes, lake sediments, and dumps.

     Comprehensive models generally emphasize a single compartment such as the
atmosphere, the hydrosphere or the lithosphere; however, they treat combina-
tions of these in a very approximate manner.  A few global models have been
elaborated to describe a multiplicity of compartments for chemical propagation
on a large space scale over long periods of time.  Separate models for the
trophic transfers of the chemical substance throughout food chains may be
coupled to the comprehensive environmental fate models as illustrated in a
later subsection for aquatic ecosystems.

     Several "hybrid" comprehensive fate models have combined interactions in
                                       154

-------
                                        TXSB IN THE
                                        ATMOSPHERE
                    RAIN OUT
                  AND FALLOUT
                         DUST RESUSPENSION
                          OR EVAPORATION
TXSB
IN SOIL
                                    CODISTILLATION
         TXSB IN
      GROUNDWATER
            CODISTILLATION
             AND AEROSOL
             RESUSPENSION
                                                     RAINOUT
                                                   AND FALLOUT
       NOTE:
         DEGRADATION OF TXSB CAN OCCUR
         IN ANY COMPARTMENT, BUT THE
         ARROWS ARE OMITTED FOR CLARITY
                                                    TXSB IN OCEAN
                                                     MIXED LAYER
SEDIMENTATION
                                          TXSB IN
                                         ABYSS AND
                                         SEDIMENTS
                                                       UPWELLING
                                                      RESUSPENSION
                                                     AND DISSOLUTION
FIGURE 5-3.  PATHWAYS FOR TOXIC SUBSTANCES  (TXSB)  IN THE PHYSICOCHEMICAL
           ENVIRONMENT SYSTEM (from Eschenroeder  et al.  1978).
                                   155

-------
the physical/chemical environment with those in the biotic compartments.   Among
the earliest of these is the one by Randers and Meadows (1971)  which used the
Forrester system dynamics simulation format to study the movement of DDT
throughout the global environment.  This model employed linear  representations
of all processes by requiring half-life specifications for the  dynamics
coefficients.  Woodwell, Craig and Johnson (1971)  reported a compartment
systems simulation of transport of DDT in the biosphere in order to trace its
progress through food chains on a global scale.  Cramer (1972)  modeled the
global circulation of DDT using a combination of physical compartments and
biological compartments to determine the exposure and fate of the material.   A
comprehensive unification of physical and biotic models was presented in
generic form by Gillett et al. (1974) in an EPA report.  Three  main units of
the environment are considered:  atmospheric/terrestrial, fresh water-aquatic
and estuarine marine.  Each of these main subdivisions was further
compartmented into structural elements.  Haque and Freed (1974) survey the
data requirements of models relative to laboratory measurements.  The input
parameters, termed chemodynamic properties, were identified and discussed.

     Transport models in combination with simple mass balancing or mass
accounting have provided an interim approach to assessing the fate of chemicals
in the environment.  Table 5-2 summarizes some of the studies which have
applied such an approach for rationalizing the movement of widely dispersed
pollutants.
5.7  MICROCOSM MODELS

     The philosophy of designing physical models which take the form of
microcosms is to reproduce on a small scale various combined processes of
biological systems and their environment (Gillett and Witt 1979 ).  Microcosms,
which may be terrestrial, aquatic, or both are intended to serve as surrogates
for the real world by incorporating the main aspects of pathways of toxic
chemicals through biological systems,  conflicting goals arise in the
development of miniature scale models because regulatory needs call for a
single standard microcosm which can be subjected to tests with a large number
of different chemicals, whereas laboratory research priorities require the
evaluation of many different design concepts and model systems in order to
develop an acceptable set of testing systems.

5.7.1  The Microcosm as a Screening Tool

     The main objective of microcosm technology is to produce a controlled,
reproducible laboratory system that simulates the processes and interactions
between components of natural ecosystems.  In the laboratory model the
environmental variables such as temperature, humidity, light, and water balance
are under the control of the investigator as well as the time phasing of
introduction of various biotic or nonbiotic components.  The system has
boundaries containing the components and in general is not self sustaining for
long periods of time.  The terrestrial systems considered in Gillet and Witt
(1979) are described in summary form in Table 5-3.  The references cited there
give detailed descriptions of the microcosm experiments; therefore, details are
not repeated here.
                                      156

-------
TABLE 5-2  MULTIMEDIA MODELS:  REFERENCES AND CHARACTERISTICS
DDT and
metabolites
Cadmium
Mercury
Nitrates
Nitrogen,
Phosphorus,
and Sulfur
Harrison, H.L., O.L. Loucks, J.W. Mitchell,
D.F. Parkhurst, C.R. Tracy, D.G. Watts and
V.J. Yannacone, Jr.  Systems Studies
of DDT Transport.  Science, 170:503,
1970.  (Mass accounting, mass balance,
regional, dynamic).

Woodwell, G.M., P.P. Craig and H.H. Johnson.
DDT in the Biosphere:  Where Does It
Go?  Science, 174:1101,    1.  (Mass ac-
counting, mass balance, global, dynamic).

Randers, J.  DDT Movement in the Global Environment.
In;  Toward Global Equilibrium:  Collected Papers.
D.L. Meadows and D.H. Meadows.  Wright-Allen Press,
Inc., Cambridge, Massachusetts, 1972. (Mass
accounting, mass balance, global, dynamic).

Rupp, E.M., D.C. Parzyck, P.J. Walsh, R.S.
Brook, R.J. Raridon and B.L. Whitfield.
Composite Hazard Index for Assessing
Limiting Exposures to Environmental Pollu-
tants:  Application Through a Case Study.
Environmental Science -and Technology, 12:
802, 1978. (Mass accounting, single source,
site specific, static).

National Academy of Sciences.  An Assessment
of Mercury in the Environment.  Washington,
D.C., 1978. (Mass accounting and mass balance,
static, global).

National Academy of Sciences.  Nitrates:
An Environmental Assessment.  Washington, D.C.,
1978.  (Mass accounting and mass balance,
static, not predicting changes, multiscale,
local, regional, national and global).

Scientific Committee on Problems of the
Environment. Biogeochemical Cycling
of Nitrogen Phosphorus and Sulfur.  SCOPE
7.  Swedish National Science Research Cen-
tre, Stockholm, Sweden, 1976. (Global, mass
accounting and mass balance, not dynamic).
                                      157

-------
                            TABLE 5-2  (Continued)
Freon and
Carbon
Tetrachloride
Chloroform
and Carbon
Tetrachloride
PCB's
PCB's
Neely, W.B.  Material Balance
Analysis of Trichlorofluoromethane and
Carbon Tetrachloride in the Atmosphere.
Science of the Total Environment, 8:
267, 1977. (Three compartment, global,
dynamic, mass accounting).

National Academy of Sciences.  Non-
fluorinated Halomethanes in the Environ-
ment.  Washington, D.C., 1978. (Multimedia source
inventory and an exposure inventory,  no
mathematical model, global mass balance).

Nisbet, I.C.T. and A.F. Sarofim.
Rates and Routes of Transport of PCB's in
the Environment.  Environmental Health
Prespectives, 1:21, 1972.   (Mass accounting,
continental mass balance,  incomplete data
on transfer coefficients).

National Academy of Sciences.
Polychlorinated Biphenyls.  Washington, D.C., 1979.
(Continental mass accounting, mass balance,
not dynamic, multiscale, incomplete data on
transfer coefficients).
                                      158

-------
D   •



    r-


EH  i-

fa  ~"
    EH


D  H



W  C
    2
O  <
EH
    EH
Q  E-.
K  fa1
tO  J

    H
W  O


W  Z
O  H
8
 _  o
a:  K
o  N
H  H
S1  K
EH
W  W
u  <
w
O
cc
H  H
DC  E
W  EH

U  Z
<

U
    w
    U
fa  M
W  S

££
O  U

    fa
 •  o
M
 I   W
in  £H
    o
W  [xj
ij  fa
«  fa
m
O  Q)
O  JJ
                                         — —   -H >,
                                                               m    »-    *-
                                                                                                       o  w
                                                                                                       u -H
                                                                                                       U  <$
                                                                         159

-------
•C

3
C
-H

C
0
U
M
 I
m

u
tM

k.


t-l -
o) eo
jj i) *~i
4-> 4-> rH

i-H >l -P
\ 10 0>
rH O E
•H O 0
0 0) rH
Ul -H
ft
to
0)



°\
00
CD
Vt
 W rH
W 0 4)
0) O ,Q
HOE
M VH -i
r4 W
0 O
•H 0
r o)
 *
4-> rHCM
to 3 E
DUO)
rH O -P
^ c to
0) 0 >,
FH E CO












rH
-H
0
to
•H -P
4J C
to id
•rH H
^ At
0)
.P
O
«
VH
ft)
U
•P C
C O
0) -H 0
tn g 4J m
C -P (fl
•rH ft) rH 4->
4J CO TI) 0) J3 3
C >i 3 H -H 0
(D (d r-t 4-J rH ^Q ~
•H tJ O 1 -H (0 CO
J2 C 0) 3 >i
E O -H rH D1 '•-I Id
< CC ^ ft 0) O T3







1
JO 0
CO C rH 0) W
4-> 0) *O 3 -H 0)
C 0) 3 Cf 4-) 0)
(U 1* rH 0) 5
•H U I C
,Q (Ti C 0) 0 r-
S 1 -H rH -H |
< \D ^— ft jj m

CO
4-1 Tl
C
Q) O
•rH (T.
rQ 1
e o
< kO

CO
4-1 T3
C
Q) 0


E o

•H
to id co


4-1 10 4-) CO O
C >i W >i -H
CD  W

E ft «
< p T3


JJ
£ 0)
-H e
4-* - 4-»
310 tJ4~l 'O 01 rH^H
OC4->C!034->4JCC 4-> fl3 fl

d4-J rH 0) Cn3 4->4J
0>,C 0) OtrHrCrH QJ Cng'O^^

H -H 0 0 0) fd -H -H rH 3 0 -H -H
1" ^. ^. r-,
                                        160

-------
     Microcosms permit investigation of interactions that combine physical
properties of substances with biological processes while avoiding the costs and
risks of full scale field or population system tests.  They give indices of
time and space distributions of chemical substances thereby providing broad
indications regarding the fate and effects of chemicals in the real world.
This bypasses the necessity, in some cases, for committing large scale
resources to field studies for many chemicals that may be considered.
Microcosm experiments, in ideal circumstances, provide a bridge between the
laboratory tests of physical and chemical quantities and the expectations for
behavior in the external environment.  Furthermore, they provide a complex of
system interactions in the laboratory that cannot be captured with laboratory
bench experiments and may not be under sufficient control for full scale field
studies to identify cause-and-effeet relationships.  Microcosm technology lends
insights into interactions and potential effects of classes of chemicals within
a controlled laboratory setting.  Therefore, it has the potential of providing
useful information for guiding the practical aspects of manufacturing, handling
and use of chemicals in applications.  This provides an advantage over
elementary laboratory tests that may provide only limited information for
anticipating adverse effects in the field.

5.7.2  Relating Environmental Behavior To Microcosm Data

     Since microcosms are not self-sustaining they are not capable of
demonstrating some significant ecological processes, such as succession or
multigeneration phenomena.  It still remains to make inferences from
laboratory studies on the components or the processes in order to infer what
happens in the open environment with respect to these effects.  The number of
physical and chemical properties are more directly measurable through
laboratory bench tests than through complex systems as represented by those in
laboratory microcosms.  Physicochemical models are available to extend these
fundamental measurements to concentration distribution in the environment as
described in the previous section.  Physical limitations or logistical
limitations may result in an inability to raise or maintain certain
environmental interactions of chemical contaminants.  These limitations and the
questions arising about direct scale-up impose serious requirements on design
of the microcosm experiment in view of the assumptions underlying the systems
objectives.

     In the previous section it was illustrated by example that large
differences prevail among the masses of abiotic material and biomass that can
be a site for a chemical contaminant.  For a given throughput of the chemical
substance, therefore, the concentrations can vary widely according to physical
and biological processes.  The controlling features of the distribution of
material suggest scaling rules for the design of laboratory microcosm
experiments in order to maintain the correct order of magnitude for various
controlling parameters.  For example, the ratio of biomass to the mass of water
in the system must be carefully selected in order to represent the amount of
dilution that occurs and, at the same time, model correctly the exposure of the
various organisms to the chemical.  Consequently, if an experimental scheme
results in one particular organism dominating the available supply of space or
                                      161

-------
nutrients, other effects in the system may be eclipsed by this imbalance.   As
in any scale model experiments, parameters must be derived and preserved in
order to replicate these same physical biological phenomena.

     The presence of walls also poses problems with regard to removal of
experimental artifacts.  Some chemicals may have a greater affinity to the wall
surface than to any other element of the system.  This creates a situation
which may not be representative of phenomena in the open environment.
Systematic experiments to test this limitation may be conducted by preserving
the ratios of the masses described above and varying the scale of the
experiment so that the ratio of certain exposures to the wall surface varies
over a wide range of values.  In this way the extrapolation to zero wall effect
may be possible so that this artifact can be effectively controlled or
corrected.  If microcosm data are to be used directly to infer behavior of a
chemical substance in the environment, generalized similitude rules must be
derived in the same manner as that employed for many other scale model
experiments.  In addition to this derivation the general steps of experimental
design must be taken in order to determine how many runs are needed in order to
insure accuracy within some predetermined statistical level of certainty.
5.8  A COMBINED MATHEMATICAL/MICROCOSM APPROACH

     At the workshop on terrestial microcosms (Gillett and Witt 1979),
mathematical and physical modeling were perceived to have an integral
relationship directed toward two objectives:  ~\) a pathway to microcosm
development, and 2) a means of providing methodology for testing chemicals in
the environment.  The coordinated development of mathematical models and
microcosms was considered to be an evolutionary process, first concentrating on
scientific issues and finally providing a logical framework that would enable
the estimation of fate and effects of chemicals on health and the environment.
This operational philosophy anticipates the scaling problems of microcosms
referred to above and suggests that mathematical modeling can serve to bridge
the gaps left by deficiencies in direct scaling.
5.9  EXPOSURE ASSESSMENT

5.9.1  Definition of Exposure

     For this chapter exposure is defined as the quantity of a substance which
reaches the external surface of an organism per unit time, or the concentration
of the substance in one or more ambient media.  The exposure of an organism may
fluctuate in time, and exposure of different organisms in a population may
differ.  A good exposure model should be able to predict (or explain) not only
the average exposure of a group of organisms, but also the statistical
distribution of exposure within the group and the temporal fluctuations.

     Note that exposure as defined above is not synonymous with intake or with
dose.  If an organism is exposed by only one route, exposure is related to
intake through an intake factor or absorption factor; intake is related to
                                       162

-------
dose through pharmacokinetic and metabolic factors.  If an organism is exposed
by more than one route, measurement of total intake or total dose may be
difficult or complex.  For further discussion see the chapter on Biota.

5.9.2  Types of Exposure Models

     The function of an exposure model is to relate the exposure of humans or
other target organisms to the distribution of a toxic chemical in the
environment and ultimately to its sources.  Some models are purely predictive;
given the distribution and strength of sources and the physical and chemical
properties of the  substance, the task is to predict the exposure of a target
population.  Other models are purely explanatory:  given measurements of
exposure, the task is to trace them back to their sources and to determine the
critical pathways which provide the most effective means of control.  Most
models have features of both types:  given some information on both sources
and exposure, the task is to relate them in a plausible and self-consistent
fashion.  If it proves difficult to do so, this often suggests that either the
inventory of known source or the measurements of exposure are incomplete.

     The primary purpose of exposure assessment is to determine whether or not
the exposures constitute a hazard to the target organisms.  For this reason,
there is no fixed boundary between exposure assessment and health or ecological
risk assessment.  The nature of the exposure assessment will depend upon the
goals of the particular study.  For example, if the goal of the study is to
ascertain compliance with standards, it may suffice to estimate the maximum
level to which an individual might be exposed.  The assumption is that if the
hypothetical maximally exposed individual is protected, all others will be
protected.

     In epidemiological studies, an attempt is made to correlate measured
exposure levels with observed biological effects, either within or between
populations.  For this purpose it is essential to estimate exposures of
populations and to establish gradients in exposure.

     The other major type of exposure assessment study is that designed to
investigate whether a chemical may occur at a level suspected to be hazardous.
In such studies, the measured or predicted exposure levels are compared with
those known to have biological effects, usually on the basis of prior experi-
mentation.  Such comparison is usually relatively simple if the environmental
and experimental exposures are by the same routes.  For example, most aquatic
toxicology studies involve exposure of fish or invertebrates to toxicants
dissolved in the ambient water at measured concentrations.  When an exposure
model leads to an estimate of the concentration of the chemical in ambient
water, this estimate can then be compared directly with those known to have
adverse effects.  On the other hand, if the toxicological studies have been
conducted by exposing fish to the toxicant in their food, it will be difficult
to interpret the potential hazard posed by an ambient concentration unless the
relationship between intake via the fills and via the food can be established.
                                      163

-------
     There are sometimes other factors which complicate the assessment of
potential hazards posed by exposure.  For example,  a toxic agent may exist in
the environment in a different physical or chemical form from that in which its
toxicity was measured experimentally.  For example, the toxicity of cadmium in
natural organic complexes may differ markedly from that of soluble cadium
salts.  Also, it may be difficult to assess the potential consequences of
fluctuating exposures in the environment if toxicity tests have been conducted
only at constant rates of exposure (as in cases where the consequences of
intermittant exposure of fish to a chemical in an industrial effluent are to be
predicted from measurement of toxicity under continuous flow conditions).

5.10.0  MULTIMEDIA EXPOSURES

     The most difficult problems of assessment arise when target organisms are
exposed simultaneously by several routes.  For example, humans are exposed to
carbon tetrachloride and chloroform not only in drinking water, but also in air
and food (National Academy of Sciences 1978). Table 5-4 shows estimates of
total intake of carbon tetrachloride and chloroform by these three routes:
these estimates were derived by combining measurements of exposure with rates
of intake (rates of ingestion of water and food, rate of inhalation of air).
Estimates of intake by each of the three routes were very variable because of
wide geographical variations in exposure, variations in rates of intake, and
uncertainty about the efficiency of absorption through the lungs.  This
illustrates the critical importance of making multimedia assessments of
exposure.  In this case it is probably reasonable to sum the intakes by all
routes to obtain a measure of potential hazard because after absorption these
chemicals are distributed through the body and the target organs are distant
from the points of entry.  However, in cases where toxicological information is
available for only one route of exposure and environmental exposure leads to
exposure of a different organ (e.g. asbestos ingested in drinking water) it may
be difficult or impossible to make a reliable assessment of hazard.

     In cases of this kind it may be important to calculate the "dose" to which
each tissue is exposed, and this in general requires a pharmacokinetic model of
the relationship between the exposure of the organism and the dose to the
tissue.  The intake of a chemical by an organism depends on the level of
exposure, on the physiological properties of the organs via which it is
exposed, and on the chemical and physical properties of the chemical.  The
tissue-specific dose depends on all the above properties and on in vivo meta-
bolism, transport, storage, and elimination.  The practical importance of these
considerations is that dose limits may sometimes be specified in terms of
maximum permissible tissue concentrations.  In such cases the exposure
assessment must provide comparable data or expected tissue levels.

     The necessity for a multimedia approach, even if releases are primarily
into one medium, has been demonstrated repeatedly for radioactive materials and
is illustrated by application to cadium release from a smelter complex
(Rupp et al. 1978).  Information on biological effects was related to kidney
concentrations of cadmium.  The magnitude of an air concentration that would
prevent accumulation of a toxic concentration of cadmium in kidneys depended
strongly on the fraction of food that was grown and consumed locally.  The
range of varation was about two orders of magnitude.
                                       164

-------
                       TABLE 5-4 (Adapted from NAS 1978)

         ESTIMATED INTAKE OF CHLOROFORM AND CARBON TETRACHLORIDE FROM
                        ENVIRONMENTAL SOURCES (mg/year)
      Source
                                CHCl.
                                          CC1,
Fluid Intake
Atmosphere
Food Supply
     Exposure Levels

Min.1   Typical2   Max.3

0.04      14.9      494
0.41       5.2      474
0.21       2.2       16
                                                          Exposure Levels
                      Min.    Typical2
                                                                         Max.'
                      0.73      1.78      4.0
                      3.60      4.80      618
                      0.21      1.12      7.3
  Total
0.66
22.3
984
4.54
7.70
629
 'Minimum exposure, minimum intake for fluids and foods; minimum exposure,
 49 percent absorption for atmosphere.

2Typical value for exposure and absorption.

^Maximum exposure, maximum intake for fluids and food; maximum exposure,
 77 percent absorption for atmosphere.
     In other cases, measurements and tissue concentrations in exposed
organisms may be the only data available for validation of a model.  Tn these
cases it is necessary to "calibrate" the system by measuring the relationship
between exposure and tissue burden under representative conditions.  One
example is that of PCB's, for which direct measurements of exposure to aquatic
organisms are scanty because of the difficulty of measuring concentrations in
water at parts per trillion levels.  In this case, measurements of tissue
concentrations in fish have been useful in defining geographical patterns of
concentration and in relating them to sources (Nisbet and Sarofim 1972; U. S.
Environmental Protection Agency 1976).  In another example, measurements of
concentrations of chlordane components and metabolites in human tissues were
used to estimate total exposure by all routes leading to the identification of
previously unrecognized sources of exposure (Nisbet 1977).
                                      165

-------
5.11.0  APPLICATION OF MODELS FROM ENVIRONMENTAL EXPOSURE

     In any scheme for hazard assessment of a chemical,  biological testing of
the target or receptor organism needs to proceed in parallel with the
estimation of the expected environmental exposure that will result from the
manufacture and use of a new or existing chemical.  This section will be
limited to the discussion of procedures for exposure assessment but the need
for the parallel effort on effects testing should not be overlooked.

     No single model can be selected as the best possible for all purposes.
The choice of an appropriate model for exposure assessment will depend on the
nature of the chemical, the environmental system, receptor of interest,
validation potential, cost, and the objectives for which a model will be used.
Traditionally models have been selected to predict exposure to an existing
pollutant for which monitoring data were available and the adverse effects of
which had been identified.  The compartment of the environment on which to
focus, the spatial and temporal scales of interest, and the critical pathways
have been generally known at the start of the modeling effort.  The case
studies on existing pollutants proved a useful guide for the development of
models for classes of compounds with similar physicochemical or chemodynamic
properties.  The challenge is to provide a systematic methodology for
predicting exposure levels that will complement and in time possibly replace
the combination of scientific judgement and monitoring information presently
used to evaluate environmental hazards.  The approach needs to be
interdisciplinary, including models for the different compartments of the
environment and for transfer across their interfaces.

     It is recognized that great uncertainty is associated with the predictions
of environmental transport models.  Field validation of environmental models
for the major classes of chemicals is needed before models are used in making
decisions on the safe use of chemicals in the environment.  At present, the
uncertainty in the transport parameters is such that there is no assurance that
increases in model complexity will yield improvement in estimates of exposure.
Nevertheless, to the extent that the more complex models simulate the processes
occurring in nature more closely, it is expected that these will eventually
prove to be the more predictive models.  The discussion below of a hierarchical
structure for model application is based on the assumption that ongoing and
proposed research on model refinement and validation will reduce the major
sources of uncertainty in model predictions.

     In order to eliminate the unnecessary generation of physicochemical and
biological test data on all chemicals, a hierarchy of tests is proposed as
shown schematically in Figure 5-4.

     At the first level of testing, a simple partioning profile based on the
physicochemical properties (i.e., vapor pressure, water solubility, and
molecular weight) should be made to estimate whether a chemical will be
preferentially located in the air, water, or sediment compartments.  The
analysis would involve an equilibrium calculation of distribution between the
three media, neglecting all degradation or transformation reactions, in order
to identify the medium of probable interest.
                                       166

-------
   C/)LU

   LU CC.

   I-Z>
LU
                     1
               OLUC3
LU

>

LU
     LU

     >
     LU
                                                                                                                          W   •

                                                                                                                          1-7
                                                                                                                          O  K
                                                                                                                          K  U
                                                                                                                          H


                                                                                                                          2  E
                                                                                                                          H  O
                                                                                                                             H
                                                                                                                          Q  E-'
                                                                                                                          K  t£
                                                                                                                          EH  2
                                                                                                                          <  E-
                                                                                                                          S  Z
                                                                                                                          H  W
                                                                                                                          EH  U
                                                                                                                          W  E
                                                                                                                          Id  O
                                                                                                       O  W
                                                                                                           EH
                                                                                                       2  U
                                                                                                       O  ta
                                                                                                       W  [L
                                                                                                       H  fc
                                                                                                                         A,
                                                                                                                         s  <

                                                                                                                         R  ^
                                                                                                                         o  <
                                                                                                                             a

                                                                                                                          •  EH
                                                                                                                         EH  CO
                                                                                                                         2  K
                                                                                                                         W  EH
                                                                                                                         ^  2
                                                                                                                         W  CQ
                                                                                                                         D
                                                                                                                         O  B
                                                                                                                         S  EH
                                                                                                                         u  w
                                                                                                                         K  -Z
                                                                                                                         H  O
                                                                                                                         H  H
                                                                                                                         ac  EH
                                                                                                                            EH
                                                                                                                            Z
                                                                                                                        Cd  O
                                                                                                                        g  O

                                                                                                                        o
                                                                                                                        H
                                                    167

-------
     The second level of testing addresses more specific questions of
concentration in the compartments of main concern.  For this level preliminary
estimations of a materials balance are required.  This will include estimation
of how the material will be used and how it will be manufactured in addition to
a preliminary estimation of proposed growth patterns.

     Once the use pattern is known it should be matched with the environment of
interest.  At this stage three standard environmental models are proposed.

1.  Air.  A standard volume of air is calculated based on the geographical  site
where the material will be produced.  Alternatively, a local diffusion model
can be employed for an appropriate time scale of interest.  This parcel of  air
may be described using a simple dispersion equation and assuming a mixing depth
of 1 km.  Into this parcel of air the emission rate for a 30 day period is
inserted.  Alternatively, the air in a box can be treated on a steady flow
basis assuming a mean wind speed of, say, 1 meter per second.  The mass balance
on this flow is used to obtain the estimated environmental concentration (EEC) .
This calculated ambient concentration is then compared with levels at which
toxicological effects can be anticipated.  If the EEC is below the no effect
level then one proceeds to the next level.  If the EEC is at or above the no
effect level then transformation studies in Level 3 must be undertaken.

2.  Water.  If the main emissions are to water then the use of a simple
dispersion model for four main water categories is indicated.  These types  are
rivers, lakes, estuaries, and coastal zones.  Using the proper transport
parameter of the selected water body the emission rate is added and an ambient
concentration is estimated.  This again leads to the decision box where the EEC
is compared with the no effect level at Level 3.

3.  Soil.  If the rate of clearance is greater than the rate of emissions then
obviously no environmental problem will result.  Only when the rate of
clearance is slow and the rate of addition is high is there a need for concern.

     If the EEC exceeds the no-effect concentration then more sophisticated
models involving various transformation schemes must be utilized (Level 3).
While it is a relatively simple matter to set up the materials balance
equations depicting transformation, it is much more difficult to identify the
parameters for the ecosystem of concern.

     The above hierarchal structure is proposed to illustrate the need for
developing a screening process to permit concentration of resources for the
measurement of chemodynamic properties and for biological testing on the
chemicals that pose the greatest problems.  In the time-frame of the Workshop
only the skeletal framework for such a hierarchy could be developed.
5.12.0  RESEARCH NEEDS

     Many of the research needs for multimedia exposure models have been
described in other chapters of this report dealing with development and
                                       168

-------
validation of compartment models.  Additional research requirements specific to
multimedia models involve the definition of transport coefficients between
media and the development of optimized strategies for use of multimedia models.

     Transport terms which may be casually treated or overlooked in single
compartment models include (1) the resuspension of particulates from locations
of initial deposition, (2) the exchange rate between suspended and settled
sediment in water, (3) adsorption on atmospheric particulates, (4) rain-out of
particulate and gas phase pollutants, a.nd (5) the clearance of chemicals from
soil.

     Multimedia exposure levels involve expenditures for computational efforts
and input parameters considerably in excess of single compartment models.
Clearly, there is need for developing methodologies for determining when single
compartment models suffice or when single compartment models can be linked
through simple input-output terms.  If simple hierarchal strategies such as
those described in the previous section are to be developed there is a need to
define scales of space and time in a systematic manner.  In view of the limited
experience with the use of multimedia models, even for radionuclides, much
field work remains to be done to determine specific conditions under which
multimedia models are suitable and to determine potential errors with their
use.  For chemical species undergoing little transformation or a simple first
order degradation law this effort would benefit from collaboration with the
National Laboratories.  Problems of chemical transformation not encountered
with radionuclides need further attention.  Of particular concern are those
where laboratory measurements do not simulate field experience, e.g.,
photolysis of compounds which can be sensitized in the field by other chemicals-
and microbial degradation.
5.13.0  SUMMARY AND RECOMMENDATIONS

5.13.1  Utility of Models

     A.  The most extensive use of transport and fate models to date has been
transport, distribution, and fate of radionuclides in the environment.   These
models have been used to predict exposures and doses to target populations.
Most of the available models for radionuclides have been concerned with
atmospheric transport and with external dosimetry and internal dosimetry via
inhalation.  Models of aquatic and terrestrial transport have been less fully
explored.  Comparatively few multimedia models have been described.

     B.  Development of multimedia models for toxic chemicals in the
environment has generally lagged behind that of models for radionuclides.  Most
models have been concerned with criteria air pollutants, with waterborne
pollutants in rivers, or with pesticides.  Recently, some multimedia models
have been developed for chemicals such as mercury, PCB's, and chloroform.
                                      169

-------
     C.  Problems involved in applying available models to chemicals in the
environment include the following:

     • The large number of potential pollutants, many of
       which are produced and released in small quantities
       at a few locations;

     • The lack of important information on sources,
       environmental behavior, and degradation;

     • The lack of understanding of factors controlling
       intermedia transfer;

     » Environmental transformations of one chemical
       into another.

5.13.2  Present Capabilities

     A.  Estimates of exposure can be made to within
         an order of magnitude for chemicals which
         are already in production and on which
         information on distribution in the environment
         and on transport properties is available.

     B.  Models can be expected to yield only the
         crudest of estimates of exposure for irew
         chemicals.  These estimates may still be
         useful, however, for identifying the major
         routes of exposure and for preliminary
         identification of potentially hazardous chemicals.

     C.  Modeling, monitoring, and scientific judgment
         based on experience are all necessary in exposure
         assessment.
5.13.3  Recommendations For Use Of Models

     A.  The purpose for which a model is to be used should be specified.
The complexity and scale of a model should be matched to the importance of the
problem and to the availability of data on model input parameters.

     B.  Models should be used for the following purposes:

         (i) Establishment of priorities for further study
             and for regulatory action;

        (ii) Preliminary assessment of risk;

       (iii) Identification of critical pathways and of the
             relative merits of various control options;
                                      170

-------
        (iv) Identification of important gaps in data,  and

         (v) Design of monitoring programs.

     C.  A hierarchical system of exposure models should be used to improve
efficiency and to reduce costs:

     •  Simple models should be explored first, and the results from the
        preliminary models used to guide the selection of models of
        increasing detail and complexity;

     •  The gathering of input data should match the needs of the model
        at the appropriate point in the hierarchy;

     •  Each stage in the development of a model should guide
        the collection of new data as input to the next stage.

     D.  The sequential development of exposure models should be paralleled by
sequential development of testing for effects in order to complete the
assessment and to maintain balance in the approach.

5.13.4  Research Needs

     A.  More information is needed on the location, time distribution, and
magnitude of releases of chemicals into the environment.

     B.  More research is needed into intercompartmental transfer mechanisms,
which are less well-understood than transfer processes within compartments.

     C.  Proposed models should be tested in several critical ways:

         (i) Models developed for one chemical or class of chemicals
             should be tested with another;

        (ii) Models developed on a small scale (e.g., microcosms),
             should be tested with data on the distribution of chemicals
             over a larger scale;

       (iii) Models should be tested for their sensitivity to variation
             in parameters such as degradation rates or intermedia
             transfer coefficients.

     (D)  Measurement programs both in the laboratory and in the field should
be conducted for the express purpose of model evaluation.
     The Exposure Assessment and Modeling Group participants are listed
                                 in Table 5-5.
                                      171

-------
               TABLE 5-5.  EXPOSURE/MODELING GROUP PARTICIPANTS
Non-Agency Participants

Allan Eschenroeder
A. D. Little Co.
Cambridge, MA

Michael Gilbertson
Environmental Protection Services
Environment Canada
Ottawa, Ontario
Canada

W. Brock Neely
Dow Chemical USA
Midland, MI

Ian Nisbet
Massachusetts Audubon Society
Lincoln, MA

Joseph Perkowski
Petro-Canada
Calgary, Alberta
Canada

Adel Sarofim (Chairman)
Massachusetts Institute of
  Technology
Cambridge, MA

Phillip Walsh
Oak Ridge National Laboratory
Oak Ridge, TN
EPA Representatives

Joseph Behar
Environmental Monitoring Systems
  Laboratory
Las Vegas, NV

Rizwanul Haque (Lead)
Office of Air, Land, and Water Use/
Office of Research and Development
Washington, D.C.

Raymond Lassiter (Co-Lead)
Environmental Research Laboratory/
Office of Research and Development
Athens, GA

David Mage
Environmental Monitoring Systems
  Laboratory/
Office of Research and Development
Research Triangle Park, NC

John Milliken
Industrial Environmental Research
  Laboratory
Research Triangle Park, NC
                                       172

-------
                               ACKNOWLEDGEMENTS
     The Exposure Assessment and Modeling Group, in reaching its recommenda-
tions on the research needs for exposure assessment and modeling, was
greatly assisted by reviews of the models developed for radionuclides
prepared by Hoffman et al. (1977a, 1977b) and on chemicals in the environment
prepared by Eschenroder et al. (1978).  Sections 5.5 and 5.7 of this chapter
have drawn heavily on these sources.
                                      173

-------
                                REFERENCES
Abrott, T. J., M. J. Barcelona, W. H. White, S. K.  Friedlander,  J.  J.  Morgan,
     K. A. Bell, C. I. Davidson and A. H. Miguel.  Human Dosage/Emission
     Source Relationships for Benzo(a)Pyrene and Chloroform in the  Los Angeles
     Basin. California Institute of Technology Report to the U.S. Environmental
     Protection Agency on Contract Number 68-03-0434, 1978.

Ahlstrom, S. W. and H. P. Foote.  Transport Modeling in the Environment Using
     the Discrete-Parcel-Random-Walk Approach.  In: Proceedings of  the
     Environmental Protection Agency Conference on Modeling, Cincinnati, Ohio,
     April 1976.

Alonso, A.  NUBE - A Digital Code to Evaluate the Hazards of Different Types
     of Reactor Accidents.  In: International Symposium on Fission  Product
     Release and Transport under Accident Conditions, April 5-7, 1965, Oak
     Ridge, Tennessee.  USAEC Report CONF-650407, Vol. II, pp. 1080-1090, Oak
     Ridge National Laboratory, 1965.

Anno, G. H., T. P. Wilcox and J. K. Witthaus.  RISC: Radiological Inhalation
     Safety Code for IBM 7090, Trans. Amer. Nucl. Soc., 6:106, 1963.

Armstrong, N. E. and E. F. Gloyna.  Radioactive Transport in Water.  Numerical
     Solutions of Radionuclide Transport Equations and Role of Plants in Sr-85
     Transport.  EHE-12-6703, CRWR-23, ORO-14, University of Texas  at Austin,
     January 1968.

Ausmus, B. S., D. R. Jackson and P. Van Voris.  The Accuracy of Screening
     Techniques. In:  Proceedings of the Symposium on Terrestrial Microcosms
     and Environmental Chemistry, Corvallis, Oregon, June, 1977, J. W. Gillett
     and J.M. Witt, eds.  National Science Foundation, Washington,  D.C., 1979.

Baker, D. A., G. R. Hoenes and J. E. Soldat.  FOOD - An Interactive Code to
     Calculate Internal radiation Doses from Contaminated Food Products,
     USERDA Report BNWL-SA-5523, Battelle Northwest Laboratories, February
     1976.

Beall, M. L., R. G. Nash, Jr. and P. C. Kearney.  Agroecosystem -A Laboratory
     Model Ecosystem to Simulate Agricultural Field Conditions for  Monitoring
     Pesticides. In:  Environmental Modeling and Simulation, W. Ott,  ed., pp.
     790-793.  EPA-600/9-76-016, U.S. Environmental Protection Agency,
     Washington, D.C., 1976.
                                      174

-------
Bell, G. D. and J. Houghton.  Risk Evaluation for Stack Releases.  British
     Report AHSB(S)R 173, United Kingdom Atomic Energy Authority, Risley,
     England, 1969.

Binford, P. T., J. Barish and F. B. K. Kara.  Estimation of Radiation Doses
     Following a Reactor Accident.  USAEC Report ORNL,-0486, Oak Ridge National
     Laboratory, February 1968.

Binford, F. T., T. P. Hamrick and B. H. Cope.  Some Techniques for Estimating
     the Results of the Emission of Radioactive Effluent from ORNL Stacks.
     USAEC Report ORNL-TM-3187, Oak Ridge National Laboratory, October 1970.

Blaine, R. A and E. L. Bramblett.  AISITE II, A Digital Computer Program for
     Investigation of Reactor Siting. NAA-SR-9982, Atomics International, 1964.

Bond, H., B. Lighthart, R. Shimabuku and L. Russell.  Some Effects of Cadmium
     on Coniferous Forest Soil/litter Microcosms.  EPA-600/3-75-036, U. S.
     Environmental Protection Agency, Corvallis, Oregon, 1975.

Bonzon, L. L. and J. B. Rivard.  Computational Method for Calculation of Radio-
     logical Dose Resulting from Hypothetical Fission Product Release.  USAEC
     Report SC-RR-70-338, Sandia Laboratories, New Mexico, 1970.

Boone, J. S., B. L. Parks and B. G. Kniazewycz.  RUBY, A Program for the
     Calculation of Activity Releases and Potential Doses from Pressurized
     Water Reactor Plant.  Carolina Power and Light Company, Raleigh, North
     Carolina, 1975.

Booth, R. S.  A Systems Analysis Model for Calculating Radionuclide Transport
     Between Receiving Waters and Bottom Sediments. In:  Environmental Toxicity
     of Aquatic Radionuclides:  Models and Mechanisms. Proceedings of the 8th
     International Conference on Environmental Toxicity, June 2-4, 1975,
     University of Rochester, Rochester, New York, pp. 133-164.  Ann Arbor
     Science Pub., Inc.,  Ann Arbor, Michigan, (also available as
     ORNL-TM-4751), 1975.

Booth, R. S. and S. V. Kaye.  A Preliminary Systems Analysis Model of Radio-
     activity Transfer to Man from Deposition in a Terrestrial Environment,
     USAEC Report ORNL-TM-3135, Oak Ridge National Laboratory, 1971.

Booth, R. S., S. V. Kaye and P. S. Rohwer.  A Systems Analysis Methodology for
     Predicting Dose to Man from a Radioactively Contaminated Terrestrial
     Environment. In:  Proceedings of the Third National Symposium on
     Radio-ecology, Oak Ridge, Tennessee, May 10-12, 1971, pp. 877-893. USAEC
     Report CONF-710501, 1971.

Bramati, L., T. Marzullo, I. Rosa and G. Zara.  VADOSCA:  A Simple Code for
     the Evaluation of Population Exposure Due to Radioactive Discharges. In:
     Proceedings of the Third International Congress of the International
     Radiation Protection Association, Washington, D.C., September 9-14,
     1973, pp. 1072-1077.  USAEC Report CONF-730907-P2, 1973.
                                      175

-------
Caltech, Department of Environmental Engineering.  Mass Balance Determinations
     for Pollutants in Urban Regions.  EPA Environmental Monitoring Series/
     EPA-600/4-78-046, 1978.

Cavagnaro, D. M.  Atmospheric Modeling of Pollution, volume 3,  1977-June
     1978 (A Bibliography with Abstracts).   National Technical Information
     Service, Springfield, Virginia, NTIS/PS-78-063/8WP, 1978a.

Cavagnaro, D. M.  Atmospheric Modeling of Air Pollution, volume 3,  1977-June
     1978 (A Bibliography with Abstracts).  National Technical  Information
     Service, Springfield, Virginia, NTIS/PS-78-063/8WP, 1978b.

Charak, I.  Atmospheric Dispersion of Fission Products Including the Effects of
     Building Dilution and Radioactive Decay.  Nucl. Appl., 3:283,  1967.

Chester, R.  0.  Biological Dose and Radiological Activity from Nuclear Reactor
     or Nuclear Weapon Fission Products.   USAEC Report ORNL-4996,  Oak Ridge
     National Laboratory, Oak Ridge, Tennessee, 1974.

Clarke, R. H.  A Multi-Compartment Model for Studying the Dose-Intake
     Relationship for Radioactive Isotopes.  British Report RD/B/N2291, Central
     Electricity Generating Board, Berkeley Nuclear Laboratories, 1972.

Clarke, R. H.  The WEERIE Program for Assessing the Radiological Consequences
     of Airborne Effluents from Nuclear Installations. Health Phys., 25:267,
     1973.

Coates. R. L. and N. R. Horton.  RSAC - A Radiological Safety Analysis Computer
     Code.  USAEC Report IDO-17151, Philips Petroleum Company,  1966.

Cole, L. K., R. L. Metcalf and J. R. Sanborn.  Environmental Fate of
     Insecticides in a Terrestrial Model Ecosystem.  Intern. J. Environ.
     Studies, 10:7, 1976.

Cole, L. K.  and R. L. Metcalf.  Terrestrial/Aquatic Laboratory Model Ecosystem
     for Pesticide Studies.  In: Proceedings of the Symposium on Terrestrial
     Microcosms and Environmental Chemistry, Corvallis, Oregon, June 1977,
     J. W. Gillett and J. M. Witt, eds.  National Science Foundation,
     Washington, D.C., 1979.

Cooper, R. E.  RADOS, A Code to Estimate Gamma Dose from a Cloud of Radio-
     active Gases.  USAEC Report DP-1098, Savanah River Laboratory,  1967.

Cooper, R. E.  EGAD - A Computer Program to Compute Dose Integrals from
     External Gamma Emitters.  USAEC Report DP-1304, Savannah River Laboratory,
     1972.

Cooper, R. E.  A Computer Code to Provide Emergency Response Capabilities
     to Estimate Whole Body Gamma Dose. In: Savannah River Laboratory
     Environmental Transport and Effects Research Annual Report FY-1975,
     (compiled by T. V. Crawford).  USERDA Report DP-1412, Savannah River
     Laboratory, 1976.
                                      176

-------
Cooper, R. E.  WRED, A Sitting Code to Estimate Dose Probability Distribution
     from Measured Meteorology Date.  Health Phys., 16:735, 1969.

Copenhaver, E. D., ed.  NSF-RANN Trace Contaminants Abstracts, 1(6).  Toxic
     Meterials Information Center, Oak Ridge National Laboratory, Oak Ridge
     Tennessee, 1974a.

Copenhaver, E. D., ed.  NSF-RANN Trace Contaminants Abstracts, 2(1).  Toxic
     Materials Information Center, Oak Ridge National Laboratory, Oak Ridge,
     Tennessee, 1974b.

Copenhaver, E. D., ed.  NSF-RANN Trace Contaminants Abstracts, 2(2).  Toxic
     Materials Information Center, Oak Ridge National Laboratory, Oak Ridge,
     Tennessee, 1975a.

Copenhaver, E. D., ed.  NSF-RANN Trace Contaminants Abstracts, 2(3).  Toxic
     Materials Information Center, Oak Ridge National Laboratory, Oak Ridge,
     Tennessee, 1975b.

Copenhaver, E. D., ed.  NSF-RANN Trace Contaminants Abstracts.  Toxic Materials
     Information Center, Oak Ridge National Laboratory, Oak Ridge, Tennessee,
     1975c.

Copenhaver, E. D., ed.  NSF-RANN Trace Contaminants Abstracts.  Toxic Materials
     Information Center, Oak Ridge National Laboratory, Oak Ridge, Tennessee,
     1976a.

Copenhaver, E. D., ed.  NSF-TENN Trace Contaminants Abstracts, Toxic Materials
     Information Center, Oak Ridge National Laboratory, Oak Ridge, Tennessee,
     September 1976b.

Copenhaver, E. D. and B. K. Wilkinson.  Transport of Hazardous Substances
     through Soil Processes.  Volume 1:  Arsenic, Beryllium, Cadmium, Copper,
     Cyanide, Lead, Mercury, Selenium, zinc, and Others. ORNL-EIS-74-70,  Part
     1. Ecology and Analysis of Trace Contaminants Program, Oak Ridge National
     Laboratory, Oak Ridge, Tenn.,  1974a.

Copenhaver, E. D. and B. K. Wilkinson.  Transport of Hazardous Substances
     Through Soil Processes.  Volume 2: Pesticides.  Ecology and Analysis of
     Trace Contaminants Program, Oak Ridge National Laboratory, Oak Ridge,
     Tenn., 1974b.

Corrill, L. S., E. D. Copenhaver and H. V. Leland.  Heavy Metals and Other
     Trace Elements in Aquatic Environments; citations (with abstracts)  for the
     literature.  ORNL-EIS-74-69. Ecology and Analysis of Trace Contaminants
     Program, Oak Ridge National Laboratory, Oak Ridge, Tennessee, 1974.

Couchman, M. L., A. W. DeAgazio and Y. S. Kim.  NURSE-1—A Nuclear Rocket
     Safety Evaluation Code for the Control Data 3600.  USAEO/NASA Report
     NUS-180, Nuclear Utility Services, Inc., 1964.
                                      177

-------
Cramer, J.  Model of the Circulation of DDT on Earth.  Atmos.  Environ.,
     7:241, 1972.

Cruz, F. D.  ESDOR Computer Code Program for an Estimate of the Radiation
     Dose in the Evaluation of Risks for Nuclear Power Plants  and Radioactive
     Facilities. Spanish Report SS/01/71, Junta de Energia Nuclear,  March,  1971
     (in Spanish); also, ORNL-tr-2628,  (in English),  Oak Ridge National
     Laboratory, Oak Ridge, Tennessee,  1973.

Culkowski, W.  M. and M. R. Patterson.  A comprehensive Atmospheric Transport
     and Diffusion Model.  ORNL/NSF/EATC-17, Oak Ridge National Laboratory,
     Oak Ridge, Tennessee, 1976.

Daniels, D. G., J. C. Sonnichsen and R. T. Jaske.  The Estuarine Version of
     the COLHEAT Digital Simulation Model.  BNWL-1342, Battelle Northwest
     Laboratories, 1970.

Draggan, S. The Microcosm as a Tool for Estimation of Environmental Transport
     of Toxic Materials.  Intern. J. Environ. Studies, 10:65,  1976.

Duguid, J. O.  and M. Reeves.  Material  Transport Through Porous Media:   A
     Finite-Element Galerkin Model.  USAEC Report ORNL-4928, Oak Ridge
     National Laboratory, Oak Ridge, Tennessee, 1976.

Eckert, R. J.   A Fortran Program for the Calculation of the Potential Hazards
     from an Accidental Fission-Product Release-MOl42.  USAEC Report WAPD-TM-
     348, Bettis Atomic Power Laboratory, 1964.

Ellison, T. W. and A. B. Dunham.  A CDC 6600 Program for Calculation of
     Inhalation and External Dose for Establishing Reactor Siting Distance
     Requirements.  USAEC Report KAPL-M-6592 (TWE-1), Knolls Atomic Power
     Laboratory, 1967.

Engleman, R. J. and W. E. Davis.  Low-Level Isentropic Trajectories and  the
     MIDAS Computer Program for the Montgomery Stream Function.  USAEC Report
     BNWL-441, Battelle Northwest Laboratories, 1968.

Eschenroeder, A., L Nelken and W. Lyman.  Models for Environmental
     Concentrations of Toxic Chemicls,  Report prepared for this workshop;
     Arthur D. Little, December, 1978.

Fletcher, J. F. and W. L. Dotson, compilers.  HERMES - A Digital Computer Code
     for Estimating Regional Radiological Effects from the Nuclear Power
     Industry.  USAEC Report HEDL-TME-71-168, Hanford Engineering Development
     Laboratory, 1971.

Friedlander, S. K., H. Brown, G. Bugliarello, L. B. Dworsky, J. J. Harrington,
     T. Hatch and J. Wei.  Technological Trends.  In: Man's Health and the
     Environment — Some Research Needs.  Report of the Task Force on Research
     Planning in Environmental Health,  chapter 10.  National Institute
     of Environmental Health Sciences,  U. S. Government Printing Office,
     Washington, D.C.,  1970.
                                      178

-------
Gile, J. D. and J. W. Gillett.  Terrestrial Microcosm Chamber Evaluations of
     Substitute Chemicals.  Proceedings of the Symposium on Terrestrial
     Microcosms and Environmental Chemistry, Corvallis, Oregon, June, 1977.
     National Science Foundation, Washington, D.C., 1979.

Gillett, J. W. and J. D. Gile.  Pesticide Fate in Terrestrial Laboratory
     Ecosystems.  International Journal of Environmental Studies, 10:15, 1976.

Gillett, J. W., J. Hill, A. Jarvinen and W. Schoor.  A conceptual Model for
     the Movement of Pesticides through the Environment.  EPA-600/3-74-024,
     U. S. Environmental Protection Agency Ecological Research Series, 1974.

Gillett, J. W. and J. M. Witt, eds.  Proceedings of the Symposium on Terres-
     trial Microcosms and Environmental Chemistry, Corvallis, Oregon, June,
     1977.  National Science Foundation, Washington, D.C., 1979.

Gogolak, C. V.  Comparison of Measured and Calculated Radiation Exposure from a
     Boiling Water Reactor Plume.  USAEC Report HASL-277, Health and Safety
     Laboratory, U. S. Atomic Energy Commission, 1973.

Harnden, D. S., ed.  NSF-RANN Trace Contaminants Abstracts, Toxic Materials
     Information Center, Oak Ridge National Laboratory, Oak Ridge, Tennessee,
     1977.

Hague, R. and V. H. Freed.  Behavior of Pesticides in the Environment:
     Environmental Chemodynamics.  Residue Reviews, F. A. Gunther, ed.
     Springer-Verlag, New York, 1974.

Hedemann, J. P.  Comparison of Mathematical Models for Calculation of External
     Gamma Doses Originating from Releases of Radioactivity to the Atmosphere.
     (In Danish).  Danish Atomic Energy Commission, Risoe Research
     Establishment, 1974.

Heffter, J. L., A. D. Taylor and G. J. Ferber.  A Regional-Continental Scale
     Transport, Diffusion and Deposition Model.  NOAA Technical Memorandum
     ERL ARL-50, Air Resources Laboratories, 1975.

Heller, F., W. Schikarski and A. Wickenhauser.  MUNDO - A Digital Code for
     Computing Accident Dose Rates in Reactor Environs.  German Report KFK 653,
     EUR 3684d, Institut fur Angewandte Reaktorphysik, Karlsruhe, F. R.
     Germany, (in German); also, EURFNR-461, United States (in English) -
     Euratom Fast Reactor Exchange Program, 1967.

Hendrickson, M. M.  EXDOSE, A Computer Program for Calculating the External
     Gamma dose from Airborne Fission Products.  USAEC Report BNWL-811,
     Battelle Northwest Laboratories, 1968.

Hine, G. J. and G. L. Brownell.  Radiation Dosimetry, p. 693.  Academic Press,
     Inc., New York, 1956.
                                      179

-------
Hoffman, F. O., C. W. Miller, D. L. Shaeffer and C. T.  Garten, Jr.  A
     Compilation of Computer Codes for the Assessment of Radionuclides Released
     to the Environment.  Nuc. Saf., 18:343, 1977a.

Hoffman, F. O., C. W. Miller, D. L. Shaeffer, C. T. Garten, Jr., R.  W. Shor
     and J. T. Ensminger.  A Compilation of Documented Computer Codes
     Applicable to Environmental Assessment of Radioactivity Releases.
     ORNL/TM-5830, Oak Ridge National Laboratory, Oak Ridge, Tennessee, 1977b.

Houston, J. R. and K. R. Heid.  PUDEQ:  A Computer Code for Calculating Dose
     Equivalent from Internal Deposition of Plutonium at Hanford.  USERDA
     Report BNWL-B-450, Battelle Northwest Laboratories, 1975.

Houston, J. R., D. L. Strenge and E. C. Watson.  DACRIN - A Computer Program
     for Calculating Organ Dose from Acute or Chronic Radionuclide Inhalation.
     USAEC Report BNWL-B-389, Battelle Northwest Laboratories, 1974.

Hubschmann, W. and D. Nagel.  ISOLA II - A Fortran IV Code for the Calculation
     of the Long-term Alpha and Beta Dose Distributions in the Vicinity of
     Nuclear Installations.  German Report KFK 2210, Gesellschaft fur
     Kernforschung m. b. h., Karlsruhe, K. R., Germany (in German);  also,
     ORNL-tr-4295 (in English), Oak Ridge National Laboratory, Oak Ridge,
     Tennessee, 1975.

ICRP.  Recommendations of the International Commission on Radiological Pro-
     tection,  Report of Committee 2 on Permissible Doses for Internal
     Radiation.   ICRP Publication 2, Pergamon Press, London, 1959.

ICRP.  Recommendations of the International Commission on Radiological
     Protection,  Report of Committee 4 on Evaluation of Doses to Body Tissues
     from Internal Contamination due to Occupational Exposure.  ICRP
     Publication  10, Pergamon Press, London, 1968.

ICRP.  Recommendations of the International Commission on Radiological
     Protection,  Report of Committee 4 on the Assessment of Internal
     Contamination Resulting from Recurrent or Prolonged Uptakes.  ICRP
     Publication  10A, Pergamon Press, London, 1971.

ICRP, Task Group  of Committee II.  Task Group on Lung Dynamics for Committee II
     of the ICRP. Health Phys., 12:173, 1966.

Kenfield, G. P.,  W. R. Lahs, W. B. Sayer, R. M. MacAdams and N. A. Harris.
     Curie-Dose-Thunderhead, A Digital Computer Program for External and
     Internal Radiation Dose Calculation.  USAEC Report NAA-SR-8884, Atomics
     International,  1965.

Killough, G. G.,  D. E. Dunning, Jr. and J. C. Pleasant.  INRE II: A Computer
     Implementation of Recent Models  for Estimating the Dose Equivalent to
     Organs of Man from Inhaled or Ingested Radionuclides.  NUREG/CR-0114,
     ORNL/NUREG/TM-84, Oak Ridge National Laboratory, Oak Ridge, Tennessee,
     1977.
                                       180

-------
Killough, G. G. and L. R. McKay.  A Methodology for Calculating Radiation
     Doses from Radioactivity Released to the Environment.  USERDA Report
     ORNL-4992, Oak Ridge National Laboratory, Oak Ridge, Tennessee, 1976.

Killough,, G. G. and P. S. Rohwer.  INDOS - Conversational Computer Codes to
     Implement ICRP-10-10A Modes for Estimation of Internal Radiation Dose to
     Man.  USAEC Report ORNL-4916, Oak Ridge National Laboratory, Oak Ridge,
     Tennessee, 1974.

Killough, G. G., P. S. Rohwer and W. D. Turner.  INREM - A Fortran Code
     Which Implements ICRP 2 Models of Internal Radiation Dose to Man.
     USAEC Report ORNL-5003, Oak Ridge National Laboratory, Oak Ridge,
     Tennessee, 1975.

Kim, Y. S.  NUS Effective Energy Program, NUS-156.  Nuclear Utility Services,
     Inc., 1963.

Lanes, J.  A Procedure to Determine the Relative Extent of Reactor Fission
     Product Releases Under Various Meteorological Conditions.  UUSAEC Report
     HASL-148, Health and Safety Laboratory, 1964.

Lange, A.  ADPIC;  A Three-Dimensional Computer code for the Study of Pollutant
     Dispersal and Deposition Under Complex Conditions.  UCRL-51462, Lawrence
     Livermore Laboratory, 1973.

Lee, E., R. J. Mack and D. B. Sedgley.  GADOSE and DOSET:  Programs to
     Calculate Environmental Consequences of Radioactivity Release.  USAED
     Report GA-6511, General Atomic, 1966.

Lehmann, E. J.  Sediment Water Interaction and Its Effect upon Water Quality,
     A Bibliography with Abstracts.  NTIS/PS-77/0021/4WP, National Technical
     Information Center, Springfield, Virginia, 1977.

Lehmann, E. J.  Water Quality Modeling - Hydrological and Limnological Systems,
     vol. 2, 1975 - June, 1977, A Bibliography with Abstracts.  National Tech-
     nical Information Service, Springfield, VA.  NTIS/PS-78/0694/OWP, 1978a.

Lehmann, E. J.  Water Quality Modeling - rtydrological and Limnological Systems,
      vol. 3, July, 1977 - June,  1978, A Bibliography with Abstracts.  National
      Technical Information Service, Springfield, VA.  NTIS/PS-78/0650/8WP,
      1978b.

Lichtenstein, E. P., T. W. Fuhremann and K. R. Schulz.  Translocation and
     Metabolism of   C-phorate as Affected by Percolating in a Model Soil-Plant
     Ecosystem.  J. Agric. Food Chem., 22:991, 1974.

Lighthart, B., H. Bond and M. Richard.  Trace Element Research Using
     Conferous Soil/Litter Microcosms.  EPA-600/3-33-091, U.S. Environmental
     Protection Agency, Corvallis, Oregon, 1977.
                                      181

-------
Lighthart, B., H. Bond and B. G. Volk.  The Use of Soil/Litter Microcosms
     With and Without Added Pollutants to Study Certain Components of the
     Decomposer Community.  Proceedings of the Symposium on Terrestrial
     Microcosms and Environmental Chemistry, Corvallis, Oregon, June 1977,
     J. W. Gillett and J. M. Witt, eds.  National Science Foundation,
     Washington, D. C., 1979.

Luna, R. E and H. W. Church.  DIFOUT, A Model for Computation of Aerosol
     Transport and Diffusion in the Atmosphere.  USAEC Report SC-RR-555,
     Sandia Laboratories, 1969.

MacDonald, H. F.  Assessment of Environmental Hazards Following Nuclear Reactor
     Accidents.  In: International Symposium on Rapid Methods for Measuring
     Radioactivity in the Environment, Munich, Germany, July 5-9, 1971,
     pp. 43-54.  USAEC Report CONF-710705, IAEA-SM-148/38, 1971.

MacKay, T. F. and R. F. Ely, Jr.  Computation of Radiological Consequences
     Using INHEC Computer Program.  GAI-TR-101P, Gilbert Associates, Inc.,
     Reading, Pennsylvania, 1974.

Martin, J. A., Jr., C. B. Nelson and P. A. Cuny.  AIREM Program Manual:  A
     Computer Code for Calculating Doses, Population Doses and Ground
     Depositions Due to Atmospheric Emissions of Radionuclides.
     EPA-520/1-74-004, U.S. Environmental Protection Agency, Office of
     Radiation Programs, 1974.

Mills, M. T., R. C. Dahlman and J. S. Olson.  Ground Level Air Concentrations
     of Dust Particles Downwind from a Tailings Area During a Typical
     Windstorm.  USAEC Report ORNL-TM-4375, Oak Ridge National Laboratory,
     Oak Ridge, Tennessee,  1974.

Moore, R. E.  AIRDOS - A Computer Code for Estimating Population and Individual
     Doses Resulting from Atmospheric Releases of Radionuclides from Nuclear
     Facilities.  ORNL-TM-4687, Oak Ridge National Laboratory, Oak Ridge,
     Tennesse, 1975.

Nash, R. G., M. L. Beall, Jr. and W. G. Harris.  Toxaphene and 1,1,1-trichloro-
     2,2-bis-p-chlorophenyl)ethane (DDT) Losses from Cotton in an
     Agroecosystem Chamber.  J. Agric. Food Chem., 25:336, 1977.

National Academy of Sciences.  Nonfluorinated Halomethanes in the Environment.
     National Academy of Sciences, Washington, D.C., 1978.

Nisbet  , I. C. T.  Human Exposure to Chlordane, Heptachlor, and Their
     Metabolites.  Unpublished report to U. S. Environmental Protection Agency,
     1977.

Nisbet, I. C. T. and A. F.  Sarofim.  Environmental Health and Perspectives,
     1:21, 1972.

Nowicki, K.  Estimation of  Population Doses from a Nuclear Power Plant During
     Normal Operations.  Danish Report RISO-M-1808  (In English).  Danish
     Atomic Energy Commission, Risoe, Research Establishment,  1975.

                                       182

-------
Onishi, Y., P. A. Johanson, R. G. Baca and E. L.  Hilty.  Studies of Columbia
     River Water Quality, Development of Mathematical Models for Sediment
     and Radionuclide Transport Analysis.  USERDA Report BNWL-B-452,
     Battelle Northwest Laboratories, 1976.

Otter, J. M. and P. A. Conners.  Description of the COMRADEX-III Code.
     USNRC Report TI-001-130-053, Atomics International, 1975.

Pasquill, F.  The Estimation of Diffusion Coefficients from Meteorological
     Data.  Meteorol. Mag., 90:33, 1961.

Plato, P. A., D. F. Menker and M. Dauer.  Computer Model for the Prediction
     of the Dispersion of Airborne Radioactive Pollutants.  Health Phys.,
     13:1105, 1967.

Plato, P. A., D. F. Menker and M. Dauer.  Some Examples and Limitations of the
     RADS and ARADS Computer Programs.  Health Phys., 16:383, 1969.

Randers, J. and D. Meadows.  System Simulation to Test Environmental Policy:
     A Sample Study of DDT Movement in the Environment.  In: Toward Global
     Equilibrium - Collected papers, D. Meadows,  ed.  Wright-Allen Press,
     Cambridge, 1971.

Reeves, M. and J. O. Duguid.  Water Movement Through Saturated-Unsaturated
     Porus Media:  A Finite-Element Galerkin Model.  USAEC Report ORNL-4927,
     Oak Ridge National Laboratory, Oak Ridge, Tennessee, 1975.

Reeves, M., P. G. Fowler and K. E. Cowser.  A Computer Code for Analyzing
     Routine Atmospheric Releases of Short-Lived Radioactive Nuclides.
     USAEC Report ORNL-TM-3613, Oak Ridge National Laboratory,  Oak Ridge,
     Tennessee, 1972.

Richardson, L. C.  User's Manual for the FORTRAN Version of RSAC, A Radiolo-
     gical Safety Analysis Computer Program, USAEC Report IDO-17261, Phillips
     Petroleum Co., 1968.

Ross, R. H., ed.  NSF-RANN Trace Contaminants Abstracts.  Toxic Materials
     Information Center, Oak Ridge National Laboratory, Oak Ridge, Tennessee,
     Last Issue (71977) .

RSIC.  Computer Code Collection, ACRA-II-Kernal Integration Code Estimation
     of Radiation Doses Caused by a Hypothetical Reactor Accident.  CCC-213,
     Radiation Shielding Information Center, Oak Ridge National Laboratory,
     Oak Ridge, Tennessee, 1974.

Rupp, E. M., D. C. Parzyck, P. J. Walsh, R. S. Booth, R. J. Raridon and
     B. L. Whitfield.  Composite Hazard Index for Assessing Limiting Exposures
     to Environmental Pollutants:  Application Through a Case Study.  Environ.
     Sci. Technol., 12:802, 1978.
                                      183

-------
Sagendorf, J. F.  A Program for Evaluating Atmospheric Dispersion from a
     Nuclear Power Station.  NOAA Technical Memorandum ERL ARL-42,  Air
     Resources Laboratory, 1974.

Sato, K., Y. Ito, I. lyori and A. Yamaji.   A Program for Computing External
     Dose Rate from a Radioactive Cloud and Numerical Examples for Nuclear
     Ships (In Japanese).  Senpaku Gijutsu Kenkyujo, 5:55, 1968.

Shin, C. S. and E. F. Gloyna.  Radioactivity Transport in Water-Mathematical
     Model for the Transport of Radionuclides.  USAEC Report EHE-04-6202,
     CRWR-18, University of Texas at Austin, 1967.

Slade, D. H., ed.  Meterology and Atomic Energy 1968.  TID-24190, U.S. Atomic
     Energy Commission, 1968.

Soldat, J. K.  Modeling of Environmental Pathways and Radiation Doses from
     Nuclear Facilities.  USAEC Report BNWL-SA-3939, Battelle Northwest
     Laboratories, 1971.

Soldat, J. K., N. M. Robinson and D. A. Baker.  Models and Computer Codes for
     Evaluating Environmental Radiation Doses.  BNWL-1754, Battelle Northwest
     Laboratories, 1974.

Specht, E., C. Martin, J. Otter and R. Hart.  Description of the COMRADEX-II
     Code.  USAEC Report TI-001-130-048, Atomics International, 1975.

Stallman, F. W. and F. B. K. Kam.  ACRA -A Computer Program for the Estimation
     of Radiation Doses Caused by a Hypothetical Reactor Accident.
     ORNL-TM-4082, Oak Ridge National Laboratory, Oak Ridge, Tennessee, 1973.

Start, G. E. and L. L. Wendell.  Regional Effluent Dispersion Calculations
     Considering Spatial and Temporal Meteorological Variations.  NOAA
     Technical Memorandum ERL ARL-44, Air Resources Laboratories, 1974.

Steyn, J. J. and Y. S. Kim.  Users Manual for the Code EXGAM: A Code to
     Calculate Gamma Photon Dose from an Airborne Radioactive Release.
     USAEC-NASA Report NUS-329, NUS Corporation, 1966.

Strenge, D. L. DACRIN - A Computer Program for Calculating Organ Dose from
     Acute or Chronic Radionuclide Inhalation:Modification for Gastro-
     intestinal Tract Dose.  USAEC Report BNWL-B-389 SUPP, Battelle Northwest
     Laboratories,  1975.

Strenge, D. L., M. M. Hendrickson and E. C. Watson.  RACER - A Computer Program
     for Calculating Potential External Dose from Airborne Fission Products
     Following Postulated Reactor Accidents.  USAEC Report BNLW-B-351, Battelle
     Northwest Laboratories, 1971.

Strenge, D. L. and E. C. Watson.  KRONIC - A Computer Program for Calculating
     Annual Average External Doses from Chronic Atmospheric Releases of
     Radionuclides.  USAEC Report BNWL-B-264, Battelle Northwest Laboratories,
      1973.
                                       184

-------
Strenge, D. L., E. C. Watson and J. G. Droppo.  Review of Calculational
     Models and Computer Codes for Environmental Dose Assessment of Radioactive
     Releases.  BNLW-B-454, Battelle Northwest Laboratories, 1976.

Strenge, D. L., E. C. Watson and J. R. Houston.  SUBDOSA - A Computer Program
     for Calculating External Doses from Accidental Atmospheric Releases of
     Radionuclides.  USERDA Report BNWL-B-351, Battelle Northwest Laboratories,
     1975.

Thykier-Nielsen, S.  Models for Calculation of External Gamma Doses and
     Inhalation Doses from Releases of Radioactive Isotopes to the Atmosphere.
     (In Danish).  Danish Report RISO-M-1725, Danish Atomic Energy Commission,
     Risoe, Research Establishment, 1974.

Travis, J. R.  A Model for Predicting the Redistribution of Particulate Con-
     taminants from Soil Surfaces.  USERDA Report LA-6035-MS, Los Alamos
     Scientific Laboratory, 1975.

Trubey, D. K. and S. V. Kaye.  The EXTREM III Computer Code for Estimating
     External Radiation Doses to Populations from Environmental Releases.
     USAEC Report ORNL-TM-4322, Oak Ridge National Laboratory, Oak Ridge,
     Tennessee, 1973.

Turner, W. D.  The EXREM II Computer Code for Estimating External Doses to
     Populations from Construction of a Sea-Level Canal with Nuclear
     Explosives.  USAEC Report CTC-8, Union Carbide Corporation, 1969.

Turner, W. D., S. V. Kaye and P. S. Rohwer.  EXREM and INREM Computer Codes
     for Estimating Radiation Dose to Populations from Construction of a
     Sea-Level Canal with Nuclear Explosives.  USAEC Report K-1752, Oak Ridge
     National Laboratory, Oak Ridge, Tennessee, 1968.

Tveten, U.  Areas Affecteed by Ground Deposition of Cs-137 and Description of
     the Computer Code ARCON.  In: Proceedings of a Symposium on Siting of
     Nuclear Facilities, Vienna, Austria, December 9-13, pp. 497-515.
     IAEA-SM-188/35, 1974.

U. S. Environmental Protection Agency.  Compilation of Air Pollution Emission
     Factors.  AP-42, Office of Air Quality and Planning Standards, U.S.
     Environmental Protection Agency.  Research Triangle Park, North Carolina,
     1976.

U. S. Environmental Protection Agency.  Non-Point Water Quality Modeling in
     Wildland Management: A State-of-the-Art Assessment, Vol. 1.  Office of
     Research and Development, U. S. Environmental Protection Agency,
     Athens, Georgia, 1977.

Veverka, O., V. Valenta and K. Vlachovsky.  METEO-1, A Programme for Calcula-
     tions of Atmospheric Dispersion of Activities from Nuclear Power Plants.
     Czech. Report ZJE-166, Skoda Works, Plzen, Czechoslovakia, 1975.
                                      185

-------
Voilegue, P. G.  Calculation of Organ and Tissue Burdens and Doses Resulting
     from an Acute Exposure to a Radioactive Aerosol Using the ICRP Tank Group
     Lung Report on the Human Respiratory Tract.  USAEC Report IDO-12067,
     Idaho Operations Office, 1968.

Voilleque, P. G.  AERIN, A Code for Acute Aerosol Inhalation Exposure
     Calculations.  Health Phys., 19:427, 1970.

Watts, J. R.  Modeling of Radiation Doses from Chronic Aqueous Releases.
     Presented at Health Physics Society Annual Meeting, San Franciso,
     California, June 27-July 2, 1976.  USERDA Report DP-MS-75-126, Savannah
     River Laboratory, 1976.

Wenzel, D. R.  A Computer Code to Calculate Doses Resulting from Release of
     Radionuclides to the Atmosphere.  Presented at Health Physics Society
     Annual Meeting, Houston, Texas, July 7-11, 1974.

Willis, C. A., G. A. Spangler and W. A. Rhoades.  A New Technique for Reactor
     Siting Dose Calculations.  Health Phys., 19:47, 1970.

Winton, M. L.  A Compilation of Computer Codes for Nuclear Accident Analysis.
     Nucl. Safety, 10:131, 1969.

Winton, M. L.  Computer Codes for Analyzing Nuclear Accidents.  Nucl. Safety
     12:461, 1971.

Winton, M. L.  Computer Codes for Analyzing Nuclear Accidents.  Nucl. Safety,
     15:535, 1974.

Wong, P.  A FORTRAN Program to Determine Absorbed Organ Dose Due to Inhalation
     of Radionuclides.  USNRDL-TR-67-7, u. s. Naval Radiological Defense
     Laboratory, 1967.

Woodwell, G., P. Craig and H. Johnson.  DDT in the Biosphere: Where Does It Go?
     Science, 174:1101, 1971.
                                       186

-------
             6.0 REGULATORY ASPECTS OF TRANSPORT AND FATE RESEARCH
6.1  INTRODUCTION

     Within the past three decades there has been enormous growth in both the
number and amount of chemicals used by society.  These chemicals are used as
Pharmaceuticals, food additives, pesticides, and industrial chemicals.

     The manufacture and use of these chemicals has provided society with
better health and a higher standard of living.  However, the proliferation of
chemical usage has elicited concern over the long range effect of these
materials on man and his environment.  A part of this concern is the
realization that many of the chemicals have become widespread environmental
contaminants.

     In the production and use of many chemicals, it is nearly impossible to
avoid the escape of at least small amounts of the material to the environment,
either by inadvertence or design.  A closed system is not employed in most
instances of manufacture and use; thus, conditions exist for chemicals to
escape through air and water effluents.  In other situations, the chemical may
intentionally be released into the environment where it may be transported by
air or water to sites well removed from the original source.  In the process of
transport, the substance may have considerable impact on environmental quality,
various species of biota, and may involve contact with humans.

     The amount of chemical that escapes is a function of not only the quantity
produced and used, but also of the manner in which the chemical is handled by
man.  Prudent practices can reduce the escape of the chemical and reduce its
impact on the environment.  Similarly, practices followed in manufacture,
handling, and use of chemicals can reduce the hazard to man.  Since any
chemical in sufficient amounts may be hazardous, the objective in developing
appropriate practices and strategies in handling chemicals is to minimize the
hazard by control of the quantity that may escape into the environment.

     Because of the possible hazards in chemical transport, more must be known
about the fate of chemicals entering the environment.  Over the years, a body
of information has been developed demonstrating the relationship of the
properties and reactivity of the chemical to fate in the environment.  With
this information it is possible to utilize appropriate physicochemical
parameters to gain some insight on the transport and fate of a given chemical
in the environment.
                                      187

-------
     While a significant body of information has been developed regarding
chemical transport and transformation, further elucidation of principles in
this area is required for rational regulation of manufacture, transportation,
use, and disposal of chemicals.  Of particular interest are those principles
that may indicate the utility of "benchmark measurements" and the value of
"bench-mark compounds" as a. guide to the fate of new substances.  At present,
there are a number of physical parameters of a compound that may be measured
and an equal number of processes that may be studied to elucidate transport and
transformation .  These studies, particularly if conducted in the environment,
may require years before definitive answers are obtained.  In the interest of
protecting man and the environment through rational regulation of chemicals
while simultaneously allowing society to benefit from their development and
use, reliable laboratory tests providing rapid assessment of transport and fate
are needed;  Such tests would permit the introduction and utilization of new
chemical substances in a reasonable period of time and at a reasonable cost.
Similarly, the information gained would provide a basis for developing
strategies for minimizing environmental pollution and exposure as well as
tactics for safe disposal of the material after use.

     In order to prevent significant adverse effects on man or the environment
and to provide orderly and expeditious introduction and use of economically
important chemicals, thorough research should be supported to develop new and
improved test protocols to evaluate transport and transformation of chemicals
in the environment.  This research should develop tests giving the most
reliable predictions while avoiding the accumulation of irrelevant data.  These
protocols should utilize the minimum number of tests that will produce the
desired correlation and reliability, and which allows an assessment of
transport and transformation.  Such information would aid in the rational and
effective implementation of laws and regulations.
6.2  LEGAL REQUIREMENTS FOR CHEMICAL TRANSPORT AND FATE DATA AND THEIR USE IN
     REGULATION

6.2.1  Toxic Substances Control Act (TSCA)  PL94-469

     TSCA does not explicity require the obtaining of chemical fate information
but the law contains several features which implicitly lead to that requirement.

     Section 2b of TSCA states that "It is policy...that...adequate data
should be developed with respect to the effect of chemical substances and mix-
tures on health and the environment and that development of such data should be
the responsibility of those who manufacture and those who process such chemical
substances and mixtures".

     In Section 3(5) the term environment is defined as including water, air
and land, and the inter-relationships which exist among and between water, air,
land, and all living things.

     Section 4 of the Toxic Substances Control Act (TSCA) is principally
concerned with existing chemicals which will be listed on an inventory being
prepared by EPA.  The section states that if the administrator finds that the
                                       188

-------
manufacture, distribution, processing, use or disposal of a chemical substance
or mixture may present an unreasonable risk of injury to health or the
environment, and that there are insufficient data and experience to reasonably
determine or predict such risk, and that testing is necessary to develop such
data, or that there is, or may be significant or substantial human exposure,
the administrator shall by rule require testing to develop relevant data.

     Section 5 of TSCA requires EPA to control new chemicals or significant new
uses of existing chemicals but does not grant to EPA authority to require
testing of a new chemical.  However, EPA is required to assess the risk posed
by new chemicals to health or the environment.  This risk assessment will be
based partly on the requirement that manufacturers submit all relevant test
data in their possession or control to EPA which has 90 days to perform this
risk assessment (extendable to 180 days).  If within that time period EPA makes
a determination that the proposed use, manufacturing, processing, distribution,
or disposal activities will present an unreasonable risk to health or the
environment, EPA can regulate those activities.  If EPA finds that there is
insufficient data to assess the risks posed by the chemical, and the use,
manufacturing, etc., of the chemical may present an unreasonable risk to health
or the environment, or there will be substantial exposure to man or the
environment, EPA can regulate the chemical until the necessary data are
developed.

     Because of the difference in wording with respect to existing and new
chemicals, the Office of Toxic Substances (OTS) will be formulating rules for
testing existing chemicals and presenting guidelines for the testing of new
chemicals.  The guidelines will provide some counsel to manufacturers of new
chemicals regarding EPA's position on the kinds of information necessary to
assess the potential for unreasonable risk.

     With respect to the information which the Agency would like to have in
order to make such an assessment, there should be little difference between
existing and new chemicals.  One key determination will be the expected
or estimated environmental concentration (EEC).  One of many elements which
determine EEC is the fate of the chemical in the environment(s) under consi-
deration.  Fate, in turn, may be viewed as consisting of transport and persis-
tence possibilities.  An analysis of potential transport mechanisms leads to
the conclusion that certain physical and chemical data should be considered.
Depending on the nature of the specific chemical, such data may include solubi-
lity in water, the vapor pressure, the octanol/water partition coefficient, pH,
density, particle size, and other data as discussed in the proposed Section 5
guideline drafts.  Certain data may be used to predict volatility and soil/
sediment adsorption possibilities.  EPA would like to have information on bio-
degradation, photolysis, hydrolysis, abiotic oxidation/reduction, and dissocia-
tion possibilities in order to predict persistance.

     The rules and guidelines will specify the methodology EPA believes
appropriate to obtain the kinds of data discussed.  These data will be
used in making detailed assessments of the potential risks.  Although it is
believed that such data are essential in assessing risk, fate data and esti-
mates are only part of the many factors which must be considered in the
assessment.
                                      189

-------
     Careful consideration of the test plan and the sequencing of tests
should reveal relatively few chemicals that will need to undergo the entire set
of tests.  Factors which will influence the decision on which tests to run will
include the known physical and chemical properties of the chemical and the ways
in which the substance is or will be manufactured, processed, transported,
used, and disposed.

     In the course of developing testing guidelines, EPA has considered a num-
ber of tiered testing systems which have been proposed by various interested
parties and some which have been developed within OTS.  Generally, these test
schemes tend to contain basic physical and chemical properties and short-term
persistence and acute toxicity tests in the "lowest" or "first" tier and
longer-term, more costly tests in the higher tiers.  The higher tier tests
might include the use of radiolabeled test material, the determination of
metabolic intermediates, and some sort of mass balance in biodegradation test-
ing.  Another example of a higher tier test would be the determination of the
"effective vapor pressure" and volatility rate of a chemical mixed in a well-
defined soil under controlled soil moisture, air humidity, air flow, and tem-
perature conditions.

     The first tier tests are often called "screening tests" and while they
provide a predictive first approximation, more elaborate studies conducted in
the higher tiers may be required.  There is a need to develop satisfactory de-
cision criteria for progressing from one tier to the next.  Defining the nature
and sensitivity of decision triggers will have a great deal to do with the
expense of testing a substance as well as with' the desired degree of certainty
that the system will yield a low percentage of false negatives and false
positives.

6.2.2  Federal Insecticide, Fungicide and Rodenticide Act (FIFRA)

     On July 3, 1975 EPA promulgated final registration regulations in 40 CFR
Part 162 Subpart A.  These regulations established the basic requirements for
registration of a pesticide product.  Section 162.8(b) (3) (ii) Environmenta1
Chemistry lists the kinds of data the Agency would generally require to
register a pesticide intended for "outdoor application", as that term is de-
fined in 40 CFR Sl62.3(cc).  The term includes any pesticide intended for
outside application or whose consequences would be used to define what chemical
substances remain in the environment after a pesticide is used, how long and at
what levels the residues persist in the environment, and where the residues are
likely to be found in relation to where the product has been used.  The kinds
of data required in Section 162.8(b) (3) (ii) to evaluate a pesticides'
environmental chemistry characteristics include, but are not limited to: field
stability data on the active ingredient(s) indicating the dissipation time and
its modes of degradation and/or metabolism; persistence, degradation and
accumulation data for target and/or non-target species for the active
ingredient(s), metabolite(s), or degradation product(s); and mobility data for
the active ingredient(s), metabolite(s) or degradation product(s) including
volatility and leaching properties.  Environmental chemistry information in
support of safe methods for the disposal of the pesticide formulation and
pesticide containers is also required under this subsection.
                                      190

-------
     Under the authority of Section 3(c) (2), 8 and 25(a)  of FIFRA, as amended,
the Agency published proposed Guidelines on June 25, 1975  and again on July 10,
1978 which described with more specificity the kinds of data which must be
submitted to satisfy the requirements of the registration  regulations.  The
proposed Guidelines specify the conditions under which each particular data
requirement is applicable to a pesticide product, the standards for acceptable
testing, stated with as much specificity as the current scientific disciplines
can provide, and the information required in a test report.  The appendicies to
the Guidelines provide useful information and references for designing test
protocols and, in some cases, examples of acceptable protocols for conducting
the required testing.

     Environmental Chemistry data requirements specified in Section 163.62 as
proposed Guidelines in the Federal Register, July 10, 1978 greatly expand upon
but generally parallel the basic registration requirements outlined in Section
162.8(b) (3) (ii) of the FIFRA Section 3 Regulations for Environmental Chemis-
try.  Section 163.62-12, which is reserved in the current  Guidelines, will also
contain requirements for environmental chemistry data which would be used with
other data to establish safe reentry intervals. Section 163.63-13 will be re-
vised at some future time to establish specific data requirements concerning
the disposal and storage of pesticides.

6.2.3  Regulatory Aspects of Transport and Fate of Chemicals in the Environment
       as Related to Resource Conservation and Recovery Act (RCRA), 1976

     The Solid Waste Disposal Act as amended by the Resources Conservation and
Recovery Act of 1976 (P.L. 94-580 [October 21, 1976]) provides (1) for
improvement of practices in solid waste disposal to protect public health and
environmental quality, (2) for control by regulation of hazardous waste from
the point of generation to the point of disposal, and (3)  establishes resource
conservation as the preferred solid waste management approach.

     The definition of solid waste (Section 1007 [27]) encompasses garbage,
refuse, sludges, and other discarded materials including liquids, semi-solids,
and contained gasses (with a few exceptions) from both municipal and industrial
sources.  Subtitle C of the act creates a "cradle-to-grave" management control
system for hazardous waste, a subset of all solid wastes.   Solid waste which is
excluded from subtitle C will be subject to the requirements of subtitle D of
the Act, under which open dumping is prohibited and environmentally acceptable
practices are required.

     Subtitle C consists of seven sections that establish a hazardous waste
program.  Section 3001 requires EPA to define criteria and methods for identi-
fying and listing hazardous wastes.  Those wastes which are identified or list-
ed as hazardous are then included in the management control system constructed
under Section 3002-3006 and 3010.

     Section 3002 addresses the standards applicable to waste generators and
announces the mechanics of the manifest system to track waste transported from
the point of generation to its ultimate disposition. Section 3003 authorizes
                                      191

-------
standards for transporters of hazardous waste, to assure that such waste is
carried carefully.  The Agency has attempted to coordinate closely with Depart-
ment of Transportation regulations.

     Section 3004 addresses standards affecting owners and operators of hazar-
dous waste treatment, storage, and disposal facilities.  These standards define
the levels of human health and environmental protection to be achieved by
these facilities and provide the criteria against which EPA (or state) offi-
cials will measure applications for permits.  Facilities on a generator's pro-
perty as well as off-site facilities are covered by these regulations and do
require permits; generators and transporters do not otherwise need permits.

     Section 3005 regulations set out the scope and coverage of the actual per-
mit including the granting process for facility owners and operators.  Require-
ments for the permit application as well as for the issuance and revocation
process are defined by regulations under 40 CFR Parts 122, 124, and 128.
Section 3005(e) provides for interim status during the time period that EPA or
the states are reviewing the pending permit applications.  Special regulations
under 3004 apply to facilities during the interim status period.

     Section 3006 requires EPA to issue guidelines under which states may seek
both full and interim authorization to carry out the hazardous waste program in
lieu of the EPA-administered program.  States seeking authorization in accor-
dance with Section 3006 guidelines need to demonstrate that their hazardous
waste management regulations are consistent with and equivalent to EPA regula-
tions under sections 3001 -3005.

     Section 3010 requires any person generating, transporting, owning, or
operating a facility for treatment, storage, and disposal of hazardous waste to
notify EPA of this activity within 90 days after promulgation or revision of
regulations.  The notification should identify and list a hazardous waste sub-
ject to Subtitle C regulation and the waste may not be legally transported,
treated, stored, or disposed after the 90 day period unless this timely noti-
fication has been given to EPA or the authorized state during the above 90 day
period.  Owners and operators of inactive facilities are not required to
notify.

     Most of the hazardous waste regulations discussed above have been proposed
in the Federal Register and should be final by January 1, 1980.  Sections 3001,
3002, and 3004 were proposed on December 18, 1978, Section 3003 on April 28,
1979, Section 3006 on February 1, 1978, and Section 3010 on July 11, 1978.
Section 3005 will be proposed and Section 3006 reproposed under authority of
the Clean Water Act.

     In order to develop and implement regulations, research should be applied
to some of the unknowns regarding transport and fate of a variety of hazardous
wastes destined for treatment, storage, or disposal on land.  Research is
needed on the movement of chemical pollutants through all media and especially
clay soil, the mechanisms of adsorption and desorption, the release of vola-
tiles from the soil and vapor phase transport, the effects of disposal on the
hydraulic conductivity of soil, and techniques for monitoring leachate and
groundwater at land disposal sites.
                                      192

-------
6.2.4 Safe Drinking Water Act of 1974 (PL93-523)

     Under this act, the administration is required to regulate any substance
that may have an adverse effect on man.  Regulation may be accomplished either
by setting a standard or by requiring treatment to remove the hazard.
Synthetic organic chemical contaminants in drinking water originate from
organic chemical contaminants derived from chlorination practices at the water
plant (Trihalomethanes or THM's), while organic chemical contaminants in the
water source derive from direct or indirect industrial discharges, agricultural
sources (pesticides), and runoff.

     The proposed regulations address these sources separately.  Trihalome-
thanes are to be controlled by means of establishing a Maximum Contaminant
Level (MCL) of 0.10 milligrams per liter (mg/1) (100 micrograms per liter).
Synthetic organic chemicals of industrial origin are to be controlled by treat-
ing the water in granular activated carbon (GAC) or the equivalent.  The time
schedule for installation and operation of GAC in affected systems (those that
do not receive a variance) is 5 years from promulgation.

     With respect to chemical fate (transport and persistence) information, the
Safe Drinking Water Act and its administrators focus on the assurance that
certain undesirable constituents do not appear in drinking water above defined
maximum concentration levels.

6.2.5    Clean Water Act of 1977 (PL95-217)

     This act is an amended form of the Federal Water Pollution Control Act,
which was last amended in 1972.  The objective of the act is "to restore
and maintain the chemical, physical, and biological integrity of the Nation's
waters."  The act covers a multitude of water-quality related subjects, in-
cluding research, grants for construction of treatment works, standards and
enforcement, permits and licenses, and many other general provisions.  The act
directs EPA's Administrator to "prepare or develop comprehensive programs for
preventing, reducing, or eliminating the pollution" of ground and surface
waters in the United States.  As part of these programs, the Administrator is
required to "conduct and promote the coordination and acceleration of research,
investigation, experiments, training, demonstrations, surveys, and studies re-
lating to the causes, effects, extent, prevention, reduction, and elimination
of pollution."

     The terms "transport" and "fate" of chemicals do not appear in the act.
Nevertheless, transport and fate information are probably necessary for EPA to
fulfill the requirements of several of the sections of the act.  These sections
are discussed below.

6.2.5.1  Section 307 - Toxic and Pretreatment Effluent Standards

     Section 307 (a) (1) requires the Administrator to publish a list of toxic
pollutants.  He also has the authorization to add or remove pollutants from
                                      193

-------
this list but must "take into account the toxicity of the pollutant,  its per-
sistence, degradability, the usual or potential presence of the affected orga-
nisms in any waters, the importance of the affected organisms, and the nature
and extent of the effect of the toxic pollutant on such organisms." Since
transport and fate information is necessary to determine persistence  and degra-
dability of the pollutants, this information is necessary whenever the Admini-
strator adds to or removes from the toxics list.  Currently the list  contains
sixty-five compounds, substances, and groups or classes of compounds.

     Section 307 (a)(2) provides EPA the opportunity to require effluent limi-
tations on a pollutant-by-pollutant rather than industry-by-industry basis as
provided elsewhere in the act (S.301).  When requiring these effluent
limitations, the Administrator is directed to consider the same things as under
Section 307 (a)(1) including persistence and degradability.  Therefore, some
transport and fate information is needed for effluent limitations under Section
307.

6.2.5.2  Section 304 - Information and Guidelines

     Under this section, the Administrator is required to develop and publish
water quality criteria which shall reflect the "latest scientific knowledge...
on the concentration and dispersal of pollutants, or their by-products, through
biological, physical, and chemical processes."

6.2.5.3  Section 403 - Ocean Discharge Criteria

     Section 403 (c)(1) requires the Administrator to promulgate guidelines for
determination of the degradation of the waters of the territorial seas, the
contiguous zone, and the oceans.  Among other things, these guidelines are to
include "the transfer, concentration, and dispersal of pollutants or their by-
products through biological, physical, and chemical processes," and also pollu-
tant persistence.

     There are also several other sections of the act where transport and fate
information may be useful in determining potentials for water quality deterior-
ation, although the language of the act does not specifically call for it.  In
Section 311, Oil and Hazardous Substance Liability, transport and fate data are
used in support of regulation to control industrial effluents and to identify
candidates for the toxic pollutant list under Section 307  (a)(b) of CWA (1977).
Transport and fate data are also used in screening candidate chemicals to
identify "red flags" used in risk assessments to determine if a candidate
meets requirements of the list.

     To determine regulatory options for pollutants not controlled by BAT deve-
loped under Section 301 of CWA (1977) [Section 307 (a)(2)], transport and fate
data are used in the early warning (Action Alert) phase which identifies "red
flags", if any, and gives presumptive indicative of risk.  The data are also
used in risk assessment to get quantitative determination  of risk and identify
critical pollutant sources which "drive" that risk.

     As part of predictive exposure quantification, transport and fate informa-
tion is used in conjunction with source identification to  predict quantities
                                       194

-------
lost to the environment and the environmental compartment to which the chemical
is lost.  These results are subsequently confirmed by monitoring.   Finally, the
data are needed in conjunction with monitoring to help identify unsuspected
sources when monitoring is the first indicator of potential risk ("red flag"),
as part of observed exposure quantification.

     Other areas not directly limited to aquatic parameters where  transport and
fate data are needed include source ranking, identification of other media if
regulation beyond CWA must be considered [Section 307 (a)(2)], and completion
by multimedia mass or materials balance.

     These fate (transport and transformation) data are used in one aspect of
the CWA to provide one basis to quantify exposure for specific quantitative
risk assessment, and to permit system identification to the most important
regulable sources of pollutant subject to that risk assessment.  The latter is
the more directly regulative of the two basic uses of job data.
6.3  NECESSARY DATA FOR ESTIMATING EXPECTED ENVIRONMENTAL CONCENTRATIONS
     AND/OR EXPOSURES

6,3.1  Introduction

     The subject chemical and the various media of the environment in which the
chemical exists are two interacting factors to consider in the transport and
transformation of chemicals in the environment.  Each medium has its own physi-
cal, chemical and biological properties to be considered for interaction with
other media.  These interactions are facilitated by weather (wind, sun, rain-
fall, etc.) which transports and disperses or concentrates both the chemical
and the media in varying proportions to each other.  Concentration of chemicals
in the environment are rarely at a steady state for long periods of time so
that hazard assessment must be related to changing concentrations and their
toxicity to organisms.

6.3.2  Quantity of Chemical

     The total quantity of a chemical released to the environment is important
in respect to its distribution in areas of different size (National Academy of
Sciences 1975b; National Institute of Environmental Health Sciences 1977).
The chemical may affect, or be affected differently by, the surrounding media
if it saturates or does not saturate the media.  Water runoff of a chemical
from small areas into streams may be quantitatively different from runoff from
large areas.  Since some chemicals have many uses, all of the uses and eventual
disposition of the chemical, whether alone or in a product, must be accounted
for and environmental concentrations predicted for various media.  This is
particularly true with persistent compounds since even compounds of low
volatility can be transported significantly in air or water.
                                      195

-------
6.3.3  Physical Dilutions of Chemicals Applied and Transported to
       Various Segments of the Environment

     Knowledge of the potential for physical dilution of chemicals in a given
receiving medium is important, especially if little or no degradation or trans-
fer occurs in the original medium (Hamaker 1972).  Dilutions are significant
because of the resulting concentration of chemical and the intensity of the
effect produced by it.  Because the range, area, weight, volume of soil, water,
air, and organisms are difficult to visualize, a summary of data on representa-
tive-sized samples of these media on earth are needed to estimate comparative
and representative dilution factors.  These dilution factors in water are
applicable to various ecosystems such as large and small bodies of water which
are flowing or still, or land areas of various sized drainage basins with
varying climates, soils (or lack of soils) and soil coverage.  Important
factors are air movement, evaporation rates of chemicals, and the water and
particulate content of air and their clearance rates from air due to
precipitation.

     Data on air and water turnover rates for various areas of the United
States, for example, include average annual values for water runoff on land of
various slopes, elevation, and vegetative coverage, wind speeds, and monthly
temperatures.  The representative values of these measurements need to be
worked into a skeletal chart for ballpark references to be used in calculating
and modeling environmental concentrations.

     Volatility, condensation, and freezing are changes in water which inter-
relate with the transport of water soluble compounds.  Rainfall, wind movement,
and temperature are all important weather factors which help determine trans-
port of chemicals between various media such as air-water, water-air, soil-air,
and air-soil, etc.  Data on the long distance transport of chemicals by wind or
by rivers carrying treated soil particles require knowledge of the range of
size, volume, and weight of the particulate load carried by such transportation
means.

     All chemicals may exist in gas, liquid, and solid phases in various ratios
which are temperature-, pressure-, and solubility-dependent.  These phases are
always striving for a balance in complex biological systems such as pictured
below:

                                   air (gas)
                                    	    soil, animals,

                                             ^    plants (solid)

     Within biological systems, equilibrium  is rarely if ever reached since
organisms must use energy to replace cells,  grow in size or numbers, respire,
and excrete.  Therefore, organisms demand a  constant flow in respiration,
                                      196

-------
nutritional, and excretion systems.  Major liquid flow systems in plants and
animals are supplied by water externally.  Internally, they are supplied by
water in the form of blood in animals and sap in plants.  Uptake and distribu-
tion of chemicals in biological systems, whether within or between organisms
and their external environments, involves several phases which can be simplis-
tically pictured as:

               adsorption                             absorption
         (external contact with                (internal penetration
         cell, organ, or body      •*—-	  through barriers of skin,
         surface or microorganism, 	—^  leaf, root, stomach, cell,
         animal or plant)                      etc.)
                                 flow systems
                               (bodies of water,
                               blood, sap, air)

     The tendency of a compound to adsorb onto a given surface is related to
its inherent chemical and physical properties and is reproducible under stand-
ardized test conditions (National Academy of Sciences 1975b; National Insti-
tute of Environmental Health Sciences 1977).  Chromatographic analytical
methods are used by chemists in separating and identifying compounds from a
mixture of chemicals.  Similarly, the separation of sediments and dissolved
materials in natural waters is analogous to a giant chrornatographic system in
the way in which it separates compounds by adsorption on various natural
surfaces.  The ratios of separation of a given chemical between liquids,
solids, and/or gases have been described as partition or distribution
coefficients or as adsorption coefficients between solids and liquids or
solids and gases.  The properties of chemicals which are key indicators for
estimating typical distribution coefficients are solubility (especially in
water and fat solvents), vapor pressure and the interfacing surface energy
forces, whether mutually attractive or repellent (anionic, cationic, pH,
polarity, and other electrical forces).

     The laws of physics and chemistry can be used to predict volatility, and
air-water, water-solid, and air-solid partition relationships under ideal con-
ditions.  Under practical conditions, the same compound may volatilize readily
from glass, poorly from wood, and decompose on copper surfaces.  Thus, while
surface layers of molecules may act according to these laws, adsorption and
absorption may play a large part in determining what percentage of the dose of
a chemical applied to the surface of a solid object is available for exchange
to air or water.  Once absorbed in a solid, the chemical may be "metered" back
out to the surface at a given rate dependent on the interaction of the physical
and chemical properties of the chemical and substrate and various transport
mechanisms. Aeration rates of water due to water turbulence and wind are
important in chemical dilutions.
                                      197

-------
6.3.4  Degradation

     Degradation of chemicals usually occurs at different rates in different
media and under different climatic conditions.  Half-life determinations or es-
timations are needed for representative media such as water, soil, animals,
plants, and microorganisms (Hamaker 1972).

     Degradation of chemicals occurs in the animate and inanimate world.
Animate metabolism differs from the inanimate degradation by the catalytic
action of enzymes, cellular growth, and rapid fluid turnover rates via blood or
sap and by excretion.  Metabolism takes many routes, often different in diverse
organisms.  Degradation in the inanimate world is greatly affected by climatic
factors such as air, water, sun (photolysis, temperature), and the physical
factors causing turnover cycles of air, water, and soil.  The actions of these
climatic factors have infinite variations and combinations.  Nevertheless,
these are minimum, average, and maximum values which can be calculated from
standard measurements.

     The degradation of chemicals can take place in many ways, depending on the
above variables, and reactivity of the subject and the chemical itself, such as
hydrolysis, oxidation, dehydrochlorination, dechlorination, alkylation, conden-
sation, ring fusion, ring breakage, pyrolysis, photolysis, etc.  Knowledge of
the structure and reactivity of a chemical helps to predict the major route(s)
of degradation.  The fact that degradation routes are often multiple may
complicate calculation of half-lives.

6.3.5  Estimation of Concentrations

     Estimation of concentrations of a given chemical to be applied to a given
environment at a given time must take into account the complicated combinations
of use, quantity, dispersion via physical dilution and transportation, parti-
tioning between various environmental segments as related to climate, the natu-
ral history of organisms, and the action of man.  These factors are all impor-
tant in the estimation of the fate of chemicals in the environment and expected
concentration and time of exposure of organisms to them.

     A comparison of laboratory experimental data, theoretical data, and field
experimental data must be a continuing exercise to strengthen and substantiate
useful correlations, from which to make data predictions and for use in models.

6.3.6  Useful Measurements

     The usefulness of physical-chemical properties for assessing the probable
transfer and transformation of chemicals in the environment depends on accurate
interpretation of the test data.

     Table 6-1 presents a list of parameters often found to be of value in
assessing transport and fate but it is not  intended as a check list of tests
equally useful under each law for each chemical.  Rather, the table demon-
strates the type of information that has been found to be valuable in charac-
terization of the chemical, propensity of the chemical to be transported, and
                                       198

-------
TABLE 6-1.   TYPES OF  DATA  OF VALUE  IN ASSESSING  TRANSPORT AND FATE OF  CHEMICALSt.
                                 FIFRA
                                              TSCA
                                                           CAA
                                                                       CWA
                                                                        RCRA
EIementaI
   Compos!tion

Structure

MP

BP

Subl . Point

Vapor Pressure

Sol .
   H20
   Org. Sol.

Kp

Sp Gr

Ki

PH

Hydro I

Photo I

Metab
    Terr.
    Aguatic

Leach i ng

Ads./Desorb

Oxdn.

Field Tests

Residues
    Fish
    Crop

RXN Rate

Fallout/Rainout/Deposition

Particle Size
                                                                                     +4-4-
                                              + f+4-

                                              f+4-4-

                                                 4-

                                                4-4-
                                  ++4-4-
                                               ++++

                                                + 4-4-

                                                +++
                                                                       +++
                                                                      t-H-f
                                                                      t-t-tt
           tPSAC  I973, NAS-B I975, NIEHS I977,  ACS  I97B, Goring  197?, Hamaker I972,  Freed,
            Chiou and Hague I977.

            Potable water regulators concerned  with  specific  ingredients in water and
            specifying limits (maximum allowable concentrations) in the water.  Examples
            are coliforms, certain metals,  certain chlorinated organics, chemical fate has
            not been a direct concern.
                                                  199

-------
indicate some of its reactions as related to persistence in the environment.
The number of +'s indicate the possible utility of the measurement for certain
chemicals under given conditions.  The relative utility of these properties
depends on the circumstances of use and exposure in the various environmental
segments in which the fate of the chemical must be determined.   When such
values have been obtained and ranked in comparison with other chemicals,  a
reasonable estimate of the routes and rates of transformation and fate may be
ascertained.

     Estimates of environmental concentrations and fate of chemicals in the
environment are important parameters necessary to regulate chemicals in
relation to their use and toxicity.  Presently, estimates of the transport and
partitioning of chemicals in the environment are based on physicochemical
parameters.  However, estimates of the chemical fate of the substance is more
difficult because of our lack of precise information on the interaction of
environmental forces and processes bringing about chemical alteration.

     Concentrations may be accurately assessed on empirical data such as vapor
pressure, water solubility, soil adsorption, coefficients, etc., but more
research is required to attain the same level of ability to predict the fate  of
a chemical under the varied environmental conditions.  For example, it is a
highly complex problem to attempt to estimate the mass balance of a chemical  in
water containing sediments or the runoff concentration from soil on which
chemicals are heavily or lightly adsorbed.  This arises from the multi-step
processes of chemical transfer and reactions encountered in a complex system.

6.3.7  The Role of Model Ecosphere in EPA Regulations

     The environment is a multiphase system involving air, biota, soil/sedi-
ment, and water.  It appears cost-effective to develop model systems that take
into account the interactions of chemicals within all of these compartments.
For example, a volatile chemical may be deposited on land, released to the air,
and washed by precipitation into the water where it may accumulate in aquatic
organisms; some of the latter may then enter the human food chain, and complete
the pathway of the ecosphere.  The problems of studying how the various envi-
ronmental compartments interface should be the subject of intensive research by
EPA.  There is a recognized need for establishing whether an individual com-
partment1 s model can be used as input for all other compartments of the multi-
media model or whether all compartments must be modeled as a unit.

     The relationship between mathematical models and microcosms is unclear as
to whether they are complementary or redundant.  Whatever approach is followed
requires validation by actual environmental monitoring studies.  For modeling
to be an effective tool, EPA must make statements regarding the level of pre-
cision required for the data output.  The modeler, on the other hand, must
establish amd make known the effect of the quality of the input data on the
ability of his model to achieve the EPA's goals.  Research is also needed to
delineate and describe the minimum number of critical parameters needed to do
an effective modeling exercise.  The need for meteorological data should not be
ignored.  Similarly, cognizance must be taken toward limiting or augmenting
processes, factors and responses (i.e., sensitivity analysis).
                                      200

-------
     The use of validated models for analogous compounds, thereby eliminating
the need for extensive testing on those compounds, needs to be explored.
Criteria for the applicability of these model systems needs to be developed
recognizing that a single model probably does not serve all compounds (i.e.,
organic versus inorganic, low molecular weight organics versus polymers).

     Strong cooperation is recommended between program offices of EPA in estab-
lishing the most suitable data requirements for developing models to serve
their respective program needs keeping in mind the need to strive for uniform-
ity of approach. It must be recognized that the only output that can be
expected from any of these models is the expected environmental concentrations
(EEC) and that coupling with toxicological data is needed for risk analysis.
Simplified models based on a limited number of parameters are needed to quickly
screen for problem chemicals.  The identification of trigger mechanisms and
discipline of criteria to short-circuit the collection of unnecessary data
should be an integral part of the research to establish these simplified models
or screening systems.
6.4  THE EXTENT TO WHICH THE SAME CHEMICAL FATE PARAMETERS CAN BE USED BY
     DIFFERENT EPA OFFICES

     One of the major goals of the EPA is to prevent chemical pollutants from
causing adverse effects on man and the environment (Maugh 1978).  Assessing
the potential hazard of a compound requires information about toxicity and
predicted exposure levels of the substance to non-target organisms.  Environ-
mental fate and transport tests currently under development will be used to
predict what will happen to a chemical once it enters the environment«  These
tests will ultimately allow the prediction of exposure levels based on
environmental mobility and degradation routes.

     The Toxic Substance Control Act (TSCA) and the Federal Insecticide,
Fungicide, and Rodenticide Act (FIFRA) address the problem of environmental
fate and transport most specifically.  Both acts have requirements and sug-
gested procedures for generating environmental fate and transport information.
The Clean Air Act (CAA), Clean Water Act (CWA), and Resource Conservation and
Recovery Act (RCRA) do not require such data but must be concerned with pre-
dicting what might happen to a harmful chemical entering the environment
instead of only assessing the consequences after the fact.  Therefore, chemi-
cal fate and transport parameters should be of universal concern.  Although
some EPA offices may require more detailed information (more intensive testing)
than others, they all share a common goal of accurate estimates of exposure
levels from transport and fate data (Culleton 1978).

     The different EPA offices should be encouraged to use the same chemical
fate and transport parameters wherever possible.  An EPA-wide testing scheme
may be the most efficient approach for determining these parameters.  Basic
chemical information may be obtained from initial screening tests.  At this
initial level, standardized test methodologies should be used to provide a
uniform set of information for all compounds.  The screening tests should
                                      201

-------
evolve through use by all offices.  This evolutionary process,  however,  needs
establishment of guidelines and scientific judgment on what level of sophisti-
cation is needed at the screening level to satisfy broad needs.

     As more advanced information is required, standardized tests may not be
possible because each act may require a specific type of information and all
tests may not be equally applicable to all compounds.  It may be possible to
eventually evolve an EPA-wide progressive testing scheme based on informational
requirements.  Such a scheme would allow each office to specify the amount and
type of information required to satisfy its needs in predicting environmental
fate and transport.  For example, in dealing with solid waste disposal,  RCRA
may require extensive characterization of the transport and fate of a chemical
in soils, while CAA may involve little, if any, soil testing.  The same  para-
meters and screening tests for soils should be applicable to both FIFRA  and
RCRA.  Although one office may require more detailed transport and fate  para-
meters than another, the same testing schemes and methodologies should be
applicable to both.  The same basic information should be required to estimate
the fate and transport of a chemical in water regardless of which EPA act or
office requires the information.  A generalized tier scheme would allow  chemi-
cals to progress to the more detailed tests required by the more stringent act.

     There are a number of advantages to be gained by adopting one set of
screening tests to estimate environmental fate and transport parameters  of
chemicals.  The EPA could more efficiently review the tests on a periodic
basis to insure that they are based on the best available methodology.
Each EPA office would not have to develop and-validate its own tests. Com-
panies could more efficiently generate transport and fate parameters if  they
utilized common testing procedures at the initial screening level.  The  bench-
mark approach to evaluating test data would be easier to implement if a  basic
set of fate and transport data was available  for a number of chemicals.   The
most significant environmental fate and transport routes can be easily identi-
fied from a complete base set of data.

     The parameters required by each office should be based on the use,  toxi-
city, manufacture, and volume of a new product.  Products with low toxicity,
intended for limited use, and with a low production volume, should not be re-
quired to undergo a full battery of tests.  The cost of generating a complete
and detailed set of transport and fate parameters would severely limit the
development of new, low volume specialty chemicals.  Sales of a highly
effective, limited use product may not be sufficient to cover the cost of
testing.
6.5  QUALITY ASSURANCE CONCEPTS

6.5.1  Introduction

     Quality Assurance (QA) and Quality Control (QC) are terms frequently used
as synonyms.  It is proposed, for purposes of this report, that they be cast as
                                      202

-------
the overall and specific components of systems to provide a high degree of con-
fidence in test and experimental data (Kanzelmeyer 1977).  Quality Assurance
is a total system organized and designed to provide for accurate and precise
results.  Quality Control is specific action related to components of the
system providing checks, controls, replicates, standards and calibrations
suitable for statistical estimates of confidence in data elements.

     Projecting guides will be used by a number of different facilities en-
gaged in a variety of projects while QA Guidelines should cover all basic fac-
tors.  However, the guidelines should be broad and flexible.  The goal is to
produce good, valid results.  The means to the goal should not be confused
with, or substituted for the end result or primary purpose:  valid, accurate
data.  No one specific instrument, procedure, or requirement should be mandated
unless it is the sine qua non for high quality results.

     While the following QA components are most frequently applied to analyti-
cal chemistry, the basic elements are applicable to other areas and tests.
Overall, the basis for most environmental work lay in analytical tests; thus,
the emphasis on analysis.  For example, in physicochemical parameters such as
solubility, vapor pressure, Koc, hydrolysis, photolysis and leaching the com-
ponents in 6.5.2.1 through 6.5.2.5, and 6.5.2.7 are applicable with specific
modifications to meet the specific parameter.

6.5.2  Quality Assurance Components - For Analyses

6.5.2.1  Personnel Requirements

     In any quality assurance program, selection of personnel is a most impor-
tant function.  Administrators, directors, and supervisors should evaluate
personnel on a continuing basis to determine whether they are thoroughly
trained to perform the duties of the job they are assigned.  The evaluation
should include academic achievement, work experience, on-the-job training,
workshops, and special courses.  Certain levels of proficiency should be main-
tained through training programs for all employees due to the importance that
their initial and continuing levels of competence enable them to perform at an
acceptable level.  However, overspecification and restrictive directions as to
details must be avoided.

6.5.2.2  Procedures

     All technical methods and laboratory procedures should be well-written and
updated.  A procedure manual should be maintained in the area where the testing
is performed.  As a general outline, the following are normal sections of a
procedure.

     1.  Purpose

     Principle use to include type of sample, purpose for testing, or an
     experiment and reason for procedure.
                                      203

-------
2. Summary

Include basic technique, range in concentration for which test is
applicable, problems, interference, source of error and reference to a
standard method which is equal or equivalent.

3. Safety

Include precautions for the particular method.

4. Apparatus

Listing to include type of instruments, wavelength specifications, ranges
or any parameters critical or specific to the method being described.
Illustrations are sometimes easier and clearer than lengthy descriptions.

5. Reagents, Standards, Media

Listing and/or preparation to include brand names where critical to
procedure, purities, critical weights, nearest measurement units to which
standards must be weighed or measured, concentrations, etc.

6. Calibrations and Standardization

Procedures to include graphs or tables if needed, temperature checks,
limits of method if applicable.

7. Test Procedure

Detailed step-by-step procedure should be written exactly as performed;
quality control measures can be included here or in a separate section.
However, allowance must be made for exercise of scientific judgment and
non-essential specifics avoided.

8. Calculations

All formulas must be listed as used and all variables described or
defined.

9. Interpretation

Include critical values, significance of values, follow-up indicated by
results.

10. Precision and Accuracy

Information regarding precision and accuracy of the method described
should be  included, i.e., published values determined by past experiences,
internal experiments and collaborative testing.  The source of the
information should be noted.
                                 204

-------
     11. References

     Include sources of background information on the method itself, and any
information that substantiates use of the method.

     Although the above list appears all-inclusive, it serves only as a guide
in preparing written procedures.  Not all items are applicable to all proce-
 'ures and t:  >y may be combined or rearranged as necessary.  The procedure
nanual sho' 1  t be reviewed at least annually or at any time method changes or
 •rv'Ti-  -,    ur.  All approved changes or revisions must be documented.

         KI,--   ;nts and Standards

                        altr.% pure grades.  Thf3 American Chemical Society Committee on
         j    ~agents has established criteria for certain chemicals which they
       AH'  ^r.ade or 'ACS) grade.  These chemicals are of very high purity and
should IK used in al  analyses in the laboratory, unless otherwise stated.  The
labels on the container should be checked and the contents examined to verify
that :.'"e purity of the reagents meets the needs of the particular method
i
     I'- \r     must always be ">repo.red and standardized with special care
a gains  rt    -le primary standards.  They must be restandardized or prepared
fresh rs n    ,?ary to retain their stability.  All reagents, standards and
  •neral chen -^ciLs should be labeled as to date of receipt, date opened, and
rtX^ : :  ^ i or   =te.

     s"ollow   j are a few guidelines in the preservation and storage of reagents
  ,d sto... lard  :

     1. Aqueous solutions should be stored in tightly closed glass or plastic
        bottles.

     2. Colored aqueous solutions which are light sensitive should be kept in
        dark brown or amber bottles.

     3. If reagents or standards require refrigeration, make labels or
        indicate on bottle that sample is to be refrigerated.

     4. NEVER introduce pipettes into reagent or standard bottle.

     5. Reagents and standards perform best when freshly prepared.  Laboratory
        preparation of reagents and standards should indicate on the label the
        date of preparation and the initials of the person who prepared it.  In
        some cases, it is a good idea to include the components of the
        solutions.  Whenever special precautions are necessary, they should be
        mentioned under the reagent preparation and should be included on the
        label.
                                      205

-------
     Water quality or purity is another parameter that should be given con-
siderable attention in any analytical work.  Distilled or demineralized water
is used in the laboratory for dilution purposes, preparation of reagents,  and
rinsing glassware.  High purity water has been defined as water that has been
distilled and/or deionized so that it will have a specific resistance of
50,000 ohms (2.0 micromhos conductivity) or greater.  Although ordinary dis-
tilled water is adequate for many analyses, specific methodology may require a
further refinement or conditioning of the purified water.

     Laboratory reagent water should be checked periodically for resistivity,
trace metals, bactericidal properties, or any parameters which may influence
test results.  The procedures for the quality checks should be documented.

6.5.2.4  Laboratory Apparatus and Instruments

     Familiarity with the equipment is necessary in order to produce valid data
and involves not only the capabilities but also the limitations of the test
equipment.  Much of this information is received from the vendor when equipment
is purchased.  However, the analyst has the primary responsibility for main-
taining the equipment.  A preventive maintenance program should be established
to insure peak performance of all equipment.  This preventive maintenance
program should be documented by keeping a ledger and/or visible posting on
the equipment, the dates of preventive maintenance, as well as the due dates.

     Each piece of measuring or test equipment should be calibrated on a sched-
uled basis using appropriate standards, as well as approved published prac-
tices.  Regardless of the type of calibration or materials used, an effective
quality assurance program requires accuracy levels of those materials that are
consistant with the method of analysis.  All of this information should be
documented.

6.5.2.5  Sample Collection and Preservation

     The quality of the sample in any experiment will control the results of
the test.  It is unfortunate that engineering miracles in automation and newer
analytical methodology are generally unable to differentiate between a properly
collected sample and one which is improperly collected.  Therefore, it is
essential that guidelines be instituted for the collection, preservation and
transport of samples to the laboratory.  The guidelines should include the
following:

     1. Type of specimen needed

     2. Minimum sample size

     3. Proper method or technique of collection including specific storage
        instructions

     4. Container to be used - with amount and type of preservative, if any

     5. Holding time.
                                      206

-------
A rejection policy should also be established specifically for those cases
which do not meet the above criteria.

6.5.2.6  Analytical Quality Control

     Good facilities, competent personnel, an up-to-date procedure manual, high
purity reagents and standards, and the best in calibrated equipment do not
automatically provide valid and reliable data.  The basic requirements in any
quality assurance program have to include the above conditions, however, the
main ingredient used in determining validity and accuracy of results is analy-
tical quality control.

     Analytical quality control incorporates not only proved and acceptable
methods, but analysis of a control sample (internal quality control or statis-
tical quality control) and confirmation of the ability of a laboratory to pro-
duce acceptable results by requiring analysis of unknown reference samples
(external quality control, proficiency testing, or laboratory evaluation).

     Since most analytical errors in the laboratory are classified as deter-
minate (traced to a particular problem) or indeterminate (statistical errors),
it is possible to assess statistical data graphically by use of control charts.
In order to monitor daily operations for accuracy and precision, data must be
accumulated.  Basically, the procedure is as follows:

     1. Document all control values for a relevant period of time.

     2. Determine the mean and standard deviation.

     3. Reject values falling outside of ± 3 standard deviations.

     4. Recalculate the mean and standard deviation.

     5. Calculate the mean ± 2 standard deviations.

     6. Establish the range of acceptable values.

Particular attention has to be given to the understanding and interpretation of
these data and an action policy should be developed to determine the course of
action taken when values are out of control.  Some form of comparison or tabu-
lation should be kept to evaluate results and detect trends.

     Graphic display of control values provides a visible means of determining
control limits, but it is not the most desirable method of quality control.
One of the most efficient means of quality controlling analytical procedures is
computerization.  Many existing information processing systems are capable of
producing such statistical parameters as frequency, range, mean, standard de-
viation, and coefficient of variation.  Computers are able to store vast
amounts of information with relative ease of retrieval.  Many are designed to
document or display control values on a regular basis and provide useful infor-
mation when values are out of range or show a particular shift or trend.  Comp-
uterization with automated and semi-automated equipment can effectively
                                      207

-------
document and assess those statistical parameters necessary for the evaluation
of the testing systems.  Even though the computers will perform much of the
data analysis and print results in report form, this information has to be
interpreted by laboratory personnel.

     An external system of audits and checks is another aspect of quality con-
trol involving both real and simulated specimens.  Proficiency testing can
assist the laboratory in determining whether the values it reports are compar-
able to those of other labs when using the same or similar methods.  In many
laboratories, in-house values are subject to subtle changes from time to time
and these changes are not recognizable without the use of some external source
of reference.  Well-run proficiency testing programs serve to complement the
daily quality control program and also provide motivation for better laboratory
performance.  Proficiency testing samples should not be treated as special
tests, but should be handled as a part of the routine work load.  If the
laboratory considers its routine performance so poor as to require special
handling, then it has more problems than could be corrected by a proficiency
testing program.

     There are a number of areas in the laboratory where the final decision of
an analysis or series of analyses is based on the education and experience of
the analyst and not on a quantitative measurement.  Because of the subjective
nature of the work, sound scientific judgment, consultation with colleagues and
supervisors, and a modification of techniques and procedures may often be re-
quired.  The final results in many instances are based on comparisons.  Condi-
tions leading to the final results should be controlled and documented. The
quality control aspects of non-quantitative data justify careful observation
and proper documentation of observations in order to support conclusions.

6.5.2.7  Definitions (Mandel 1977)

     Quality Assurance - Program designed to assure required accuracy and
precision of results.  Included are selection of proper methods, tests or oper-
ations, quality control, selection of limits, evaluation of data, and qualifi-
cations and training of personnel.

     Quality Control - Specific actions required to provide data for the
Quality Assurance program.  Included are standardizations, calibration, repli-
cates and check samples as part of a specific test procedure or specific pro-
cedures for a given area or function which supports one or more test methods.

     Accuracy - The closeness of a test result to the expected value.

     Precision - Reproducibility of test results.

     Standard Deviation - A measure of the average dispersion from the arith-
metic mean value:
                          SD =        E(X  - X)2
                                         N - 1
                                      208

-------
                          Where:

                           I = "Sum Of"

                          X-| = Value of Single Determination

                          X  = Arithmetic Mean of N Determinations

                          N  = Number of Determinations

     Arithmetic Mean - The sum of all observations divided by the number of
observations:

                            _   Sum of Observations
                            X = 	

                                No. of Observations

     Coefficient of Variation - Standard deviation expressed in terms of the
percentage of the mean:
                                CV = -    X 100
                                     Mean

     Range - The interval between the highest and lowest figures in a related
series of data.

     Frequency - Magnitude of occurence.

     Determinate Errors - Those errors which originate from sources such as
malfunctioning instruments, incorrect operations, inappropriate temperature
setting, glassware, contaminated chemicals, and human errors.

     Indeterminate Errors - Those errors which are inherent in any testing
system and are not traceable to any particular problem.

     Primary Standard - A substance whose concentration or purity is known -
usually 99% or better.

     Linear Regression - Statistical technique used to determine the relation-
ship between two variables, one of which is known with a certain degree of
accuracy and the other is predicted by fitting the variables to a line equa-
tion rather than plotting and visually estimating that equation.
6.6  CONCEPT OF BENCHMARK CHEMICALS

     Only in recent years have serious attempts been made to correlate labora-
tory observations on the fate and transport of chemical substances to their
actual behavior in the environment.  In part, the delay in undertaking such
                                      209

-------
studies was due to the complexity and variability inherent in the biochemical
systems under investigation and a lack of standard methodology that could yield
a more unified, scientific approach.  The lack of a unified approach has limit-
ed the ability of environmental chemists and regulatory agencies to confidently
predict and correlate the environmental fate and transport of one chemical
substance with that of another.  In the case of pesticides, which are used as
examples since they have been more thoroughly studied, meaningful environmental
chemistry should be conducted to provide relevant information on their behavior
and significant dissimilation products in the environment.  This will allow
judgment to be made of their potential impact on the environment when they are
to be used on a large scale.

     Prior to the introduction of an experimental pesticide on a large scale,
it is suggested that a judgment on its potential fate and transport in the
environment be constructed by comparing results generated by suitable pro-
tocols for the experimental pesticide with those obtained for a "benchmark"
pesticide chemical currently or previously in large scale use.  It is under-
stood that the environmental behavior of these benchmark chemicals has been
observed or measured in numerous laboratory and field studies and it has been
judged acceptable or unacceptable relative to specific types of end uses.  Such
an approach has been briefly described by Hamaker (1972) and Goring (1972) for
pesticide degradation in soil and for all other aspects of pesticide behavior
in the environment, respectively.

     Using the example of an experimental pesticidal chemical, the sequence of
events in the "benchmark" chemical concept may be described in the following:

     1. Define the most important properties of the pesticide that determine
        its behavior in the environment, e.g. volatility, Kp, adsorption,
        reaction rates.

     2. Develop laboratory methods to accurately and precisely measure these
        properties.

     3. Measure these properties for the major "benchmark" chemicals currently
        being used or for those "benchmark" chemicals that had been used on a
        very large scale for many years and for which a large body of
        information is available.

     4. Establish the relationship between the laboratory  (greenhouse and
        small-scale field studies included when available) measurements and
        field behavior for the "benchmark" pesticides.

     5. The potential environmental fate and transport of an experimental
        pesticide chemical could then be predicted by comparing the data with
        those accumulated for  "benchmark" chemicals.  This may be done by the
        use of mathematical models.

The utility of this "benchmark" approach is that it provides a rational matrix
of information on the environmental behavior of pesticides that increases in
scope, depth and usefulness as each new experimental pesticide is incorporated
                                      210

-------
into its proper place in the matrix.  This concept of "benchmark" chemicals may
be fine-tuned relative to its structure-activity and predictive value as more
information becomes available on the various aspects of environmental fate and
transport for specific chemicals.

     A list of suggested "benchmark" pesticides is given in Table 6-2.   Selec-
tion of these pesticides was based both on wide usage over a long period of
time and on an overall range of their properties that encompass the extremes of
behavior likely to be encountered.  In a few cases, chemicals were selected for
their unique physicochemical properties or their environmental behavior
patterns.  It is hoped that additional "benchmark" pesticides will be added to
this list as the need arises.  The final list should be representative of other
pesticide classes and uses such as fumigants, aquatic pesticides, and
repellents.  This concept is not limited to a given class of chemicals such as
pesticides but may also include all environmental chemicals including the
naturally-occurring compounds.

     In support of the "benchmark" concept, a suggested classification matrix
for pesticides is provided in Table 6-3.  It is important not to consider this
matrix as all-encompassing but merely as illustrative of the kind of experi-
mental data which should be considered in predicting the fate and transport of
chemicals in the environment.  For a more complete understanding of the
environmental fate and transport of chemicals, one should add data requirements
to improve the predictability of the benchmark concept using various
tiers of testing as required by appropriate criteria.  For example, the
described experimental parameters are not applicable to cationic chemicals.

     Although this section does not include the effect of chemical substances
on the target and non-target organisms, the benchmark chemical concept may also
incorporate toxicological parameters to be used as a tool to predict potential
structure-toxicity hazards.

     It is hoped that continual improvements in the data presented in tables
6-2 and 6-3 will be achieved by the promotion and constant validation of this
approach via data generated by all scientists, e.g., industrial registration
data, data generated by regulatory agencies, and published literature reviews
and information.  In order to fill the data gaps and to improve the usefulness
of this approach, additional research should be conducted (research grants,
contracts by government agencies) on chemicals which may have little or no
economic significance.

6.6.1  Utilization of the Benchmark Approach

     The following elaboration of the process of developing the benchmark con-
cepts and data indicate activities that reflect the utility of the approach:

     1. Determine the most important properties of chemicals substances that
        best determine their environmental fate and transport.

     2. Determine which of these properties can be measured by conducting
        laboratory and greenhouse experiments.
                                      211

-------
        TABLE 6-2 SUGGESTED BENCHMARK CHEMICALS
Chemical Class
Pesticides
Aliphatic acids

Aniline-based

Benzoic Acids

Carbamates

Dinitroanilines

Fumigants

Organochlorines

Organoraetallics

Organophosphs.es



Phenols

Phenoxyalkanoic acids

Phthalamides

Picolinic acids

Thiocarbamates

Triazines

Ureas

Miscellaneous
Dalapon

Chlorpropham, Alachlor

Dicamba, Chloramben

Carbaryl, Carbofuran, Methomyl

Trifluralin

DBCP, 1,3-D, Methyl bromide

DDT, Lindane, Dieldrin

MSMA

Parathion, Malathion, Diazinon,
   Methyl Parathion, Phorate,
   Fensulfothion, Chlorpyrifos

Dinoseb

2,4-D, 2,4,5-T

Captan

Picloram

EPTC

Atrazine

Diuron, Fluometuron

Dichlobenil, Benomyl
 Common Names
                          212

-------
c    —    —-
o
V
                             c
                             c
                                                         c
                                                          I
                                                         o
                                                          i
                              I

                            N^
                             C
H !
£-j
t \
a:
*n
u ! x
D i £*
w ffi
E-> >
W
w
g
D «
to j
K
1
• 1
n !
i !
vO 1
i
M '
3 o
ffl 1 . —
F^ i -t-
^ I _ (D •-
ID « 0 t-
j L' - - C

o o >- o
O — .C (D
— E D- l_
O d) (D
— S2 L JT
OD O O O


X
A








^_,
E
ex
CL
4-
C >•
 —
Q- -Q
TD —
C O
— in

* — i_
— (D
O +-
01 ID

<
1
C


1
c
A









^
x'
£
^
x.^
0)
L.

in
(B
L
Q.

L.
O
Q.
(D

CD
CN



C
O
A








CN

V
^
in
>^
ID
TJ


.'— ^-^
in X
>- CL

O 4-
1- ID

I

O
fx:



O
c
A








IN
V,
+
^
01
>^
(D
-o ^^
>-- (D
•c
— L'
10 —
>~ -^

o c
4- Q.
O
JT
CL

Q
C
i —


C
A












^
O

(U

L
. —
<:

u
 —

c
^
1
1

                                213

-------
     3.  Collect and review all available research data  (literature  review,  data
        generated by the EPA and data,  if available,  in EPA  files)  on the
        environmental fate and transport of  chemicals that have been
        introduced, used,  or are naturally present on a significantly large
        scale over a long period of time. This  part of  research should  collate
        the laboratory (greenhouse)  data and the  field  data  separately.

     4.  Determine if existing laboratory or  greenhouse  methods accurately and
        precisely indicate the environmental fate and transport parameters.
        Additional research may be necessary to improve the  experimental
        methodology.

     5.  Select benchmark chemicals based on  their structure-reactivity, wide
        and long-term usage, and for unique  characteristics.

     6.  Establish the relationship between the  laboratory (greenhouse)
        measurements and the environmental behavior for the  selected benchmark
        chemicals.

     7.  Using the previously obtained structure-reactivity relationships,
        calculate the expected environmental fate and transport of  a new
        chemical substance under investigation  and then,  through  appropriate
        regulatory treatment, propose its "expected environmental profile",
        (i.e. first or second tier assessment of  expected environmental fate
        and transport).

     8.  The true validation of the structure-reactivity benchmark chemicals
        concept is realized when the new chemical is  subjected to field testing
        under actual use conditions.

     9.  These new field data on the new chemical  should then be fed back to
        confirm the predictive value of the  benchmark chemicals concept.

    10.  Benchmark chemical concepts may be extended to  structure-toxicity
        relationships where the necessary data  are available.

6.6.2  Limitations of Benchmark Chemical Concept

     The structure-reactivity concept emphasizes  pesticides  and within  that
class of compounds (based on use) cationic compounds  are  exceptions because of
their unique chemistry.  Because of this exception and  because the  concept  has
not been fully validated, there is reason to believe  that it may  not be useful
in every case.  Therefore, it is mandatory to make certain that scientific
judgment prevails at all levels of assessment (environmental fate and/or risk
assessments) while dealing with a new compound.

     The benchmark concept gives primary considerations to pesticidal
chemicals.  It should be noted that due to differences  in the properties and
quantity of chemicals regulated under different laws, the utility and
application of the benchmark concept will differ  under  the various  acts such as
the Clean Air Act, Clean Water Act, and Resource  Conservation of  Recovery  Act,
as compared to that for TSCA and FIFRA.

                                      214

-------
     In attempting to predict the environmental behavior of a new compound by
the use of the benchmark concept, one must use a probability approach coupled
with mathematical modeling.  The benchmark concept and mathematical modeling,
until further refined and validated, cannot be expected to precisely predict
the environmental fate of an experimental chemical.  Instead, it serves to pro-
vide a profile of transport and fate behavior and to focus attention on more
important and pertinent tests that must be conducted to more accurately eluci-
date the fate and transport of a chemical in the environment.
6.7 DISCUSSION

     There are a number of federal laws dealing with pollution that involve
regulation of chemicals (Federal Water Pollution Control Act, Clean Air Act,
Resource Conservation and Recovery Act, Federal Insecticide, Fungicide and
Rodenticide Act, and Toxic Substances Control Act).  With respect to chemicals,
the basic objective in each of these acts is to prevent any significant adverse
effect on man and the environment which requires attention be given to manu-
facture, transport, use, and disposal of chemicals.  Once in the environment,
the chemical may be transported by natural processes and may directly contact
man or other biota.  The amount and distance transported under given conditions
will be dependent upon the persistence of the chemical or its fate in the en-
vironment.  Therefore, regulation of chemicals to achieve the objective of
human and environmental protection requires adequate knowledge of transport and
fate of the chemicals.  This section of the report is concerned with research
needs on transport and fate of chemicals to allow effective and rational im-
plementation of laws.  In addition, there is need for concern for development
of testing protocols, quality assurance of tests, and appropriate application
of tests to acquire relevant information.

     Though a number of laws are concerned with the regulation of chemicals,
the extent of this concern and requirement varies markedly from law to law.
The Toxic Substances Control Act and the Federal Insecticide, Fungicide, and
Rodenticide Act deal with chemicals from manufacture to ultimate disposal while
the Drinking Water Act and others have a more narrow spectrum of interest.
However, to the extent that regulation is required under an act, basic informa-
tion on transport and fate is necessary for proper implementation of the regu-
lations.  Fundamental information that indicates the propensity of the chemi-
cal for transport and persistence should be identified by appropriate research.
The present development of the field of chemodynamics indicates that such
physicochemical parameters as solubility, partition and distribution coeffi-
cients, and certain fundamental reactions may serve as a basis for predicting,
at least to a first approximation, the probable transport, fate, and environ-
mental distribution of chemicals.

     The large number of chemicals with which the laws must be concerned pre-
cludes the detailed individual attention and data collection that would come
from field trials and monitoring data.  Therefore, chemodynamics and the
"benchmark chemicals" approach utilizing laboratory measurements are needed and
                                      215

-------
appear feasible (Goring 1972).  Additional  research  should  be  conducted  by EPA
to ascertain whether there may be broader application of this  approach to  give
a first level assessment of transport and fate.

     In order to prevent significant adverse effect  on man  and the  environment,
the same amount of information will not be  required  on all  chemicals.  By
virtue of the properties of t.;e cht;i- ..sal, methc " of  use, and other  considera-
tions, it will be possible to identity low  hazard compounds .   The  most  econo-
iT_cal coi:rse of action would be to develop  a multi-tier system of data
requirements.  Tttu,,, the first tier of requirements  may be  for basic physical-
chemical data obtained in laboratory studies with appropriate  interpretation
and predic  ons derived from the dav*.  Using  t e benchmark chemical techni-
 'ue  estim,.  3 of prob^ole biological   rtivlty jnd persistence may  indicate
  ••re elabor^ ;e biological testing :. s    es^ary.  Deper; :ing  upon the physical
as  I bir logical properties of the c>   -ou  *,  the amount ~o be produced, and  the
metho- of use, rcore elabor ~e ret,,, • enu     wou] 1 be  established for the  sue-
 -essiv  tiers of testing.  The ult'  t° - ^al would be the development of a
 . atheistic \~L model that would give   s  r._   ~_e     -na' -s of  the transport,
 .stribution, and fate of such a ciiem: al.  Ho*; er,  -urren*- ma"Her',atical
..-jde"  have not been developed that permit  sue  , ssesoi.ents an  ^uch models
ohouiv- not be expected to be forthcoming  in the future without further
resear \.

     In <_ ^vising appropriate regulocoi  a-jtion, the  abilit-  •   identi  '  and
   ~ure tae subject compound i'  essenu_^...  Therefore, t'-    velopment of
ar,a.j.ytical methods of appropriate sensitivity, precision .   reliability is
Important.  The same may also be said of  the various other  tests to determine
the physical and biological properties of the  compound.  It is important,
therefore, that such tests and analytical methods be developed and  a suitable
protocol of quality assurance be a part of  the program.
6.8  RECOMMENDATIONS

     A vigorous program of research  on  transport  and fate  is  needed to regulate
chemicals and prevent significant  adverse  impact  on  man  and the environment
while allowing for the orderly and expeditious  development of needed chemicals.
The development of the appropriate basic information for interpreting and
prediciting the transport and fate of chemicals is needed  for implementation of
the current and future laws  and  regulations.

     There are also a number of  natural and anthropogenic  chemicals in the
environment.  These may mask or  interfere  with  the detection  or monitoring of
some chemical of  future interest.  The  background levels of a number of
chemicals should  be identified and quantitated.  This information is for the
purpose of knowing what is in the  environment,  the possib'e significance of *'ie
presence of a given chemical, and  to avoid confusion of  this  chemical with so.ne
similar but new product.
                                       216

-------
     Research should be supported to develop further basic tests to provide
improved scientific information on transport and transformation and expand the
basis on which predictions may be made.

     Appropriate generic test protocols as well as benchmark chemical approaches
should be developed and implemented in order to deal with the large number of
chemicals covered by the various acts.

     The role of biota including man should be recognized as a factor in
transport and fate and cognizance be taken of the role of the interaction o
physiological, genetic, and population composition changes in the fate of t; *
chemicals.

     Appropriate tier test protocol development should be a priority activity
to facilitate assessment of chemical hazards.  The tier test protocols should
take cognizance of the amount of chemical, toxicity, the manner of use, and
propensity for transport.  Such tier testing would allow the most attention to
be given to the materials posing the greatest hazard to man and the environment.

     Where mathematical models are to be used for a basis of regulation,
research should be undertaken to develop more accurate and reliable holistic
models.  These models should take into account the multimedia aspects of
transport and fate and should involve "sensitivity analyses" and temporal and
spatial relationships.  Valid multimedia exposure models or linked single
compartment models are needed to assist in rapid, consistent exposure
assessment of new chemicals.

     Methodologies are needed for screening new chemicals for risk assessments
based on environmental transport and fate considerations.  EPA will have a
limited amount of time to assess the risk posed by proposed use, manufacturing,
processing, distribution, and disposal activities and needs decision criteria
by which it can focus resources on potentially high risk chemicals.

     A qualitative ranking or profile of the relative importance of the various
environmental transport and persistence parameters is needed.  This ranking
should be based on monitoring data interpreted in comparison with results of
laboratory environmental fate studies and microcosm studies.  Testing and
risk assessment resources should focus on the most important environmental fate
parameters, emphasizing the more minor parameters only where structure-
reactivity relationships and other factors indicate the need.

     Decision criteria should be developed for testing and risk assessment.
Such testing criteria will indicate when testing is necessary and what types of
tests should be performed.  These criteria may be based on results from simple
tests, structure-reactivity relationships, etc.  The risk assessment criteria
referred to are those to be used at the end of the risk assessment process
after the screening phase.

     Development of more cost effective screening tests and more research on
structure-reactivity relationships for new chemicals needs to be undertaken.
The economics of new chemical development is such that testing costs should be
kept minimal.
                                      217

-------
     EPA requirements in regulations for test methods and procedures should be
broad and flexible.  The goal is to produce good, valid results.  The means
should not be confused with the goal or substituted for valid, accurate data
suitable for its intended purpose or use.  No one specific instrument,
procedure, or test related requirement should be mandated unless it is the sine
qua non for high quality results.

     There is an urgent need for research and development of analytical and
test methods which are suitable and practical for routine use in monitoring and
control functions for compliance with regulatory aspects.  The goal should be
availability of accepted, practical valid methods at the time maximum
contaminant levels or similar requirements are proposed.  Tin effort is also
needed to establish a reliability assessment system to consolidate the
variability of all test data, and other data into a reliability estimate of
conclusions or decisions such as the Department of Defense Reliability
Programs.

        The Regulatory Group participants are listed in Table 6-4.
                                       218

-------
                   TABLE 6-4.  REGULATORY GROUP PARTICIPANTS
Non-Agency Participants

Kris Bandal
3M Company
St. Paul, MN

Barry Bochner
Atlantic Chemical Corporation
Nutley, NJ

Virgil Freed (Chairman)
Department of Chemistry
Oregon State University
Corvallis, OR

Robert Jamieson
Proctor and Gamble Company
Cincinnati, OH

Gene Kenaga
Dow Chemical USA
Midland, MI

Paul Michael
Monsanto Chemical Company
St. Louis, MO

Al Tiedman
Consolidated Laboratories
State of Virginia
Richmond, VA
EPA Representatives

John Bachman
Office of Air Quality Planning and
  Standards/
Office of Air, Noise, and Radiation
Research Triangle Park, NC

Robert Brink (Lead)
Office of Toxic Evaluation/
Office of Toxic Substances
Washington, D.C.

Stuart Cohen
Office of Chemical Control/
Office of Toxic Substances
Washington, D.C.

Mark Seagal
Office of Water Planning and Standards/
Office of Water and Waste Management
Washington, D.C.

Gunter Zweig
Office of Pesticide Programs/
Office of Toxic Substances
Washington, D.C.
                                      219

-------
                                  REFERENCES
American Chemical Society.  Cleaning Our Environment — The Cl"  mical Basis of
     Action, 2nd edition.  American Chemical Society, Washington,  D.C.,  1978.

Culleton, B. J.  Toxic Substances Legislation:  How Well Are Laws B. Ing
     Implemented?  Science, 201:1198, 1978.

Freed, V. H., C. T. Chiou and R. Haque.  Chemodynamics:  Transport and Behavior
     of Chemicals in the Environment — A Problem in Environmental Health.
     Environmental Health Perspectives, 20:55, 1977.

Goring, C. A. J.  Agricultural Chemicals in the Environment:  A Quantitative
     Viewpoint.  In:  Organic Chemicals in the Soil Environment, C. A. J.
     Goring and J. W. Hamaker, eds., pp. 253-340.  Marcel Dekker, New Yorks
     1972.

Hamaker, J. W.  Decomposition:  Quantitative Aspects.  In:  Organic Chemicals
     in the Soil Environment, C. A. J. Goring and J. W. Hamaker, eds., pp.
     253-340.  Marcel Dekker, New York, 1972.

Kanzelmeyer, J. H.  Quality Control for Analytical Methods.  A.S.T.M.
     Standardization News, 5:20, 1977.

Mandel, J.  Statistics and Standard Reference Materials.  A.S.T.M. Standardiza-
     tion News, 5:10, 1977.

Maugh, T. H., II.  Chemical Carcinogens:  The Scientific Basis for
     Regulations. Science, 209:1200, 1978.

National Academy of Sciences.  Principles for Evaluating Chemicals in the
     Environment.  National Academy of Sciences, Washington, D.C., 1975a.

National Academy of Sciences.  Decision Making for Regulating the Chemicals in
     the Environment.  Environmental Studies Board, National Academy of
     Sciences, Washington, D.C., 1975b.
                                      220

-------
National Institute of Environmental  Health Sciences.   Human Health and the
     Environment — Some Research Needs,  2nd Edition.   Public Health Service,
     National Institutes of Health,  National Institute of Environmental Health
     Sciences, U.S. Department  of Health,  Education and Welfare,  1977.

President's Science Advisory  Committee.   Chemicals and Health.  Report of the
     Panel on Chemicals and Health of  the President's  Science Advisory
     Committee.  National  Science Foundation,  1973.
                                 REGULATORY ACTS



Clean Air Act,  1970

Clean Water Act,  1977

Federal Environmental Pesticide  Control  Act,  1972  -  Amended 1978

Federal Insecticide, Fungicide and  Rodenticide Act,  1972

Federal Water Pollution Control  Act,  1972

Occupational Safety and Health Act,  1970

Resource Conservation and Recovery  Act,  1976

Safe Drinking Water Act, 1974

Toxic Substances  Control Act, 1976

  *U.S. GOVERNMENT PRINTING OFFICE: 1982/559-092/3383
                                     221

-------