THIRD JOINT U.S./USSR SYMPOSIUM ON
THE COMPREHENSIVE ANALYSIS
OF THE ENVIRONMENT
Tashkent, USSR
October 10-14, 1977
-------
E P A-600 / 9-80-024
May 1980
Third Joint U.S./USSR Symposium on
the Comprehensive Analysis
of the Environment
Tashkent USSR
October 10-14, 1977
Office of Research and Development
U.S. Environmental Protection Agency
Washington, DC 20460
-------
PREFACE
The eighteen papers contained in these Proceedings were originally presented
in English or Russian at the Third US-USSR Symposium on Comprehensive Anal-
ysis of the Environment which was held in Tashkent USSR from October 10-14,
1977. Two additional presentations, both on mathematical models (Gringof,
D'Arge), are also included in this publication (Section V).
The Symposium was conducted under the auspices of Project 02.07-21
"Comprehensive Analysis of the Environment", one of 40 environmentally ori-
ented projects under the US-USSR Agreement on Cooperation in the Field of
Environmental Protection. This project has been mainly concerned with environ-
mental goals and elaboration of a scientific basis for achieving them.
The publication of these proceedings is in accordance with the Memorandum
from the Sixth Session of the US-USSR Joint Committee on Cooperation in the
Field of Environmental Protection, signed in Washington, DC on November 18,
1977. This document called for independent publication of the Proceedings in both
the United States and the Soviet Union.
1
-------
Delegations
to the Third Joint US-USSR Symposium
on the Comprehensive Analysis of the Environment
UNITED STATES
Mr. Carl Gerber
US Project Leader
Office of Research and Development
Environmental Protection Agency
Washington, DC
Dr. Robert Chapman
Health Effects Research Laboratory
Environmental Protection Agency
Research Triangle Park, NC
Dr. Morris F. Cranmer
National Center for Toxicological Research
Food and Drug Administration
Jefferson, AR
Dr. Ralph D'Arge
Environmental Economics Department
University of Wyoming
Laramie, WY
Ms. Elaine Fitzback
Office of Research and Development
Environmental Protection Agency
Washington, DC
Mr. Clinton W. Hall
Office of Energy, Minerals, and Industry
Office of Research and Development
Environmental Protection Agency
Washington, DC
Ms. Jeanie E. Loving
Office of Research and Development
Environmental Protection Agency
Washington, DC
Mr. Lyn Starbird
Economics Section
U.S. Embassy Moscow
Dr. Frank Wilkes
Environmental Research Laboratory
Environmental Protection Agency
Gulf Breeze, FL
Dr. Herbert Wiser
Office of Research and Development
Environmental Protection Agency
Washington, DC
USSR
Prof. Yu. A. Izrael
Soviet Project Leader
Corresponding Member of the USSR Academy of
Sciences,
Head of Hydromet Service
Dr. Yu. A. Anokhin
Institute of Applied Geophysics of Hydromet
Service
Prof. M. Ya. Antonovskiy
Central Economic-Mathematical Institute of the
USSR Academy of Sciences
Prof. K. G. Gofman
Central Economic-Mathematical Institute of the
USSR Academy of Sciences
Dr. I. G. Gringof
Director of Central Asian Regional Research
Hydrometeorological Institute of Hydromet
Service (SARNIGMI)
Dr. A. A. Gusev
Central Economic-Mathematical Institute of the
USSR Academy of Sciences
Prof. I. I. Ilyinskiy
Uzbek Research Institute of Hygiene and
Occupational Diseases
Academy of Medical Sciences of the Uzb. SSR
Dr. I. M. Nazarov
Institute of Applied Geophysics of Hydromet
Service
Dr. R. A. Ryazanova
F. F. Erisman Moscow Research Institute of
Hygiene
Public Health Ministry of RSFSR
Prof. I. V. Sanotsky
Institute of Labour Hygiene and Occupational
Diseases of the USSR Academy of Medical
Sciences
Dr. S. M. Semenov
Institute of Applied Geophysics of Hydromet
Dr. Yu. A. Starikov
Expert of the US-USSR Joint Committee on
Cooperation in the Field of Environmental
Protection
2
-------
Prof. B. A. Tashmukhamedov
Institute of Biochemisty of the Uzbek SSR
Academy of Sciences
Dr. L. M. Filippova
Institute of Applied Geophysics of Hydromet
Service
USSR EXPERTS
Mr. K. A. Alimzhanov
Uzbek SSR Gosplan,
Head of the Department of Environmental
Protection
Mr. A. G. Akhmedbaev
Uzbek SSR Council of Ministers
Head of the Department of Environmental
Protection
Dr. N. F. Beloborodova
SARNIGMI
Dr. R. M. Valiev
Uz RIP town-building
Dr. A. V. Ganiev
Uz RIP town-building
Dr. A. D. Dzhuraev
SARNIGMI
Dr. G. E. Insarov
Institute of Applied Geophysics of Hydromet
Service
Dr. V. N. Kolesnikova
Institute of Applied Geophysics of Hydromet
Service
Mr. V. M. Koropalov
Institute of Applied Geophysics
Dr. Yu. V. Levchenko
Uz RIP town-building
Dr. V. I. Minchuck
Uz RIP town-building
Dr. S. A. Nishankhodzhaeva
SARNIGMI
3
-------
CONTENTS
Preface I
List of Participants 2
1. Introductory Remarks
The Role of Comprehensive Analysis in Organizing the Optimal
Interaction of Man and Nature
Yu. A. lzrael 7
Decision-Making to Control Environmental Pollution
C. R. Gerber 12
II. General Approaches
U.S. Approaches to Environmental Pollution Problems
J. Loving, E. Fit/back 16
Circulation of Carcinogens in the Environment -
L. M. Shabad 21
A Critique of Measurement Methodologies and Data Analysis,
Storage and Reuse—
H. L. Wiser 25
Basis for the Priority List of Substances for Monitoring Environmental
Contamination and the Role of Synthetic Organic Substances —
F. Ya. Rovinskiy 28
III. Ecological Effects
Ecological Approaches to Evaluating the State and Controlling the
Quality of the Environment-
Yu. A. lzrael, I. M. Nazarov, L. M. Filippova, Yu. A. Anokhin,
V. M. Koropalov, A. Kh. Ostromogilskiy, A. G. Ryaboshapko 31
Toxicity Control in Discharge—
D. 1. Mount 48
Anthropogenic Impact on Migrating Animals—
V. Ye. Sokolov, D. S. Pavlov, V. D. Ilyichev - 51
Microcosms as Indicators of Ecosystem Stress—
F. G. Wilkes, C. W. Hall 56
Ecological Monitoring in Relation to the Problem of Regulating
Environmental Quality —
L. M. Filippova, F. N. Semevskiy, S. M. Semenov, V. A.
Abakumov, G. Ye. Insarov, M. Ya. Antonovskiy 69
Evaluation of Priority in the Order of Contaminants—
V. D. Fedorov 79
5
-------
IV. Health Effects
Fundamentals of Control over Chemical Contamination of the
Environment, Preeminence of Medical Indications over all other
Approaches to Environmental Protection
N. F. Izmerov, I. V. Sanotskiy 83
The Relationship of Clinical and Epidemiologic Studies of the
Environment to Each Other and to Public Policy Decisions
R. S. Chapman N6
Analysis of the Priority Series of Contaminants from the
Physiologist's Viewpoint
B. A. Tashmukhamedov 97
Biological Effects of Certain Organic Derivatives of Carbamic Acid
A. P. Shitskova, R. A. Ryazanova, F. F. Erisman 100
Use of Animal and Laboratory Tests to Screen for Toxic Effects
M. F. Cranmer 103
V. Socio-Economic Considerations
Current State of the Problem of an Economic and Extra-economic
Evaluation of Man's Effect on the Environment and Ways to
Further Formulate It
I. P. Gerasimov 123
Social Valuation for Environmental Pollution Control
R. C. D'Arge 126
Socio-economic Basis for the Development of Industry with Regard for
Environmental Protection
M. Ya. Lemeshev, K. G. Gofman, A. A. Gusev 139
VI. Additional Presentations
Bases for Mathematical Modeling of the Moisture, Heat and Salt
Transport Processes in Soils to Determine Pollution
I. G. Gringof, Yu. M. Denisov 147
Models of Economic Growth and the Environment
R. C. D'Arge 156
VII. Record of the Third US-USSR Symposium on Comprehensive
Analysis of the Environment 157
6
-------
THE ROLE OF COMPREHENSIVE ANALYSIS OF THE
ENVIRONMENT IN ORGANIZING THE OPTIMAL INTERACTION
BETWEEN MAN AND NATURE
[RoP vsestoronnego analiza okruzhayushchey prirodnoy sredy v organizatsii
optimal'nogo vzaimodeystviya cheloveka s prirodoy]
YU A. IZRAYEL'
INTRODUCTORY REPORT
The primary task of the "Comprehensive Analy-
sis of the Environment" project is to study the most
diverse aspects and effects of certain factors (pri-
marily) on the elements of the biosphere, and to
perform an analysis of the effects and results of this
action in order to reveal the aspects which require
the adoption of measures to reduce the effects of the
action. The results of such an analysis are neces-
sary for the organization of the optimal interaction
between man and nature.
Comprehensive analysis encompasses an enor-
mous number of interactions and relationships, and
its implementation requires that a large number of
questions be answered which refer to the compe-
tence of diverse scientific directions. This study can
be made with varying detail, but the main point is
that in a comprehensive analysis no priority aspect
of the interaction can be left out. Thus, the obliga-
tion to examine and consider (at the first stage pos-
sibly the most approximate) all the important aspects
of the interaction is the main peculiarity of a com-
prehensive analysis of the environment.
Since a comprehensive analysis of the natural en-
vironment is necessary to organize the optimal in-
teraction of man and the environment,' it is an im-
portant component of a comprehensive analysis of
the interaction between man and nature. These two
concepts are very difficult to separate.
An examination of the role of comprehensive
analysis in the implementation of the optimal inter-
action between man and the environment requires
not only the determination of this role but also the
direction of its use at different stages and the vari-
ous types of interaction between man and nature.
This type of interaction occurs primarily during
man's economic activity.
The interaction can arise as a result of the use of
the natural resources of a given region, for example,
open-cut mining and city designing and building; the
conscious transformation of nature, for example,
construction of reservoirs and the reversal of rivers;
and the arrangement of industries which discharge
wastes into the environment without significant use
of on-site resources. These described types of inter-
action are generally accompanied by adverse con-
sequences; in certain cases man may obtain a spe-
cific benefit from such changes, and in some cases
where the changes are a result of direct changes in
nature, practically no benefit, as is apparent in the
last example.
In the organization of the optimal interaction be-
tween man and the environment, the adverse con-
sequences either will cease or will be minimized. In
the process of optimization of such relationships
there emerges a second type of interaction between
man and nature which is difficult to attribute direct-
ly to his economic activity; this is a type of inter-
action which is directed towards regulation of rela-
tionships:
This is organization of observations and the study
of different factors which affect the environment
and the state of the biosphere, that is, monitoring of
the anthropogenic effect on nature. The monitoring
includes observation, evaluation and prediction of
the environmental conditions. Thus, a comprehen-
sive analysis of the state of the natural environment
is the end result of monitoring;
That is, control of the quality of the environment
which results in an optimal state with regard to the
different interests of man, for example, the state of
his health and welfare.
Strategic decision-making and the control of eco-
nomic and natural processes can also be part of this
type of interaction.
The role of comprehensive analysis of the envi-
ronment in the strategy of optimizing the interaction
between man and nature is especially important in
the organization of control over nature, including
questions on the organization of monitoring and the
regulation of environmental quality. The purpose of
7
-------
studying the state of the environment and the organ-
ization of monitoring is to answer the following
questions:
a) what is the state of the biosphere at the pres-
ent time, and what changes can be expected in the
future?
b) what are the reasons for the desirable and un-
desirable changes in the biosphere, and what are the
sources of the undesirable changes?
c) what are the loads, actions, and stresses which
are harmful and undesirable on the basis of formu-
lated or adopted criteria?
The formulation of such criteria implies an an-
swer to the question of what type of action is per-
missible and what kind of ecological reserves the
biosphere (its elements) has.
To properly regulate environmental quality and
organize control over various processes for the pur-
pose of optimizing the relationships between man
and nature, it is necessary to develop a regulation
and control strategy. This development necessitates
that the following questions be answered:
d) what environmental conditions should be ac-
cepted as high and acceptable quality?
e) what degree of environmental quality should
one strive for considering the versatile purposes
and interests of man, that is, should one strive for
high quality or just acceptable quality?
0 what actions are most expedient or are of
prime importance from the viewpoint of reducing
ecological and aesthetic damage, and will these ac-
tions possibly have negative concomitant effects?
g) where must such actions be relatively more
vigorous, and the requirements for environmental
protection relatively more stringent: on contami-
nated or uncontaminated territory?
h) what actions have the most expedient eco-
nomic and social effects?
This list of questions could continue, although addi-
tional questions would be a detailing of those al-
ready enumerated.
In order to answer the majority of the listed ques-
tions specialized research and serious discussions
are required.
We will examine the scheme (structure) of a com-
prehensive analysis of the state of the environment
as applied to organization of monitoring and the reg-
ulation of environmental quality, taking into consid-
eration the discussions at the previous two sym-
posia on comprehensive analysis [1,2]. Initially, we
will examine the plan for the first stage of compre-
hensive analysis which is necessary to answer the
question: what is the state of the biosphere, and
what does this state mean for man and the ecologi-
cal systems at the present and in the future. This
section of comprehensive analysis includes the sys-
tem of data obtained with the help of primary mon-
itoring, and as a result of calculations, mathematical
modeling, and various laboratory and full-scale ex-
periments.
The plan for the first stage of comprehensive
analysis of the state of the natural environment fol-
lows.
The element of the biosphere A is affected by dif-
ferent factors Ci, governed by sources BK. The state
of the ecosystem is characterized by the quantity An
(abiotic component), Ah (biotic component, individ-
ual biological elements), and A,. - the system as a
whole. After the action, the state of the system is
characterized respectively by the quantities A^, A,',
a;.
The sources of action can be external (in relation
to the element of the biosphere A), or internal; they
can be natural components of the system or arti-
ficial, that is, introduced from the outside. The fac-
tors of action can be: a) contamination (chemical
factor), b) irradiation (physical factor), c) mechani-
cal, d) geophysical (action of various fields and the
collection of enumerated factors), and e) biological
(for example, competition, suppression, and other
community characteristics).
Thus, the state of A £ synthesizes the effect of all
the mixed and diverse actions. In analyzing the
state of Ae the following must be considered:
1) the quantity of changes A,. -» A,'.; that is, the
dimensions of the ecological and aesthetic damage;
2) the role of each source Bk and factor Q and
their contribution to this change.
In order to study the change At. -> Ae it is neces-
sary to know: the primary and natural state of A,,
(mean and extreme values), and background state,
that is, the state governed by the background quan-
tities of both artificial and material actions. It is also
necessary to consider the change in the state not
only under the influence of individual physical
(chemical) factors of the action which can be easily
related to specific sources, but also the constant
stress effects which weaken individual organisms,
populations and the system as a whole. This must
be considered to accurately describe the state A,,,
and eliminate additional computations. To evaluate
the effect on the population, it is necessary to take
into account the distribution of organisms not only
in space and time /n(Ft, t)/, but also in susceptibility
and sensitivity to the given action, and the total
number of organisms of different populations Nm
exposed to the action.
It is necessary to present the laws governing the
spread of these factors in a uniform environment,
and the possibilities of transfer from other environ-
ments to explain the reasons for a given action. Fur-
thermore one should determine the effect of several
factors acting simultaneously on the system, the
strengthening or weakening of their action as a re-
sult of this, and the strengthening or weakening of
the total system during the simultaneous action on
different levels of the system.
-------
To evaluate the action of various factors on the
organisms, elements of the biosphere, and ecosys-
tem as a whole, it is important to show whether the
given factor has a specific threshold of action,
above which the effect of the action occurs, and be-
low which it equals zero, or if the threshold is de-
fined as a certain acceptable probability of a harm-
ful effect. Thus, the concept of threshold is particu-
larly important for evaluating effects on individual
organisms. For example, in industrial hygiene, a
fairly strict standard is set as a result of this concept
in order to secure workers against occupational in-
jury exceeding a given norm. Using the probability
approach to define the action and if possible, any
factors of the action is the more correct approach to
determine the permissible loads on communities
and systems. Such an approach will make it pos-
sible to perform a complex and comprehensive
analysis of the action and its load on the system,
and to determine the distribution with regard to or-
ganisms exposed to the action. This approach will
also make it possible to avoid a situation which per-
mits the discharge of an enormous quantity of con-
taminating substance, /30V into any natural area,
even into unique ones like Lake Baikal, where
/30 s MPC (maximum permissible concentration),
and V < volume of the given area (aqueous or
air).
For an analysis and determination of the role of
each source of action it is expedient to
a) calculate the concentrations and their action
by using the known laws governing spread and ac-
tion described by known functions;
b) use mathematical modeling to analyze the pro-
cess or parts of the process;
c) set up an experiment in the laboratory, under
full-scale conditions (in situ) to detect the action of
a given factor on an individual population, society,
and system;
d) formulate general evaluations and analyses
from the results of observations and study under
full-scale conditions.
These methods will help to implement a stepwise
synthesis of all the detected effects to obtain a com-
prehensive picture of the situation. This will lead to
a solution of the inverse problems.
The analysis of the state Ae, and determination of
the economic damage include numerous nonuni-
form factors and effects, such as the intensity of
the acting factor, for example, concentration; the
toxicity or degree of biological effect; as well as the
quantity of organisms (M) in m population which
are exposed.
Analysis of the change Ae -» Ae also includes
analysis of Ae -* Aj.\ where Ai' is the state of a given
ecosystem in the future, since integrating the analy-
ses over time is implied, that is, it is possible to pre-
dict Ae on the basis of the available data.
It is necessary to select correctly the most impor-
tant factors and effects.
In order to analyze the ecological significance of
these factors or effects, it is expedient to use criteria
developed especially for this, for example, the max-
imum permissible discharge concentrations (MPC)
in various media for individual pollutants, the maxi-
mum permissible discharges (MPD) for sources,
and the maximum permissible ecological loads
(MPEL) for an evaluation of loads on the commu-
nities or the systems as a whole.
Using various methods of analysis, the priority is
determined by:
a) utilizing the ratios for amount of change in the
initial state of the system, that is, the sensitivity of
the system;
b) how close the state A'e approached the critical;
c) examining the absolute change Ae —~ A,' ex-
pressed as the product of the average change in in-
dividual organisms N of m-populations by the num-
ber of N of all m exposed to the action.
The priority determined by the first method can
be used to substantiate the organization or to per-
fect monitoring. The priority determined according
to the criticality of the system can be used to sub-
stantiate the organization of adoption measures,
primarily, emergency measures and those for short-
term plans. The priority determined with the help of
the last method can be used mainly for elaborating a
general strategy for making decisions to develop a
long-term plan of measures. It is this method which
is particularly important in the production of eco-
nomic evaluations and the prevention of significant
economic damage.
The aforementioned first method of analysis can
be repeated, that is, conducted again on new and
refined data. This will be the next approach to a cor-
rect solution of the problem, that is, description and
evaluation of the actual state of the environment.
In addition, the conducted analysis makes it pos-
sible to pass to the next stage of comprehensive
analysis, that is, economic evaluations and the de-
velopment of a strategy for regulating environmen-
tal quality and controlling the processes to protect
nature from negative anthropogenic actions.
The conducted analysis also makes it possible to
answer a number of questions raised in the begin-
ning of the report. For example, the selection of the
strategy of expediency for the development of the
economy in an uncontaminated locality or locality
already exposed to stress can be significantly assist-
ed by comprehensive analysis of the possible
thresholds of action and the probability of various
effects of the action.
If the concept of the threshold of action is
adopted, development on uncontaminated territory
would be absolutely harmless and not doubted if the
factors of the action were below the threshold. Fur-
9
-------
ther development is prohibited in zones where the
permitted limit (reserve) to the established thresh-
old level has been reached. If the probability con-
cept is adopted, calculation of the amount of con-
tamination or action would be required from the
viewpoint of the amount in the environment, as well
as from the viewpoint of exposed individuals.
The often mentioned concept of complete prohi-
bition of economic development in uncontaminated
areas does not have sufficient scientific sub-
stantiation, and in the author's opinion, does not
withstand criticism.
The next stage in comprehensive analysis must
include consideration of economic factors. Consid-
eration of economic factors in the formula for deter-
mining damage can change the priority of measures
on environmental protection away from negative
anthropogenic actions. Besides the three main com-
ponents; intensity of the action, its toxicity or bio-
logical (ecological) effectiveness, and the number of
individuals of different populations or elements of
the biosphere exposed to the action; there is anoth-
er important factor which determines the specific
economic cost of the affected element of the bio-
sphere (damage).
The socially justified amount of outlays for envi-
ronmental protection are those in which the rates of
outlays or their total sum do not exceed the rates of
economic damage or the total sum of this damage.
However, the optimal economic situation should be
one where the sum of outlays for environmental
protection and damage from the effect on nature
will be minimized. Naturally, the optimum should
be determined for real potentialities. In certain cas-
es the optimal situation from the economic view-
point will be unacceptable due to the interests of
human health or other types of considerations.
The regulation of environmental quality must in-
clude the organization of monitoring, the implemen-
tation of comprehensive analysis of the state of the
environment, and an economic evaluation of the
damage from anthropogenic actions, as compared
to the cost of environmental protection measures.
Regulation of environmental quality must be di-
rected towards a limitation of anthropogenic ac-
tions, in the first place contamination, which have a
negative effect on nature, to complete cessation of
them in the future. As already noted, such limita-
tions must be substantiated from ecological and ec-
onomic positions.*
The following approaches can be enumerated as
possible variants of strategy for limitations. They
have already found frequent practical application in
other countries. These applications are discussed in
various works [1,2,3].
•Again we stress that in comprehensive analysis of priority
from the ecological viewpoint, it is obligatory to consider all
three components, and from the economic viewpoint, four.
a) limitation of the quantities of MPC and
MPEL;
b) limitation of the discharges of industrial and
other enterprises based on the MPC or MPKL quan-
tities;
c) complete prohibition of discharges into unique
areas or in uncontaminated regions;
d) limitations governed by the best achieved or
achievable technique or technology of production;
e) imposition of fines or taxes on any discharges
or discharges over the permissible limits;
f) limitations according to the optimal solution of
analyzing outlays and damage from contamination,
or benefits from production outlays considering the
available resources;
g) striving for discharges below the permissible
levels using the priorities determined with the help
of comprehensive analysis.
The use of each of the indicated variants "in the
pure form" has its advantages and disadvantages.
The most effective application is the use of a combi-
nation of two or more approaches. Without going
into a detailed discussion of the enumerated ap-
proaches, it can be said that to this researcher the
most expedient is the combined use of the ap-
proaches described in point "b", and partially in
"e", and also considering "d" and the obligatory
observance of point "f". This approach results in a
gradual "step-by-step" reduction in the norms for
the limitation of discharges down to the require-
ments stipulated in point "b". The first part of point
"e" apparently will attract attention in the future
for use in the regulation of the spread of harmful
substances between different countries. Point "g"
is very impressive, but it is not practical at the pres-
ent time.
The role of comprehensive analysis according to
the described scheme consists not only of a com-
petent and substantiated selection of strategies for
the quality control of the environment, but also of
the selection of a strategy for man's interaction with
nature. Currently, a fairly serious analysis and cal-
culation for individual sections are being made, but
as a rule it is antedated after a negative interaction
occurs.
Comprehensive analysis must also be an impor-
tant element in such major models as those which
describe the development of the modern world.
This question was not suitably covered in the very
extensive work of Mesarovic and Pestel [4]. Only
by conducting such a comprehensive analysis can
we calculate how close we are to a crisis situation,
what reserves remain in the biosphere, and how
much time mankind has to make effective decisions
and take measures to preserve the enormous num-
bers of populations of animals, plants, insects, and
the single, in the literal and figurative sense, popu-
lation of man.
10
-------
REFERENCES
1. Vsestoroniy analiz okruzhayushchey prirodnoy sredy. Trudy
I sovetsko-amerikanskogo simpoziuma (Comprehensive Anal-
ysis of the Environment. Works of the First Soviet-American
Symposiuml, Tbilisi, 1974, Leningrad: Gidrometeoizdat, 1975,
325 p.
2. Vsestoroniy analiz okruzhayushchey prirodnoy sredy. Trudy
II sovetsko-amerikanskogo simpoziuma [Comprehensive
analysis of the environment. Works of Second Soviet-Ameri-
can Symposiuml, Honolulu, 1975, Leningrad: Gidrometeoiz-
dat, 1976, 306 p.
3. de Nevers, N. 1977. Air Pollut. Con. Assoc. 27(3): 197-205.
4. Mesarovic, M. and E. Pestel. Mankind at the Turning
Point. Hutchinson, London, 1975.
11
-------
DECISION-MAKING TO CONTROL ENVIRONMENTAL POLLUTION
CARL R. GERBER
INTRODUCTION
To discuss decision-making to control environ-
mental pollution poses a difficult task. The problem
lies within the issue itself. For in its broadest sense
the word "environment" can encompass practical-
ly everything, including aesthetics and other sub-
jective personal factors. Even if we limit our dis-
cussion to the physical environment, we still must
deal with an extremely complex matter. Not only
must we know a great deal about basic sciences and
systems, but also about the interaction among sys-
tems; and we must operate in the real world of un-
certainties, societal values, and change. To accom-
plish this successfully we must apply some system-
atic, problem solving approach.
Because of the enormous difficulty in making de-
cisions in the environmental area, it is tempting to
throw up our hands and avoid a decision-making
process; or make a decision based on only one of
the critical factors. While this latter approach is of-
ten forced on us because of a crisis — or perceived
crisis — we cannot let it become our normal way of
operating. Nor can we assume environmental prob-
lems will go away with time, or that environmental
decisions will be easier to make. If anything, envi-
ronmental problems and subsequent decisions will
become increasingly more difficult to handle.
A major contributor to our expanding environ-
mental problems is population growth. With in-
creasing numbers of people and increasing con-
sumption per person, we are producing wastes in
quantities that are beyond the assimilative capacity
of our natural systems. In some areas, we may be
actually destroying the natural systems. Another
major contributor is the new, man-made sub-
stances, particularly synthetic organic chemicals,
that are being introduced into the general environ-
ment at an ever increasing rate, either directly or as
waste products. The persistence of many of these
new substances and the lack of information on their
effects, particularly secondary and long term ef-
fects, on our natural systems pose potentially mas-
sive problems both now and for the indefinite fu-
ture.
Thus there is no question that we must "do some-
thing" about protecting the environment from both
direct and indirect pollution. The question is rather
what to do, or more accurately how to address this
problem. Basically how can we make rational deci-
sions about environmental protection?
Pollution rarely, if ever, occurs because someone
wishes to destroy or harm the environment, rather
it occurs as an effect of another action which has a
benefit to someone or some group. The issue then is
how to make the trade-offs between competing ben-
efits or competing costs, where a cost to one group
may be a benefit to another. How do we make a
single decision based on more variables and un-
knowns than most decision makers ever face?
PROBLEM ANALYSIS
The greatest value of any decision analysis is that
it provides a systematic approach to a problem and
allows the problem to be broken down into manage-
able pieces. It also makes explicit the things being
compared, the factors being considered, and the
trade-offs being made. If all the factors could be
quantified or reduced to a common denominator,
say dollars or rubles, objective comparisons would
then be facilitated and subjective values of probabil-
ity and uncertainty would not affect the decision.
Unfortunately, such quantification is not now pos-
sible, and may never be.
Much theoretical work has been done to structure
and optimize the decision making process, most of
it originating in economic and statistical theory. Un-
fortunately, the terminologies have not yet been
clarified and many of the theories are too abstract.
The systematic approach that is most applicable
to environmental problems is risk-benefit analysis.
Here some, if not all, of the costs as well as the
benefits are not certain and one must deal with
probabilities. Probabilities here relate not only to
human health and the ecosystem, but also to the ec-
onomic and social structure. Risk-benefit analysis is
a generic term for techniques encompassing the
evaluation or assessment of risks, costs, and bene-
fits of a problem and the alternative projects or poli-
cies to solve that problem [Ref. 2, p. 2].
There are, however, several problems that arise
in carrying out formal risk-benefit decision analy-
ses, since it is difficult (or impossible) to quantify
some considerations or even determine their proba-
12
-------
bilities. Consequently, there is a tendency to work
with the aspects that are easily quantified and to ig-
nore the rest. Also many of the quantifications of
certain risks or benefits involve uncertainties and
value judgments. It may be difficult (or impossible)
to get a consensus, for example, on the value of hu-
man life. Another set of difficult questions relates to
the distribution of risks and benefits. What is the
trade-off when something is beneficial to one group
of people, but detrimental to another group? Are
the trade-offs different when the effects will be felt
by a future generation?
The U.S. National Academy of Sciences [Ref. 3,
p. 38] observes that "there is no objective scientific
way of making decisions, nor is it likely that there
ever will be. However, use of the techniques devel-
oped by decision theory and risk-benefit analysis
can provide the decision maker with a useful frame-
work and language for describing and discussing
trade-offs, noncommensurability, and uncertainty.
It can help to clarify the existence of alternatives,
decision points, gaps in information, and value
judgments concerning trade-offs." Furthermore, I
believe it should facilitate communication and
should indicate areas where research is needed.
A critical problem area for policy makers in risk-
benefit analyses is assessing the values of the gener-
al public. Current decision theory provides a frame-
work for analysis when the value structure of the
decision maker (individual or group) is known, but
does not aid in integrating the values of different in-
dividuals into a single system. Another problem is
that few policy decision makers have the time or the
expertise to carry out analyses personally. Since
there are uncertainties at every stage of an analysis,
it is important that the decision maker be aware of
the limitations that uncertainty puts on the analysis.
It is essential that there be close interaction be-
tween those who make the final decisions and the
people doing the analyses.
DECISION PROCESS
Risk-benefit analyses have been made implicitly
for centuries, and are performed by each of us in
our daily activities. They are especially applicable
as a formal decision-making tool when decisions in-
volve risks to lives, health, or the ecosystem. Basic
to these analyses is the need to define and evaluate
risk. Risk assessment, the term applied to this pro-
cess, consists of an explicit appraisal of both the
kinds and probability of threat posed by an environ-
mental hazard and to the management of these haz-
ards. An environmental hazard may be defined as
the potential threat posed to man or the ecosystem
by events originating in or transmitted by the natu-
ral or man-made environment. The consequences of
environmental hazards include threats to persons —
morbidity and mortality; threats to ecosystems; and
disruption of economic and social activities.
The first step in this process is to identify envi-
ronmental hazards. Hazard identification answers
the question of what constitutes an environmental
threat. Several methods are used to identify envi-
ronmental hazards; they include applied research,
screening, and observation. By applied research, I
mean a process of identifying or developing a stan-
dardized procedure in order to test substances, as
well as the study of effects of substances on biologi-
cal systems for their hazard potential.
Screening is a part of the hazard identification
process whereby a standardized procedure is ap-
plied to classify substances and processes for their
hazard potential. Observation, when done system-
atically through epidemiological studies, can lead to
an assessment of symptoms or consequences in re-
lation to possible causes, and thus provides a major
means of identifying the hazards of environmental
contaminants to people and ecosystems.
The fact that environmental hazards presently are
under-assessed, over-assessed, or not even identi-
fied indicates that the methodologies utilized in the
process of hazard identification must be improved
and expanded. The ideal hazard identification meth-
odology would be economical enough to identify
significant environmental hazards in an objective
manner before they actually become hazardous.
Identifying hazards, however, is only the first
step; next we need to estimate the extent of the risk
posed by these hazards. Risk estimation is a pro-
cess to measure the probability that an event of
some stated magnitude will occur. It also provides
information on the nature of the consequences
which are expected to follow the event. The accura-
cy of risk estimation is affected by the experience or
attitude of the person or persons making the proba-
bility estimates. Specifically, events which occur
frequently and have a direct impact on every day
life and livelihood are more likely to receive an ac-
curate assessment than unusual events.
MEANS OF CONTROL
Once a particular substance is recognized as pos-
ing a present or potential environmental hazard, the
decision process must then address the means that
are available to control or abate this source of envi-
ronmental pollution, and the costs and social ac-
ceptability of the control method.
The simplest means of control, in some ways, is
an outright ban or prohibition, but this is only fea-
sible when there are not obvious present or ftiture
benefits related to the substance being banned.
Rarely is this the case, so the decision analysis must
then move to considering ways of reducing the pol-
lution to acceptable levels.
13
-------
One method is through regulations or govern-
ment-set standards. Such legal limits can be based
on protection of human health and the ecosystem;
technology; or registration. Health based standards
are the most desirable in many ways for they can
relate most directly to the area of chief concern —
human health. However — given the lack of knowl-
edge of the effect of substances on humans; the re-
action of humans to pollutants, particularly the
threshold vs. no-threshold response; and the lack of
precision in determining the actual exposure hu-
mans receive — this is the least precise and most
difficult method to use to set legal limits.
A more precise approach is to set a legal limit
based on technology. In this approach, substances
that are, or may be, hazardous are controlled by
technological means before they enter the environ-
ment. Sewage treatment plants, for example, can be
required to reduce discharges of certain substances
below certain levels; or manufacturing or power
plants can be required not to release certain sub-
stances above a certain amount into the atmo-
sphere. Here the question is to what level should
pollutants be controlled. Should they be controlled
to the level of what current, proven technology can
do; or to a level that technology is believed, either
through laboratory work or small-scale operations,
to be able to do? In the former case, pollution would
be controlled to a lesser degree than if new tech-
nology were to be developed. A major drawback to
this approach is that it does not take into account
the synergistic effects of pollutants; and unless ap-
plied simultaneously across all discharges (into the
air and water, and on the land) it will only result in a
pollutant being moved from one media to another.
While its benefits and risks with regard to human
health and the ecosystem are hard to determine, the
costs of a technology-based legal limit are relatively
easy to determine.
Registration is a process and, therefore, avoids
many of the weaknesses of setting specific health or
technology-based numerical limits. By requiring
registration of potentially hazardous materials, it is
possible to gain at least a first approximation of the
amount and location of the substance. This ap-
proach allows more precision in the risk-benefit
analysis. For here, when a substance is known to be
hazardous, that fact can be specifically weighed
against its benefits; and as a result, those uses
where the benefits outweigh the risks can be ap-
proved while those where the risks are greater can
be banned. This technique, of course, does not take
into account synergistic effects and requires many
individual decisions.
An alternative, to government imposed legal lim-
its, is to use "economic" or incentive means of
achieving control. One way to apply this is to
charge people or organizations that discharge pollu-
tants into the environment. This approach is most
effective when the charges are slightly more than
what it would cost to stop polluting. A variation of
this approach is to charge the amount it takes to
clean up the pollution; yet once some pollutants are
introduced into the general environment they are
very difficult, if not impossible, to remove. If the
charge system were used, certain substances there-
fore would have to be effectively banned.
A positive incentive system is to provide funds,
technology, and/or training to specific potential pol-
lutant sources to assist in reducing or eliminating
the pollution at the source. The logical extension of
this approach is to have a centralized government
pollution control system such as wastewater treat-
ment plants.
A final aspect of a risk-benefit analysis is to deter-
mine "what society is willing to pay to avoid certain
risks, or conversely what hazards they are willing to
accept to avoid certain costs — in other words,
what type of control is acceptable. It is in this area
that the uncertainties are the greatest and the meth-
odology the weakest; yet it is the part of the whole
analysis most critical to the decision maker in
reaching a final decision in many, if not all, environ-
mental matters.
Underlying all the risk assessment activities is a
need for good data. Without good data all our deci-
sion analyses are tenuous and uncertain and our ef-
forts to control pollution are academic exercises. If
we do not know, or cannot tell, what pollutants are
emitted from what sources, where they are in the
environment, and when and how they impact hu-
mans and the ecosystem, our decision-making then
is nothing but a theoretical game. Thus, the collec-
tion, analysis, storage, and handling of data are,
collectively, a vital underpinning of our efforts to
maintain and improve the quality of the environ-
ment.
CONCLUSION
With the world's population doubling, at the pres-
ent rate, every 37 years, the present 4 billion people
will be 7 billion by the year 2000. The rate of con-
sumption per person is increasing at an even faster
rate in many parts of the world. There are 40,000-
50,000 chemicals already in existence; and new
ones are coming at the rate of about 1,000 per year.
Thus we are faced with environmental problems
that would tax even the best established and func-
tioning decision-analysis methodology.
But rather than be overwhelmed by the situation
and merely reacting as problems reach the crisis
stage, I believe we can, and must, take some specif-
ic steps. Obviously one of the simplest ways of
gaining on the overall problem, particularly with po-
tentially toxic chemicals, is to use a generic ap-
proach to studying potential pollutants rather than
14
-------
the present ad hoc chemical by chemical method;
another major step forward would be the develop-
ment of good screening techniques. Even with these
two improvements, we would probably be faced
with more problems than we could solve immedi-
ately; and this requires that we formulate or set
some priorities.
My own set of priorities in attacking the mounting
problem is to handle pollution problems that affect
human health first; those that threaten to disrupt
ecosystems next. Human welfare would be third on
my list, while specific but non-disruptive ecosystem
changes would be fourth. Others, I am sure, would
have a different set of priorities, but clearly more
discussion is needed in this area. For unless we can
establish some overall set of priorities, we may see
one group, region, or country attacking one set of
problems only to be impacted by the failure of an
adjacent unit to do likewise.
In addition to the need to establish a set of nation-
al and international priorities, there are other more
specific hurdles we must work to overcome if we
are truly going to maintain and improve the quality
of our environment. Most fundamental is the lack of
good basic data and understanding of the processes
and systems we must deal with in assessing envi-
ronmental impacts; even more frustrating, because
it should be easier to obtain, is the lack of good in-
formation on environmental trends both in local
areas and on a global scale. And finally one of the
most difficult problems we face, even if we use risk-
benefit decision analysis, is how to trade-off present
benefits against potential future hazards. This is
particularly difficult when we must impose a cost on
the public to pay for benefits they will never see;
and which we can give no absolute assurance will
ever occur.
I do not have the answers to these perplexing
questions or the solution to these problems, but I do
believe it is through meetings such as this and con-
tinuing dialogues between all interested and in-
volved parties that the issues can be clarified. And
from such clarification and good analyses, answers
will come.
REFERENCES
1. National Academy of Engineering, Committee on Public En-
gineering Policy, Perspectives on Benefit-Risk Decision
Making, 1972.
2. Van Horn, Andrew J. and Wilson, Richard, The Status of
Risk-Benefit Analysis, BEAG, December 1976.
3. National Academy of Sciences, Committee on Principles of
Decision Making for Regulating Chemicals in the Environ-
ment, Decision Making for Regulating Chemicals in the En-
vironment, 1975.
4. Scientific Committee on Problems of the Environment, Risk
Assessment of Environmental Hazards, Robert Kates,
(SCOPE 8 Report), 1976.
5. Clark, Elizabeth M. and Van Horn, Andrew J., Risk-Benefit
Analysis and Public Policy: A Bibliography, 1976.
6. National Academy of Sciences, Principles for Evaluating
Chemicals in the Environment, 1975.
7. Daly, Phyllis A., "Environmental Decision Making and the
Use of Risk Assessment," unpublished paper, 1977.
8. Tihansky, Dennis P. and Kibby, Harold V., "A Cost-Risk
Benefit Analysis of Toxic Substances," J. Environ. Sys.,
Vol. 4(2). Summer, 1974.
9. Otway, Harry J. and Palmer, Philip D., "Risk Assessment,"
Futures, April 1973.
10. National Academy of Sciences, Study Panel on Assessing
Potential Ocean Pollutants, Assessing Potential Ocean Pol-
lutants, 1975.
11. Rowe, William D., An Anatomy of Risk, John Wiley and
Sons Inc., 1977.
12. Raiffa, Howard, Decision Analysis, Addison-Wesley, 1965.
13. Harris, Robert C., "Suggestions for the Development of a
Hazard Evaluation Procedure for Potentially Toxic Chem-
icals," MARC Report No. 3, A Research Memorandum,
1976.
14. Lawrence, William W., Of Acceptable Risk Science and the
Determination of Safety, William Kaufman, Inc., 1976.
15
-------
U.S. APPROACHES TO ENVIRONMENTAL POLLUTION PROBLEMS
JEANIE E. LOVING AND ELAINE M. FITZBACK
SUMMARY
There are three major approaches to abating and
controlling environmental pollution: by individual
pollutant or class of pollutants; by industry, tech-
nology or process which creates the pollutant; and
by geographical region. Because there are advan-
tages and disadvantages to each approach, a combi-
nation is needed to permit comprehensive analysis
and management of the environment. This paper re-
views some aspects of these approaches, all of
which are used in the United States.
INTRODUCTION
In the field of environmental management, there
is general agreement that the overall goal, simply
stated, is to protect human health and ecological
systems from being adversely affected by con-
taminants. There is also general agreement that the
actions taken to achieve this goal must strike a bal-
ance between adequate protection and the econom-
ic and social costs of such protection.
One of the difficulties facing decision-makers in
solving any contemporary problem is how best to
approach the issue in question. Everyone would
agree that all the pertinent facts must be weighed
objectively. But pollution control can sometimes
become an emotional issue. The definition of the
word "pollution" in Webster's English Dictionary
is "making something unclean or contaminated."
There are emotional implications even in the defini-
tion of the word. People can understandably be-
come emotional when water supplies are dirtied by
toxic chemicals and human waste products, and the
air is sullied by industrial and automotive emis-
sions. People can also become understandably emo-
tional when the cost of cleaning up the environment
significantly affects their own incomes.
Yet pollution, is after all, an example of an ele-
mentary concept of physics; the law of con-
servation of mass and energy. When materials are
taken from nature for human use in various ways,
they are modified, used as products or energy, for
example, and returned to nature. Most often, they
are returned in forms which are different from those
found naturally. It is the material returned, its relo-
cation and concentration when returned, which
may damage the environment. Everyone is familiar
with" the problems that industrial effluent discharges
can bring to water, emissions of sulfur and nitrogen
oxides can bring to air, and pesticides can bring to
aquatic and terrestrial ecosystems. Further, pro-
cesses or technologies for solving such problems
may create new ones, also related to the modifica-
tion and redistribution of materials. On the other
hand, each of the activities which produce environ-
mental problems can have its benefits.
Undertaking the task of balancing hazards, costs,
and benefits of man's activities for effective envi-
ronmental management is not simple and can be ap-
proached in a variety of ways. This paper reviews
three main approaches, all of which are used in the
U.S. simultaneously or in combination. The ap-
proaches are: by individual pollutant; by industry or
technological process; and by geographical region.
Each approach has its advantages and disadvan-
tages, some of which are indicated as follows.
SOME ADVANTAGES AND DISADVANTAGES
OF EACH APPROACH
Dealing with environmental contaminants on an
individual or "pollutant-by-pollutant" basis is the
simplest approach. Using this approach, an individ-
ual substance is traced from its entry into the envi-
ronment, through the environment to man. Its ef-
fects on ecosystems at critical points are assessed,
as are concentrations to which humans are ex-
posed. The effects on health at varying concentra-
tions and durations of exposure are examined, and
exposure-response curves are evaluated.
The pollutant-by-pollutant approach permits con-
trol of the pollutant at its origin, particularly where
primary sources of the pollutant are fairly readily
identifiable. From a research standpoint, this ap-
proach is also the simplest to undertake.
An illustration of this approach would be found in
dealing with carbon monoxide. Most anthropogenic
CO is emitted into the environment in the exhaust
from automotive vehicles. Although CO is dis-
persed in the atmosphere, there are a sufficient
number of vehicles and sufficient motor transport
in U.S. urban centers to maintain airborne concen-
trations well above natural background levels. In-
haled CO forms carboxyhemoglobin in the blood,
16
-------
decreasing the blood's capacity for facilitating oxy-
gen and carbon dioxide exchange. The adverse
health effects associated with evaluated carboxy-
hemoglobin levels have been sufficiently well docu-
mented to cause concern about human exposure to
the levels of CO found in urban air. Thus, in order
to protect health, the source of the CO is regulated.
The recent U.S. experience with Kepone, a toxic
and potentially carcinogenic insecticide, is also an
example of the pollutant-by-pollutant approach.
Kepone, or chlordicone, is a chlorinated organic
compound, similar to aldrin and dieldrin. It was dis-
covered in the James River and Chesapeake Bay in
1975. The substance was first detected in the Hope-
well, Virginia municipal-treated effluent and was
traced to a small chemical plant in that city. Con-
tamination was found in the air, ground and build-
ings of the plant. Symptoms such as nervous trem-
ors, chest pain, and eye trouble were observed in
many of the company's workers or their families.
By the time the plant was closed in July, over
100,000 pounds of kepone had already been dis-
charged in the James River. On the basis of data
provided by the Environmental Protection Agency
(EPA) and the Virginia State Water Control Board,
the Governor of Virginia closed the James River to
fishing. The producers of the kepone were later con-
victed of polluting under the Federal Water Pollu-
tion Control Act.
The research program emerging from this in-
cident is a good example of a joint laboratory, field,
and modeling effort carried out on a federal, state,
and local level. Investigations into health and envi-
ronmental effects of the discharge have been
coordinated by a task force established in Decem-
ber 1975. Over the last two years, there has been a
continuous monitoring program to detect levels of
kepone in finfish, shellfish, soil, sediment, and wa-
ter samples in the James River and Chesapeake
Bay. Various studies on the effects of kepone on
estuarine animals, its bioaccumulation, uptake from
sediments, and fate and degradation have been car-
ried out; mathematical models of fate and transport
of kepone are being developed; and measures
for removing or stabilizing kepone have been
studied.
The Kepone incident also prompted the Virginia
State Legislature to pass a Toxic Substances Infor-
mation Act, the first of its kind passed by a state in
the U.S. It required industry to report on the chem-
icals used in its manufacturing processes. The goal
of this program is a computerized data information
system which will provide ready access to a geo-
graphical inventory of chemical substances. If a
spill occurs in the future, using this system the State
of Virginia will be able to quickly pinpoint the
source and take ready action to ameliorate the situ-
ation.
The Kepone episode, although constituting a
grave health problem, is an example of a relatively
simple control situation in which a single compound
could be traced to a single major source. However,
the selection of an option for control using the indi-
vidual pollutant approach becomes difficult, when
the substances enter the environment from numer-
ous or pervasive sources. This is the case, for ex-
ample, with agricultural run-off. In the U.S. this
type of phenomenon is called a "non-point" source
problem.
Moreover, contaminants entering the environ-
ment from either non-point or specific or "point"
sources may interact as they reach living things,
which may result in biologic synergism or inhibition
with respect to the effects produced. If some of the
interactants are from non-point sources, and some
are from point sources, the control strategy must be
complex, and the pollutant-by-pollutant approach
may not be correspondingly sophisticated enough.
It is also known that as contaminants move
through the environment, they may undergo a vari-
ety of chemical transformations. Each new sub-
stance formed may exert its own unique effect on
humans and on ecosystems. This then means that
options for control must also consider the chemical
by-products of the original contaminants. To ade-
quately generate and evaluate the data necessary
for decisions concerning the control of each and
every contaminant and its transformation products
would require an enormous investment of people,
time, and money. It is apparent, therefore, that the
individual pollutant approach must be supple-
mented by other means of coping with the com-
plexities that pollution control entails.
Approaching pollution control according to the
technological, industrial, or other process which
produces contaminants is a second alternative way
of confronting environmental management issues.
With this approach, the various processes for ex-
tracting the earth's materials and using them to
manufacture products are assessed in terms of their
health and environmental effects. Processes which
generate power, for example, and technologies for
treating and disposing of industrial and municipal
wastes are evaluated in this manner.
This technology or process approach strikes at
the core of the environmental management problem
— that is, at the consequences of large-scale redis-
tribution of the earth's materials and energy. This
approach can also encompass both point source and
non-point source problems.
But an element of the pollutant-by-pollutant ap-
proach is nevertheless needed. When evaluating the
health and environmental impact of an air or water
effluent from a technology or process, something
must be known about the individual components of
the mixture. This is at least in part because it may
17
-------
be unnecessarily costly to control the total effluent
from a given technology, when perhaps only a few
of its components are known to significantly exert
adverse health or ecological effects. There may be
less costly ways to modify the process so as to re-
duce or eliminate the discharge of the most dam-
aging substances. Further, to conserve natural re-
sources, it is important to know whether there are
individual compounds of value in the discharge
which can be recovered and economically recycled.
From a research standpoint, this second ap-
proach is useful in that it reduces to some extent the
burden of investigating each and every pollutant. It
can foster scientific consideration of interactive ef-
fects. Used with the pollutant-by-pollutant ap-
proach, it can produce valuable research analyses
— if the chemical and physical behavior of a com-
pound in the environment and its effects on living
things are known, then useful inferences about simi-
lar compounds can be drawn when they are present
in different types of effluents.
The technology or process approach has many
other advantages not enumerated here, but it alone
does not quite yield all of the necessary answers.
Because ecosystems vary morphologically, func-
tionally, and dynamically from one place to anoth-
er, a given effluent or discharge will impact one geo-
graphical area in a certain way and another area in a
different way.
Approaching pollution control by the third alter-
native, that is by geographic region, takes this obvi-
ous fact into consideration. This approach provides
a framework for analysis of control options which
can insure that the nature and degree of control
needed for one sector of the environment is not un-
necessarily imposed on another. The geographic ap-
proach for control is quite compatible with current
research approaches, particularly in the ecological
sciences. For example, divisions along the lines of
aquatic and terrestrial research are traditional, with
subdivisions of each varying with geographic lo-
cale.
Because contaminants produced in one area can
be translocated long distances, something must be
known about the sources of contamination ob-
served in a given region. Thus, pollution control in
one region may necessitate control of a source in
another. Most often, the original source will be a
technology or process such as the examples pre-
viously given, and the need to supplement a geo-
graphical approach with a technology-based or pro-
cess approach should be apparent.
The control of sulfur dioxide is an example of the
need for regional considerations. S02 has been reg-
ulated in the U.S. for a number of years; but is has
been learned that S02 can be converted to sulfate as
it is transported over long distances. The sulfate
species formed may be potentially more dangerous
to health than S02, depending on their concentra-
tions and the duration of human exposure. The sul-
fates may also contribute to acid precipitation in
areas quite remote from original sources of S02. If
the presence of sulfates is to be reduced in a given
location, the analysis required to do so must include
consideration of the sulfate precursor and its
sources, including their location.
Each of the three approaches, used alone, cannot
provide the total framework needed for comprehen-
sive analysis and management of the environment.
A combination of all three offers a more satisfactory
conceptual way to ferret out the sources of con-
taminants, to determine which are of most concern,
and to estimate where they are likely to have the
most adverse effects.
In the U.S., as in other countries, there is a series
of legislation or enforceable laws requiring and au-
thorizing pollution control. These laws also deline-
ate the respective roles which the various federal,
state, and local agencies play in the National effort.
The laws provide for the agencies implementing
them to achieve the needed combination of ap-
proaches for solving pollution problems.
EXAMPLES OF MAJOR ENVIRONMENTAL
LEGISLATION
EPA's mission evolves primarily from ten basic
acts of Congress: The Clean Air Act (CAA); the
Federal Water Pollution Control Act (FWPCA);
the Safe Drinking Water Act (SDWA); the Solid
Waste Disposal Act (SWDA) and its amendment
the Resource Conservation and Recovery Act of
(RCRA); the Federal Insecticide, Fungicide and
Rodenticide Act (FIFRA) and its amendment the
Federal Pesticide Control Act of (FEPCA); the
Toxic Substances Control Act (TSCA); the Public
Health Service Act (PHSA); the Noise Control Act
(NCA); the Marine Protection Research and Sanc-
tuaries Act (MPRSA); and the National Environ-
mental Policy Act (NEPA). Brief highlights of some
of these Acts are as follows.
The Clean Air Act authorizes the EPA to set na-
tional ambient air quality standards for each air pol-
lutant for which air quality criteria are available, for
new and modified stationary sources, and for haz-
ardous air pollutants. Although each State has pri-
mary responsibility for the air quality in its geo-
graphical region, it must prepare implementation
plans for review by EPA. The Act also provides for
the regulation of individual pollutants emitted from
automotive vehicles.
Authority for EPA's water pollution control pro-
gram is embodied to a large extent in the Federal
Water Pollution Control Act. The Act authorizes
EPA to set effluent limitations and water quality
standards, to issue, or have the States issue, waste-
water discharge permits, and to distribute grants for
18
-------
constructing municipal pollution control facilities.
The Act also directs EPA to investigate practicable
means to treat municipal sewage, improved ways to
identify and measure the effects of pollutants
created by new technological developments, and
methods to evaluate the effects on the water quality
of augmented streamflows to control pollution. The
Safe Drinking Water Act directs EPA to establish
national drinking water standards on an individual
contaminant basis — primary standards for the pro-
tection of public health and secondary standards re-
lating to taste, odor and appearance of water.
The Federal Environmental Pesticide Control
Act is a revision of the Federal Insecticide, Fungi-
cide, and Rodenticide Act. The latter required that
before a pesticide could be transported across State
boundaries, it had to be registered and certified ef-
fective against the pests indicated on its label, and
safe for humans, crops, animals and the environ-
ment. FIFRA also stated that the pesticide in ques-
tion could not leave harmful residues on food or
feed. In addition to these requirements, FEPCA re-
quired proper application of pesticides and extends
registration and regulation to all pesticides, includ-
ing those used in a single state. Pesticides must be
classified for "general" use or "restricted" use. If
classified as "restricted," they must be used under
the supervision of certified applicator. EPA is also
authorized to establish tolerance levels for pesticide
residues on food and animal feed. Pesticide manu-
facturing plants must be registered and must submit
information on types and amounts of pesticides pro-
duced to EPA. EPA also has the authority to issue
orders to stop the sale and use of a pesticide and for
their removal from market shelves.
The basic purpose of the Toxic Substances Con-
trol Act passed in October of 1976 is to protect hu-
man health and the environment from the hazards
of certain chemical substances. When there are in-
sufficient data available, TSCA directs to EPA to
require that testings be carried out on a chemical
substance or mixture to develop data regarding
health and environmental effects. It establishes a
committee to make recommendations on a priority
list of substances known to cause or contribute to
cancer gene mutation or birth defects. When a
chemical is listed, EPA requires testing on it within
the year. TSCA also charges EPA with the task of
regulating hazardous chemical substances, that is,
prohibiting their production or limiting the amount
produced, based on scientific data. Manufacturers
of new substances are required to notify EPA 90
days prior to the production of chemicals.
The Resource Conservation and Recovery Act
requires that EPA publish criteria for identifying
hazardous wastes and issue (or have the States is-
sue) permits for treating, storing and disposing of
hazardous wastes. In addition to addressing future
open disposal on land of hazardous wastes, RCRA
directs EPA to promulgate guidelines for solid
waste management systems and provide technical
and financial assistance to State and local govern-
ments for developing and implementing their sys-
tems.
The Marine Protection Research and Sanctuaries
Act provides for the regulation of the disposal of
chemicals or other materials, including high level
radioactive wastes, at sea. A permit must be ac-
quired from EPA — or in the case of dredge spoils
— from the Corps of Engineers. EPA is also direct-
ed to designate disposal sites and assess penalties
for improper disposal.
HIGHLIGHTS OF RESEARCH APPROACHES
The same U.S. laws which provide for pollution
abatement and control authorize research so that
control actions taken are informed ones. These laws
thus inherently recognize that science must be the
driving force behind pollution abatement. As the
Federal organization most responsible for pollution
control, EPA draws from research conducted in
many government, private, and academic groups.
EPA has its own research program as well. Many of
the Soviet participants in this symposium have
worked with EPA scientists under the U.S./
U.S.S.R. Agreement on Cooperation in the Field of
Environmental Protection.
EPA's research program is specifically set up to:
• identify contaminants
• discern their environmental transport, inter-
actions and fate
• document existing and potential levels of ex-
posure
• discern the health and ecological responses to
pollutant exposures
• develop new or refined means for pollution con-
trol
• determine economic costs and benefits of pollu-
tion control and
• document health and ecological benefits of con-
trol
Such information must answer questions about:
which pollutants or groups of pollutants or process-
es should be controlled, the extent to which control
is needed, and the control options which afford the
most adequate health and environmental protection
at the lowest cost to society.
EPA's research seeks to define, analyze, and
solve existing pollution problems, while helping to
prevent the advent of new ones as much as pos-
sible. To do so requires a combination of the ap-
proaches reviewed in this paper. A description of
the entire program is well beyond the scope of the
present discussion. Some examples, however, may
convey the flavor of the way in which EPA's re-
search uses these approaches.
19
-------
Pesticide research is conducted on an individual
pollutant basis. Because each compound must be
registered, its potential health and ecological impli-
cations must be known. The studies conducted are
largely toxicological ones, with development of
methods for identifying the compounds in tissue
and environmental samples being an important fea-
ture of this work. Over the past few years, the re-
search emphasis has shifted from environmentally
and biologically persistent compounds such as DDT
to less persistent ones such as the carbamates. The
hazards of the former types of compounds are fairly
well known and regulatory measures have initiated
a shift away from their manufacture and use in the
U.S.
Contaminants in drinking water are also ad-
dressed on an individual pollutant basis. Health ef-
fects and control technology research are designed
to furnish health data for deriving maximum per-
missible contaminant levels in water supplies and
providing available technology to help insure that
the levels are not exceeded.
Control technology research is conducted with
respect to air pollutants as well as those in water.
The approach in control technology research is ori-
ented toward mining, manufacturing, and industrial
processes as examples. EPA has a research pro-
gram focused specifically on the environmental im-
pacts of various processes and technologies for pro-
ducing energy. Another example of a technology-
based approach lies in the research being conducted
on means of treating, disposing of, or re-using in-
dustrial and municipal wastes. The relationship of
this approach to the pollutant-by-pollutant ap-
proach can be illustrated in research being con-
ducted on waste disposal or re-use. An important
aspect of this research is the identification of indi-
vidual contaminants and their possible hazards to
determine if the alternative methods of disposal or
re-use must be modified.
The regional approach can be typified by a long-
term study on regional air pollution which has fairly
recently been completed. A variety of individual
pollutants were measured in the St. Louis, Missouri
area with a view toward understanding their behav-
ior in the atmosphere over that region. A sequel
program has been initiated which concentrates on
the long-distance transport and transformation of
sulfur oxides. Other examples of the regional ap-
proach include projects which will contribute to ef-
fective land-use planning and river basin manage-
ment. Studies on pollution problems in the Great
Lakes region are an important part of EPA's re-
search effort. This research includes modelling of
the Lakes and developing capabilities for predicting
the ultimate impact of pollutants on the Lake eco-
systems.
Perhaps the most timely example of how the indi-
vidual pollutant, industrial or technology, and re-
gional approaches must and can be effectively com-
bined is in toxic substances research. There is a
clearly emerging awareness in the United States of
the need to understand the health and environmen-
tal consequences of the thousands of chemicals
present in and annually entering the environment. It
is impossible to achieve a scientific understanding
of the hazards of each compound and its degrada-
tion products. But a thorough knowledge of individ-
ual yet representative compounds can yield impor-
tant predictive clues for similar compounds and
give some bases for establishing priorities for re-
search as well as control. It is also important to un-
derstand the complex effluents from various indus-
tries, technologies, or processes with regard to such
factors as dispersion, and qualitative and quan-
titative changes with time in order to determine
where regulation can be most useful. If the behavior
of mixtures of chemicals in different regions is un-
derstood, then the ways and levels in which the
mixtures reach humans can be estimated.
The research tasks posed by toxic substances
pollution are formidable. In simply determining
ecological implications of toxic substances, not to
mention discerning their health hazards, the neces-
sary research will require a sophisticated combina-
tion of approaches. Some of the areas where eco-
logical research will be fruitful and where combined
approaches will be useful can be illustrated in some
of the research needs which have been identified.
1.) A hierarchy of rapid, reliable and low-cost
techniques is needed for screening potential tox-
icities of individual compounds or groups thereof.
2.) Methods are needed for bioassays on-site near
complex industrial or municipal effluents. 3.) Re-
fined and field-validated microcosms are needed for
improved assessments of the effects of either indi-
vidual compounds or complex effluent mixes on dif-
ferent types of ecosystems in different geographic
areas. 4.) Improved ways of detecting the signifi-
cance of complex discharges as well as individual
pollutants to ecosystem dynamics are also needed.
CONCLUSION
The present review, although cursory, is intended
to illustrate how a combination of major approaches
to pollution problems is necessary to comprehen-
sive analysis of the environment and effective envi-
ronmental management. Dialogue about {he best
ways to cope with pollution problems can be of
great value — indeed, many volumes have been
written on the subject and more can be expected in
the future. The subject is an exciting one, and
surely will continue to stimulate the thinking of
many people in many countries.
20
-------
CIRCULATION OF CARCINOGENS IN THE ENVIRONMENT
[O Tsirkulyatsii kantserogenov v okruzhayushchey srede]
L. M. SHABAD
World-wide attention is focused on the problems
of environmental contamination. However, there
are comparatively few who realize that carcinogen-
ic substances, that is, substances which produce
cancer or malignant tumors compose part of this en-
vironmental contamination.
The concept of chemical carcinogens originated
on the basis of observations of so-called occupa-
tional cancer. These observations led to the con-
clusion that malignant tumors can arise in man as a
result of certain chemical agents in the environ-
ment. Many experimental studies have established
chemical carcinogens. Chemical carcinogens have
been found among many different chemical com-
pounds, but primarily among the polycyclic aromat-
ic hydrocarbons (PAH), nitroso compounds (NC),
cyclic amines (CA), and aminozao compounds
(AAC). The chemical carcinogens can be divided
into four groups:
1. Highly carcinogenic — Eliciting tumors in man
and animals; 2. Strongly carcinogenic for laboratory
animals — potentially dangerous for man; 3. Weakly
carcinogenic; 4. Doubtfully carcinogenic.
The primary method for determining and eval-
uating the strength of the carcinogenic action of
chemical substances is tests on animals; mainly,
mice, rats, and hamsters. Currently, rapid methods
are being developed which are based on the muta-
genic effect or other changes in the biological and
biochemical systems. A number of physicochemical
methods are used in the search for, and quantitative
analysis of, known carcinogenic substances in dif-
ferent complex mixtures, for example, in con-
tamination of the atmosphere, soil, and reservoirs.
Thus, for example, in the USSR accurate methods
have been developed for quantitative spectral fluo-
rescent analysis of PAH on the basis of the
Shpol'skiy effect. Shpol'skiy and his colleagues
(1952) showed that PAH in normal paraffin solu-
tions at low temperatures (77°K) are distinguished
by quasilinear luminescent spectra. On the basis of
these observations A. Ya. Khesina formulated a
technique in our laboratory which is convenient for
quantitative analysis of PAH in extracts of various
industrial products, dusts, soots, soil samples, res-
ervoir contamination, exhaust of internal com-
bustion engines and other media.
Using these methods, for about 25 years our labo-
ratory has been studying carcinogenic PAH in the
environment. This report is based mainly on the re-
sults of studies by the author and his colleagues
over this period.
The primary sources for chemical carcinogens
which enter the environment are heating systems;
certain industrial enterprises; and engines of cars,
aircraft, and river and sea vessels.
Among the PAH, the most widespread in the en-
vironment is benzpyrene (BP).
BP — PAH indicator. As a result of a large num-
ber of studies, we are convinced that BP can be con-
sidered an indicator substance for the entire group of
PAH, since in those areas where BP is found, as a
rule, there are also a number of other PAH. Among
these BP is, however, the strongest carcinogen.
Our recent studies on various PAH in automobile
exhaust (Shabad, Smirnov, Khesina, Hunigen, el
al., 1977) indicated that there is a correlation be-
tween their presence and the BP content. Knowing
the quantity of BP found in the samples one can
with sufficient reliability judge the level of a number
of other PAH.
Generally, the same number of tumors were ob-
tained. In a comparison of the results from smearing
the skin of mice with benzene solutions of pure BP
and extracts from the soot of automobile (Gurinov
et al., Zabezhinskiy), aviation (Smirnov), or ship
engines (Klubkov) with the same BP concentration
but also containing other PAH.
BP is very stable and universally spread in man's
environment. Its importance as in indicator of car-
cinogenic PAH was internationally acknowledged
at the symposium "Atmospheric Contamination
and Cancer" in Stockholm (March, 1977) in a pre-
sentation by Soviet researchers.
Thus, many studies which serve to outline the cir-
culation of PAH can be limited to a determination of
BP.
Contamination of the Atmosphere. It was in-
dicated above that the main sources of environmen-
tal contamination with chemical carcinogens, that
21
-------
is, heating systems, industrial enterprises, and in-
ternal combustion engines, discharge into the atmo-
sphere. A number of carcinogens which are carried
in small dust particles and aerosol droplets can be
transported great distances. A large fraction of
these contaminants are probably neutralized by ul-
traviolet radiation and ozone. PAH of larger par-
ticle size, and possibly other carcinogens as well,
for example, NC, settle on the earth. They are thus
transferred into the soil, water, and plants. In par-
ticular, the territory surrounding petrochemical en-
terprises and airports is subject to soil con-
tamination.
Many researchers have found that carcinogenic
hydrocarbons contaminate urban atmospheres
(Shabad et al., 1949, 1954; Waller, 1952; Kotin,
1954, and others). The quantity of PAH, and in par-
ticular BP, is greater in major cities and industrial
zones than in small population areas and agricultur-
al regions, and in general corresponds to the fre-
quency of lung cancer.
Contamination of Soil. In 1959 we reported (Sha-
bad and Dikun) that definite amounts of BP were
found in soil samples of a large city (Leningrad).
This was also confirmed in the United States (Blu-
mer, 1961) FRG (Bornefif, 1962), France (Mallet,
1962), Czechoslovakia (Zdrazil and Picha, 1966),
and by many of our researchers (Shabad, Il'nitskiy,
Kogan, Smirnov, Shcherbak, and Yan'sheva).
Contamination of Reservoirs. A number of stud-
ies (Shabad, Il'nitskiy, et al.) have examined PAH
contamination of reservoirs. The PAH level in res-
ervoirs of agricultural areas is considerably lower
than in those of industrial areas. PAH in rivers is
significantly lower than in large cities, instead of
higher. PAH is accumulated in reservoirs in the bot-
tom sediment, in algae, and in fish.
Contamination of oceans, seas, and rivers with
petroleum products has attracted universal atten-
tion. Even Thor Heyerdahl reported that he found
massive oil contamination in the open ocean. How-
ever, until now there have been comparatively few
systematic studies of the PAH quantity in reser-
voirs and of their sources. As a result of the work of
Borneff et al.. Mallet et al. and our work with
Il'nitskiy and Klubkov, it has become clear that the
main sources of PAH contamination of reservoirs
are: 1. waste water of coke, oil refining, and certain
other industries; 2. torrential run-off of large cities;
3. exhaust of ship engines; and 4. oil residues which
fall into the water during accidents or as a result of
washing of tankers. The presence of PAH, and in
particular BP in the exhaust of outboard boat mo-
tors has been shown by Klubkov in direct tests. It
was found that the boat motor discharged about 500
fjLg of BP per hour into the water. Smearing the skin
of mice with soot from the exhaust pipes of such
motors indicated its carcinogenicity as practically
100%.
The quantity of BP in the water, vegetation, and
bottom sediment of those reservoirs where motor-
boats are constantly circulating is considerably
higher than in those where water transport is forbid-
den. The same recommendations which refer to au-
tomobile engines, in principle, can promote a reduc-
tion in the quantity of BP in the exhaust of boat and
liner engines. The waste water which contains PAH
must be under constant control and be subject to
the appropriate treatment before being discharged
into open reservoirs. The same technological ad-
vancements which reduce the PAH discharges into
the environment can promote a decrease in the
amount of PAH in waste waters. Thus, detection
and study of the sources of PAH enable us to pro-
pose a whole series of measures to promote hygien-
ic prevention of cancer.
The studies we conducted make it possible to
study the unique circulation of PAH in man's envi-
ronment. Carcinogens discharged into the atmo-
sphere circulate, are partially exposed to degrada-
tion, and partially settle on the earth and in reser-
voirs. These substances are then redistributed.
They are accumulated, for example, in the bottom
sediment of reservoirs and are collected by certain
aquatic organisms. Thus, for example, considerable
amounts of carcinogenic BP bioaccumulates in mol-
luscs, for example, oysters and mussels; in fish, for
example, flounder and sturgeon; and so forth.
In the soil BP penetrates into the deep layers all
the way to the subsoil waters and also bioaccumu-
lates in plants. Plants bioaccumulate BP to varying
degrees. Thus, for example, comparatively signifi-
cant quantities of BP were found in potatoes, cer-
tain other vegetables, and in sunflower seeds which
were grown on soils strongly contaminated with
PAH (Grimmer, Fritz, Shabad, et al.). At the same
time no significant quantities were found in grains
(Shabad et al.).
As the carcinogens circulate in the environment
they not only accumulate in certain areas but are
also destroyed. Thus, for example, degradation oc-
curs in the soil and in waste water under the influ-
ence of bacteria (Poglazova, Meysel', Shabad, Khe-
sina); in the tissue cultures, including human fibro-
blasts (Belitstsiy et al.); in animals, for example,
cows (Gorelova and Cherepanova); and in man
(Shabad and Dikun). Degradation of carcinogens is
apparently a consequence of the action of non-
specific oxidases induced by them.
The degradation of PAH in soil generally is not
complete. The degree of degradation is dependent
on the type of attenuating chain reaction and as a
result a certain background level is created. Ac-
cording to our data, based on the study of many
hundreds of soil samples of different regions of the
USSR, the background level can be up to 5 /u.g/kg of
dry soil. This background exists, apparently,
-------
throughout the USSR and, probably, throughout
the planet. This widespread background level is due
to discharges from such sources as aviation, and the
possibility of the natural origin of chemical carcino-
gens, in particular BP.
BP can, for example, be formed from a carbon-
containing substrate in the germination of seeds
(Vettig, Gerbert, Shabad, and Khesina, 1976). As
shown by the results of the expeditions to Kam-
chatka, Sakhalin, and the Kuril Islands (Il'nitskiy,
Belitskiy, Shabad, 1975, 1976), BP was found in the
volcanic ash, lava, soil, vegetation, and thermal wa-
ters of the volcanic zone (volcanoes Burlyashchiy,
Tolbachik, Tyatya, and others), which indicates its
natural abiotic origin. The detection of BP in the
deep layers of the soil in the permafrost zone in-
dicates the existence of this substance throughout
several tens of thousands of years.
As seen from the above, the natural origin of car-
cinogenic PAH, and in particular BP, has been
proven. This partially explains its background level
in the environment. However, it cannot explain
those great quantities of carcinogenic substances
which are observed in the environment of large
cities and industrial areas. According to Weinstein,
in the United States 1300 tons of BP are discharged
into the environment annually, and according to
Suess, about 5000 tons annually throughout the
world. These contaminations depend on the activity
of man and are an undoubted pathogenic and carci-
nogenic factor for the population. As mentioned
previously, 250 or even 25 BP is sufficient to pro-
duce a malignant tumor in laboratory mice.
The etiological importance of chemical carcino-
gens, and in particular PAH, in human carcinogen-
esis has been proven by observations of occupa-
tional malignant tumors and the increase in frequen-
cy of lung cancer in inveterate cigarette smokers, as
well as a number of other, more isolated incidences.
It is sufficient to indicate the sharp increase in lung
cancer morbidity in the twentieth century, particu-
larly in large cities and industrial centers.
The study of carcinogenic substances has opened
a number of paths to the prevention of malignant
tumors. The detection of sources of their discharge
must lead to a perfection of technology and produc-
tion, including technology without waste products,
which would eliminate or at least reduce the dis-
charges of carcinogens into the environment. Hy-
gienic prevention must result in the end or reduc-
tion of human contact with carcinogens.
The most important law is the dependence of the
carcinogenic effect on the dose and duration of ex-
posure. On this basis specific hygienic limits can be
set, that is, MPC (maximum permissible concentra-
tion) or MPD (maximum permissible dose) for car-
cinogens. They are based on a study of carcinogens
in the environment, the results of tests on animals
with different doses of carcinogens, and epidemic
logical studies of specific contingents of the popu-
lation. In the USSR, the MPC's have been con-
firmed for BP in the atmosphere of populated areas
(0.1 /xg/100 m3), working areas (15 /ig/100 m3), and
in reservoirs (0.005 /J.g/1).
Based on this information, one of the main tasks
of monitoring harmful substances in man's environ-
ment is monitoring of carcinogenic PAH. Experi-
mental results show that BP can serve here as an
indicator substance. BP monitoring is of particular
importance since the maximum permissible concen-
trations in different spheres of the environment
have been established for the first time for this sub-
stance.
Work on studying the spread and circulation of
PAH can serve as an example for future research on
other carcinogenic substances, for example, nitroso
compounds. Special attention must be given to the
natural chemical carcinogens, for example, prod-
ucts of mold fungi (alphatoxins and fusariotoxins)
and certain higher plants, such as ragweed, and cer-
tain ferns and palms. The spread of such natural
carcinogens is usually limited; therefore, they are
found only in specific localities. However, it must
be remembered that under some conditions they
can enter food products.
As is apparent from the aforementioned, the pos-
sibility of environmental contamination with chem-
ical carcinogens cannot be discounted. Their con-
trol is an essential part of preventing morbidity due
to neoplasms. The possibility of morbidity pre-
vention at the present time has been proven by ob-
servations of large migrant populations. In Japan
the morbidity of stomach cancer is especially high,
but it is reduced in the Japanese that move to the
United States. The frequency of lung cancer is very
high in England. However, the morbidity of this
type of cancer is diminished in the English who
have moved, for example, to Australia. Thus, it ap-
pears that the morbidity of malignant neoplasms is
variable; it depends on the set of environmental
conditions — atmosphere, diet, and other factors.
Our main task is to reduce the population's con-
tact with chemical carcinogens to minimal doses,
which will make it possible to "postpone" malig-
nant tumors to the 120-150-year-old age group, that
is, prolong cancer beyond the limits of the longest
human life.
REFERENCES
1. Vettig, K., G. Gel'bert, L. M. Shabad, and A. Ya. Khesina,
1976. "Analysis of benzpyrene in plant seeds before and after
germination." Vopr. onkol. 22(12):Jl-54.
2. Il'nitskiy, A. P., G. A. Belitskiy, and L. M. Shabad, 1975.
"Carcinogenic polycyclic aromatic hydrocarbon benzpyrene
in volcanic discharges." Doklady Akademir Nauk SSSR
225(1):214-216.
23
-------
3. Poglazova, M. N., el al., 1971. "Metabolism of benzpyrene
by microflora of various soils and individual species of micro-
organisms." Doklady Akademir Nauk SSSR 198(2): 1211.
4. Poglazova, M. N., el al., 1972. "Destruction of benzpyrene
by microorganisms in waste waters." Doklady Akademir
Nauk SSSR 204<1):222.
5. Shabad, L. M., 1966. "Question of hygienic standardization
of carcinogenic substances." Gig. i san. 11:18.
6. Shabad, L. M., 1973. O tsirkulyatsii kantserogenov v okru-
zhayushchey srede [Circulation of carcinogens in the environ-
ment]. Moscow: Meditsina.
7. Shabad, L. M. and P. P. Dikun, 1959. Zagryazneniye atmos-
fernogo vozdukha kantserogennym veshchestvom 3.4-benz-
pirenom [Contamination of the atmospheric air with the car-
cinogenic substance 3,4-Benzpyrene], Leningrad: Medgiz.
8. Shpol'skiy, E. V., 1959. Uspekhi fizicheskikh nauk. 68:51.
9. Shpol'skiy, E. V., A. A. Il'ina, and L. A. Klimova, 1952.
Doklady Akademir Nauk SSSR 87:935.
24
-------
A CRITIQUE OF MEASUREMENT METHODOLOGIES AND DATA
ANALYSIS, STORAGE AND REUSE
HERBERT L. WISER, Ph.D.
We have the capability to measure and identify
trace constituents in air, water, soil, tissue, and liv-
ing organisms to many orders of magnitude greater
sensitivities than heretofore; e.g., measurements to
a few parts per trillion permit identification of or-
ganic compounds such as chloroform in drinking
and surface waters. Does this discovery mean that
man is beginning to pollute his drinking water sup-
ply with chloroform, or does it mean that chloro-
form has always existed in those waters but we
were unable to detect it at those small concentra-
tions? Or are there other explanations? Whatever
the explanation, those chemicals, now detectable,
known to be toxic, perhaps carcinogenic, must now
be monitored in water supplies throughout the Na-
tion. Thus there is another parameter to be mea-
sured frequently and in many locations and the in-
formation stored for future trend analysis.
In air, photochemical oxidants are cause for con-
cern. It is insufficient to measure only the oxidant
concentrations; we must also measure precursors,
solar radiation flux, and other interdependent pa-
rameters in order to adequately understand the pro-
cess from source(s) through transformations to ulti-
mate impact in order to determine what control ac-
tions will be effective. This requires measurements
at frequent time intervals of additional parameters
and storage for analysis.
So we have arrived at a phase in our tech-
nological society where rapid response instruments
and automation systems are providing large quan-
tities of data at vast rates. Perhaps use of computer
storage is necessary for immediate R&D purposes,
but it raises questions of data storage facility in-
undation and future data utilization.
The Remote Air Monitoring System (RAMS) of
the Regional Air Pollution Study in St. Louis
(RAPS) has collected, over a three year period, data
with automated instrumentation for a number of
pollutants (plus supportive meteorological data)
over a vast network covering the St. Louis Region.
The RAMS system collected ambient data from
25 stations on a minute by minute basis, 24 hours
per day. These computer-collected minute values
represent averages based on 120'/2-second in-
strument readings. Each of the 25 RAMS stations
produces a complete data record from its 2530 in-
struments every minute. Each data record contains
the station identification number (101-125), the data
record time including year, day, hour, and minute,
the minute average values, and the status words for
that minute. These data records are transmitted
from the RAMS stations to the central computer.
During each 24 hour period there is the potential
for 36,000 RAMS data records (25 stations x 1,440
minutes/day). A full data tape for every day was
sent to the RTP Environmental Sciences Laborato-
ry for analysis. Was the measurement frequency of
one reading per minute for each parameter at every
station necessary? It was believed necessary for re-
search purposes by modelers and theoreticians at
the time the RAPS experimental program was de-
signed. I shall now present an example of a comput-
er data bank usage for another primary purpose.
STORET, a storage and retrieval data bank, now
contains about 40 million data bits from more than
250,000 water quality sampling locations through-
out the United States. This data is collected at vari-
ous frequencies depending on location and purpose,
following recommended and established sampling
and analytical methods. 95% of the observa-
tions involve only 17% of the 1700 stored parame-
ters. The seldom used 83% of the parameters, rep-
resenting 5% of the data, may seem like a waste of
computer capacity and cost. Is therefore such a
large storage and retrieval system necessary? The
users overwhelmingly shout "YES," because por-
tions of the seldom used 5% of the observations are
considered necessary for individual users' analysis
and decision-making. The users are primarily those
who input data, water quality assessment analysts,
water quality control managers and policy and deci-
sion-makers.
The sensible solution appears to be to retain and
store in the modern library, i.e., in computerized
data banks, all good data that might be needed by
other scientists, environmental assessment and
quality managers, or policy decision-makers at
some future time. The problem arises in defining
"good" data as well as identifying other valuable
information to be stored. The person obtaining the
experimental data of course believes his data, as
25
-------
well as his experimental methodology used to ob-
tain the data, was good. A user's definition of good
quality data, however, depends on the user's need
and objectives. A question frequently asked is,
"Why isn't computer stored data used?" It is and it
isn't. Researchers may not, but environmental quality
analysts and managers do make use of available
stored data. Environmental quality managers use com-
puter-stored data extensively for following pollution
trends, estimating future pollution trends, studying
the improvement of environmental quality after
specified pollution control measures were taken, us-
ing baseline and trend data for decision-making on
plant siting, etc.
A major issue against using computer-stored data
has been the quality of the data as perceived by crit-
ics of the system. The retriever's main concern is
for data quality, hence his or her doubt, hesitation,
and perhaps non-retrieval of data that is unfamiliar.
Perhaps the reason the fears regarding the quality of
the data have not been put to rest is that quality
perception depends on the observer. "Beauty is in
the eyes of the beholder." Actually, different re-
quirements among user classes is the reason. A sur-
vey indicated that primarily those who put data into
the bank tend to retrieve their own data and closely
related data stored by others and admit that pooling
of data contributes significantly to the value of the
data base. Research requirements for precision, sen-
sitivity, and certain parameters in use of stored data
often exceeded the precision, etc. of the intended
use and therefore the quality of the stored data. But
environmental quality assessment and management
persons appear to be satisfied with monitoring data
of the quality stored in computers.
Environmental quality assessment functions are
concerned with all aspects of collection and analy-
sis of samples and the reporting on the finding and
trends of such analysis. STORET is widely used as
a reliable tool in compiling, presenting, and summa-
rizing data in such a manner that the user technical
staff uses its time for analysis. In actual practice,
the easy accesses to previously collected and ana-
lyzed data serve as bases for new analyses of prob-
lem areas and even suggestions for improvement.
Environmental quality management is applied
technology and as such, is supported more by expe-
rience than by scientific procedures. Data-handling
techniques and data requirements for management
activities are much less precise and rigorous than
for assessment activities.
The management of pollution control programs is
essentially the responsibility of those charged with
making decisions. Knowledgeable analysis of data,
especially of trends within a computer data bank,
may provide insight into progress being made by
pollution control programs.
Again addressing data quality, major deficiencies
with respect to quality of the data are frequently of
mechanical origin; that is, errors caused during data
input process. For example, a large percent of er-
roneous values in the station identification data and
station type codes have been encountered. Such
deficiencies detract significantly from the usability
of the data.
The quality of parametric data stored is relatively
good. Individual input/users have borne the respon-
sibility for maintaining the quality of the data.
Sometimes their back data already in the system is
upgraded. It should be noted that the data stored was
secured for particular studies, giving little thought
to possible use of the information by others for dif-
ferent purposes.
More often, a paucity of data was a greater prob-
lem to the researcher. The level of precision for the
applied areas of water quality management was not
as critical a factor as the availability of data. If no
data or insufficient data is readily available, deci-
sions (permits, etc.) that must be made are made
without the benefit of data.
Information is usually stored in computer data
banks in the raw or corrected numerical value for
that parameter as measured by an acceptable or
preferred analytical method. However, error bars
are usually not included. This may be a drawback
for other users of the data in the computer. This is
especially the case where long term trends change
slowly. The confidence in stored data is further di-
minished when instruments are changed and not
noted in the computer.
Generally, though, I believe there is ineffective
reuse of stored data because of lack of knowledge
that the bank exists and what is in it, inadequate
data, inadequate quality (whether caused by errors
in coding or by inadequate experimental methodol-
ogy).
Operation of a computer data bank is primarily a
data management function that should be designed
for programs and activities it is assigned to support.
It does not have a mission of its own. 1 believe im-
proved systems optimally designed for user classes
will evolve, as more potential users become more
educated.
In closing, I shall present two examples where
STORET was valuable for water quality manage-
ment decision-making. Similar experiences have
occurred in air quality management practices.
1. Low pH's measured in the St. Mary's River Ba-
sin, Georgia, were attributed to natural background
and, therefore, were asked to be considered as a
basis for revision of standards. STORET retrievals
for data spanning many years were run for pH data
for the total region. An analysis indicated that while
there were some low pH values, they occurred only
during January. Thus the U.S. EPA was able to
hold the line and not make an exception. Since stan-
26
-------
dards are not applicable during times of natural
stress, if the low pH had been natural during the
whole year, the standards would have had to be
changed. If the standard had been lowered, the area
might have been considered a prime area for indus-
try to locate because of an existing exemption to
standards.
2. A case where an exception to a standard was giv-
en involved a long range project covering permits,
basin plans, and standards. A plant's effluent was
polluting a river under its normal operating condi-
tions, believing it was lowering the BOD but not in a
severe manner. A mathematical model based upon
data stored in STORET, showed that almost no dis-
solved oxygen could be maintained in a stream at
the 7-day, 10-year low flow at corresponding maxi-
mum temperature when a certain plant discharged
the BOD allowed under the issue guidelines. How-
ever, the plant had the capacity to hold 51 million
gallons of effluent. The model predicted that with-
holding discharges during periods of low river flow
and discharging during periods of high river flow
was a feasible solution based on the data available
in STORET. Since stored data covered the period
from 1965 to 1972, there was sufficient confidence in
the model to grant an exception. The management
of plant effluent has followed this model. The com-
pany has continued to monitor, and data will be in-
put to STORET to determine the prediction's accu-
racy. The alternative to this decision, in the absence
of historic data, would have been to close the plant
in order to save the river.
Thank you for the opportunity to address you.
27
-------
BASIS FOR THE PRIORITY LIST OF SUBSTANCES FOR
MONITORING ENVIRONMENTAL CONTAMINATION AND THE
ROLE OF SYNTHETIC ORGANIC SUBSTANCES
[Obosnovaniye prioritetnogo perechnya veshchestv dlya monitoringa
zagryzneniya okruzhayushchey sredy i rol'sinteticheskikh organicheskikh
veshchestv]
F. YA. ROVINSKIY
In the development of a program for observation
of the contamination level in areas of the environ-
ment, considerable complications arise in the for-
mation of a list of priority indices for environmental
quality.
Two fundamentally opposite tendencies are the
basis of the problem. One tendency is the constant
need to expand the list of observed substances. This
is related not only to an expansion of the list of dis-
charged substances, but also to the continuous
growth of knowledge in the area of the composition
of discharges and their toxicity, and to the growth in
the number of standardized indices. In the USSR
over 150 substances are now standardized for the
air of cities, over 500 substances for reservoir wa-
ter, and over 60 substances are standardized for
fishing criteria. Conversely, the list of observed
substances cannot expand endlessly, since this
would be an expensive and inoperative system.
Time will only compound this problem.
Special attention should be given to synthetic or-
ganic substances, many of which are not inherent to
nature. Nature is not able to degrade these, which
leads to their continuous accumulation in the envi-
ronment.
In the future the solution to this problem may be
reached in great extent by the introduction of com-
plex indices of environmental quality, that is, types
of hydrobiological indices for water quality. The
use of such indices is limited. By using such indices
the answers to the question of whether the quality
of a given environment is satisfactory or unsatisfac-
tory, can be obtained, but it is doubtful whether an
answer can always be obtained to the question of
what the exact reasons for the emergence of the
trouble in a given area of the environment are. Be-
cause this question cannot be answered, the mon-
itoring system cannot be used to make recommen-
dations for the elimination of the causes and con-
sequences of environmental contamination.
Compromise approaches are the current solution
to the given problem. Considering the main tasks of
the national monitoring system in the formation of a
priority list of observed substances, it is expedient
to start with the prevalent substances, their tox-
icity, and their significance in local discharges. Var-
ious approaches are possible, of which the follow-
ing are the primary ones.
First, substances which are discharged in large
quantities and therefore cause widespread con-
tamination of the environment must be studied.
Characteristic representatives of this type sub-
stance are, for example, sulfur dioxide, dust, car-
bon monoxide, and certain other urban atmospheric
pollutants; petroleum products, phenols, deter-
gents, and other water pollutants; and pesticides in
soils and sediments.
Second, the most toxic substances must be stud-
ied, that is, those which have low MPC (maximum
permissible concentration) values.
An analysis of MPC tables indicates that most
substances with air quality standards have a MPC
of 5-100 jig/m3. However, a number of toxic com-
pounds of arsenic (except hydrogen arsenide), lead
sulfide, hexavalent chromium, as well as a number
of organic substances, for example, acetophenone,
hexamethylenediamine, mesidine, meta- and para-
chlorophenylisocyanate, and styrol. There is also a
small quantity of substances with a MPC less than 1
/xg/m3, for example, metallic mercury (0.3 Mg/m3),
lead and its compounds (0.7 /xg/m3), and methyl-
mercaptan (0.009 /ig/m3).
A similar pattern also occurs for reservoir water.
Most standardized substances have a MPC from 100
to 1000 /ug/1, however, many toxic substances have
an MPC of 1-2 fj.g/1. These include inorganic com-
pounds of selenium and mercury; a number of organic
substances such as Aldrin, ortho- and paradichloro-
*Here and further the MPC values are given converted for
metals.
28
-------
benzene, methylbenzoate, sodium butylxanthate,
phenol, thiophos, and phozaJone. A small number of
substances, for example, salts of beryllium, diethyl
tributyl tin methacrylate, have a MPC within 0.1-0.2
fig/\. And finally, for a large number of inorganic
and organic substances such as sulfides, active
chlorine, Altax, benzopyrene, mercaptobenzothia-
zole, tetraethyl lead, and certain pesticides (Simazin
and its hydroxy compound), their "absence" in wa-
ter has been established as the norm. According to
fishing standards, the "absence" has been estab-
lished for a significantly wider range of substances,
including, for example, DDT and a number of other
pesticides.
Developments in the area of MPC establishment
are not the only reason for the constantly increasing
list of toxic substances. Scientific studies are con-
tinuously expanding the range of substances known
to have carcinogenic and mutagenic action. If one
considers that the joint presence of two or more
substances often increases the toxicological action,
then the list of substances subject to control can sig-
nificantly expand even now, and will tend to further
expand as advances are also made in this area.
Third, observations must encompass substances
which are known to be present in the discharges or
effluent in a given region. It is not obligatory for
these substances to belong to the aforementioned
groups; however, they must have fairly great local
importance in order to implement their control.
Monitoring of the common substances is to a sig-
nificant degree a solved problem; the list of such
substances has been formulated, methods are avail-
able; therefore, here the main efforts are directed
towards the perfection of monitoring, and in partic-
ular, its automation.
The monitoring of toxic and locally significant
substances requires great attention. Assuming that
this category should include substances with very
low MPC values, the existing list of MPC's has been
analyzed to select substances for toxicity criteria.
However, this criterion alone is not sufficient. The
toxic substance should be included in the mon-
itoring program only if it is present at the given ob-
servation point.
In addition to toxic substances, there are sub-
stances which are considerably less toxic, but have
fairly important local significance, and therefore are
included in the priority list of observations in a giv-
en region. Thus, for example, the presence of non-
ferrous metallurgy industry requires the monitoring
of such metals as lead, copper, zinc, tin, and others;
the presence of oil refineries requires monitoring of
a number of organic substances (alphamethylstyrol,
acetaldehyde, pyridine bases, synthetic aliphatic
acids, formaldehyde, and others.)
We will examine one of the possible approaches
to an evaluation of the priority of contaminating
substances. It will be examined in the example of
materials in the network monitoring contamination
of water, land, and urban air. For these purposes
we introduce the priority coefficient:
where Q and Cj are the annual average concentra-
tion of i- and j-substances in the air, respectively.
MPCi and MPCj are the maximum permissible
concentrations corresponding to the i- and j-sub-
stances, respectively.
This coefficient simultaneously takes into ac-
count the toxicity and relative contribution of the i-
substance in the contamination of water or air in the
given region. Here selecting the correct reference
substance is important. The following main criteria
are vital in the selection process: sufficient preva-
lence and reliable monitoring.
Petroleum products satisfy these conditions as
reference substances for water, land, and urban at-
mospheric dust.
Analysis of the data on air pollution in the cities
made it possible to obtain the following series of ft
coefficients:
phenol
hydrogen nitrogen ammonia sulfur
sulfide dust dioxide gas dioxide
2.8 1 0.9 0.7 0.4
For surface waters the coefficients are arranged
in the following series:
SPAV,
chromium,
petroleum ammonium, zinc,
products DDT iron nitrates copper
1 0.34 0.25-0.27 0.17 0.02
In different regions, the priority series for water,
as well as for air, change somewhat. For each water
area or city, a series of priority indices can be ob-
tained. Depending on the cost of monitoring or oth-
er considerations, this series can be limited to a spe-
cific number of substances which also comprise the
program of observations for the given area. An
analogous approach can also be used for sea water,
soil, and other media.
In analyzing the indicated data one should first
consider that they are derived from the national
monitoring system. The advantage of these data is
the great statistical reliability for the substances
which are measured. At the same time a number of
substances, many of which are synthetic organics,
either are measured in a limited number of water
areas of cities, or have not yet been measured. Nev-
ertheless, if one agrees with the aforementioned ap-
proaches to analysis of the priority of contaminants,
then one can conclude that monitoring of the syn-
thetic organic contamination in water areas, has
high priority.
29
-------
Problems can arise in the creation of a priority list
for monitoring outside the urbanized regions, that
is, background monitoring of regional and base sta-
tions, and biospheric preserves. The criterion of
priority based on MPC in this case loses its meaning
and such properties of substances as prevalence,
stability, ability to migrate and spread great dis-
tances, and finally, the ability to damage the level of
the ecosystems or biosphere as a whole, become of
primary importance. For the global system of mon-
itoring the environment (GSME), the list of sub-
stances is more limited than for the national mon-
itoring systems; nevertheless, such synthetic organ-
ic substances as DDT and other organochloride
compounds occupy common high priority positions
in the program of background monitoring.
REFERENCES
1. Bespamyatnov, G. P., K. K. Bogushevskaya, A. V. Bespa-
myatnova, et al., 1975. I'redvino-dopusiimyye ktmtscnirutsii
vrednykh veshcheslv v vozdukhe i vode [Maximum per-
missible concentrations of harmful substances in the air and
water], Leningrad: Khimiya.
30
-------
ECOLOGICAL APPROACHES TO EVALUATING THE STATE AND
CONTROLLING THE QUALITY OF THE ENVIRONMENT
[Ekologischeskii podkhod k otsenke sostoyaniya i regulirovaniyu kachestva
okryzhayushcheyi prirodnoi sredij
YU. A. IZRAEL, I. M. NAZAROV, L. M. FILIPPOVA, YU. A. ANOKHIN,
V. M. KOROPALOV, A. KH. ASTROMOGILSKIY, and A. B. RYABOSHAPKO
Introduction
The development of a scientifically-based pro-
gram for controlling the quality of the environment
requires an evaluation of those contamination levels
which are considered permissible with respect to
their effect not only on man but also on plants and
animals, that is, biogenesis as a whole. The results
of such an evaluation should be a statement about
exactly what quality of the environment we should
strive to achieve, both from the standpoint of the
direct effect of the action on live organisms and
from the standpoint of the remote consequences
that may arise [8,12].
A system of sanitary-hygienic standards has been
set up to evaluate the permissible level of con-
taminants that act on man, using the sanitary-hy-
gienic standardization practice in the USSR for the
maximum permissible concentration (MPC) of envi-
ronmental contaminants. Small scale studies were
done to determine the effect of maximum per-
missible loads of contaminants upon natural and
anthropogenic ecological systems and the animal
and plant populations comprising these systems.
These studies were only for small population num-
bers which have a professional significance, for ex-
ample, a number of MPC's have been worked out
for fish hatcheries. The work on standardizing the
action of contaminants on forest ecosystems is just
beginning. However, the necessity of developing a
group of studies to evaluate the permissible loads
on natural systems, that is, ecological standard-
ization is already quite obvious.
In contrast to the sanitary-hygienic standard-
ization of the contaminant concentration in the at-
mosphere which acts to protect man from their
harmful effects, ecological standardization is direct-
ed at providing that quality of the environment in
which the normal development and function of nat-
ural and also anthropogenic, ecological systems is
possible. The aim of ecological standardization is
the development of standards for the permissible
concentration of harmful materials in natural media
(atmosphere, natural waters, soils, and also in the
biota) which correspond to the permissible stress
reaction of the natural systems. These ecological
standards should be the basic quality criteria for the
environment and should be used in a system of eco-
logical monitoring to evaluate its state under condi-
tions of anthropogenic influence. The value of the
permissible load on the ecosystem should be deter-
mined by a system of permissible concentrations of
the contaminants, for which the individual values
may change against a background of the load al-
ready present in the given natural system.
The development of only single ecological stan-
dards for the quality of the environment is, how-
ever, not sufficient. The final and most important
step in ecological standardization is the develop-
ment of ecologically-based norms for the per-
missible discharges for contaminants on a local, re-
gional, and global scale, using the data from the first
step. The observation of these norms, by a con-
trolling-monitoring system for the sources of con-
tamination, should ensure that the norms for the
permissible action of contaminants and the norms
for the permissible deviation of the ecosystem
structural and functional indices from the back-
ground will not be exceeded. This final step, which
is lacking at this time, is the real basis for lowering
the contamination level of the environment and for
achieving the desired environmental quality [9].
The necessity of studies to evaluate the quality of
the environment to provide for the normal develop-
ment and functioning of ecological systems is deter-
mined by the marked differences in the sensitivity
of man and different representatives from the plant
and animal world to the same actions. This dif-
ference in sensitivity explains why sanitary-hygien-
ic standard for the MPC which protects man from
31
-------
the action of some harmful substance is often not
sufficiently rigorous to protect some plants and ani-
mals, and the ecosystem as a whole. The fact is that
forms which have a very high sensitivity to such an
anthropogenic action may have a deciding role in
the stability of an ecological system, and in this case
even slight actions can have serious consequences
for this ecosystem.
On the other hand, an approach to setting up the
permissible norms for the action on the ecological
system by determining the maximum permissible
load on natural objects should differ from the sani-
tary-hygenic approach, since it does not allow any
action or any damage with respect to an individual
organism. From the standpoint of the protection of
ecological systems, the loss of a single individual is
not dangerous under the conditions that the stability
of the system, its variety, and productivity are re-
tained. Thus, in contrast to the hygienic norms, the
ecological norms cannot be localized or discrete.
In order to develop a system of ecological norms,
two levels of action must be considered for factors
harmful to natural systems: a level of critical action
which begins with the death or irreversible degrada-
tion of the given system or some type of prolapse,
and a level of permissible action which differs great-
ly from the critical level. Thus, two types of indices
must be developed as the result of ecological stan-
dardization: the index of the critical load and the
index of the permissible load on the system. It ap-
pears that the requirement for quality of the sur-
roundings should be different for the unique, natu-
ral, and anthropogenic divisions which were sug-
gested [9] for ecological systems. Thus, for unique
natural objects, for example, national forests, in
which the main problem is to retain the genetic
stock of the ecosystem, none of its components can
experience an anthropogenic action which exceeds
the permissible load, that is, a level of relative
harmlessness. In the case of natural ecosystems
which are not unique natural objects, the critical
load may be approached for individual populations,
and as a consequence, certain populations which do
not have a particular ecological or economic value
may be replaced by others. For anthropogenic eco-
logical systems, for example, cities, navigable riv-
ers, canals, and agricultural areas, it is sufficient to
retain only those populations which are necessary
to man. Therefore, critical loads are permissible in
this case on the forms and elements in the ecosys-
tem. It is understood that the MPC for man must
always be observed.
In light of these ideas, the question arises as to
which levels (intensities) of the responding reac-
tions of the biota should be made the basis for the
ecological norms for permissible actions. The norm
for the permissible concentration of the con-
taminant obtained by the sanitary-hygienic ap-
proach differs greatly (sometimes by several orders
of magnitude) from the LCM value which also is not
applicable here. There have been a number of stud-
ies in which Kml = 0.1 - 0.05 is introduced as a
safety factor in the evaluation of the ecological
norm for certain populations, that is, the per-
missible concentrations of the contaminants pro-
posed by the authors which act on these popu-
lations should be 10-20 times less than the LC50 val-
ue. Such an approach, on the whole, should be
considered reasonable, beginning at a certain eco-
logical level. At the lowest levels of the nutrition
chain, the critical and permissible loads may be
identified if the population at this level is not
unique.
Thus, the following requirements can be formu-
lated as the basis of the ecological norm:
a)
r Y,(R)nn „ m„(R)dR ^
Kml(AXso)
m
for each population m, beginning with the level mo;
Y,(R)
and b)
)n
can reach unity for a very limited number of orga-
nisms in each population. Here, Y, is the concentra-
tion of the ith ingredient, nm(R) is the standardized
distribution of the organisms, m- is the population
in the area, and (AK<£)m is the critical load (concen-
tration) on the population.
It is apparent that the ecological norms for the
concentration of harmful substances in the environ-
ment will differ greatly in many indices from the
sanitary-hygienic norms because, as shown pre-
viously, the sensitivity of man and the animal and
plant population to the anthropogenic action differs.
Thus, it follows that the ecological norms cannot be
the same for any type of ecosystem or for any phys-
ical-geographical conditions.
Let us discuss an extremely important fact. The
present-day sanitary-hygienic norms determine the
maximum permissible concentrations for man of the
harmful materials in individual media, for example,
the atmosphere, and potable water. They do not
take into consideration the effects of chemical and
biological accumulation of the harmful materials in
nonpermissibly high concentrations as the result of
their transmission into a different medium, for ex-
ample, from the atmosphere into water, and from the
water into the biota. They also do not take into con-
sideration their accumulation in food chains or their
conversion to more toxic forms during migration. The
sanitary-hygienic norms in their present form are
basically the same for city conditions, for populated
places, and for areas of water consumption, that is,
for those objects in which the secondary natural
32
-------
processes are subordinate. In evaluating the action
of harmful substances on the ecosystem, it appears
that those factors which are not of great significance
in sanitary-hygienic standardization acquire an im-
portant role, and even a deciding role, in ecological
standardization. It also appears that it is not the ini-
tial concentration of a harmful substance in a medi-
um that has the primary effect, but its transmission,
accumulation, and transformation in critical parts of
the ecosystem which result in an altered concentra-
tion for the substance and the appearance of sec-
ondary products with other toxic properties.
In light of the foregoing, a set of studies on the
nature and rules for the propagation, accumulation,
destruction, and transformation of contaminants in
ecosystems and, their conversion from one medium
to another on local, regional, and global scales, be-
comes very important in ecological standardization.
Only the study of the fate of a contaminant from the
source of its discharge through its physical, chem-
ical, and biological transformations and reactions
with other varied factors in the environment to the
point at which it enters a living organism and acts
on it, can help develop scientifically-based ecologi-
cal standards for the permissible action on the living
component of the biosphere.
Most harmful substances are present in the envi-
ronment in the form of complex mixtures and com-
pounds containing a large amount of different in-
gredients. Thus a comprehensive analysis of the be-
havior of harmful substances must be made a
requirement. This requirement should include a
study of their combined action on all elements in
nature—plants, land and water animals, and the abi-
otic component of the biosphere—in order to deter-
mine the values of the maximum permissible loads
on natural objects (systems) as a whole [10].
Our experiments and those of other researchers
have shown [1], that the possibility of harmful sub-
stances accumulating in significant quantities in in-
dividual zones of the natural medium and being con-
verted into new, more toxic forms as the result of
geophysical, geochemical, and biological processes
is a deciding factor in standardizing their action on
natural systems. This work is devoted to the study
of this aspect and can be illustrated by using such
contaminants of the environment as sulfur dioxide,
mercury compounds, and pesticides as examples.
When sulfur dioxide and mercury compounds are
discharged from elevated sources, for example,
high smokestacks, they are spread over large dis-
tances (hundreds and thousands of kilometers),
contaminating large masses of air, making their at-
mospheric transmission become an international or
even global problem. Although the concentration of
these substances in the air near the surface rapidly
drops to permissible levels (according to existing
MPC norms) because of atmospheric scattering and
conversion processes which take place, a careful
analysis of their migration and conversion chains in
natural media, taking into account their secondary
accumulation and possible action on living orga-
nisms, leads to unexpected conclusions. Thus, con-
centration levels of these atmospheric substances
which are several times lower than the MPC norms
may not be permissible from the standpoint of eco-
logical systems. A similar conclusion was arrived at
from the study of the migration processes of pesti-
cides in the environment.
The following examples are set up on the basis of
a single scheme in which the source for the in-
troduction of sulfur dioxide, of the mercury com-
pounds, and of the pesticides, respectively, is ana-
lyzed. The migration paths and the conversion of
these substances in the environment are followed,
and the values of the permissible concentrations
which arise in individual natural media are eval-
uated.
I. Statement of the Problem
Let us consider some ecosystem (biogenesis)
consisting of N components (subsystems). An indi-
vidual population, a community, parts of a geophys-
ical medium, etc., may be the components of the
ecosystem.
The following notations will be used:
Xi is the vector of the state of the ith component
of the ecosystem.
Cy is the concentration of the jth harmful sub-
stance in the i,h component of the ecosystem
(i = 1, • • •, N; j = 1 ¦ ¦ • m).
The number of components N of the ecosystem,
that is, the dimensions of each of the vectors Xi, is
determined by the complexity of the ecosystem
being considered and of each of its components,
and by the degree of detail which is desired.
The concentration of the harmful substances
could have been included as the abiotic component
of the vector of state Xt. However, for our pur-
poses, it is convenient to consider only those con-
taminants whose concentration in the ecological
system is due, either mainly or completely, to an-
thropogenic activity. Examples of such con-
taminants are pesticides, heavy metals, sulfur diox-
ide, petroleum products, and others.
Let:
X = (Xi • • • X„) be the vector of state of the
entire ecosystem;
Ci = (Cu • • • Cim) be the concentration vector
of the contaminants in the i"1
component.
C = (Ci • • ¦ Cm) be the concentration vector
of the contaminants in the
ecosystem as a whole.
33
-------
Given the above, the dynamics for the state of the
ecosystem and the migration and conversion of the
contaminants may be described by the equations:
X, = F.(X, C, Q)
Cu = u de-
scribes the concentration of the jth contaminant in
the ilh component of the ecosystem due to exchange
processes among the components and the decompo-
sition and chemical conversions of the con-
taminants.
Q = (Q, • Qn) is the vector for the external
actions, not related with the
contaminants.
This may be either natural actions or anthropo-
genic actions, for example, a change in temper-
ature, water content, rate of catching fish, cutting
and planting trees, and other varied actions.
E( = (En • • • EiM) is the anthropogenic in-
troduction of contaminants into the ith component
of the ecosystem. The contaminants may enter the
atmosphere, the soil, or the natural waters as a re-
sult of anthropogenic activity.
The function /3|(Xi) takes into account the change
in the concentration in the ith component per unit of
mass of the contaminant which enters the com-
ponent. The harm done with respect to the ith com-
ponent of the ecosystem, will designate by D|(Xi,
Ci), and is determined both by the current state X|
and the concentration of the contaminant Ci in it.
For example, let us assume that the ith component is
an ecosystem of some water reservoir and that
there is an anthropogenic introduction of con-
taminants into this reservoir. This ecosystem is
harmed by various factors, for example, eu-
trophication, different types of deposits. In addi-
tion, the quality of the water can be lowered from
the standpoint of future community or industrial
water supplies. The possible approaches for esti-
mating the harm because of the action on the sur-
rounding medium are given in [8].
The problem of standardization consists of deter-
mining those actions whose damage to each com-
ponent of the ecosystem does not exceed a given
value Dt, that is,
D,(X,, Ci) s D, (1.2)
The contamination of the environment is related
to the fact that, at the present time, the production
of the majority of useful products is unavoidably as-
sociated with the formation of wastes which enter
the environment. The amount of these wastes de-
pends, first of all, on the production technology.
For a given technological process, the amount of
waste is almost proportional to the amount of the
product which is produced. Therefore, for a given
technology, the discharges {Et} and the action Q are
closely related to the profitability of the product
that is produced. Let V(Q, E, • • EN) be the profit
function for the existing technology. Our problem is
to maximize the profitable effect, that is,
f T
V(Q, E, • • • EN)dt => max (1.3)
. 0
while fulfilling the limitations in (1.2). It is obvious
that:
Eij^O (1.4)
We will use values of X and C, in the absence of
anthropogenic activity, as the starting conditions
for the system (1.1). It is easy to see that for a given
technology an increase in the output of the profit-
able product will be associated with an increase in
the waste and an increase in its introduction into the
environment. Therefore, within fixed limits, an in-
crease in Eij is associated with an increase in V.
Consequently, the function V will be a monotonic-
ally increasing function with respect to each argu-
ment Ejj in some vicinity of zero; that is, for:
0
-------
tion of the contaminant for which harm due to con-
tamination does not exceed the permissible concen-
tration.
In the following, our approach will be illustrated
using such contaminants as mercury, sulfur diox-
ide, and pesticides.
In addition, in all of the cases we will consider the
entrance of the contaminant into a single medium.
We will also assume that the circulation of these
substances can be described by a linear system of
differential equations:
where C = (C,
C = HC + E (1.7)
. CN) is the concentration vec-
tor for the contaminants
in the natural media.
For simplicity, we will only consider the propaga-
tion of a single contaminant; therefore, the second
subscript j will be dropped, a = II An II is the matrix
for the coefficients of transmission between media.
E is a vector, each of whose components is equal to
Ei = K) ¦ /3t • E. Here, E is the entry of a con-
taminant. As in the foregoing, # is a normalizing
factor. Ki is the amount of total discharge which en-
ters into the ith medium. It is determined by the
technology. This will be discussed in more detail for
each of the contaminants.
The discharge E will be regarded as constant;
therefore, only the steady state of the system C° will
be considered. Assuming that the harm is propor-
tional to the concentration, we maximize the values
of E:
E => max
for the limitations:
-a-'KE s CD (1.8)
where CD is a vector whose ith component is the
concentration for which the contamination harm of
the ith medium does not exceed the maximum per-
missible value.
Usually, the maximum permissible concentra-
tions in the given medium will be used for (CD)t-
The problem (1.8) is easily solved by:
E0p = min
(CD)i
(-«-' K),J
(1.9)
Here, the vector K is determined from the condi-
tion E = K ¦ E.
By calculating the value of the concentration Q,
taking into account the value of E0p from equation
(1.9), we get:
C. = (_a_1K)E0p
(1.10)
From (1.10) we can determine the limiting concen-
tration for each medium.
The following contaminants will be considered
consecutively to illustrate this approach: mercury
compounds; pesticides; and sulfur dioxide which is
propagated in such natural media as the atmo-
sphere, natural waters, and soils. The sequence for
considering these contaminants is related with their
most important characteristic, that is, their "life-
time" in the environment.
Thus, the mercury compounds can be retained
(and are retained) in the environment, particularly
in rocks and soils, for an almost infinite period, of-
ten exerting a fatal action on the living component
of the biosphere.
Pesticides, which are also usually stable sub-
stances, nevertheless decompose gradually. Typi-
cal values of the "lifetime" for many pesticides are
values on the order of several years.
Finally, the processes for the neutralization of
sulfur dioxide in the atmosphere are characterized
by values on the order of several hours.
As will be shown below, the "lifetime" is one of
the most important characteristics which determine
the specific features of the problem of environmen-
tal contamination by mercury, pesticides, and sul-
fur dioxide.
2. Mercury
The anthropogenic introduction of mercury exists
in all media: the atmosphere, soil, and natural wa-
ters.
Mercury enters the atmosphere as the result of its
direct discharge during the combustion of fuels, for
example, coal and oil; the roasting of ores; and in
the direct production of mercury.
TABLE l
Type of activity
World
Entry into
Total
Medium
production,
atmo-
entering
(main
lCt/yr
sphere,
t/10"t
entry)
Combustion of
3 • 103
2
6000
Atmosphere
solid fuels
Combustion of
2 • 103
1
2000
Atmosphere
oil
Treatment of
—
—
1000
Atmosphere,
ores
hydrosphere
Application of
Soil,
chemical
100
4
400
atmosphere
fertilizers
(pesticides)
The sources for the modern introduction of mer-
cury into the environment during production pro-
cesses and from the use of mercury in the basic
branches of industry are shown'in Table 1.
The data given in Table 1 can be used to set up
ideas about the order of magnitude of the discharges
of mercury into the environment and about the rela-
tive distribution of these discharges among the geo-
physical media. We can see that the main part of the
35
-------
TABLE 2. MODEL PARAMETERS
No.
Parameter
Notation V alues used
1. Time for extracting mercury from t, 3- 10~* yrs
the atmosphere
2. Time for extracting Hg from soil r,a 360 yrs
into the atmosphere
3. Time for extracting Hg from soil r,v 850 yrs
into surface waters
4. Time for extracting Hg from rwl) 35 yrs
surface waters into the ocean
5. Natural sources of Hg in soil qN 6-3-103T/yr
6. Fraction of Hg deposited on dry f 0,8
land
7. Fraction of anthropogenic Hg
included in global cycle
Mercury concentration in the
absence of anthropogenic sources
8. In the atmosphere
9. In the soil
10. In surface waters
0-1
Qa°
QS
3 10*T
3,6 ¦ 10"T
1,5 • IV T
used to evaluate the state of the media, that is, the
atmosphere and natural waters. We will carry out
the discussion in the following way. In the absence
of anthropogenic sources, the natural concentration
of mercury in the atmosphere is equal to Qa, and
the corresponding concentration near the earth is
Ca — IO B g/m3. If the mercury concentration in the
atmosphere is QA, then the corresponding concen-
tration near the earth is
r ^ caQa.
A Qa '
analogously, Cs = and CB = Cqo">
where C8 and CB are the concentrations of the mer-
cury in the soil and in water, respectively. Thus, the
system of equations (2.1-2.3) assumes the form:
CA = K
Cs = f
EahtCa
Qa°
Cs
Ca + Cs
Ta 7 a
— • ^ (2.4)
C§
-cs
+
(1
qNCs
tsa Ca° ' Qs
TSW }
K)EahtCs0
SA
+
(2.5)
CB = Cs
— Cb IA wo (2.6)
Ql
Cg
eg
Following the approach described above, we will
use Cj&pc = 3.10-7 g/m3 [6] as the limitation on the
concentration in the atmosphere near the earth.
The MPC's have not been established for soils.
Therefore, we will not set any limits on the mer-
cury concentration in the soil.
According to equation 2.3, we will use CUM = 0.5
^g/liter as the concentration which should not be
exceeded in water. We will evaluate the maximum
permissible discharge as:
Ea = min
'MFC
Ql
c?
- a'sqN
CumQb 32
a.i2qs
PO
a"* + al2(l ~ k) ' otuk + aM(l - k)
where a11 are the elements of the matrix A-1. A is
the matrix for the system (2.1-2.3):
Ea = (3 -r 4) • 104 tons/year.
From this the maximum mercury concentration
in the atmosphere near the earth, for which the av-
erage mercury concentration in water is ~ 0.5 /zg/
liter, will be equal to ~ 10-8 g/m3. This is approxi-
mately 30 times less than the established MPC for
the atmosphere.
For soils, the corresponding concentration is
equal to:
D§p = {a21* + a22d - «)} Ea + a22qN
which corresponds to values which exceed the pres-
ently accepted background concentrations in the
soil by only 1.5 times.
However, an important property of the system
being studied should be pointed out. We will call
this "growth inertia." As shown in Figure 1, even
after the discharge into the atmosphere is discontin-
ued, the concentration of mercury in natural wa-
ters, which results from this discharge, will contin-
ue to increase for a long time. For the evaluation of
the maximum permissible mercury concentration in
the atmosphere on a regional scale, we will consider
the model for the circulation of mercury in the re-
gion. The scheme for this model is given in Figure 2.*
The equations for the model are as follows:
CA = -CA|f + f
\ Tv ta
1 — (X
+ Cs + xqAHT (2.8)
CB =
Cs — -Cs
+ -^Ca
Ta
Cs
Tsa
_L + U_
^SA 7"SW
' *)qAHT
(2.9)
1
7"SW
\Tcei
+ +
ceg tBDO/ T A
^CA
Cceg , A _ MtCb _ Cc
. ^ceg _
rWO
tBO
+ (2.10)
(2.11)
'Translator's note: Figure 2 omitted in foreign text.
36
-------
anthropogenic discharges of mercury enters direct-
ly into the atmosphere.
Through numerous studies, ideas have been de-
veloped on the behavior of mercury in geophysical
media; the processes for the conversion of mercury
and its compounds in the atmosphere, soil, and hy-
drosphere; and the paths for its migration between
media [2].
Mercury is released from rocks and soils either in
the process of erosion or as the result of the activity
of living organisms.
Metallic mercury is relatively soluble at almost all
values of oxidation-reduction potential and pH
which are encountered in nature. Its solubility in
water depends greatly on the presence of oxygen in
the water.
The mercury ion can form compounds with both
organic and inorganic substances. The accumula-
tion and conversion of mercury in living organisms
is very important in this.
According to modern concepts, the main product
from the biological metallization by mercury are
mono-and dimethylmercury. If dimethylmercury is
formed, it readily evaporates into the atmosphere
because of its low solubility in water and high vol-
atility. There it decomposes to elemental mercury
due to ultraviolet rays. If the monomethylmercury
is formed, it is leached from the deposits into the
water and then accumulates in living organisms.
The ratio between the amounts of mono- and di-
methylmercury that are formed is pH dependent.
For high values of pH [8-9], almost all of the
methylated mercury is in the form of dimethylmer-
cury. At lower pH levels, the product is monometh-
ylmercury.
Although there are rather good qualitative con-
cepts about the possible conversions of mercury in
natural media, the quantitative characterization of
these conversions requires further study.
Mercury vapors which enter the atmosphere dur-
ing natural processes behave differently in the at-
mosphere from the vapors which enter from indus-
trial discharges. A large amount of industrial aero-
sols have a much shorter lifetime than natural
atmospheric aerosols. Thus, the mercury in indus-
trial discharges, by settling on the coarsely-dis-
persed fraction of the aerosol present in the dis-
charges, falls on the soil near the plant. The lifetime
of this part of the mercury in the atmosphere is mea-
sured in hours [1]. Its amount in the discharges,
equal to (1 - k), depends on the dispersion compo-
sition of the aerosol in the discharges, and it de-
creases as the discharges are purified of the aero-
sols.
The harmful effect of mercury and its compounds
on man's health is due to its toxic, genetic, and te-
ratogenic effects. The cases of mass poisoning be-
cause of the use of fish and other seafoods with high
concentrations of mercury as food and because of
the use of grain treated with alkylmercury prepara-
tions are widely known.
As we can see from Table 1, the main anthropo-
genic introduction of mercury into the environment
is through vapors entering the atmosphere. How-
ever, the main danger of these discharges is not the
poisoning of the atmosphere. Mercury that is de-
posited onto the soil from the atmosphere is washed
from the soil and can enter water reservoirs, where
it is most dangerous. Highly-toxic organic mercury
compounds can be formed in an aqueous medium
through the action of microorganisms, for example,
bacteria and phytoplankton.
Mercury compounds are capable of accumulating
in the hydrobiospheres, particularly in fish. Here
they can present a direct threat to man's health. The
CMpc for fish, which is used in the U.S.A., Canada,
and in other countries, is 0.5 ppm, calculated using
the dry weight. The bioaccumulation coefficient for
fish is at least 103. The mercury concentration in
water for which the CMpc is reached in fish can then
be evaluated at 0.5 ^.g/liter, which is an order of
magnitude less than CMpc = 5 /xg/Iiter for reservoirs
meant for sanitation. Obviously, such concentra-
tions are approached if we take into account the
possibility that part of the mercury is converted into
highly toxic organic compounds for which, as a
rule, the maximum permissible concentration is 0.1
jig/liter [4],
In the calculations given below, 0.5 /xg/liter is
taken as the concentration which must not be ex-
ceeded in water.
A model was developed earlier for the global cir-
culation of mercury [1], which reduces to the fol-
lowing system of differential equations;
Qa = KEaht
ta tsa
(2.1)
Q„ = f 2a + qN
Ta
Ql
rSA
-(-+-) +
\Tsa 1sw/
(1 - K)EAht (2.2)
Qb
Oil
Tsw
Qb
TW0
(2.3)
Qa. On. and Qb arc the total amounts of mercury in
the atmosphere, soil, and surface waters on dry
land. Qb includes not only the dissolved form of
mercury, but also the mercury which is in the form
of suspended particles and that in bottom deposits.
These model parameters were determined from
data on mercury concentrations in natural media,
and are given in Table 2.
In order to evaluate the permissible discharges
according to the above scheme in equations (2.1-
2.3), we will change to concentrations which can be
37
-------
/t*KT
J I I I I I I I 1 I I I I,
7" WO T1 t
Figure 1. Dynamics for the anthropogenic concentration ol mercury in soil (p) and in surface waters.
where CA, Cs, CB, Cceg[2/Me] are the average con-
centrations of mercury with respect to area in the
atmosphere, in the soil, in the water, and in bottom
deposits, respectively.
Tv —time of ventilation of the region;
ta —lifetime of the aerosol in the atmosphere;
a —ratio of the surface area to the area of the
region as a whole;
tsa —time for extraction of mercury from the
soil into the atmosphere;
Tsw —time for extraction of mercury from the
soil into the reservoir;
Tceg —time for deposition of suspended parti-
cles;
tBdo—time for water exchange;
sd2 —portion of the aerosol fraction of the
mercury in the atmosphere;
fii —fraction of the mercury in the form of
suspended particles;
k —fraction of the anthropogenic mercury
which enters the atmosphere;
Qaht —discharges of mercury (in g/m2 yr);
Tuoi—lifetime for the mercury in bottom de-
posits.
The model parameters, rv, t,.,,g, xBi,(), sit, fx, and a
depend on the following actual conditions: the state
of the atmosphere, the intensity of water exchange
in the reservoir, the reservoir depth, the amount of
suspended particles in the reservoir, the area of the
water basin, and the reservoir surface.
The calculations were made using the following
values of the parameters: rv = 10 days, rBDO = 10
years, - 0.5, = 0.8, 0.01 < a s 0.2, 10 days
tSEd - 30 days, k = 0.1.
Considering that not all of the reservoirs have
fishing value, we will use a MPC of 5 peg/liter [4] as
the limiting concentration for the water of sanitation
reservoirs in calculations on a regional scale.
The calculations show that, with rare exceptions,
the reservoir is the limiting unit. The values of the
maximum concentration in air are shown in Figure 3
(in units for the accepted CMpc for air) for which the
maximum permissible concentration for sanitation
reservoirs will not be exceeded. It can be seen that
these concentrations are always less than the CMW
for air.
Finally, for a given region, the model parameters
must be refined for the actual conditions. Never-
theless, the given example shows the importance of
38
-------
Depth
<30 m
Depth
<10 m
10-1 -
10-2
0.1 0.2 a
Figure 3. Dependence of the maximum concentrations in the air on a = S^/S^
the described approach to standardizing the anthro-
pogenic discharges of mercury into the atmosphere
on a regional scale.
3. Pesticides
At the present time, hundreds of different chem-
ical compounds are being used as pesticides. Many
of these have deleterious effects on the biota. In ad-
dition, some pesticides, primarily the organochlor-
ides are very stable under the influence of biological
and chemical reactions, and therefore persist in the
environment for many years. The danger of pesti-
cides to natural populations is due to both direct
toxic, genetic (mutagenic and cytogenetic) actions
and to the secondary effect of restructuring and de-
stroying the biological equilibrium within an eco-
system. The greatest ecological harm comes from
stable poisonous chemicals, that is, the organ-
ochlorides which accumulate in large quantities
even when their concentration in the medium is
small. This accumulation is highest in the vital or-
gans of predatory fish and birds which complete the
39
-------
food chain. As a result, even when the lower levels
of the food chain do not experience any noticeable
harmful action, the death or deterioration of the
population which completes the food chain may be
observed, due to a lowering of the reproductive ca-
pacity and viability of both those experiencing the
action and their offspring [18].
The main problem which arises in connection
with the daily use of hundreds of thousands of tons
of stable pesticides is determining the final distribu-
tion of these substances among the various ele-
ments of the biosphere and their subsequent effects
on the biota and on man. The harm caused by the
contamination of the environment depends on the
physical-geographical conditions of the given re-
gions, as well as the quantitative and qualitative
characteristics of the substances which are dis-
charged. That is, for the total of the given parame-
ters, the degree of harm may vary depending on the
conditions in the region. Consequently, critical situ-
ations due to pesticide use may initially arise in
those regions which have low rates of pesticide de-
composition in different media [19], as well as in-
tensive use.
In this study, an evaluation of the scattering of
pesticides in a central Asian region is made. Large-
scale use of the pesticides, a high air temperature,
and a high degree of soil erosion by wind and water,
favors the intensive redistribution of the pesticides
among the various natural media [3].
The transmission of pesticides between the atmo-
sphere, soil, and water was studied. Since the pesti-
cides are not used over the entire area of the region,
the soil in the model is divided into two parts: soil I
— fields treated with pesticides, and soil II — the
rest of the area in which pesticides are not used.
The system of differential equations which describe
the circulation of the pesticides among the atmo-
sphere, soil, and water can be written in the follow-
ing form:
C, =
kP
sts + nisv-s
c = n - «)f
8 mT
VjTpa
V.fiC.
m'^TAS
Q
Ti
+
Ci
where
a(2) _ -FACa , hIbCb _ C'p
ml'iAs m's'xEP rs
VjfjjCs ^ m'n'C'" + m'l'CT
~ p
Tp
<2)
(3.1)
mirAs
niBTsw
Cb
IB
ca, C<£>, C<|\ CB-
- concentration of the pesticides
(mg/m3, mg"/kg, mg/kg, mg/li-
ter), respectively, in the atmo-
sphere, soil-I, soil-II, and wa-
ter:
Va—volume of the atmosphere;
m',5', m'ii', mB—-mass of soil-1, soil-II, and sur-
face waters, respectively;
P—intensity of pesticides use;
k— fraction of the pesticides that
enters the atmosphere during
treatment of the fields;
(1 - k)—part of the pesticides that falls
directly onto the field during
treatment;
/,,/¦>,./i-part of the area occupied, re-
spectively, by fields treated
with pesticides, by a water sur-
face, and by fields not treated
with pesticides;
ra— lifetime (total) for the pesticides
in the atmosphere;
rAs— time required for extracting the
pesticides from the atmosphere
onto the underlying surface;
tpa—time required for extracting the
pesticides from the soil into the
atmosphere;
tsw— time required for extracting the
pesticides from the soil into
surface waters;
rs—total lifetime for the pesticides
in the soil;
rEp—time for the extraction of the
pesticides from surface waters
into the soil;
tb— total lifetime for the pesticides
in the surface waters.
In order to evaluate the limiting medium and the
limiting concentrations in the media, for which the
MPC is not exceeded according to the approach de-
veloped in section 1, the following quantity must be
maximized:
(1 - k)P => max
for the limitations:
O < C? < CT
(3.2)
(3.3)
where
Cf —steady solution to the system (3.1);
CT —maximum permissible concentration in the
i,h medium.
HCCH (hexachlorocyclohexane) and DDT,
which was widely used earlier and is now prohibited
in the USSR, are used as the model pesticides.
These pesticides were selected, first of all, because
they are persistent in the environment and, second,
they are deleterious to living organisms and, third,
their circulation among various media has been
studied more completely than that of other pesti-
cides. The starting values for the model parameters
40
-------
were found on the basis of our experimental data
and that of other authors [7,5,17], The values of the
parameters used in the calculations are given in
Table 3. As a result, it was found that for both DDT
and HCCH, the limiting medium is the soil or the
atmosphere. The calculated values of the limiting
concentrations for the soils in treated fields and for
the atmosphere are shown in Figure 4 for different
values of K = (0 - 1). The relationships that were
obtained can be explained in the following way. If at
the moment of treating the field the part of the pesti-
cide which enters the atmosphere of the valley lies
within the range of 0 to K« (where Ko is the value of
K for which the soil and atmosphere become the
limiting media simultaneously), then the limiting
medium for the given pesticides is the soil. How-
ever, if the amount of the pesticides which enters
the atmosphere increases during the treatment, that
is, K > Kg, then the limiting medium becomes the
atmosphere. Consequently, increasing the concen-
tration to the MPC level in the atmosphere (for 0 <
K < Kj) or in the soil (for K# < K < 1) results in the
MPC being exceeded, respectively, in the soil or in
the atmosphere. The pesticide concentrations in the
TABLE 3
Parameter values
parameters
DDT
HCCH
y.
45 • 10" m3
45 • 10l! m3
m"'.
2.4 • 10" kg
2.4 ¦ 10" kg
m"',
6.3 • 10" kg
6.3 ¦ 10" kg
mB
5 • 10" kg
5 10" kg
Tas
(1-6) days
(10-30) days
T.
(6-30) days
(6-30) days
TPA
4yrs
1,3 yrs
Tsw
30 yrs
30 yrs
TeP
lyr
lyr
Tb
1, 3 mos.
20 days
Tp
lyr
0, 5 yr
/.
0.27
0.27
f,
0.70
0.70
fa
0.03
0.03
water in this case will remain below the MPC used
in the USSR. Even though the concentration of the
pesticides in water is less than the MPC, it can harm
the aqueous biota because of bioconcentration of
the pesticides at different food levels. This aspect of
the problem will not be considered in this example.
¦^a.op
\
\
\
DDT
0,1 0,2 0,3 0,4 Ko 0,6 0,7 0,8 K
"»,op
\
/
LINDANE
/
/
y
/
/
1 N
0,1 Ko 0,3 0,4 0,5 0,6 0,7 0,8 K
Figure 4. Calculated values of the limiting concentrations of the pesticides in the soil (1) and in the atmosphere (2). K is
the fraction of the pesticides which enters into the atmosphere. (3), (4) are the calculated values of the pesticide
concentrations in the soil and in the atmosphere, respectively, under the condition that when 0 s K < Ko, the
concentration of the pesticides in the atmosphere is equal to the MPC, and for Ko < K < I, the concentration
of the pesticides in the soil is equal to the MPC.
41
-------
4. Sulfur dioxide
The world-wide anthropogenic emission of sulfur
into the atmosphere is mainly due to the operation
of power plants (55%), nonferrous metallurgy plants
(15%), and ferrous metallurgy plants (10%) [26J. In
all cases, the sulfur which is present in fuels or ores
is oxidized to S02 and discharged into the atmo-
sphere in this form.
Ever-increasing discharges of sulfur dioxide into
the atmosphere are found everywhere in the world.
In 1960, its emission was evaluated at seventy mil-
lion tons, in 1975 at 120 million tons, and by the
year 2000 it will reach 280 million tons [26].
The sulfur dioxide is extracted from the atmo-
sphere rather intensively during its transmission by
air currents. The time scale for the processes of ex-
tracting S02, in contrast to mercury and the pesti-
cides, is characterized in terms of hours, and there-
fore the problems of air contamination by sulfur
dioxide is chiefly local and regional in character.
What are these problems, and how can we stan-
dardize the introduction of anthropogenic sulfur
dioxide into the atmosphere?
First, we must consider the main features for the
behavior of sulfur dioxide in the atmosphere. The
intensive extraction of sulfur dioxide from the at-
mosphere is associated with three basic mecha-
nisms: chemical conversions, washing out with rain
and cloud droplets, and absorption by the under-
lying surface. The first mechanism deserves greater
attention. Sulfur dioxide is a very reactive gas
which is readily soluble in water. The sulfur dioxide
can be oxidized to sulfur trioxide (SO,) during its
transmission in the atmosphere. This can combine
with molecules of water to form sulfuric acid. The
main reaction in cloud droplets and rain is also the
oxidation of S02 to the hexavalent state of sulfur.
The oxidizing agent is usually dissolved oxygen,
and the catalysts are the salts of such metals as iron,
and manganese [13]. Sulfuric acid is again formed as
the result of the oxidation. This oxidation of S02
must take place unusually rapidly. This is because
only traces of sulfurous acid are found in freshly-
fallen rain and the sulfur is mainly in the form of
sulfuric acid and its salts. Therefore, rain washes
S02 from the atmosphere and drops sulfuric acid
and its salts on the underlying surface.
Many substances of natural and anthropogenic
origin, which are present in the atmosphere and are
dissolved in cloud droplets and rain, can react with
the sulfuric acid to neutralize it. These reactions re-
sult in the formation of stable sulfate compounds
which do not undergo further conversions.
The given reactions mainly take place in the liq-
uid-drop phase. As the drops evaporate, solid aero-
sol particles are formed. The sulfate particles are
then extracted from the atmosphere by dry settling
processes and the washing out of deposits.
Generalizing the foregoing, we can picture the be-
havior of the sulfur dioxide by means of the follow-
ing block scheme (Figure 5). All of the sulfur com-
pounds shown in Figure 5 have some action on the
environment.
The sulfur dioxide itself can affect both animal
and plant life. Respiratory diseases in man and ani-
Figure 5. Block model for the behavior of sulfur compounds in the atmosphere
42
-------
mals may be caused by relatively high concentra-
tions of sulfur dioxide in the air. A noticeable res-
piratory effect occurs in continuous exposures to
concentrations greater than 100 pig/m3 (15J. Certain
types of plants are also very sensitive to sulfur diox-
ide. The bryophyta, and lichens have the greatest
sensitivity. Among the trees, Conifers are the most
sensitive to sulfur dioxide. Latent damage is ob-
served in some types of pines at medium concentra-
tions of 20 /xg/m3 [15,27]. Damage to the leaves at
the cell level is observed in deciduous trees at con-
centrations on the order of 100 jug/nr' [27]. It has
been estimated that the discharge of one ton of sul-
fur dioxide per year into the atmosphere results in
the death of 0.05-0.1 hectares of forest [6].
The reaction products from the conversion of the
sulfur dioxide have a much greater effect on the en-
vironment. The following effects have been ob-
served: an increase in the number of cases of
respiratory diseases in man and animals, the death
of vegetation and the suppression of its growth, an
increase in the corrosion of materials, the de-
struction of lime and marble structures, the souring
of soils and enclosed reservoirs, and changes in the
optical characteristics of the atmosphere.
Deleterious effects from suspended sulfates begin
to be observed at concentrations of 6-10 /ig/m3 [31].
At a medium concentration level of 10-12 fig/m3, a
6% increase in respiratory ailments was noted at an
air temperature of 1-10° C, and a 32% increase for
temperatures above 10° C [21].
The action of sulfur-containing substances on
plants may have either a direct or tangential charac-
ter. High concentrations of sulfuric acid in fog drop-
lets can severely damage leaves; however, the side
effects of the action show up on a much greater
scale, that is, the nutrient properties of the soils be-
come poorer when the acid rain falls on it. This is
the case for the soils of the northwest and central
chernozem region of the USSR. Usually, such soils
are already acid, and the addition of excess acid
with the precipitates greatly lowers their fertility.
In a number of regions, the acidification of fresh
waters is a serious problem. The consequences are
felt particularly strongly during the thawing period
when the large amount of acid which has accumu-
lated in the snow in the course of the winter enters
into the reservoirs. For the majority of fish species,
this period coincides with the spawning period.
Many lakes and rivers in Norway and Sweden have
lost their industrial fishing value because of the
acidification of the surface waters [14,15].
As we can see, the "spectrum" of possible ac-
tions of the sulfur compounds is very broad. Re-
turning to the question posed at the beginning of
this section, on how to standardize the sulfur diox-
ide discharges into the atmosphere, we have found
that standardizing these discharges strictly on the
requirements of observing the sanitary-hygienic
norms for the MPC is not sufficient. In fact, the
norms for sulfur dioxide and sulfuric acid concen-
trations in the air were established separately with-
out taking into consideration their possible mutual
conversion. Usually, no norms are given for the
concentration of sulfates in the air. It is obvious that
if the norm for sulfur dioxide discharge is calculated
on the basis of the MPC values for sulfur dioxide
alone, then this may result, first of all, in not ob-
serving the norm for the concentration of the sec-
ondary products in the air, for example, sulfuric
acid, and second, it may result in a marked acid-
ification of the atmospheric precipitates and, as a
consequence, a serious problem of soil and surface
water acidification. Therefore these facts must be
taken into consideration to standardize discharges
into the atmosphere.
We will consider the case of an equilibrium sys-
tem in which the area of the source is rather large,
there are no side effects, and the emission of sulfur
in the form of S02 is compensated for by runoff
mechanisms. If we assume that all of these mecha-
nisms obey a first-order kinetic rule, then the con-
centration of sulfuric acid and sulfates in the atmo-
sphere is related with the sulfur dioxide concentra-
tion by the following relationships:
Ns°2K3 = Nh*so<(K4 + k5 + K6)
Ms
M
h„so4
NS0sK3 = Nv
(K4 + K» + Kfi)(K7 + K8)MS0*
K6MMcS(,«
(4.1)
where N1 is the concentration of the jth compound in
the atmosphere; M1 is the molecular weight of the jth
compound.
We will calculate the values of the coefficients Ki
and the values of NH«S0« and NMeS0< for the minimum
and maximum values of Kt.
The absorption of S02 by different types of un-
derlying surfaces, including European soils, has
been studied in a number of works [23,29,30]. The
coefficient K, is usually given in the form:
K - ^
H
(4.2)
in which Va is the linear rate of absorption, and H is
the height of the distribution layer. We can assume
that the value of V„ can change within the range
from 0.005 to 0.01 m/sec [23,24]. The height of the
distribution layer depends on the meteorological
conditions and, for European conditions, the aver-
age value can be taken as equal to 1000 m
[22,32,33]. Consequently, the value of K, lies in the
range of 0.018 to 0.036 hr_1.
43
-------
The washing out of S02 and fine precipitate parti-
cles can be described by the empirical expression
[20,33]:
k2 = K5 = K8 = AfVT (4.3)
where A is the Langmuir coefficient, f is the dura-
tion of precipitation, and T is the average intensity
of the precipitation. The value of A can be taken as
equal to 10-4 sec-1 [24,28]. The value of f was calcu-
lated for conditions in Sweden and England and is
equal to 0.1 [24,33]. However, this value does not
characterize the average European conditions, and
for regions far from the shore it will be much lower.
In the calculations, we will assume that f is in the
range from 0.06 to 0.1. As in [33], the value of T is
assumed to be equal to 1 mm/hr. Using these f and T
values, a yearly precipitation of 520 to 880 mm is
obtained, which actually corresponds to the real
conditions in various regions of Europe. Thus, the
values of K2, K5, KH lie in the range from 0.022 to
0.036 hi-1.
The rate of chemical conversions of S02 in the
atmosphere depends on the presence of oxidizing
components and catalysts in the air, the intensity jpf
the solar radiation, and other factors. Values of Ra
range from 0.02 to 1 hr-1 [25], In [ 11 ], it was shown
that the values of K:) vary with the transmission
time from 0.20 hr-1 in the first hours to 0.02 hr
after two days of transmission. Nevertheless, for
estimating calculations we will assume that K3 =
0.07 - 0.15 hr-'.
Particles of H2S04 and MS04 in the atmosphere
are very small in size [13,16]. Their average diame-
ter is — 1 /x. For particles of this size the settling
rate is 0.005 m/sec which, for a height of H = 1000
m for the distribution gives a value of K4 = K7 =
0.018 hr-1.
Almost nothing is known about the rate of atmo-
spheric neutralization of H2S04 to sulfates. Starting
pH
l/i
"O
c
3
a
o
o
3
O
o
E
o
0)
c
o
w. C
=3 3
d) 2!
1.0 K~-*max
\
\
I SO2
M
\
\
\
0.5H \
V
\
y HiSO*
A
I \
i \
•v
MeSO*
20
-i 1 1 1—
600
-T- __
Transmission time, hrs
-i 1 1 1 1 1 1 1
1000 1500
Transmission distance, km
10
20
30
40
lo"
60 70
Transmission time, hrs
-1
500
-J 1 1 1 1 r
1000
1500
Transmission distance, km
Figure 6. Change in the concentration of the sulfur compounds and pH with time for the maximum values of the coef-
ficients Ki + K*
44
-------
PH
4"
Kj min
SO 2
V
20
—i r~
40 60
Transmission time, hrs
\
\
W2SO4
V—.v.
/ \
/ \
/
-1—1 1—1 1—1—1—1 1 | 1 1 1—1—|
500 1000 1500
Transmission distance, km
MeSO*
.../. ..
—I
10
~20
—1—
30
—1—
40
~50 60 TO
Hold time, hrs
-1 1—
1000
~" 1
1500
Transmission distance, km
500
Figure 7. Change in the concentration of sulfur compounds and in the pH with time for the minimum values of the coef-
ficients + Kt
with the fact that from 50 to 90% of the sulfur in the
precipitation which falls on central Europe and
southern Nonvay is in the form of sulfuric acid [17],
the value of R„ is estimated to be in the range from
0.005 to 0.02 hr1.
As we have already pointed out, the example
being considered may be realized above a large,
constantly active source. In this case, a large num-
ber of contaminants will be present constantly in
the atmosphere. Consequently^ under Jhese condi-
tions the maximum values of R3 and Ke should be
used, that is, 0.15 hr-^ and 0.02 hr-1, respectively.
We will assume that K4 and R7 are equal to 0.018
hr-1 and that Ks and KB vary in the range from 0.022
to 0.036 hr-1. Calculations of the equilibrium con-
centrations of HjSO and of the sulfates for S02 con-
centrations of 50 fig/m3 obtained the following val-
ues:
CJS£°« = 155MKrH2S04/M3
CES°« = 191MKrHeSCVM3
C®?r = 56MKrS04-/M3
C®°.r = 94MKrSOr/M3
In the Soviet Union, the mean daily MPC for S02
in populated areas is equal to 50 p,g/m3, and for
HaS04 it is 100 /ig/m3. Consequently, in our case if
the norms for the MPC of S02 are observed, the
H,S04 concentration exceeds the MPC by a factor
of 1.5-2. Values for the maximum permissible con-
centration of sulphates have not been established;
however, according to the data in [31], the thresh-
old for action on man can be considered to be a con-
centration of 10 /xg/m3. Consequently, the equilibri-
45
-------
um concentration of the sulfates will exceed the
threshold value of a factor of 5-10.
As we have already indicated, the acidifying ef-
fects of the precipitates greatly influence the ecolog-
ical systems. If we consider that the acidification of
the precipitates occurs because of the washing out
of S02 at a rate of K2 and H2S04 at a rate of K4, then
the specific flow of sulfuric acid will lie in the range
of 45-84 g H2S04/m2 ¦ yr, and the value of the pH for
the precipitation will be an unusually low ~ 2.72.
It is clear that such precipitations over wide terri-
tories are not permissible. It would lead to changes
in the natural geochemical processes and would re-
sult in colossal economic and ecological harm. Con-
sequently, it is impossible to set up norms for the
atmospheric S02 concentration for widespread ter-
ritories based on the existing sanitary-hygienic
norms for MPC.
Further, let us consider the case for the emission
of S02 by individual sources and the transmission of
the polluted air masses. The dynamics for such a
system, according to the block scheme given in Fig-
ure 5, may be described by a system of differential
equations:
NS()2 = -(K, + K2 + K3)Ns°2
NV\ = KaN80. - (K4 + K5 + K6)N"2S04
N MeSL>4 = R6N"2S(J4(R7 + K8)NMeS04 (4.5)
The results of solving the system of equations
(4.5) for the maximum and minimum values of the
coefficients K, K8 are shown in Figures 6 and 7.
The dependences for the concentration of sulfur in
the atmosphere in the form of its various com-
pounds on the time of transmission into the atmo-
sphere or the distance for the transmission are illus-
trated. For this it was assumed that the average rate
of transmission is 30 km/hr. It can also be seen from
these figures, that the maximum H2S04 concentra-
tion in the atmosphere is reached after 8-14 hours at
a distance of 240-420 km from the source.
Let us assume that, as in the case considered ear-
lier, the concentration norm for S02, which is equal
to 50 /ig/m3, is observed in an atmospheric layer
1000 m above the source. Even if we do not take
into account the effects of the horizontal and verti-
cal dilution, the maximum value of the concentra-
tion will not exceed 27-30 /ig/m3; that is, it is equal
to ~ 0.3 of the MPC for HjjS04. However, the value
of the pH for the precipitation will be very low. The
calculated values of the pH are also shown in Fig-
ures 6 and 7 as a function of the time or distance of
transmission. Even at a distance of 600 km from the
source, the value of the pH for the precipitation will
be lower than permissible and will lie in the range of
3.8-4.2. The foregoing means that if the sanitary-hy-
gienic norms are observed for the S02 concentra-
tion above the source, the direct influence of this
source, caused by acidification of the precipitation,
will extend great distances from the source, compa-
rable with the size of most European countries.
REFERENCES
1. Abramovskii, V. P., Yu. A. Anokhin, V. A. lonov, I. M.
Nazarov, and A. Kh. Ostromogilskii. 1976. Globalnyi balans
I predelno-dopustimye vybrosy rtuti V atmosferii. (Global
balance and permissible emissions of mercury into the atmo-
sphere). Pages 35-48 in: Vsestoronnii Analiz Okruz-
hayushchei Prirodnoi Sredy. (Proceedings from the Second
US-USSR Symposium). Gidrometeoizdat, Leningrad.
2. Anokhin, Yu. A., G. N. Voronskaya, I. Ya. Nikolishin,
A. V. Saliyev, and A. Kh. Ostromogilskii. In print. Glo-
balnyi balans rtuti V biosfere. (Global balance of mercury in
the biosphere). Review. VNIIGMI MTsD.
3. Atabayev Sh. T. 1972. Pestitsidy 1 gigiyena vneshnei sredy V
usloviyakh zharkogo klimata. (Pesticides and the health of
the environment in hot climates). Tashkent, Meditsina. p.
144.
4. Bespamyatnov, G. P., K. K. Bogushevskaya, A. V. Bespa-
myatnova, Yu. A. Krotov, L. L. Zelinskaya, V. F. Plekhot-
kin, and G. G. Smirnov. 1976. Predelno-dopustimye kont-
sentratsii vrednykh veshchestv v vozdukhe I vode. (Per-
missible Concentrations of Harmful Substances in Air and
Water). Khimiya. p. 456.
5. Bozilov, V. A., V. M. Koropalov, S. M. Malakhov, I. M.
Nazarov, and A. I. Osadchii. 1976. Atmosfernyi perenos
DDT. (Atmospheric transfer of DDT). Pages 2-5 in US-
USSR Symposium on the Migration and Transformation of
Pesticides in the Natural Environment, October 20-27. Tbi-
lisi.
6. Boltneva, L. I., T. I. Sisigina, and I. M. Nazarov. 1976. O
prognoze porazheniya rastitelnosti vybrosami pro-
myshlennykh predpriyatii V atmosferii. (A Forecasting of
plant disease by industrial emissions into the atmosphere).
Pages 61-63 in Pasteniya I promyshlennaya sreda. Nau-
kovaya Dumka Izd., Kiev.
7. Filippova, L. M. 1976. Geneticheskiye aspekty opredeleniya
dopustimykh nagruzok. Genetic aspecls of determining per-
missible loads). Pages 136-143 in Vsestoronnii Analiz Ok-
ruzhayushchei Sredy. (Proceedings of the Second US-USSR
Symposium). Gidrometeoizdat, Leningrad.
8. Goncharuk, E. I., V. I. Tsinryan, K. S. Stefanovskii, and
V. I. Perelygin. 1975. Prognozirovaniye stoikosti pestitsidov
V vode, pochve I rasteniyakh. (Forecasting the persistence
of pesticides in water, soil and plants). Gigiyena I Sanitariya.
10:18.
9. Izrael, Yu. A. 1976. Dopustimaya antropogennaya nagruzka
na okruzhayuschuyu prirodnuyu sredy. (Permissible anthro-
pogenic loading in the natural environment). Pages 10-19 in
Vsestoronnii Analiz Okruzhayushchei Prirodnoi Sredy (Pro-
ceedings of the Second US-USSR Symposium). Gidro-
meteoizdat, Leningrad.
10. Izrael, Yu. A. and L. M. Filippova. In print. Otsenka 1 puti
dostizheniya zhelayemogo kachestva okruzhayushchei
sredy. (Analysis and means to achieve a desired environ-
mental quality). In Monitoring the state of the environment.
Gidrometeoizdat, Leningrad.
11. Izrael, Yu. A., Yu. A. Anokhin, A. Kh. Ostromogilskii,
F. N. Semevskii, S. M. Semenov, and V. N. Kolesnikova.
1976. Imitatsionno-balansovaya matematicheskaya model
regiona oz. baikal kak instrument vsestoronnego analiza,
dolgosrochogo prognozirovaniya I opredeleniya dopusti-
mykh nagruzok vozdeistviya narodno-khozyaistvennoi
deyatelnosti na kachestvo okruzhayushchei sredy I sos-
toyaniye ekologicheskikh sistem. (Imitation-Balance Mathe-
matical Model of the Lake Baikal Region As an Instrument
of Comprehensive Analysis, Longterm Forecasting and De-
46
-------
termination of Permissible Loadings From Agricultural Ac-
tivities on Environmental Quality and the Condition of Eco-
logical Systems). Pages 246-265. In Vsestoronnii Analiz Ok-
ruzhayushchei Prirodnoi Sredy. (Proceedings of the Second
US-USSR Symposium). Gidrometeoizdat, Leningrad.
12. Korolev, S. M., A. G. Ryaboshapko. 1977. Parametry at-
mosfernogo perenosa zagryaznyayushchikh veshchestv S
sushi na okean. (Parameters of atmospheric transfer of pol-
lutants from land to the ocean). Page 24 in Zagryazneniye I
okhrana okruzhayushchei sredy. VNI1GMI MTsD. Ek-
spress-lnformatsiya, Obninsk.
13. Yunge, Kh. 1965. Khimicheskii sostav I radioaktivnost at-
mosferi. (Chemical composition and atmospheric radio-
activity). Page 69 in Mir, Moscow.
14. 1974. Acid precipitation and its effects in Norway. Ministry
of Environment. Olso, Norway. 16 p.
15. 1971. UN Conference on the Human Environment. Air pol-
lution across national boundaries. The impact on the envi-
ronment of sulfur in air and precipitation — Sweden's case
study. Stockholm. 48 p.
16. Brosset C., K. Andreasson and M. Ferm. 1975. The nature
and possible origin of acid particles observed at the Swedish
west coast. Atmos. Environ. 9:631-642.
17. Cramer, J. 1973. Model of circulation of DDT on earth. At-
mos. Environ. 7:241.
18. Edwards, C. A. 1972. Environmental toxicology of pesti-
cides. Plenum Press, London. 432 p.
19. Riswanul H., and V. H. Freed (eds.) 1975. Environmental
dynamics of pesticides. Plenum Press, New York-London.
387 p.
20. Fisher, B. E. A. 1975. The long range transport of sulfur
dioxide. Atmos. Environ. 9:1063-1070.
21. French, J. G. 1975. Effects of suspended sulfates on human
health. Environ. Health. Perspect. 10:35-37.
22. Garland, J. A., and J. R. Branson. 1976. The mixing height
and mass balance of S02 in the atmosphere above Great
Britain. Atmos. Environ. 10:353-362.
23. Garland, J. A .etal. 1974. Deposition of gaseous sulfur diox-
ide to the ground. Atmos. Environ. 8:75-79.
24. Hogstrom, U. 1974. Wet fallout of sulphurous pollutants
emitted from a city during rain or snow. Atmos. Environ.
8:1291-1303.
Junge, C. 1972. The cycle of atmospheric gases — natural
and manmade. Quart. J. Royal Met. Soc. 98:711-729.
26. Kellogg, W. W. 1972. The sulfur cycle. Science. 175:587-
596.
27. Knabe, W. 1976. Effects of sulfur dioxide on terrestrial vege-
tation. Ambio. 5:213-218.
28. Makhon'ko, K. P. 1967. Simplified theoretical notion of con-
taminant removal by precipitation from the atmosphere.
Tellus. 19:467-476.
29. Owers, M. J., and A. W. Powell. 1974. Deposition velocity
of sulfur dioxide on land and water surfaces using a 35S trac-
er method. Atmos. Environ. 8:63-67.
30. Payrissat, M., and S. Beilke. 1975. Laboratory measure-
ments of the uptake of sulfur dioxide by different European
soils. Atmos. Environ. 9:211-217.
31. U.S. Environmental Protection Agency. 1975. Position pa-
per on regulation of atmospheric sulfates. EPA-450/2-75-007.
78 p.
32. Rodhe, H. 1972. Measurements of sulfur in the free atmo-
sphere over Sweden, 1969-1970. J. Geophys. Res. 77:4494.
33. Sciven, R. A., and B. E. Fisher. 1975. The long range trans-
port of airborne material and its removal by deposition and
washout. Atmos. Environ. 9:49-68.
34. Whelpdale, D. M., and R. W. Shaw. 1974. Sulfur dioxide re-
moval by turbulent transfer over grass, snow, and water sur-
faces. Tellus. 26:196-205.
47
-------
TOXICITY CONTROL IN DISCHARGES
DONALD I. MOUNT, PH.D.
The history of water pollution control strategies
in the United States during the twentieth century
can be divided into three phases. Prior to recogni-
tion of the importance of water-borne diseases such
as typhoid, there was little concern for what was
put into rivers and streams because the population
was so small and the dilution was sufficient so that
wastes that were discharged did not pose more than
local nuisance problems. Industrial discharges were
smaller and fewer than today and synthetic organic
chemicals were not important. Most discharges
were chiefly domestic sewage or other biologically
degradable organic materials. Persistent materials
were discharged in insignificant quantities.
After water-borne diseases were controlled by
disinfection of water supplies, there was a relatively
long period of time when concern centered around
damage to water use as a result of municipal and
industrial discharges. While human health was giv-
en nominal importance, the successful control of
water-borne diseases by disinfection produced a
feeling of security. Fish kills and other obvious
gross damage to aquatic systems received greater
attention. This was the period of time during which
the saprobic system approach was in vogue and
much biological effort was directed toward the iden-
tification of indicator organisms that could pinpoint
damage caused by sewage outfalls. The emphasis
through this period was more on better waste as-
similation by the surface waters, rather than better
waste treatment.
Toward the latter half of the 1960's public atten-
tion was focused on preserving ecosystems and in
1965, the Water Quality Act was passed which re-
quired establishment of water quality standards.
Protection of all water uses was the objective but
because aquatic life requirements are restrictive,
standards for fish and other organisms received the
most attention. The principal thrust of activities fol-
lowing that Act, was not so much human health but
rather the protection of aquatic organisms. There
was much emphasis on demonstrating harm from
what were considered to be excessive levels of con-
taminants and there was a growing recognition that
the establishment of legal blame and subsequent en-
forcement was difficult under this procedure. Fur-
thermore, allocation of the assimilative capacity of
the receiving water to numerous dischargers was
complex.
Such experiences led to a new philosophy which
is currently embodied in legislation passed in 1972,
whereby emphasis was changed from impairment of
water use to one of uniform treatment standards.
This approach has typically been called the "Tech-
nology-Based Approach" and is predicated upon
the deficiencies that were identified under the pre-
vious water quality criteria approach. In the tech-
nology-based strategy, treatment requirements are
prescribed, based on either the best practical treat-
ment or the best available treatment, and plants are
required to meet a minimum treatment requirement
whether or not there is demonstrated evidence that
the receiving water will be damaged if such treat-
ment is not achieved.
The minimum treatment philosophy derives im-
petus from experience of past years in which dam-
age to the receiving water was not demonstrable for
given discharges, even though one could be certain
damage was real. Also, the best practical treatment
was considered to be the best that could reasonably
be required whether or not better treatment was
needed. ,
These technology-based treatment standards
have been developed, not considering the receiving
water or its assimilative capacity, but rather consid-
ering only the available treatment technology, its
cost, and its relationship to production volume. Ac-
cordingly, the permissible level of phenol a steel
mill can discharge in its waste is based upon the
tons of steel that it produces and the cost of treat-
ment. Concurrent with the development of the tech-
nology-based strategy was an increasingly strong
opinion on the part of many groups in the public
sector as well as scientists, that ecosystems, no
matter how slightly changed, would ultimately be
damaged unless they were maintained in their origi-
nal form. As a result, there was also a shift in the
definition of unacceptable harm from use impair-
ment to any detectable change in the system. This
changed definition hastened the movement to insti-
tute minimum treatment requirements whether or
not damage could be established.
The architects of the technology-based approach,
however, realized that there had to be some consid-
48
-------
eration given to the assimilative capacity of receiv-
ing waters and that in instances best practical treat-
ment might not be adequate to protect the receiving
water. A category called "water quality limited
streams" was defined for those waters for which
best practical treatment was insufficient. That is,
additional treatment beyond the minimum treat-
ment requirements would be needed in order to pro-
tect the receiving water. These water quality limited
streams are simply those ones where assimilative
capacity of the stream will be exceeded even if the
minimum treatment requirements are instituted.
It is important to note that the uniform minimum
treatment requirements are uniform in the sense
that the amount of pollutants discharged is a func-
tion of the production capacity of the plant, rather
than uniform in regard to the extent of damage in
the receiving water. At the present time, while wa-
ter quality criteria based on damage to water use are
still in force, the water pollution control program in
the United States is focused more on minimum re-
quired treatment of wastes with considerably less
attention to the impact on receiving waters. This
shift from water quality criteria (maximum per-
missible concentrations) to technology-based stan-
dards, requires that biologists and others concerned
with the impact of discharges on aquatic ecosys-
tems reorient their thinking and revise their ap-
proach in such a way as to fit in with the tech-
nology-based strategy.
While the differences may be subtle, they are
very important and lead to a different toxicology ap-
proach to fit the technology-based regulatory strate-
gy. Unless aquatic toxicologists perceive this dis-
tinction and fit their approach into this strategy, the
minimum treatment required may not be sufficient
to protect ecosystems. If such were to happen, the
protection of aquatic communities would be depen-
dent on a case-by-case evaluation that is time con-
suming, costly in resources and dependent on harm
to the receiving water before remedial action could
be taken. The advantages of adequate minimum
treatment of all effluents to reduce toxicity would
not be gained.
While biologists and toxicologists disagree with
the technology-based approach on the grounds that
some places will be required to install more treat-
ment than necessary and others not enough, the fact
remains that the rate of increase of pollutants and
discharges is far outstripping the rate at which prop-
er controls for existing discharges can be deter-
mined. As pointed out earlier, demonstrating use
impairment is cumbersome. Various estimates put
the number of new chemicals being produced each
year in the range of one to two thousand and clearly
the rate at which permissible concentrations for
these materials are being established is much slower
so that we will continue to fall further behind if we
persist in the pollutant-by-pollutant approach in
each discharge. Furthermore, as is recognized by
all of those working in aquatic toxicology, we are
not giving sufficient attention to mixtures of pollu-
tants that occur in effluents.
Perhaps the strongest reason to join the tech-
nology-based approach is the opportunity to get
some degree of treatment on all wastes so that the
collective effects of numerous discharges will be
lessened and the demonstration of harm in the re-
ceiving water will not have to be made for each dis-
charge on a case-by-case basis. There are also other
advantages to the technology-based approach for
protecting communities. If treatment is required on
all wastes so as to reduce toxicity, waste treatment
research will focus on the removal of toxicity, a fo-
cus long overdue and nearly absent at the present
time. Historically in the USA, treatment of domes-
tic wastes has sought to achieve principally reduc-
tion of BOD, and suspended solids. The treatment
of industrial wastes has unfortunately followed the
same course in many instances, even though these
measures of treatment efficiency are not particular-
ly pertinent to many industrial wastes, especially
for some such as the organic chemicals industry. In
municipal wastes of today there are many chemicals
present now that were not in such wastes 15 or 20
years ago because of area-wide waste collection
systems. The number of chemicals found in many of
our waste discharges today is of such magnitude
that specific, analytical measurements are virtually
impossible or at least prohibitively costly. And so
they are not done. Furthermore, we recognize more
and more the need to assess the impact of single
pollutants in the presence of many other con-
taminants and when one considers the almost in-
finite number of mixtures that result from variable
discharges among many industries, one realizes that
the aquatic toxicologist must employ surrogate
measures that deal with large groups of compounds
at once rather than only a single chemical.
An approach which is now being examined in the
United States is one in which the amount of toxicity
removed by modern treatment technology in vari-
ous industrial categories is specified and instituted
as a treatment requirement on an industry-wide
basis. This is not to say that the goal is to achieve
effluents from treatment plants for all industrial cat-
egories that have no toxicity, but rather to specify
how much of the toxicity must be removed in order
to meet the minimum treatment requirements. The
analogy between this approach and the removal of
biological oxygen demand is very close. In both in-
stances, we are using organisms to measure the ef-
fect which we want to quantify rather than an index
(e.g., chemical concentration vs. toxicity) and in
both cases, the measure is a surrogate one which is
a result of the response of a group of organisms to
49
-------
an unknown group of pollutants. There is no clear
assurance in such an approach that the available
treatment technology will be sufficient to remove
enough toxicity to protect all receiving waters.
However, this deficiency is no more important than
is true of many other treatment specifications which
are identified in the minimum treatment require-
ments. The objective is to provide a way to reduce
the discharge of toxic materials without the need for
a case-by-case demonstration of use impairment be-
fore action can be taken. Furthermore, as is espe-
cially well exemplified in power plant discharges,
the effects of each discharge may be relatively small
and difficult to quantitate. If all plants are required
to provide minimum treatment, then the multiple ef-
fects of numerous discharges into one river or
stream can be substantially reduced, a goal virtually
unattainable if damage must be demonstrated at
each outfall.
Quantitating toxicity removal in treatment stan-
dards is not a precise measurement and many scien-
tists find it unacceptable. Neither is a BOD mea-
surement, however. On the other hand, those who
have carefully examined the problem of industrial
effluents are impressed by the tremendous range in
constituents among the thousands of effluents that
are discharged into receiving waters each day. We
must recognize that one cannot do precise tox-
icology tests if the effluents being studied are not
highly consistent qualitatively and quantitatively
from day to day. Highly variable effluents with
varying discharge quantities over time are the rule
rather than the exception for most industrial opera-
tions and tests must be devised to deal with a large
percentage of the effluents in order to gain a valid
picture of the problems. This means, therefore, that
those tests which are used to measure effluents such
as toxicity, will have to be simple tests that can be
performed quickly and cheaply on a large percent-
age of the effluents.
At the present time there are screening tests
available to measure acute toxicity that meet these
requirements. There are not yet proven methods to
measure residues, chronic toxicity, teratogenicity,
carcinogenicity or mutagenecity of such materials.
A promising test (Ames test) is now available as an
early indicator of potential carcinogenicity, but
many more and better tests are needed to measure
those more subtle, long-term effects. In the mean-
time, it should be the goal of aquatic toxicologists to
institute a minimum treatment requirement requir-
ing a reduction in acute toxicity in order to bring
about some degree of toxicity removal until more
sensitive and more refined tests can be developed.
Such screening tests will give erroneous results, in
some instances. It is difficult, if not impossible, at
the present time for such tests to take into account
chemical/physical changes that will take place in ef-
fluents upon dilution in the receiving water. As
pointed out above, at the present time we are not
able to predict well chronic effects or carcinogenet-
ic effects of chemicals by their acute toxicity.
If incorporation of toxicity removal into mini-
mum treatment requirements can be achieved,
treatment will be designed to remove it and much
progress towards protecting aquatic ecosystems
will be achieved. With the highly toxic chemicals in
use today, especially given the persistence of some,
BOD is clearly not an adequate way to measure
treatment adequacy. Specific limits on a chemical-
by-chemical basis is too cumbersome. Because we
wish to protect organisms from toxic effects of ef-
fluents, required removal of toxicity in a tech-
nology-based strategy makes eminent sense.
50
-------
ANTHROPOGENIC IMPACT ON MIGRATING ANIMALS
[Antropogennoye vozdeystvive na migriruyushchikh zhivotnykh]
V. YE. SOKOLOV, D. S. PAVLOV, and V. D. IL'ICHEV
Migration is the movement of animals from one
place of habitation to another. The adaptive impor-
tance of similar migratory movements is to guaran-
tee favorable conditions for the existence and re-
production of the species population. Due to their
migrations, populations inhabit regions during spe-
cific stages of their life cycle which are not suitable
for their constant habitation. The biological impor-
tance of individual migrations of animals varies.
There are two types of migrations: aperiodic and
periodic. A single migration cycle of animals is typi-
cally composed of food, wintering, and reproduc-
tive periodic migrations.
Migration is a global phenomenon which is char-
acteristic of different levels of ecological organiza-
tion. Bacteria, lower and higher plants, and in-
vertebrate and vertebrate animals all migrate in dif-
ferent periods of their life cycles. Among the
animals, migrations are clearly manifest in insects,
crustaceans, lampreys, fish, birds, and mammals.
Billions of organisms and millions of tons of living
material are subject to periodic seasonal and diurnal
movements. In order to evaluate the scales of these
phenomena, only the following example will be
cited. Up to 10 million migratory birds were record-
ed within a 100 km radius at one moment during
migrations.
Besides active migrations, passive migrations al-
so have enormous importance in the animal world.
Passive migrations are made primarily by the use of
atmospheric and oceanic currents and host trans-
port. For example, many parasites migrate on a
body or within the body of other organisms, thus
developing and transferring from one host to anoth-
er, and from one medium of habitation to another.
Migrating fish, birds, and mammals transfer para-
sites that belong to various classification groups—
the simplest—(Protozoa), flukes (Tremotoda), tape-
worms (Cestoda), nemathelminths (Nematoda),
leeches (Hirudinea), parasitic crustaceans (Crus-
tacea), blood-sucking insects (Insecta), and others.
The development of ground and air transporta-
tion as a result of mankind's activity has significant-
ly increased a new form of animal movements-
movements with the aid of transportation. Thus, for
example, animals which attach to the underwater
portion of ships, to buoys, and in general to floating
objects, such as Balanus, and the mussel—Mytilus,
are transferred with them over enormous trans-
oceanic distances. Although this form of movement
is also close to passive migrations, it can hardly be
grouped with them.
Migrations of animals are a multiple-plan phe-
nomenon which affects the most diverse and vital
aspects of organisms. This report will discuss only
those aspects of these relationships which relate to
human activity. The interrelationships of migrating
animals and man are complex and many-sided. Ani-
mal migrations are related to various aspects of hu-
man activity, for example, trade, plant protection,
environmental protection, public health, protection
from biological injuries, and others. At the same
time it is quite natural that human activity has an
enormous effect on migrating animals. An analysis
of this anthropogenic action on migrating animals is
the topic of this report.
Fishing, hunting, and catching of aquatic in-
vertebrates and marine animals is one of the most
ancient forms of human action on migrants. These
trades rely on the fact that many animals such as,
fish, crabs, birds, land and water mammals migrate
in great numbers. The peculiarities of population
ecology permit a certain number of the animals to
be eliminated by these trades without causing undo
harm to the population. Exceeding this norm in
predatory and unreasonable trade results in a sharp
decrease in the population number, and in a deep
depression in the species population. Examples of
such a detrimental effect are Atlantic herring
(Clupea harengus), cod (Gadus morhua), whales
(Cetacea), and other animals. A traditional and also
long active factor is the destruction of birds during
their migrations. In the past giant nets were placed
in specific places on the path of migration along the
Caspian shore. Today this type of hunting has been
prohibited. Something similar, the use of bird lime
as a method of entrapment, has been practiced until
now in Italy. The main catch is tiny sparrows
(Passeriformes) which have been traditionally pro-
tected in nearly all other countries of Europe.
51
-------
Other powerful factors which affect animal migra-
tions are pesticides, oil, and other types of con-
taminants. When migrating under the influence of
this factor, the animals select the least contami-
nated sections of water and land. Thus, migratory
paths are altered locally or over a great distance de-
pending on the degree of contamination. The sites
for reproduction of the migrants and the final points
of their migrations are also altered. In a number of
cases contamination is the cause of a drop in the
number of migrants. For example, supertankers
which are especially prone to catastrophes due to
their giant dimensions, and consequently, poor con-
trollability and lower strength, have already caused
a reduction in the number of migratory ducks (Ana-
tidae), seagulls (Laridae), small fish (Alcidae), and
cormorants (Phalacrocoracidae). The oil slick from
the "Torrey Canyon" tanker catastrophe killed
over 30,000 birds on the shores of England alone.
Oil extraction on the continental shelf also poses a
considerable threat to many migrating animals.
Hydraulic works are another powerful anthropo-
genic factor which affects migrating animals. Dams
have disrupted the paths of migration of sturgeon
(Acipenseridae), salmon (Salmonidae), herring
(Clapeidae), and other fish on the majority of the
world's rivers. The spawning grounds have proved
to be inaccessible to the sires. Death of fish occurs
during the sloped migration over dams and turbines.
Here the greatest losses occur in the high-pressure
hydrosystems, as a result of the barotrauma of the
fish air bladder during exit from the reservoir into
the tail water of the dam. Mass death of fish, pri-
marily the fry, occurs in various water intake
works, in particular during irrigation. As a result,
ever greater attention is being directed towards pro-
tecting fish from entering these works in our coun-
try, as well as in the United States.
Regulation and reduction of the flow of large riv-
ers, and salinization of certain inland seas result in a
cessation of migrations, a decrease in the popu-
lation number, and practically the complete dis-
appearance of whole populations of anadromous
and semianadromous fish. As an example one can
cite the drastic drop in population of the white
salmon (Stenodus leucichthys leucichthys) and the
anadromous herrings (Caspialosa) in the Caspian,
the pike perch (Lucioperca lucioperca) and bream
CAbramis brama) in the basin of the Sea of Azov,
and others.
Large and small reservoirs which are created dur-
ing hydraulic engineering construction are rapidly
adopted by migrating birds as a site for wintering or
regions for rest during migrations. Thus, for ex-
ample, the development of giant reservoirs in Cen-
tral Asia created good conditions for the wintering
of birds from the water-swamp complex. The final
points of migration of a number of species, includ-
ing economically important ones have been dis-
placed to these reservoirs. European and Asiatic
populations which had previously wintered in other
places are now concentrated at the new Central
Asiatic wintering places. All the major reservoirs
have been "adopted" by migrating birds as soon as
ecological conditions suitable for them have
emerged, that is, during the one-two year period af-
ter filling.
The total area of artificial reservoirs is now hun-
dreds of thousands of hectares. Of special impor-
tance are such major reservoirs and overflow lakes
as Kelifskiye (Turkmenistan), Arnasayskiye (Uzbe-
kistan), Kayrakumskiye (Tadzhikstan), and others
on which up to 1 million birds hibernate.
It should also be noted that the reservoir-coolers
of nuclear power stations and hydroelectric power
stations have changed the migratory cycles of a
number of water fowl. A number of water fowl have
shifted their wintering area to the north and winter
on the heated reservoirs. Certain species, for ex-
ample, mallards (Anas platyrhynchos) have devel-
oped sedentary or semisedentary populations on
these non-freezing reservoirs.
On the whole, the creation of artificial reservoirs
can be considered favorable for migrating water or
near-water fowl and the predators that feed on
them. On the other hand, disruption of the ecologi-
cal "balance" by the existing reservoirs often has
adverse consequences for the same migratory birds.
Thus, for example, the drainage of swamps led to
the disappearance of the Kolkhidskiye winterings of
birds, while the shoaling of the Caspian led to the
depletion of the Gasan-Kuliyskiye winterings. In
1958 six million water fowl wintered in Kirov Bay,
in 1968 their number had dropped to 1.5 million, and
currently to just 0.3 million. The cause of this dras-
tic drop is agricultural melioration and a drop in the
level of the Caspian, as well as a reduction in the
area of reservoirs and swampy land.
Migrating animals are also strongly affected by a
change in the landscape and an overall expansion of
the territory occupied by crops, that is, reduction in
forest areas, plowing up of pasture land, develop-
ment of virgin lands, confiscation of lands for indus-
trial complexes, ground trunk lines of gas pipelines,
railroad and highway embankments, irrigation
channels, and other alterations of land use. These
actions have a definite and sometimes fatal effect on
terrestrial animals. The most tragic example is the
African elephant (Loxodon africanus). The damage
by the elephants to their own biotopes in East Af-
rica was to a considerable degree related to the ces-
sation of their former seasonal movements. The ele-
phants in Kenya and Uganda had migrated east to
the coast of the Indian Ocean. The development of
the territory by the Europeans, mass cutting of for-
ests and the creation in their place of plantations of
52
-------
coconut palms and sisal in the coastal.zone, the pas-
sage of railroads, and the emergence of large settle-
ments in previously uninhabited localities made
such migrations impossible. Apparently similar mi-
grations also occurred in many African Ungulata.
The plowing of pasture land here, as in other areas,
results in the disappearance of these animals.
In a number of cases the physiological loads on
the migrants increase as a consequence of man's
transformation of the environment. For example, as
a result of the cutting down of forests in southern
Spain and on the southern Balkan peninsula these
territories were not suitable resting areas for forest
sparrows preparing to cross the Mediterranean Sea.
They were forced to begin the "sea crossing" with-
out having reached these territories and thus in-
curred an increased physiological load.
But for a number of animals, in particular birds,
examples can be cited of a completely opposite ef-
fect of the expansion of the crop landscape on mi-
grations. The cutting of large areas of forests and
increased plowing in the forest zone have long since
elicited the movement to the north of such birds as
the field lark (Alauda arvensis) of the ordinary len-
ticular bob (Carpodacus erythrinus) and a number
of other birds. Increased numbers have been
noted as a result of range expansion into the
forest zone along the cut areas. In summary, the
in the composition of forest species, the portion of
forest biotopes in the total land area and other alter-
ations.
The action of man on migrants is so extensive
that in a number of cases, for example, in the con-
struction of giant cities or heated reservoirs, it re-
sults in the attraction of animals to the anthropogen-
ic landscape, their refusal to migrate, and their
transformation into sedentary populations. This
phenomenon is well known for many birds. Mal-
lards winter on the non-freezing reservoirs of vari-
ous cities. Recently, a large number of rooks (Cor-
vus frugilegus) and starlings (Sturnus vulgaris) have
started to hibernate in Moscow. In Prague and Ber-
lin populations of black thrush (Turdus merula)
have long since become sedentary.
As a consequence of the expansion of crop terri-
tory, the migratory routes to an ever greater degree
are lying in regions developed by man. In this re-
spect different structures and pipelines have a sig-
nificant effect on the migrants, in particular terres-
trial animals. For the northern deer (Rangifer ta-
randus) that make regular seasonal movements
(from northern Siberia to the forest zone for winter
or from the Kola peninsula to the mountain tundra
for winter), this type of structure often becomes an
almost unsurmountable barrier and results in a
change in the migratory paths.
On Taimir, the gas pipeline Messoyakha-Noril'sk
and railroad Noril'sk-Dudinka-settlement Talnakh
blocked the path of the spring migration of northern
deer in 1967-1969. For more than a month the ani-
mals were delayed by the gas pipeline and railroad
until certain individuals found a detour east of
Norilsk, and the other deer followed them. By 1970
this route had become a traditional one for the pop-
ulation; the deer had learned to bypass the obstacle.
An analogous problem is now present in the Ca-
nadian tundra due to the construction of oil pipe-
lines. Canadian zoologists are attempting to formu-
late methods for changing the migratory paths of
caribou by constructing models of the oil pipelines
on their migratory paths.
The expansion of Moscow, Leningrad and Rya-
zan' have resulted in the obstruction of the migra-
tory paths of moose which often appear on the city
streets. The separation of sheep pasture lands in
Kalmyk by wire fences has led to a disruption in the
migratory paths of saigas. The animals have learned
to react correctly to the fence; they do not fight it
but follow it for a dozen kilometers until they find
an opening.
Birds are also, although to a much smaller de-
gree, under the influence of ground structures. Dur-
ing nighttime migratory flight against the wind,
birds occasionally run into cables when flying at low
altitude. Such phenomena are more often observed
in the mountains because of their more variable at-
mospheric situation, but they are not a rarity in the
lowland areas.
Separately standing structures can also affect mi-
gration if they are located in the path of an intensive
migration of birds, for example, on the sea shores,
on capes extending far into the sea, and on islands
among vast straits or narrow sea passages. As early
as the end of the last century it was noted that light-
houses attract land birds flying at night over the sea
(the famous Gel'goland lighthouse). Structures such
as high television towers are likely even more dan-
gerous for birds, if they are located in the path of
intensive or concentrated migration. On the basis of
a study of the regular accumulations of birds found
in the morning under the television towers near the
city of Tallahassee, Florida, a complete picture of
bird migration was developed, that is, migration dy-
namics, species composition, and other factors.
Thousands of airplanes hit migrating birds annu-
ally in the world. Of these, roughly each tenth colli-
sion is accompanied by serious consequences. Sim-
ilar "meetings" of birds with airplanes increase in
frequency during their migratory period, and dam-
age not so much the populations of migrating birds
as they present a threatening danger to man.
Migrating birds are a powerful biotic (ecological)
factor. Birds, because they are a constant, some-
times short-term ecosystem element present in
large numbers and great biomass, can have an
enormous influence in various ecosystems. There-
53
-------
fore, anthropogenic changes in the nature of
migrations can entail disruptions in the functioning
of the ecosystems on an evolutionary scale. The
study of migrations would help solve a number of
important problems in the biosphere, population
ecology, and environmental protection. An
enormous role in this must be played by new meth-
ods, that is, radar, biological telemetry, satellite te-
lemetry, and others.
Radar technology has played a great role in the
study of bird migrations. With its help, the basis of
phenomena which were completely new and un-
known until now have been established, and as a
result views on migration have been reexamined.
Aviation ornithology, with the help of radar, can be
used in fog or at night to reveal accumulations of
birds which are a danger to airplanes. Plastic bands,
"collars," have contributed greatly to the study of
bird migration since the volume of obtained infor-
mation has increased enormously. If the return was
\% using conventional banding, then using collars
brings the percentage of returns to 80% since they
make it possible to observe the same bird many
times without killing or disturbing it.
Radar tracking of wild animals makes it possible
to study their movement under natural conditions.
American researchers using this method obtained a
significant amount of biological data which would
be practically impossible to obtain otherwise. We
used the method of radar tracking to observe the
Siberian roe deer (Capreolus capreolus pygargus)
on the South Urals and discovered that the animals
of the studied population were sedentary and ter-
ritorially conservative. This fact is very important
since until recently the Siberian roe deer, in con-
trast to the European, migrated extensively. The
extent of certain migratory paths of the Siberian roe
deer reached 400 km. There are two apparent rea-
sons for the cessation of these migrations: intensive
urbanization in the migratory regions and a signifi-
cant reduction in the number of Siberian roe deer.
In the USSR radar tracking was also used to
study sturgeon (Acipenseridae) migrating in rivers.
These studies, which were intensive in the sections
near the dams, made it possible to select the most
favorable sites for the placement and the routines of
the fish-passage structures designed to admit mi-
grants through the dams into the headwater.
Currently, the problem of migration is not only
the study of the laws governing the spatial move-
ment of animals, but also the use of these laws to
control these aspects of migration which affect the
national economy. Trade, public health, transporta-
tion, plant protection, and other branches of the na-
tional economy greatly depend on how comprehen-
sive and complete the study of the migration of indi-
vidual species of animals is. Of great practical
importance, for example, is the migration offish. To
predict the trade and search for fish, it is necessary
to know the sites of the migratory accumulations
and the factors which produce these accumulations.
Fish migrating from internal reservoirs to spawn
must be admitted through the dams at the correct
time. The passively migrating fry which slide down-
stream must be protected from falling into the water
intake structures in which millions of specimens die.
The study of periodic migrations of animals can
be used to predict and control certain biological
processes. The solution to the task of controlling
animal behavior will arm us with new means for
combatting migrating insect (pests) blood-sucking
insects (carriers of serious diseases); will complete-
ly reorganize trade; will protect airplanes from col-
liding with migrating birds; and will provide many
other benefits. Even now we are obtaining practical
benefits from the introduction of these methods.
For example, the use of acoustic repellents against
migrating starlings (Sturnidae) on the vineyards of
Kazhakhstan produces a pure profit of 25,000 rubles
for every 100 ha of vineyard. As is known, starlings
in Europe and Asia can reduce the grape and fruit
harvests by up to 70%, and their number is steadily
rising. At the same time it is impossible to take mea-
sures to limit their number, since starlings are the
natural and reliable press predator of locusts.
Chemistry is acquiring great importance in the de-
velopment of methods to control animal behavior,
since chemical orientation is widespread in the ani-
mal world and plays a great role in migrations.
Since migration cannot occur without spatial ori-
entation of the animal, orientation is a necessary
component of the migratory process. Adaptive ap-
proaches have led to a new concept of orientation.
This concept incorporates the use of many refer-
ence points by the animals, as well as analyzer sys-
tems which interact with each other in a complex
and hierarchical manner.
A special aspect of the practical meaning of mi-
grations has developed out of the growing state
awareness of this problem. Migrating animals do
not know the state boundaries, and therefore the
success of studying them, and their protection and
use, greatly depends on international agreements
and international cooperation. Study of the migra-
tions of fish, birds, and cetaceans has already been
reflected in corresponding international agree-
ments. Intergovernmental Soviet-Japanese conven-
tions have been signed for protection of migratory
birds, and conventions of the IDA (International
Development Association) for protection of reser-
voirs used by migratory birds. Specific results have
already occurred through Soviet-American cooper-
ation. Using plastic collars, Soviet and American
scientists have studied the migratory paths of the
unique population of white geese (Chen caerules-
54
-------
cens) on Wrangel Island, which winter on the Ameri-
can continent. In the United States hunting of white
geese is very intensive, and even the geese which
have migrated from Wrangel Island are shot. Band-
ing of geese on Wrangel Island has shown that geese
from the USSR dominate in eastern Alaska and the
entrance to the Yukon, while the "American"
geese join them later. The proposals by the scien-
tists to change the periods and quotas for white
geese shooting on the wintering areas make it pos-
sible to preserve the population of geese on Wran-
gel Island.
International committees are also operating to
band birds and to prevent the danger of airplanes
hitting birds. Individual symposiums are convening
on the migrations of insects and fish, but an inter-
national center for the migration of all animals has
not yet been created. The need for the creation of
such a center constantly becomes greater.
This discussion has shown that the anthropogenic
action on migrants occurs at different stages of the
migratory cycle, and therefore migrating animals
can serve as indicators of the state of the biosphere
on range stretching hundreds and thousands of kilo-
meters, and often encompassing the territory of a
number of states. This is precisely why it is impor-
tant to control the condition of migrating popu-
lations. As shown above, the organization of such
control to a large extent depends on international
agreements and cooperation.
55
-------
MICROCOSMS AS INDICATORS OF ECOSYSTEM STRESS
FRANK G. WILKES, PH.D. and CLINTON W. HALL
OVERVIEW
In exchange for the benefits of increased tech-
nology, society has been required to accept a cer-
tain amount of environmental degradation. In order
for society to rationally determine an acceptable en-
vironmental "price" for technological progress, it
is necessary to identify and measure the associated
environmental damage. Since ecological systems
are continually being subjected to a wide variety of
stresses, only a portion of which are a result of hu-
man activities, this cost-benefit comparison be-
comes difficult.
The major objective of any ecological effects re-
search and development (R&D) program should be
to determine the environmental cost side of the
cost/benefit equation, that is: i) to identify the envi-
ronmental impacts associated with particular hu-
man activities; ii) to quantify those effects; and iii)
to determine whether they are ecologically signifi-
cant. Once these objectives are met, society can
compare the environmental costs and benefits asso-
ciated with particular actions, and define acceptable
trade-offs.
Determination of the significance of a given envi-
ronmental impact requires discrimination between
change*, induced by anthropogenic causes and those
occurring naturally. Ecosystems are dynamic en-
tities subject to naturally occurring changes. More
often than not, environmental stresses result in a
change of the rate of certain naturally occurring
processes. The problem, then, is one of detecting
the changes resulting from man's activities over and
above natural processes.
In the formulation of environmental policy that
reflects both informed scientific judgment and cur-
rent societal values, it is necessary to define the de-
gree of potential impact associated with a certain
activity and its significance upon an ecosystem.
Traditional environmental assessment studies can-
not always provide this type of information. This
discussion will examine a number of available re-
search tools that are used to evaluate the adverse
ecological impacts that often result from man's ac-
tivities. Their respective abilities to provide some of
the necessary information upon which environmen-
tal policy decisions can be made will be considered.
In particular, the focus will be on selected micro-
cosm research techniques that appear to satisfy
many of the information requirements of the U.S.
Environmental Protection Agency's (EPA) regula-
tory programs.
ECOSYSTEM INTEGRITY
Just as the whole organism is more than its com-
ponent cells and tissues, so the whole ecosystem is
more complex than the sum of its constituent parts.
In analyzing an ecosystem, both structural and
functional traits must be examined. Structural as-
pects (e.g., species composition, population diver-
sity, standing biomass) represent ecosystem organi-
zation, while the functional processes (e.g., nutrient
cycling, sediment transport, energy flow) describe
essential ecosystem services. In assessing the ef-
fects of stress in an ecosystem, both structural and
functional perturbations must be considered.
Biological impacts can be examined in the con-
text of organism, population, and community lev-
els. At the organism level, shortening of the life
span or death due to an environmental insult consti-
tutes a significant impact. An example of a signifi-
cant organism-level impact is fish mortality at an in-
take screening structure or by entrainment in a ther-
mal or chemical plume. Although significant at the
organism level, such impacts are not of con-
sequence to the populations of these organisms un-
less the mortality level exceeds the resiliency pro-
vided by the compensatory responses of the popu-
lations. Therefore, the death of one fish is a
significant impact at the organism level, but mortali-
ty of even a large number of fish may not be a signif-
icant impact at the population level if the loss of
these fish is made up by compensating mechanisms
of that population. At the population level, an ex-
ploitation rate (total of commercial, sport, and any
other) might not be considered significant unless it
is great enough to cause a large continuing decline
in population size.
Community-level impacts are expected to follow
from population-level impacts. Major shifts in the
relative abundance of a given species can alter in-
56
-------
ter- and intra-species relationships that can have an
impact on the community as a whole. However, de-
pending on the community structure and dynamics,
it can be speculated that even total disappearance of
a species may not result in any consequential
changes in the functional integrity of a community,
especially if replacement species are readily avail-
able. Studies at the organism level provide insight
into the mechanisms of damage and species toler-
ance from a given environmental impact, but popu-
lation-level studies are a must for estimating the
number of organisms that may be removed or ex-
ploited without significantly impacting the popu-
lation. It is often necessary to examine at least three
levels of organization (e.g., individuals, popu-
lations, communities) simultaneously to understand
the structural and functional relationships within an
ecosystem.
Ecological Displacement
Once a change in structure or function has been
reliably determined, it is necessary to estimate its
significance in terms of the overall ecosystem. Sev-
eral factors warrant consideration when one wres-
tles with the question of significance; namely, what
is the duration, magnitude, and reversibility of the
potential impact, and what is its probability of oc-
currence.
A number of significant impacts could likely be
avoided if we were able to predetermine the capa-
bilities of an ecosystem to resist and/or recover
from change. The ability to routinely determine, a
priori, a system's response/recovery capability is
not currently available. A variety of approaches can
be used to imply ecosystem vulnerability to stress,
but few are widely applicable. There are several
points, however, that provide assistance in deter-
mining a system's assimilative capacity, namely:
• How much stress can the ecosystem sustain be-
fore it undergoes structural or functional
change (i.e., what is its ability to assimilate the
stress)?
• To what degree is this change in ecosystem
structure or function reversible or elastic (e.g.,
how much stress can the ecosystem accommo-
date, and still be able to return to normal)?
• How many times can the ecosystem recover af-
ter insult (e.g., how resilient is it)?
All of these factors are important in determining
the significance of any ecological change. How-
ever, explanations regarding the biological signifi-
cance of an impact are usually subjective and vary
widely, depending on the backgrounds, affiliations,
and biases of the individuals making the evaluation.
For example, to one group of people killing even a
few fish at a power plant would constitute a signifi-
cant impact, whereas to another group an impact
may be considered significant only if almost total
disappearance of a species due to plant-induced
mortality occurs. In reality, the level of significant
impact may be somewhere between these two ex-
tremes. The problem is to narrow the range be-
tween the two extremes and determine the level
that in reality would constitute a biologically signifi-
cant impact.
As can be seen, there are major discrepancies
within the scientific community as to the definition
of the ecological significance of a particular envi-
ronmental impact. Part of this incongruence results
from the application of ecological testing protocols
without proper recognition of their inherent limita-
tions, and part probably reflects the variations in
current societal values. For example, an impact
could be considered significant, if it exceeds a given
standard or regulation (regardless of its biological
significance), or if it affects a legally recognized or
protected component. It is appropriate, then, to dis-
cuss a number of currently available ecological test-
ing techniques, and to acknowledge their advan-
tages and disadvantages from the environmental
policy-maker's perspective.
INDICES OF ECOLOGICAL RESPONSE
Several research techniques are available to in-
dicate ecosystem status. They generally can be
classified into three broad categories:
• Laboratory studies — where single or multiple
plant and/or animal species are evaluated under
closely monitored environmental conditions.
• Field studies — where various sized plots in a
natural ecosystem are evaluated, with little or
no attempt to control environmental condi-
tions; also included are the site surveys where a
"before/after" collection of ecological data is
gathered and compared.
• Simulation/modelling — where computer tech-
niques are utilized to express the ecological
system in mathematical terms for further evalu-
ation, or physical models of a particular system
are constructed, on a much smaller scale, to ex-
amine the physical processes involved.
The choice of testing method depends on the infor-
mation needs, the level of detail required, and avail-
able resources, among others. Each method has
been used, with varying degrees of success, to pre-
dict or to define the ecological impact resulting from
man's activities. However, all are not capable of de-
fining the ecological effect to such degrees as to al-
low consistently some reasonable judgment of bio-
logical significance.
Given the Environmental Protection Agency's
regulatory activities, some of the available testing
alternatives are more attractive than others, be-
cause they are capable of providing sufficient infor-
mation upon which to decide significance. It should
be recognized that a combination of testing tech-
57
-------
niques would offer the most comprehensive charac-
terization of an ecological impact. To select the
most effective testing combinations, one must rec-
ognize inherent limitations of each technique. A
number of research techniques are presented and
discussed from a policy-making perspective.
Phy sical/C hemical C har acterization
Past EPA regulatory actions addressing point
source effluent discharges have utilized a variety of
physical and chemical measurements (e.g., BOD,
dissolved oxygen, suspended solids) to predict the
impact of such discharges on the aquatic ecosys-
tem. Such techniques ignore several critical points
in assessing ecological impact; namely:
• they are indirect measurements of discharge
impact, and are not very useful in predicting
ecological damage.
• actual pollutant concentrations in the receiving
waters are not determined.
• these techniques provide little or no indication
of ecological significance of the impact.
Laboratory Toxicological Studies (Single/Multiple
Species)
A variety of laboratory toxicological tests are
available for numerous plant and animal species.
Response indicators range from acute lethality to
behavioral indices. Single or multi-species tests can
be conducted, with ancillary pollutant metabolism
and fate studies provided. There are several in-
stances where their utility in defining policy options
is unimpressive; namely:
• Response indices do not define structural or
functional ecosystem damage, but rather im-
pacts on individuals or selected species.
• Toxic response in a single species, although im-
portant, is not necessarily representative of the
entire community response.
• Higher order impacts on the ecosystem are usu-
ally ignored.
Microcosms
Microcosms (i.e., miniature ecosystems which
contain selected structural and functional com-
ponents of an in situ system) are an attractive meth-
od of defining ecological response to stress since the
available techniques are capable of monitoring a va-
riety of indices, including species, population, and
community changes, structural/functional relation-
ships and assimilative capacity of the system. A mi-
crocosm is usually economical to establish and
maintain, is representative of the parent ecosystem
and repetitive trials can provide statistically reliable
results. Major concerns in policy decisions include
the following:
• Microcosms can provide statistically reliable,
representative indications of ecosystem dam-
age, at several structural levels (e.g., species,
population, community).
• Microcosms can be used to define secondary
impacts, including the bioaccumulation/degra-
dation of pollutants.
• Microcosms can be used to define secondary
impacts, including the bioaccumulation/degra-
dation of pollutants.
• Microcosms can be used to predict ecosystem
response, elasticity, and resilience.
• Use of microcosms in studying low-level,
chronic effects is often difficult, due to com-
ponent generation times within a microcosm.
• Some ecosystem relevance may be lost as a re-
sult of the "scale-down" factor.
Field Studies
Field studies can range from small experimental
plots to biomes. Actual ecosystem responses can be
monitored, but are often difficult to relate causally
to stress because of the inability to control con-
founding variables. Field studies are very difficult to
experimentally manipulate because a host of envi-
ronmental factors are beyond the experimenter's
control. Most studies are quite expensive and time
consuming to conduct. Several policy implications
must be recognized:
• Field studies provide a realistic indication of
ecosystem response, when proper indices are
monitored.
• Simple "before/after" baseline studies, can ac-
cumulate vast amounts of data that are often
meaningless unless subjected to very sophisti-
cated analysis.
Simulation/Modelling
Use of simulation models is important in terms of
provision of sharp focus on expected impacts and
collection of relevant information for analysis of im-
pacts. They permit large-scale problems to be ap-
proached from a systems standpoint. The predictive
models can evaluate assumptions where no hard
data are available, provide synthesis, and predict
future impacts. Constraints of time, money and the
possibility of irreversible damage resulting from a
wrong choice usually precludes prior experimenta-
tion to test the consequences of alternative policy
choices. Validity of assumptions used in the models
is critical. One must recognize, however, that what
is important biologically is not always mathemati-
cally tractable, and long-term predictions can be
very unrealistic. Major policy questions include:
• Interpretation of model's results is often diffi-
cult and very dependent upon the quality of in-
put information.
• Unanticipated events within model may have
large, unexpected effects.
58
-------
• Many ecological interactions are not easily rep-
resented by mathematical formulas.
• Model validation is often difficult and subject to
little agreement.
• Utility in examining complex systems where di-
rect experimentation is impossible or impracti-
cal.
EXAMPLES OF MICROCOSM APPROACH
Scientists in the Environmental Protection Agen-
cy are finding that microcosms offer significant ad-
vantages in assessing the ecological significance of
environmental stress.
Microcosms have been defined as miniature eco-
systems which contain those components and pro-
cesses necessary to investigate specific origins,
flows, fates, and/or effects of materials in the envi-
ronment.
The better a microcosm addresses the objectives
for which it was designed, the more useful the re-
sults will be in advancing knowledge of that process
or ecosystem. Thus, the specific kinds of micro-
cosms needed are those that are representative of
the processes and types of natural systems of inter-
est. These microcosms should contain the process-
es and components of the system that are thought to
influence the dynamics that are being investigated.
If features such as structure, stability, etc., are sus-
pected of playing a role in a particular response,
then they must be included as part of the micro-
cosm.
Thus, microcosms allow for the investigation of
particular processes or mechanisms which occur in
the environment, i.e., a particular subunit of an eco-
system. While EPA scientists are utilizing the mi-
crocosm approach to study ecosystem effects in
freshwater, terrestrial and marine/estuarine sys-
tems, the following examples will be limited to es-
tuarine ecosystems presently being studied at
EPA's Environmental Research Laboratory in Gulf
Breeze, Florida.
Predator/Prey System
Estuarine ecosystems contain many important
species interactions which are vital to the structural
and functional integrity of the system. Many of
these processes may be vulnerable to stress and
may be altered significantly by pollutants thus caus-
ing the ecosystem to deviate from its stable state.
Because they play a major role in determining com-
munity structure and species diversity, many preda-
tor/prey relationships fall in this category of vulner-
able ecosystem processes. A test or method which
would demonstrate and quantify the effect of a pol-
lutant on an important estuarine predator/prey rela-
tionship would therefore be valuable in predicting
the potential impact of a contaminant on an estua-
rine ecosystem. Such a test, designed to determine
the effect of an organic pollutant on a specific es-
tuarine predator/prey relationship, is described
herein.
It has been demonstrated that methyl parathion
impairs the ability of grass shrimp Palaemonetes
pugio to escape predation by the gulf killifish Fun-
dulus grandis. It is also known that the juvenile
sheepshead minnow Cyprinodon variegatus, anoth-
er major prey of F. grandis, is much more tolerant
of organophosphate pesticides, based on 96 hr
LC50 values, than are grass shrimp. In a multi-prey
system the predator F. grandis may be expected to
consume a higher proportion of affected or impaired
species in the presence of an organophosphate pes-
ticide. An experiment was conducted, therefore, to
determine if low concentrations of methyl parathion
might affect the rate of predation and prey prefer-
ence of F. grandis when provided with two prey, P.
pugio and C. variegatus, simultaneously.
After acclimation in the laboratory for 14 days, 24
adult P. pugio and 24 juvenile C. variegatus were
exposed to 3 concentrations of methyl parathion in
160 liter aerated aquaria. Following 24 hr exposure,
a single F. grandis was introduced into each tank.
The number of individuals of each prey species re-
maining in the aquaria were counted each day for
five days following introduction of the predator.
The ratio of P. pugio to C. variegatus surviving af-
ter each day of predation at each toxicant concen-
tration was calculated.
The results of this experiment are shown in Fig-
ure 1. The ratio of shrimp to fish in the control
aquaria increased rapidly during the five day test,
indicating a strong preference of F. grandis for C.
variegatus. At a concentration of 0.021 /x.g/1, the ra-
tio of shrimp to fish increased until day 4 after
which it leveled off, indicating that the predator had
switched from fish to shrimp. A similar switch oc-
curred at 0.119 ng/\ and at 0.475 fig/I on days 3 and 2
respectively, indicating that the higher the methyl
parathion concentration, the more rapidly the dele-
terious effect on the shrimp escape behavior. After
five days, the methyl parathion concentration re-
sulted in increased preference of F. grandis for P.
pugio relative to C. variegatus.
These results indicate that sublethal concentra-
tions of methyl parathion can alter the relative pro-
portions of prey in a predator's diet. Given a choice
of P. pugio and juvenile C. variegatus, both ex-
posed to methyl parathion, adult F. grandis in-
creased the proportion of P. pugio in their diet over
time. This change in predator preference became
more pronounced with increasing concentration of
methyl parathion, and was apparently caused by a
decrease in the ability of P. pugio to avoid pre-
dation.
Although it is difficult to extrapolate results ob-
tained in this laboratory experiment to what might
59
-------
take place in nature, these results suggest that low
levels of methyl parathion in an estuary could result
in a reduction in number of crustacean individuals
and an increase in number of smaller fishes includ-
ing juveniles. The increase in fish density and the
resultant increase in competition could alter growth
and survival patterns of ecologically and com-
mercially important fish species. Such a long-term
detrimental impact cannot be predicted from the re-
sults of this test. The method can, however, be used
to screen the effects of chemicals on this particular
ecosystem process. It will provide information
which, while not of a magnitude and significance
necessary to form the complete basis for decisions
on the environmental compatability of potential pol-
lutants, may, when added to the results of other
studies, aid in the total assessment of the ecological
damage which might be expected when certain
chemicals are introduced into the estuarine environ-
ment.
Benthic System
One of the major concerns of regulatory agencies
is the determination and prediction of the effect of
pollutants on benthic organisms. Such information
is needed to determine both the potential impact of
chemicals which move from the water column into
sediments and of dredged material deposited on the
bottom.
Marine polychaetes produce distinct character-
istic features on a substrate surface as a result of
their activity. For example, the lugworm, Arenieda
cristata, an important benthic organism because it
reworks the sediment, produces funnel shaped de-
pressions on the substrate surface as a result of its
normal feeding processes. A decrease in the num-
6.0-
Z
o
Q
O
2
0C
£L
>
o
CO
LU
h—
UJ
z
o
2
LU
<
_l
<
0.
CONTROL
0.024 ug/1
0.119 ug/1
T
2
T
4
T
5
DAYS OF PREDATION
RATIO OF TWO PREY SPECIES REMAINING AFTER
PREDATION BY FUNDULUS GRADIS IN DIFFERENT
CONCENTRATIONS OF METHYL-PARATHION
FIGURE 1
60
-------
.1 '¦ • !• . . . r, > ,i
: .'.'.Vr -vr-V- • V-)
• * * .** •* * % ^ *, * • f *¦ j *
..•!
A. TIMER
B. CAMERA
C. LIGHTS
D. WATER
E. SEDIMENT
BEIMTHIC SYSTEM APPARATUS
FIGURE 2
ber of feeding funnels produced by the lugworms
would indicate an interruption in this activity. Ex-
periments have been conducted to develop a meth-
od of monitoring changes in these surface features
as a function of environmental stress.
Tests were conducted in two static, aerated, 125
liter aquaria (Figure 2). Each aquarium received 25
centimeters of clean sand and 72 liters of filtered
seawater.
Lugworms of similar size were placed in each
tank and seventy grams of ground seagrass was
then added forming a dark mat approximately 3 mm
thick over the sediment surface. After the system
had acclimated, the test compound, methyl para-
thion, was introduced into one tank.
Changes in the surface features produced by the
lugworm were monitored utilizing time-lapse pho-
tography. A photograph of the substrate surface
was taken at 12 hour intervals for 144 hours. The
amount of surface disturbance was determined at
each 12 hour interval. The cumulated surface area
disturbed was plotted against time to provide a sub-
strate modification rate for both exposed and con-
trol lugworms. Since both tanks were treated identi-
cally, any significant difference in the rate of sub-
strate modification could be attributed to the test
compound.
The results of exposure of the lugworms to three
concentrations of methyl parathion are presented in
Figure 3. At 25 and 75 ug/1 there was no significant
difference between the substrate disturbance rate of
the exposed and the control lugworms. At 150 ug/1
the feeding activity of the exposed lugworms was
significantly lower than that of the control.
61
-------
300
200
100
0
300
200
100
0
400
300
200
100
0 ¦
25/jg/l METHYL - MARATHION
O
X
„-o
8--'
75/jg/l
1 1—
108
A-
. _o-
—I—
120
1
132
T
84 96 108 120 132 144
TIME (hours)
ITRATION OF M ETH Y L-P A R ATHIO N ON LUGWORM ACTIVITY
FIGURE 3
62
-------
Based on these results, a water concentration be-
tween 75 and 150 ug/l, or lower, would interrupt the
normal cyclical pattern of the lugworm resulting in a
decrease in its substrate reworking activity. This
could reduce the lugworm mediated exchange be-
tween the water and sediment resulting in an in-
creased residence time in the water for the pollu-
tant. Reduced feeding activity of the lugworm could
also ultimately result in the death of these orga-
nisms. Such an event would affect the overall trans-
port of nutrients and pollutants through the system
as well as alter food chains of which the lugworm
forms a part.
Whatever the environmental significance of this
deviation from normal activity caused by pollutant
stress, it is clear that this technique can demon-
strate a behavioral effect on an important benthic
organism at sublethal concentrations. Such a test,
which demonstrates the response of an organism to
low levels of pollutant, will be of value in providing
a very sensitive means of determining the potential
impact of substances introduced into the estuarine
environment on an important ecological process.
Eco-Core Systems
It is well recognized that the main agents for re-
turning organic carbon compounds to the carbon
cycle are the bacteria and fungi. It is these groups of
organisms that effect mineralization of compounds
of natural occurrence and might be expected
to degrade those of synthetic origin. Therefore,
microbial considerations are essential in the prop-
er functioning, response, and recovery from en-
vironmental perturbations in natural and artificial
ecosystems.
Model systems have been developed for studying
microbial interactions which closely mimic natural
conditions. The Eco-Core system is an artificial lab-
oratory microbial microcosm that evaluates both
the degradative potential of indigenous micro-
organisms and the effects of chemical perturbation
on microbial ecology.
The Eco-Core (Figure 4) utilizes sterile glass cy-
clinders to extract sediment and water cores from
the environment or other microcosm systems. A
number of cores can be run simultaneously, includ-
ing replicates and associated controls. Radiolabeled
pollutants are used to facilitate analysis. Water col-
umn degradation products are continuously mon-
itored by thin-layer chromatography and autoradi-
ography; volatile products, including HC02 and or-
ganics, arc continuously scrubbed from air exiting
the systems. Effects of the toxicant on the microbial
ecology of the cores are monitored by heterotrophic
plate counts and total C02 evolution. Changes in
microbial physiological indices induced by pollu-
tants are obtained by selective media plating from
the water column.
The Eco-Core system is a unique technique for
assessing biodegradation of xenobiotics utilizing
natural assemblages of microorganisms. It offers
advantages of determining: i) interactive forces con-
tributing to the metabolism of the test compound, ii)
changes in microbial populations induced by the
toxicant, iii) ease of handling, and iv) replication of
systems with minimum effort. However, because
the core is a static system, it has the disadvantage of
possible artifactual changes induced by accumulat-
ing metabolites. Good reaction rates may be better
obtained from systems that resemble the dynamic
nature of an aquatic system.
Continuous Flow Systems
The application of continuous flow systems to mi-
crocosm studies allows the dynamic nature of an
ecosystem to be investigated in the laboratory. In
terms of fate studies, the use of continuous flow
systems will provide the best information on the
rate at which a pollutant will be degraded, the ex-
tent to which it will be degraded and the effects of
various environmental parameters on that degrada-
tion process. Relating of such data to processes ac-
tually occurring in a natural aquatic environment is
considerably more valid with continuous flow sys-
tems than it is with static systems. Ideally, the best
kinetic information will come from a system in
which a particular process under study is rate-limit-
ed by as few factors as possible. Although this is
somewhat unreal in terms of a natural aquatic envi-
ronment, it does allow one to determine which ki-
netic factors are the most important and how they
interrelate with an overall more complex process.
The "Eco-Core" system previously discussed is a
method which describes the end-product of a total
degradation process: i.e., it reveals the final out-
come of a series of interrelationships between com-
ponent parts. Continuous flow systems, on the oth-
er hand, make it possible to study the individual
components and intermediate processes which lead
to specific end-products.
Two types of continuous flow concepts will be
discussed here; i) a small scale system (500 ml reac-
tor vessel volume) which permits a study of the de-
tails of microbial transformation processes and ii) a
large scale system (40 liters reactor vessel volume)
which will contribute information on the fate of a
pollutant as it is mediated by animal metabolism
and various forms of marine biota.
In an exemplary investigation to determine the
usefulness of the small scale continuous flow sys-
tem to study a complex microbial process, the mi-
crobial breakdown of the pesticide methyl para-
thion has been examined. The continuous flow sys-
tem is diagrammed in Figure 5. It is designed to
accommodate multiple stages (to allow a series of
degradation processes to be studied as a methyl
63
-------
SAMPLING
ECO-CORE MICROCOSM
FIGURE 4
-------
SINGLE FEED, DUAL STAGE CONTINUOUS CULTURE SYSTEM
FIGURE 5
-------
parathion (MPS) supplemented (50-500 ppb) arti-
ficial seawater. Radiolabeled substrate is used to
monitor the degradation rates, and degradation
products were quantitated by this layer chromatog-
raphy, autoradiography and gas chromatography.
The radioactive substrate was added to the system
either as a spike into the primary reactor vessel (to
monitor wash out kinetics) or as a continuous input
from a reservoir (where its quasi-steady state kinet-
ics are followed).
The large scale continuous flow systems are de-
signed to study the fate of a pollutant as a con-
sequence of marine macro biota particularly as it
applies to the metabolic capabilities of various ani-
mal populations. The important aspect here, how-
ever, is to work with a continuous flow system
which will give a good budget analysis of a pollu-
tant's fate; i.e., it will give you information on con-
centration of the pollutant and its degradation prod-
ucts in all compartments, water, air and sediment.
Such a system has been designed (Figure 6); a re-
actor vessel containing 30 liters of seawater cov-
ering a 9 cm thick layer of marsh sediment is contin-
uously fed raw filtered seawater. All air and water
leaving the system can be analyzed and the sedi-
ment can be readily cored for analysis. To maintain
low concentrations of pollutants, radiolabeled ma-
terials are employed. Rate kinetics are obtained by
observing washout rates (i.e., a single spike of ra-
diolabeled pollutant). Once the radioactivity levels
in the water and effluent go below background on a
per ml basis, all water leaving the system is passed
through polystyrene resins to concentrate the radio-
active materials.
The dynamics of two pollutants have been mod-
eled in this system; methyl parathion, a degradable
pesticide, and Kepone, a persistent organochlorine
pesticide. Toxicant movement into the sediment
proved to be the major compartmentation phenom-
enon in each case and thus the sediment burrowing
lugworm, Arenicolu cristata was selected to assess
its effect on the fate of these toxicants. It was
shown that lugworm caused a much more extensive
dispersal of MPS throughout the sediment when
compared to a lugworm-free control tank. In the
control tank, the radiolabeled methyl parathion re-
mained in the top 5 cm for ten days. But in the ex-
perimental tank, the lugworm activity caused al-
most complete mixing of MPS throughout the sedi-
ment within four days. The lugworm also mediated
a faster breakdown of MPS, relative to the control
tank. A steady, decreasing budget of radioactivity
CONTINUOUS-FLOW MICROCOSM
FIGURE 6
66
-------
in the system suggested accumulation of unextract-
able residues in the sediment, a phenomena verified
by the other systems mentioned herein. Analysis of
extractable radioactivity demonstrated both p-ni-
trophenol and amino methyl parathion as degrada-
tion products. The extent of production of these in-
termediates however, was not significantly greater
than those produced by the microbial degradation
alone (i.e., control tank).
Continuous culture systems allow for the reduc-
tion of a complex system to compartments, each of
which can be studied separately. The possible inter-
action between these compartments can be then
considered, and the system analyzed as a collection
of individual processes connected to the whole by a
network of linkages. This approach provides a
means of handling a system too complex to be in-
vestigated directly.
Micro-Ecosystem
A microcosm has been developed to assess both
long and short term pollution effects on a Spartina
salt marsh community metabolism. The emphasis of
this system is on the use of an ecosystem-level pa-
rameter to measure pollutant effects rather than
changes in individual ecosystem processes. Data
from these units will be compared with data ob-
tained from natural salt marshes and will be util-
ized in the development of mathematical models of
salt marsh ecosystems.
The salt marsh micro-ecosystems consists of 4
square tanks containing intact substrate and faunal
assemblages. Natural species and nutrient recruit-
ment are accomplished through a simulated tidal
system which is in temporal synchrony with the nat-
ural site.
The community metabolism response was se-
lected to be the primary test response because it has
been shown to be a very sensitive indicator of com-
munity imbalances; because replicate laboratory
systems do not differ significantly with respect to
levels of community metabolism and, because it is
possible to automatically measure the response.
These systems have been found to incorporate
several inherent advantages. They more closely ap-
proach the natural system than any other design
with which we are familiar; plots can be sacrificed
and replaced with relative ease. Terrestrial insect
populations closely resemble the populations of the
natural marsh site and simultaneous measurements
in the field and in the enclosures allow for a crude
field calibration of the units.
To date, the community metabolism studies show
very good agreement among the replicates. A nutri-
ent flux determination technique has been devel-
oped. The ability of the marsh plots to either assimi-
late or export nitrogen and phosphorous nutrients
to the waters coursing in and out of the marshes is
measured directly. The preliminary results indicate
the marshes are "sinks" for and "sources" of N.
These studies are the first direct measurement of
fluxes of this sort for the marshes.
UTILITY OF THE MICROCOSM APPROACH
In conclusion, the use of microcosms in ecologi-
cal impact studies provides a substantial amount of
scientific information which can be used to bolster
the identification of ecologically significant change.
I would like to stress again the point that we are
striving to develop a number of tests which yield
information about different processes and mecha-
nisms within ecosystems. Tests which examine the
non-biological and the biological; the fate and the
effects of specific compounds. Not all the tests will
be applicable to all compounds, nor are they meant
to be. But, by selectively choosing which tests to
conduct, it will be possible to develop information
applicable to the compound of interest and the
problem at hand.
The benefits associated with microcosm research
techniques can be summarized as follows:
• Microcosm complexity can be tailored to the
needs of the researcher — population or com-
munity response can be obtained, rather than
species or individual responses.
• Environmental conditions can be controlled
thereby minimizing confounding variables.
• System complexity allows a broad spectrum of
response indices to be monitored; including
habitat, energy flow, food web shifts, predator/
prey relationships, structural/functional integri-
ty.
• Indirect ecological effects can be monitored
due to the complexity of the microcosm.
• Assimilative capacity of the system can be de-
termined from repetitive pollution exposure re-
gimes; system inertia, elasticity, and resilience
can be defined to a certain degree.
• Microcosms permit evaluation of subtle, inter-
active effects of pollutant combinations.
• Responses are representative of parent ecosys-
tem.
• Inter- and intra-media pollutant transport,
bioaccumulation, and degradation can be deter-
mined.
• Identification of potentially hazardous com-
pounds through use as a screening procedure is
possible.
• Multiple exposure regimes are possible.
From a policy-making perspective, the utility of
the microcosm approach becomes apparent. One
can assess the overall structure and function of the
microcosm, and using repeated stresses, one can
define that microcosm's elasticity and resilience.
Both direct and indirect pollutant effects can be de-
67
-------
termined for a variety of exposure conditions. The
task of assigning societal significance to a given pol-
lutant effect is facilitated, by providing an indication
of the extent, duration, and reversibility of the eco-
system impact. In combination with other test
methods, microcosms can extend our understand-
ing of ecological impacts, and make environmental
policy decisions more definitive.
68
-------
ECOLOGICAL MONITORING IN RELATION TO THE PROBLEM OF
REGULATING ENVIRONMENTAL QUALITY
[Ekologicheskiy monitoring v svyazi s zadachey regulirovaniya kachestva
ikruzhayushchey prirodnoy sredy]
L. M. FILIPPOVA, F. N. SEMEVSKIY, S. M. SEMENOV, V. A. ABAKUMOV,
G. Y. INSAROV M. YA. ANTONOVSKIY
I. Ecological Monitoring-Comprehensive System for
Observing, Evaluating and Predicting the State
of the Environment
In order to formulate programs for environmental
control and to regulate environmental quality, it is
necessary to have:
—an idea of what quality (state of contamination)
of the environment can be considered acceptable [2,
3,4]—that is, to have a goal for environmental qual-
ity control;
—information on the observed state of the envi-
ronment and the trends of its change;
—an evaluation of the correspondence (or non-
correspondence) of the observed and predicted
state of the environment to the acceptable (or desir-
able) state.
All three enumerated questions must be solved
by a system of ecological monitoring.
We will call ecological monitoring, according to
Yu. A. Izrayel's suggestion [3], a comprehensive
system of observations, evaluation, and prediction
of changes in the state of ecological systems and
their elements under the influence of anthropogenic
action. From this formulation, there derives a num-
ber of specific features and ecological monitoring
problems that cannot be solved within the frame-
work of already-existing specialized services and
systems of observation and control. Such special-
ized services include geophysical services, systems
of departmental control over populations of game
animals, forestry services, programs for observing
natural preserves, etc. These main features and
specific ecological monitoring problems are;
—comprehensive evaluation of the state of the
environment from observations and studies of the
physical, chemical, and biological indices charac-
terizing this state;
—detection of changes in the state of the environ-
ment resulting from anthropogenic factors (in par-
ticular, contaminants) against a background of
changes in natural systems due to natural causes;
—need for information about the inter-
relationships between components of natural eco-
systems, without which it is impossible either to ef-
fectively evaluate and predict the state of the
environment or to select optimal strategies for reg-
ulating environmental quality.
A program of ecological monitoring directed to-
wards scientifically substantiating changes in the
environment must have the information necessary
to control the quality of the environment. In our
opinion, it is expedient to divide such a monitoring
program into two independent, although inter-
related, parts:
1. Experimental and theoretical studies to estab-
lish the characteristic "dose-response" relationship
for individual contaminants or their complexes. We
can use these to determine what actions and what
responses are permissible and do not exceed the
ecological reserve of a given system [4,13]. This
step can be called ecological standardization, for as
a result of these studies, we obtain the permissible
norm for the active pollutant, or the permissible
norm for the response of the biota.
2. Complex of observations, measurements, and
studies the purpose of which is:
a) to establish the actually existing levels of envi-
ronmental contamination. These observations and
measurements we call chemical (or ingredient)
monitoring, and in spite of the existing opinion, we
consider it correct to include in the category of
chemical monitoring the measurement of contami-
nant quantities in the biota; b) the study of the trans-
fer, transformation, and accumulation of pollutants
in the food chains; the tracing of contaminants with
the help of species-indicators and species-accumu-
lators; and similar measurements and studies on the
nature and intensity of active factors with the use of
69
-------
biological subjects. The state of the environment is
evaluated here by a comparison to permissible
norms developed during ecological standardization
of the content of pollutants in the environment, in-
cluding in the biota;
b) to set up observations of the biota that make it
possible to determine the nature and intensity of
biota responses to the action, as well as to evaluate
the state of the biota by comparison with the per-
missible norms formulated during ecological stan-
dardization. This step is the actual biological mon-
itoring.
It would seem that, by knowing the norms of per-
missible actions or contaminant concentrations that
correspond to the norms of permissible biota re-
sponses, (permissible changes) that it would be suf-
ficient to develop only chemical monitoring and to
control only the levels of contamination. However,
it is not possible to make an adequate evaluation
and prediction of the state of the environment
merely from chemical indices, even with sub-
stantiated norms. For contaminant content in the
environment, biological methods of controlling the
environment make it possible to solve problems
that cannot be solved by other methods. In the at-
mosphere, soil, and water, as a rule, there are com-
plex sets of chemical contaminants that affect com-
munities of organisms in a fundamentally different
way from the way individual contaminant com-
ponents affect them. Only biological indices make it
possible to analyze combined effects of complex
sets of contaminants on a specific environment.
With biological methods, one can see the conse-
quences of a single or continuous contamination that
could be missed in chemical monitoring, since the
results of chemical analyses refer only to the mo-
ment of sample taking. Control over environmental
conditions by biological indices permits detection of
the actions preceding direct observation, since bio-
logical indices provide information derived from
continuous environmental contamination monitors.
Biological indices permit determination of the emer-
gence and action of a secondary contamination, and
interpretation of the seasonal and diurnal changes in
contaminant levels in the environment. Observa-
tions and studies of the state of the environment by
biological indices are of high priority also because
they guarantee the possibility of directly evaluating
the state of the specific living component of the bio-
sphere that has experienced harmful effects.
The problem of creating a system of ecological
monitoring is thus a complex one whose solution
requires the involvement of chemical, biological,
mathematical, geophysical, and, on a higher level,
also economical and sociological methods. Further,
we will examine a number of specific theoretical as-
pects of ecological monitoring that are of fundamen-
tal importance in light of the tasks of evaluating and
predicting the state of the environment.
II. Functional Units of Natural Ecosystems
One of the important elements of a scientifically
substantiated theory of ecological standardization is
a structural analysis of populations and ecosystems,
since different elements and subsystems of the
lations and ecosystems have a varying resistance to
different anthropogenic factors.
The first stage in an analysis of the structure of
populations and ecosystems is to determine their
elementary units. We take the cohort (age-class) to
be the elementary unit in the functional structure of
population. A cohort (age-class) is the collection of
individuals in a population that are at the same stage
of development |lj. Each cohort has a relationship
with the environment which is inherent only to its
system and the ecological relationships specific to
it.
Despite the long-dominant viewpoint that the pe-
riod of the organism's life up to sexual maturity is
pre-functional and determined by the autogenetic
process of development, each cohort fulfills in the
population a genealogical function specific to it, and
in the ecosystem a ecological function specific to it.
Cohorts that comprise one population are so dif-
ferent in their ecological functions that they belong
to various trophic levels. Differences in ecological
functions of cohorts result in ecological poly-
functionality of the population. Among the verte-
brates, ecological polyfunctionality is most pro-
nounced in fish. Differences in the diet of various
cohorts (age-classes) of fish in one population will
serve as an example. In the North Caspian Sea, of
the goby (Gobius kessler gorlapFlin), the older co-
horts are fish-eaters, the middle cohorts are mol-
lusk-eaters, and the younger cohorts are crayfish-
eaters; in the starred sturgeon (Acipenser Stellatus
Pal.) the older cohorts are fish-eaters, the middle
are worm-eaters, and the younger are crayfish-ea-
ters. In mammals, the ecological polyfunctionality
of the populations is much less pronounced than in
fish. Without dwelling here on the reasons for these
peculiarities in the various classes of vertebrates,
we will only note that they are of essential impor-
tance in questions of ecological standardization.
The entire set of organism adaptations in a popu-
lation expresses the interrelationship of the orga-
nism, population, and ecosystem at which, in the
normal range of conditions, a specific number of
each cohort in the population is maintained that is
sufficient to fulfill the genealogical and ecological
functions inherent to it.
From the viewpoint of ecological standardization
of effect, the presence within each population of a
cohort (age-class) limiting its number is extremely
important. Under the influence of anthropogenic
70
-------
factors, the limiting function of such a cohort can be
transferred to another cohort of the population
since different cohorts of the same population have
varying resistance to anthropogenic effects. Gener-
ally, the transfer of a limiting function under the in-
fluence of an anthropogenic factor from one cohort
to another is accompanied by a disruption in strati-
fication of the numbers of cohorts typical for the
population of the given species, and by a drastic
change in the number of the entire population. If the
criterion for permissible anthropogenic load on the
population is the lack of an essential reconstruction
(typical for this population) in the stratification of
numbers of cohorts, then the permissible anthropo-
genic action apparently must not exceed the limit
beyond which there follows a decrease in the num-
ber of the limiting cohort.
Because of their ecological functions in the eco-
system, cohorts form, in unity with the seasonal
geophysical and geochemical conditions corre-
sponding to them, interdependent complexes—pha-
lanxes that are the main components of the func-
tional structure of the ecosystem. Different pha-
lanxes permit different anthropogenic loads. The
most important task in ecological standardization is
the detection of the phalanxes most vulnerable to
anthropogenic actions, and the determination of
their time limits.
Theoretically, the last question is successfully
solved by biological monitoring within the frame-
work of the following scheme. Information on the
sensitivity of the biota to the selected type of an-
thropogenic action enters the mathematical model
of the examined ecosystem in the form of "dose-
effect" curves. The mathematical model, which
contains a block of exogenous succession, answers
the question on the structure of the ecosystem with
a new level of action. By increasing the intensity of
action, it can be established in the model which ele-
ment (population, cohort, phalanx, etc.) is the first
to drop out in the intensification of the action. This
is in a certain sense the element in the ecosystem
most sensitive to the given type of action. By trac-
ing this element, we find the simplest signs of
trouble in the ecosystem, signs which indicate the
occurrence of intensive phenomena of anthropo-
genic origin. This procedure will be especially expe-
dient and effective if the selected element (the most
sensitive) has an economic, ecological, or great sci-
entific importance. However, it must be noted that
the developed system of ecological monitoring must
be constructed on the basis of all the information on
the sensitivity of the biota.
III. Field of Sensitivity of the Biota and Selection of
Observation Targets in the System of Biological
Monitoring
The final goal of biological monitoring, as pre-
viously noted, is an analysis and prediction of the
change in the indices of the ecosystem that are im-
portant from man's viewpoint. Recently, it has be-
come evident that such prediction of ecosystem
changes is impossible, either by the intuitive meth-
od, the method of experimenting with real ecosys-
tems, or the method of simulating ecosystems in
laboratories. Under these conditions, mathematical
modeling acquires an important role [8], The prac-
tice of predicting with the help of mathematical
models has shown that models are naturally ex-
pressed in the form of systems of differential or fi-
nite-difference equations for the rates of growth in
the populations comprising the ecosystem. They
are expressed in such a way that the multiplication
factor is usually presented in the form of a function
of indices of the biotic and abiotic environment. We
recall that the multiplication coefficient for the pop-
ulation is called the ratio of its number in the next
generation to its number in the previous generation,
whereby the calculation is made at the same phases
of development of the individual. The multiplication
coefficient of the individual is the number of its
progeny that reach the calculated phase, in the case
of asexual multiplication, and half of this number in
the case of sexual multiplication. The multiplication
coefficient is a parameter in the formulation of the
most important ecological laws—in particular, the
principle of optimality, which is directly employed
in mathematical modeling of ecological systems
[10].
Regardless of the technique employed for collec-
tion of data, results must be presented in a form that
is convenient for further use; that is, out of the set
of possible specific population condition character-
istics, priority must be given to the multiplication
coefficient for a specific ecological background.
One of the most important questions that deter-
mines the program of ecological monitoring is the
selection of targets for observation and study. It
seems to us that if we are not considering a unique
natural subject, where our task is the preservation
of all the characteristic species in the ecosystem,
then the approach to the selection of subjects for
ecological monitoring must be as described below.
As noticed previously, prediction of future eco-
system conditions under anthropogenic influence is
best achievable by means of mathematical mod-
eling. We imagine mathematical models to be uni-
versal for a given ecosystem type (for example,
lake, forest); here, in each specific case, the ecolog-
ical parameters of such a model will be determined
by the specific species composition of the given
ecosystem. Due to the impossibility of experimen-
tally obtaining model parameters which relate the
sensitivity of all species in an ecosystem, it is neces-
sary to select certain representative species for the
analysis. For these selected species, the parameters
71
-------
of sensitivity are determined experimentally in the
laboratory, while for the other species these param-
eters are determined by interpolation. Interpolation
(linear) will be the most accurate in that coordinate
system in which the autocorrelation function of the
sensitivity field will decrease the most slowly. The
phylogenetic tree is such a system of coordinates.
Algorithms, were solved which incorporated the
resits of laboratory studies for any set of species, to
obtain optimal evaluations of ecological parame-
ters. These algorithms make it possible with reason-
able outlays of labor to obtain fairly accurate analy-
ses of the necessary ecological parameters of the
model which are representative of the specific eco-
system. The accuracy of interpolation will be higher
if the selected species are uniformly distributed
over the phylogenetic tree.
Under laboratory conditions one can satisfac-
torily model and study the "abiotic action-re-
sponse" function at the intraspecies level; one can
also study the functions of interaction of individual
links in the ecosystems, but no more. Experiments
made in the ecostats—units for maintaining con-
stant abiotic conditions, for example, temperature,
chemical composition of the environment, humidi-
ty, and others—make it possible to detect the effect
of the preset abiotic conditions on the ecological pa-
rameters of individuals which have selective value.
By varying the abiotic conditions and by fixing
the corresponding changes in biotic parameters per
example, fertility, resistance, etc., we will obtain
the possibility of gathering information on the de-
pendence of a generalized parameter—multiplica-
tion coefficient of an individual—on the abiotic pa-
rameters of the environment.
Generally, the proposed approach makes it pos-
sible to determine the set of classification units
larger than the species, that is, genus, family, etc.,
within each of which it is necessary to select one
species. The resulting set of species forms a repre-
sentative sample for whose elements the sensitivity
parameters are experimentally determined,
With free parameters when optimizing the track-
ing system, preference should be given to the spe-
cies that play an essential ecological role in the eco-
system, and to the links in the food chains which are
the most vulnerable to action, that is, those food
chains which end with these ecologically and/or ec-
onomically significant species. The ecological pa-
rameters which code information on the sensitivity
field of the biota are designed in particular for the
mathematical modeling of phenomena of exogenous
succession. After the determination of these ecolog-
ical parameters and the prediction of the state of the
ecosystem, it is expedient to concentrate the field
observations on the most threatened species. Gen-
erally, the priority in the selection of the system of
species for which the observation is made must be-
long to that set whose members are located in the
coordinates of the phylogenetic tree at the greatest
possible distances from each other. Here the field of
interpolation is expanded and information is pin-
pointed on the state of the biota as a whole.
As laboratory data are accumulated the described
system will be perfected, and the accuracy of inter-
polation of the field of sensitivity of the biota will
increase.
IV. Scheme for Modeling Exogenous Succession
It is generally known that very detailed simula-
tion models provide a good prediction of the dy-
namics for a very short time interval, while with an
increase in the interval of prediction the accuracy of
the prediction falls rapidly. An example of this is
the work of Menshutkin |6| on modeling the aquatic
ecosystem of Lake Dal'neye. It would seem that the
only method for controlling this undesirable effect is
the pinpointing of the input data and even more de-
tailing the perfecting of the model.
We however, propose that the indicated errors in
prediction occur for other reasons—due to the fact
that ecosystem models do not consider possible
changes in ecosystem structure, and in particular,
the factor of natural selection and adaption of liv-
ing organisms to anthropogenic effects on the eco-
system, and to changes in the habitat in general.
The mechanism of natural selection significantly
and fundamentally distinguishes the living com-
ponents of the ecosystem from the nonliving. This
cannot help but affect the methods of simulation
modeling of natural ecosystems. If one accepts this
viewpoint, then the drastic loss of prediction accu-
racy with an increase in the time interval becomes
understandable. Predictions can remain effective
while ignoring the qualitative reconstructions of the
ecosystems which occur as adaptations to a change
in the environment, for example, a change in the
chemical composition of water in climate, and gen-
erally in any external conditions.
There are as yet extremely few attempts to in-
troduce into the ecological models the factor of nat-
ural selection and succession, formalized in some
way. At the same time, the needs of theoretical re-
search on the one hand, and the needs of practical
modeling on the other hand require the creation of a
common approach to modeling the phenomenon of
exogenous succession, and generally to modeling of
structural reconstructions of the ecosystem.
Further we present the scheme for modeling the
response of the ecosystem to a change in the envi-
ronment, that is, the scheme of modeling exogenous
succession. We will apply this scheme to analyses
of a number of specific ecological problems.
Natural selection mechanisms act to maximize
the average multiplication coefficient of individuals,
72
-------
while the pressure from the environment, in particu-
lar, limited food supply, make this maximum equal
to a unit. This conclusion which is called the prin-
ciple of optimality and which dates back even to
Charles Darwin, was formalized and developed in
the works of Haldane [15J, Rashevsky [16], Cohen
[14], Semevskiy [10] and certain other ecologists.
Specifically this principle states that in a population
in a uniform environment there are only individuals
for whom the multiplication coefficient is at a maxi-
mum with respect to possible variations of ecologi-
cal parameters of the individual — fertility, resis-
tance, aggressiveness, mortality, etc. Due to the
Matthew-Kermak principle, the ecoparameters of
the individual are not arbitrary for form certain spe-
cial sets which we will call Matthew-Kermak sets.
To illustrate the last position we note that in biologi-
cal species existing under natural conditions, it is
impossible to increase, for example, fertility with-
out losing protectability. Otherwise natural selec-
tion would inevitably favor those individuals which
are more fruitful and more protected.
Due to the aforementioned, the common method
for introducing the factor of natural selection and
adaptation into the models must consist of the fol-
lowing.
Let Xj(n), • • •, xN(n) — numbers of communities
forming the ecosystem, while the number n in-
dicates the moment in time. Further let ai =
(a/, • ¦, a)1) — ecological parameters of i commun-
ity, and u — vector of external actions on the eco-
system. The model which simulates the dynamics of
the ecosystem will be examined in the form:
x,(n + 1) = f,(x,(n), • • •, xN(n); a,; ¦ • aN; u)x,(n).
i = 1, • • •, N (1)
where f, is the reproduction coefficient of i commu-
nity at n moment in time.
Let af, • ¦ • ad — real parameters, assuming that
in the neighborhood of these parameters the system
(1) has a stable stationery state. Then, due to the
principle of optimality formulated above, the real
parameters aj must maximize the function:
fi(x*„ • • •, xS; af, • • •, a,, • • • ,aNr; u),
i = 1, • • •, N,
where x*lt • • • xfc — stationary values of numbers
in communities corresponding to the set of param-
eters af, • •• a,5, whereby af varies according to the
appropriate Matthew-Kermak set. The quantities
ft at the point of the maximum are equal to unity.
We note that the functional ft is maximized only for
the corresponding vector parameter ai.
In general the functional (fj should be replaced
by the functional
Jim £ In f((x*i(n), • • •, x$(n);
M n = 1
&i, ¦ • •, ah • ¦ •, aNr; u),
where x*(n), • • • x$(n) — trajectory of system (1)
corresponding to the set of parameters af, • • •, a^.
Of course in a general situation the existence of
these limits must be postulated.
It is easy to see that the optimal ecological param-
eters af, • • a^r for a certain state of the environ-
ment u, generally speaking, are no longer the opti-
mal for another state u'. A new equilibrium state
corresponding to the new state of the abiotic medi-
um emerges in response to the change in the abiotic
ecosystem succession. In the model the special pro-
cedure of optimization will derive a new af, • • ¦ aw,
optimal for the new state u' of the abiotic medium.
We will dwell briefly on the validation for the
above formulated principle of optimality. Since bi-
ology is not yet a deductive science and does not
have traditions of axiomatic construction of theo-
ries, then to "prove" the principle of optimality can
only be done within the framework of some mathe-
matical model by identifying formal and pithy state-
ments. Such proof has the nature of an illustration,
and therefore we are justified in selecting the mathe-
matical model for such an illustration from the con-
siderations of simplicity and clearness.
Thus, let a certain population which occupies a
limited ecological niche reproduce in time accord-
ing to the law:
(*) yn +1 = f(yn, ar)yn
where y — population size, n — time, ar — scalar
ecological parameter, f — multiplication coeffi-
cient. We assume the generations not to be over-
lapping, and the real number of the population y* to
be a stable (according to Lyapunov) point in the
system (*). We attribute the fact that, in the process
of natural selection the individuals — carriers of pa-
rameters ar were present, to Lyapunov's stability of
the stationary point (y*, 0) in the system:
(**) yn +» = f(y« + xn, ar)y„
*n + i = f(yn + x„, ar')x„
where ar' — small variation of ar. For the stability of
point (y*. 0) under the assumption that function f is
smooth, it is necessary for both characteristic roots
of the Jacobi matrix in the system (**) at point
(y*, 0) to be no more than unity. One of the roots is
equal to f(y*, ar). Consequently, f(y*, a1") ^ 1 =
f(y*, a"). Thus ar maximizes the multiplication coef-
ficient f.
The same examination can also be made in more
complicated situations; however, there is no need
for this. It is much more important to use the prin-
73
-------
ciple of optimality as a powerful instrument of re-
search to pass directly to an investigation of ecolog-
ical problems.
In conclusion, we will merely note that, strictly
speaking, only the gene is optimal since a certain
heredity is secured only genetically. The use of the
principle of optimality to model biological divisions
larger than the species level is not completely cor-
rect. However, initial attempts to construct these
types of models indicate that the errors which de-
velop are usually negligible; moreover, the serious
inconsistencies which can develop are found in rare
situations and are prone to additional analysis.
We will examine several ecological problems
whose solution successfully employs the theoretical
ideas developed above. Very diverse examples
were consciously selected to demonstrate the po-
tentialities for applying our method to very different
problems encountered in ecological modeling.
The Principle of Limitation
Libikh's idea on the limiting factor generated an
entire trend in modeling of biosystems (Poletayev
[9]). Currently the Libikh function is widely used on
the right sides of modeling equations, (Menshutkin
[8]). Libikh's theory in the pure form is a typical
monofactorial theory which confirms that the
growth of individuals can, in the course of many
generations, be suppressed by one limiting factor.
However, many experimental and theoretical
works, and in particular, our own studies, have
shown the invalidity of the principle of limitation in
the form formulated above.
During comprehensive ecological studies in Ka-
zakhstan and other regions of the USSR, investiga-
tion of the vegetation on the basis of soil conditions
in the style of the Ramenskiy school, indicated that
the factors which determine the composition and
development of the desert, steppe, and meadow
vegetation are numerous. (for example, humidi-
fication pattern, soil wealth, soil structure, soil tex-
ture, and pressure of phytophages) and change con-
tinuously from point to point. The idea of leading
geochemical or other factors is based on illusion.
Under different conditions alteration of plants (eco-
type modification) and vegetation (change in spe-
cies composition of associations) occur which result
in the assimilation of vitally important elements into
a specific relationship regardless of these altered
conditions.
For example, in the transition from moist to dry
habitats the hygrophilous vegetation characterized
by powerful development of the assimilation appa-
ratus and reduction in the root system is replaced
by xerophilous vegetation with a strong root system
and reduced assimilation apparatus. More specifi-
cally, within the genus Asperula, the number of
stoma per mm2 on the lower surface of the leaf in-
creases from 51 in Aspvrulu odoruta in moist habi-
tats, to 442 in Asperula #lauca in dry habitats.
Thus, there is a foundation for the assumption
that monolimitation can occur in nature only for
very short time periods (less than a year and the
time of a generation). Of course, models of ecosys-
tems based on monolimitation can provide only
very short-term predictions. As shown previously,
an increase in the prediction interval sharply reduc-
es the accuracy of the prediction.
The following quantitative examination illustrates
the potential of our modeling method. We will in-
vestigate a certain T-year plant which developed
under natural conditions, and for simplicity we will
assume that its biomass consists of two parts: leaf
and root system. This assumption is equivalent to
the hypothesis of constant biomass of the remaining
parts of the plant (stem, branches . . .).
Thus, let x(t) — biomass of plant at moment in
time t, p. — fraction of biomass composed of leaves,
c:m — ratio of light energy and elements of mineral
nutrition necessary to build plant cells, I and M —
entrance of light energy and mineral nutrition per
unit of biomass of leaves and roots, respectively.
Then by using the "idea of limitation," the equation
for the growth can be written in the form:
x = Xk min
1/ix M( 1 - fj.)x
m
where k — constant coefficient of assimilation, X(t)
— fractions of assimilants for growth at moment in
time t. Due to the principle of optimality, 0
-------
der any conditions for individual organisms all envi-
ronmental factors are significant.
The question of the effect of limitation in a vari-
able medium, including the situation when various
factors limit at different moments in time (refer to
Poletayev (9|), is very complex and requires further
study. However, the above examples indicate that
the principle of limitation, even in Poletayev's
form, requires further substantiation from the view-
point of the effects of adaptation on the ecosystem
biota.
Modeling of Succession and Prediction of the
Population Number
We will examine a population of individuals with
nonoverlapping generations which occupy a limited
ecological niche, and let the equation of the popu-
lation dynamics take the form:
Xn + I = ax„ - fXn
where a — individual's fertility, b — susceptibility
to habitat (biotic and abiotic mortality factors). The
genotype of the individual will be characterized by a
pair of parameters (a, b). It is clear that for a set of
genotypes the multiplication coefficient (a-bx) can-
not be unlimited within the given life form. This im-
poses ratios on the pairs (a, b) which we will imag-
ine in the form of the inequality:
b sr \a2
where X — unchangeable parameter.
We will consider the effect on the population as
the disappearance of a certain percentage of indi-
viduals as a result of a change in the habitat, for
example, as a result of discharges of contaminants.
The equation for the dynamics of the investigated
population assuming additional mortality from ac-
tion on the environment has the form:
x„ + , = axn - fxS - UXn
We compute the dependence of coefficients a and
b of a stationary level of population number x* on
the intensity of a new factor of mortality u on the
basis of the principle of optimality, that is, the con-
dition
(a - u - fx*) = max
Calculations with regard for the alterations show
that:
a = 2(u + 1)
f = 4A(u + l)2
4\(u + 1)
while, if the alterations are not considered:
From the form of the dependence of the station-
ary population number on the intensity of the ac-
tion, it is easy to see that by ignoring the effect of
qualitative alterations we will make an erroneous
prediction of the stationary population number if
the assigned intensity of the external action on the
habitat of this population is used.
The Amount of Resistance of the Individual to
Factors of Mortality
The ecological parameters traditionally used to
quantify population control are fertility and survival
rate from different factors of mortality. In this re-
spect it is important to study the survival rate of
individuals of natural populations which is charac-
teristic of the equilibrium state in the ecosystem
(absence of succession). Research in this direction
is promising in terms of detecting the anthropogenic
factors acting on populations to which adaptation
has not yet occurred.
It turns out that justified hypotheses on the
amount of resistance to mortality factors can be
proposed with very general assumptions. This pos-
sibility arises as a result of the development of a
corresponding quantiative theory based on biologi-
cal postulates, and in particular, on the principle of
optimality.
The constructiveness of the principle of opti-
mality that we formulated earlier is determined by
the presence of information on the relationships be-
tween the ecological parameters of the individual
within the given life form. A purely qualitative hy-
pothesis is sufficient to obtain significant quan-
titative ecological results.
We will examine the population dynamics of a
certain population:
yn + i = k„y„
where y — number, n — generation number, and kn
— multiplication coefficient. Let N factors of mor-
tality be active which are characterized by station-
ary and ergodic random processes x1, • • •, xN, (dis-
ease, predators, etc.), and let the individual distrib-
ute his lifetime A for the formation of progeny ao
and protection from mortality factors at, • • •, aN,
ao + • • • + aN = A. The quantity ao • • • aN is given
in units of biomass. The biomass of progeny is con-
sidered a single biomass. It is assumed that from the
given factors of mortality the individual develops a
specialized protection. This relationship can be
stated as:
k = f0(a0)fi(x1, aO ¦ • • fN(xN, aN)
where fi — survival rate from ith factor of mortality.
75
-------
This described situation is called a simple N — fac-
tor system.
We assumed that:
1) f0, f,, • • • fN are concave functions a«, - - - aN
respectively.
2) ft(x', A) = 1 for any x', i = 1, • ¦ •, N, that is,
an individual, by using the entire lifetime, can be
absolutely protected from any mortality factor.
3) Parameters ao, • • • aN are such that the mathe-
matical expectation
Mx1, •• •, x^fln k}
is maximized when ao + • • • + aN = A.
We will not justify hypotheses 1 and 2 here since
they are of a purely biological nature, but point 3 is
the formulation of the principle of optimality in the
stochastic variant.
Within the framework of the assumptions on the
mean fertility quantities geometrical in time f0 and
the survival rate f cited above, it can be asserted
that:
~{» - ^ +"T where Q = f0(A)
1
N + 1
k 2:
(N + DN
+ 1
When the lifetime A and the possibility of protec-
tion make it possible to be protected from all of the
•mortality factors (perhaps even at the price of the
whole lifetime), the quantities f,, • • •, fN and k are
evaluated as follows:
* I N \N
f" ' ' '» N 2 (n + l) '
k ^
N
N + 1 In + 1/ (N + 1) e
which reduce the prey multiplication coefficient
produces stability of the pair "predator-victim."
This situation is quantitatively described by the fol-
lowing statements:
a) u(x) does not decrease, is continuous, u(0) =
0,
lim u(x) < 1
x -~ «
b) (0) > 1, v(x) is continuous and does not in-
crease
lim v(x) > 0
for the system
xn +, = u(xn)yn
yn + i = v(xn)y„
The evaluations cited above were obtained with
the most general assumptions about the habitat of
the population. This means, in particular, that the
question of the stability of the consumer-producer
pair receives a new interpretation. In actuality,
since the quantity Q (potential fertility) should be
evaluated in order of magnitude as 103, with Ns4,
the quantity K is always greater than unity. Con-
sequently, the number of victims of the population
cannot be stabilized by the low number of mortality
factors, and in particular, by one species of preda-
tor or parasite. Apparently, food stabilized the con-
sumer, and not vice versa as Lek assumed [7].
If the adaptations of the victim are flexible (reac-
tions to the current state of the environment, such
as migrations), then the lower limitation on the mul-
tiplication coefficient of the prey with other factors
There is a compact PC R+ xR+, for which R^xR^ is
the area of attraction.
By applying this statement to predator-prey inter-
actions, one can obtain limitations below and above
the numbers of predators and prey (parasites and
hosts).
The quantitative statements presented above are
based on the following considerations.
Let gi(a,, ¦ • • am), i = 0, * • •, N — essentially
continuous concave functions on convex compact
K in M-dimensional space RM = {a = (at, • • •,
aM)la; E R1}, twice differentiable by K.
LEMMA: F(a) = gXa)g1>(a) • • • gfWa)
with /io, ¦ • •, mn - 0, fi0+ ¦ ¦ • + fis = 1 is concave
forK.
Proof. In the case when N = 0, the lemma is evi-
dent due to the set conditions. As a first step in the
proof we obtain the lemma statement for N = 1.
Since the concavity is a one-dimensional fact, then
it is sufficient to establish it for limitations F for
straight lines intersecting K. This makes it possible
to reduce our problem to the case a = t £ [0, lj.
Further we differentiate F:
F' = (go°gW = go"" 'gf' ~ '(MoSogi + Mtgfgo)
~ MoMigo0 'g^1 ~ 2(8ig'
For complete proof of the lemma statement, it is
sufficient to present F in the form:
F =
an
go"
• gN-1
gfiN, ft = flo + • • • + _ ,,
and make use of discussions for the case N = 1 us-
ing a step of induction from N - 1 to N. The lemma
has been proven.
Theorem. Ifgi(a,, • • • a,J, i = 0, • • •, N are con-
tinuous, concave twice differentiable functions for
K, and a# = (at., • ¦ am.) is the point of the maxi-
mum on K the functions:
G(a) = go«(a) • • • gWa),
76
-------
where m0, ¦ ¦ •, mK 2 0; then:
m0
&(a*)
— max g,(a).
m0 + ' > a e k
Proof. Let i = 0, which does not limit the gener-
ality. Let us assume that m = mi + • • • + mN and
1
f(a) = (g"}.(a) • • • gMa)) »
Further, let b be the maximum point of function
go(a) for K. Since a* is, due to the imposed condi-
tion, the maximum point for functions
G(a) = gm0»(a)fni(a),
then [G(a + t(f - a*))^ s 0 assuming a* 5* f (oth-
erwise the lemma statement is evident).
We can transform the latter inequality into the
following form:
m«[g(a* + t(b - a,))Ma,)
^ -m[f(a, + t(b - a,))Igo(aJ.
Here, the function g0 is concave according to the
condition of the theorem, while the function f is
concave due to the lemma statement. Con-
sequently,
[g„(a* + t(f - a*))Jt > 8"(| I f.
[f(a, + t(f - aJ)J,
{(0 - f(aj
If - a* I
By synthesizing the last three inequalities we ob-
tain
mof(a*)(g0(f) - go(aj) < mg0(a!ti)(f(a+) - f(f))
m0(go(f) - go(a*)) £ mgo(a%)
mo
go(a*)
go(a#)
m + m0
m0
mo + • ¦ • + mN a e K
go(f)
max g0(a).
The theorem has thus been proven.
Let f0(a), • • • fN(a) be continued, concave, twice
differentiable functions for K and
K(a) = f0(a) • • fN(a).
Then, if a* is the point of the maximum K on K,
then from the theorem follow the statements:
1. If max [fi(a), • • • fN(a)] = D, max f0(a) = Q
a e K a 6 K
then
ft(a*) ' ' ' Ma*) ^
ND \N
N + 1
fo(a*)
N + 1
K(aJ
ND
DNQ
n + 1 In + 1J (N + i)i '
2. If max fi(a) = 0| with i = 1, • • • N, then
. a e K
' (a*)
K(a,) >
D,
n + r
QD] • • • Dn
(N + 1)
,N + I
In conclusion we note that the requirement used
above on the two-fold differentiation of the func-
tions has a technical nature. It can be omitted by
using the procedure of uniform approximation of
continuous functions by smooth ones.
REFERENCES
1. Abakumov, V. A. 1973. Time structure of population and
technique for predicting its number. Trudy vsesoyuznogo
Nil morskogo rybnogo khozyaystva i okeanografii (VN1RO)
p. 91:68-86.
2. Izrayel', Yu. A. 1975. Comprehensive analysis of the envi-
ronment. Approaches to determining the permissible loads
on the environment and to substantiation of monitoring.
Pages 17-25 in Vsestoroniy analiz okruzhayushchey pri-
rodnoy sredy. Tr. I Sovetkso-amerikanskogo simpoziuma.
Comprehensive analysis of environment. Proceedings of
first Soviet-American symposium. Gidrometeoizdat, Lenin-
grad.
3. Izrayel', Yu, A. 1976. Evaluation of the state of the bio-
sphere and substantiation of monitoring. Doklad AN SSSR
226p (4):955-957.
4. Izrayel', Yu. A., and L. M. Filippova. in press. Evaluation
and means of reaching the desired environmental quality. In
Trudy I sovetsko-angliyskogo simpoziuma. Monitoring sos-
toyaniya okruzhayushchey prirodnoy sredy. Proceedings of
first Soviet-British Symposium. Monitoring the state of the
environment. Gidrometeoizdat, Leningrad.
5. Insarov, G. E., F. N. Semevskiy, S. M. Semenov, and
L. M. Filippova in press. Tasks and structure of ecological
monitoring. In Problemy ekologicheskogo monitoringa i
modelirovaniya v ekologii. Problems of ecological mon-
itoring and modeling in ecology. Gidrometeoizdat, Lenin-
grad.
6. Krogius, F. V., Ye. M. Krokhin, and V. V. Menshutkin.
1969. Soobshchestvo pelagicheskikh ryb ozera Dal'nego
(opyt kiberneticheskogo modelirovaniya. Community of pe-
lagic fish in Lake Dal'neye (Experiment in cybernetic mod-
eling). Nauka, Moscow. 122 p.
7. Lek, D. 1957. Chislennost' zhivotnykh i yeye regulyatsiya v
prirody, "Number of animals and its regulation in nature.
Izd-vo inostrannoy literatury, Moscow. 380 p.
8. Menshutkin, V. V. 1974. Theoretical bases for mathemati-
cal modeling of aquatic ecological systems. Zhurnal ob-
shchey biologii. 35 (l):34-42.
9. Poletayevt I. A. 1973. Models of waiter 'predator-victim'
and some of their generalizations with the use of the Libikh
Principle. Zhurnal obshchey biologii. 34 (l):43-57.
10. Semevskaya, V. A., and F. N. Semevskiy. 1977. Question
of the stability of ecological systems of the type 'parasite-
host' and 'predator-victim' from the viewpoint of the prin-
ciple of optimality. Zool. zhurnal. 56 (l):5-9.
11. Fedorov, V. D. 1975. Kontseptsiya ustoychivosti ekologi-
cheskikh sistem. Sbornik trudov I-go sovetsko-amerikan-
77
-------
skogo simpoziuma po vsestoronnemu analizu okruz-
hayushchey prirodnoy sredy Concept of stability of ecologi-
cal systems. Pages 207-217. Collection of works of first
Soviet-American symposium on comprehensive analysis of
environment. Gidrometeoizdat, Leningrad.
13. Filippova, L. M. 1976. Geneticheskiye aspekty opredeleniya
dopustimykh nagruzok. Sbornik trudov [I-go sovetsko-
amerikanskogo simpoziuma po vsestoronnemu analizu ok-
ruzhayushchey prirodnoy sredy. Genetic aspects of deter-
mining the permissible loads. Pages 136-143. Collection of
works of second Soviet-American symposium on compre-
hensive analysis of environment. Leningrad: Gidrometeoiz-
dat, Leningrad.
14. Cohen, D. 1970. A theoretical model for the optimal timing
of diapause. Am. Nat. 104 (3):389-400.
15. Haldane, J. B. S. 1956. The relation between density regula-
tion and natural selection. Proc. Royal Sue. Ser. B. 145.306-
308.
16. Rashevsky, N. I960. Mathematical biophysics. Third rev.
ed. Univ. Press, New York. 522 p.
78
-------
EVALUATION OF PRIORITY IN THE ORDER OF CONTAMINANTS
[Otsenki prioriteta v ryadu zagryazniteley]
V. D. FEDOROV
In a strict sense, a "priority order for con-
taminants" cannot be devised for mankind as a
whole. This is because different substances and
compounds act on various ecosystems in different
ways, and different groups of people use their eco-
logical resources differently. Consequently, the
"priority order" of contaminants that represent po-
tential dangers to health could almost be said to be
individual. A special methodology is required to de-
tect and evaluate contaminant danger in individual,
specific loci. This paper presents such a method.
If harmful undesirable effects can be viewed as
independent variables (xi, x2 • • • x„), then the be-
havior of a system, which can be viewed as a
"black box," will be determined by the set of in-
dices that define the reliability of its functioning, its
fate, etc., that is, by a series of dependent variables
(yi, y2 • • • y„), or by the parameter "output" (re-
sponse function).
The true form of the function (y,, y2, • • • yn) =
f(xi, x2 • ¦ • xn) is unknown and must be found.
In order to solve such a problem, one should first
substantiate the rules for sampling a limited number
of independent variables selected from the number
determining the state of the system. Somewhat ear-
lier (Fedorov, 1976, 1977) in the development of a
strategy for biological monitoring, we proposed
such methods and also substantiated the need for
setting a multiple factor experiment to make a pre-
diction of the state of the ecosystem.
Second, one should validate the sampling rules
for a limited number of dependent variables in order
to make a sound judgment on the state of the system
in the present and future, as well as to search for a
suitable and sufficiently efficient method for eval-
uating this state.
My work includes a discussion of the positions in
the formulated problem and one of the methods of
solving it.
If it is accepted that y = f(y,, y2 ¦ • • y„), then in
the expansion of the function y = f(Xi, x2 • • • x„) in
a power series, we obtain a certain approximation
of this unknown relationship in the form of a regres-
sion equation:
y = B0 + SBtXi + SBuXiXj + 2B|jX2i + • • •
where Bj — empirical evaluations whose value de-
termines the amount of response in the change of
corresponding variables. These evaluations are
called regression coefficients.
If, under natural conditions, the action of each
variable positively did not depend on the action of
the others, then to describe the "effect-response,"
relation it would be completely sufficient to limit the
degree of the polynomial to linear members, that is,
y = B0 + SBiX,
Only for this case would it be suitable to set up a
single-factor experiment whose methodology pro-
vided for variation in only one variable with the oth-
er test conditions used as background constants. In
a word, only for the given case could the approach
typically taken by toxicologists in establishing
MPC, etc., be used.
However, the joint action of contaminants on a
natural system very often causes mutual weakening
(antagonism) or strengthening (synergism) with the
control of individual selected indices for the state of
the ecosystems (or individual target). The additivity
of the action is specific, and usually, not a particular
case of response by the system to the combined ac-
tion. Such a situation means that the specific de-
scription of function y = f(xx, x2 ••• xn) must in-
clude terms of the equation that contain the product
of the factors, that is, Bu xtxj. Their appearance in
the regression equation indicates that the depen-
dence of the output of y on a certain factor Xi simul-
taneously depends on another factor — y, in the
study of an ecosystem as a "black box." Thus, the
interaction of factors in this case indicates only the
presence of statistically meaningful deviations from
additivity with the simultaneous action of two or
several factors on the system. In order to evaluate
the interaction of factors, we must change over to a
multiple-factor experiment which provides for si-
multaneous and independent variation of the studied
variables. Plans for this type of experiment, called
factor experiments, are well known; the technique
for computing coefficients of regression has been
described in detail in a number of handbooks, and
interpretation of results is not difficult (Maksimov,
Fedorov, 1969).
79
-------
In the simplest case for two variables that have
a negative effect on the index of the state — y,,
1) the additivity of the effect can be described by
the equation:
yi = Bo ~ BiXj - B2x2
2) the phenomenon of antagonism, that is, the
weakening of negative effects of one contaminant in
the presence of another — by the equation:
y = ~ + Bj .2X,X2
3) the phenomenon of synergism, or the increase in
effect of one contaminant in the presence of another
— by the equation:
y = B0 - B,X, - B2X2 - BI>2X,X2
As an example, a fragment was examined from
one of the tests made according to the PFE plan
(expansion unknown] in which the action of zinc
and chrome (in the form of chromate-ion) on the
phytoplankton of the Rybinsk reservoir was eval-
uated according to the change in primary productiv-
ity on the third day after the addition of specific
amounts of the elements (table 1). the regression
table 1
ACTION OF ZINC AND CHROME ON PRIMARY
PRODUCTION OF PHYTOPLANKTON
No. of tests
Concentration, mg/1
Zn Cr
Primary productivity mb
9/1-day on third
day of experiment
1
0.1
0.01
59
2
1.0
0.01
18
3
0.1
0.10
27
4
1.0
0.10
14
equation that adequately describes the results of
this experiment has the form:
P* = 29.5 - 13.5 x, - 9.0x2 + 7.0 x,x2
We thus have a case of mutual weakening of the
inhibiting action of zinc and chrome. In the recently
published work of V. N. Maksimov (1977), there is
a detailed discussion about the importance of syner-
gism and antagonism as applied to the action of con-
taminants on the ecosystems. This article shows
that an interpretation of test results can be altered
only in cases similar to that examined and depend-
ing on the form of their presentation. However, as
applied to the task formulated above, it can be con-
firmed that the most important result of our experi-
ment is an evaluation of the linear effects of the
studied factors, in so far as regardless of whether or
not we know the effect of their interaction, it can be
stated that, under the specific conditions of our ex-
periment, zinc has a stronger effect on the index
that we selected than chromate.
Thus, in order to establish the priority of con-
taminants, one can recommend setting up experi-
ments using plans of the first order, by using the
coefficients of linear regression as an evaluation for
the effect of these substances. The selection of lev-
els realizable in such an experiment must be deter-
mined by the specific conditions of that ecosystem
for which the priority series is set. Previously, we
described in detail (Fedorov, 1974, 1975) the appro-
priate procedure based on results from chemical
monitoring of contaminants. According to this pro-
cedure for upper level factors in the PFE plan, con-
taminant concentrations should be selected that can
be expected in the studied ecosystem within the as-
signed time and with the existing rate of their
changes.
We will not dwell here on planning and setting up
multiple-factor experiments, since this part of the
work does not produce any basic difficulties. It
seems much more important to us to discuss the se-
lection of a certain state of the ecosystem that could
be used to make a regression analysis of the test
results, in other words, for comparison of the ef-
fects of the studied contaminants. In fact, in the
simplest example examined above, we concluded
that zinc has priority over chrome by its action on
the photosynthesis of phytoplankton. However, it
would not be surprising if, in the action on some
other indicator — for example, on the survival rate
of crustaceans — it is found that the same chrome
addition has a stronger effect than zinc. It is clear
that, for an unambiguous evaluation of the priority
order of specific contaminants in their action on a
specific ecosystem, it is necessary to have one
single indicator for the state of this ecosystem by
whose change one could judge the harmftilness of
each of the studied effects.
This index must be generalized since no one func-
tion can indicate the response of the entire ecosys-
tem to external factors. An index that unites indi-
vidual responses of the system into a single measure
can be called the desirability function. A detailed
substantiation of this approach has been given in
the already-mentioned work of V. N. Maksimov
(1977). In order to construct the desirability func-
tion, it is necessary for each individual response of
the system to have a conventional scale on which
the size of this response in the control test or a cer-
tain "normal" size (for example, the average natu-
ral size determined for the period of the experiment)
is taken as 1, while zero on this scale is the size of
the response indicating extreme trouble in the eco-
system.
If the limits of normal variability in the variables
we selected are well known — that is, the distribu-
tion law is known for these indices in the unim-
paired system — then it becomes possible to com-
bine any "extreme" normal value of each index
with the amount of desirability equal to 0.63, which
on the scale of desirability proposed by Harrington
80
-------
(1965) corresponds to the lower limit of the concept
"good." This extreme value can be found, for ex-
ample, as is done in medical practice, according to
the percentile method (Sepetliyev, 1968), which
considers the values of any response to be normal if
they are in the range of the 25th to the 75th percen-
tile. Then the specific values of these percentiles for
each response, which are compared with the in-
dicated amount of desirability, will produce a type
of calibrated curve for the transition from any real
index to the amount of its desirability.
Another approach to construction of the desir-
ability function is possible on the basis of specific
requirements made of the ecosystem by man as the
consumer of any ecosystem component. In a num-
ber of cases, such requirements have been formu-
lated so clearly thaf in the form of a GOST (all union
state standard) they have acquired the stature of a
law. Such for example is the GOST for drinking wa-
ter. For each index included in the given standard,
one can make an evaluation of desirability di that
will adopt the value 0 if this index exceeds the limits
set by the GOST, and the value 1 if it meets the
GOST requirements. If, according to the standard,
there is one specific value lj, the exceeding of which
indicates a violation of the GOST, we obtain:
d,
0, if yt > lt
1, if y, < 1,
For indices that have limitations "at the top" and
"at the bottom" M| and mi, we will have:
d,
0, if yt > M) and yt < mi
1, if mi < y < Mi
It is not difficult to see that the quantity D =
Vdj, dj • • • dn also will take the value 0 or 1, and
here it is sufficient that at least one value dt = 0,
and then D will also be equal to 0. D will equal 1
only if all dt = 1. It is thus clear that D = 0 will
designate that the studied target does not meet the
UOSl requirements, ana u = l u tms target com-
pletely corresponds to the GOST in all indices. A
more flexible scale for desirabiliy can be con-
structed similarly to the previous one if the limit
values set by the GOST are compared not with 0,
but with the number on the scale of desirability
equal to 0.37, which according to Harrington is the
upper limit of the concept "poor' or "unsatisfac-
tory."
In any case, after the transition is made in some
way from the real variables of the state to the quan-
tities of desirability dt, the generalized desirability is
determined as
D = V d,, d2
dn
that is, as the geometric mean of the partial desir-
abilities.
We will demonstrate the basic principles for using
the desirability function in a specific example. We
will use for this purpose the data from the experi-
ment with the alga Scenedismus audricauda given
in table 2.
According to the technique for constructing the
desirability function, we must compare a certain
conventional number — the quantity of desirability
with each value and for all four indices in table 2.
When the measured indices in the test are higher
than in the control, then it is unlikely that one can
propose any single rule. For example, in table 2, in
experiment No. 3, the quantity of logarithmic
growth rate slightly exceeds that in the control. If
one agrees that a small increase in the growth rate
of the laboratory culture of algae cannot serve as a
sign of the exacerbation of its condition, then this
agreement is equivalent to the affirmation that for
the quantity M in experiment No. 3 the desirability
is 1, as in the control. One must approach such in-
dices as the duration of the lag phase in a complete-
ly different manner. It is clear that its significant in-
crease in experiments No. 2 and No. 4 cannot be
TABLE 2
ACTION OF DIESEL FUEL "L" AND BEROL 198 ON ALGAE CULTURE
Scenedesmus audricauda
Concen-
tration
inmg/1
die-
No. sel
of fuel
test (x,)
ber-
ol
198
(x.)
Number
of
cells
Production
Value
of
coded
vari-
ables
ex-
peri
meat.
quan-
tity
N/n
mill/ml
de-
sir-
abil-
ity
d,
ex-
per-
de-
sira-
Du ration
Logarith-
mic growth
rate
iment. bility
quan- dt
tity
P.
«
s/ml
ex-
peri-
ment,
quan-
tity
days
de-
sira-
bility
d,
ex-
peri-
ment,
quan-
tity
M
days
de-
sira-
bility
d.
Generalized
desirability
D-
Vd, • d, d, d,
1
50
5
-1
-1
12.0
0.94
506
0.95
1
0.80
0.590
0.98
0.915
2
100
5
+ 1
-1
6.3
0.55
397
0.87
10
0.37
0.485
0.94
0.639
3
50
10
-1
+ 1
7.9
0.73
557
0.97
2
0.63
0.695
1.00
0.817
4
100
10
+ 1
+ 1
2.5
0.07
364
0.82
11
0.30
0.445
0.92
0.355
Control
15.5
1.00
600
1.00
0
1.00
0.595
1.00
1.00
81
-------
included in the category of desirable phenomena.
However, the degree of undesirability of the ob-
served increase is difficult to evaluate unambigu-
ously. From the position of the researcher who is
interested, for example, in the most rapid accumu-
lation of culture biomass, a delay in active growth
of the culture is, of course, very disagreeable. At
the same time, delay in algae growth, if it does not
entail a sharp drop in the final yield, is not com-
pletely undesirable. During the lag phase, the algae
can adapt to the harmful action or the most stable
cells can be selected, so that, for the further fate of
the population as a whole, the appearance of a lag
phase on the growth curve can even be useful. Our
selection of desirability quantities, which are also
given in table 2, clearly shows that the desirability
of a certain lag phase duration was evaluated from
the "consumer" viewpoint.
For computation of the desirabilities in the re-
maining cases, we used in the transformation usual-
ly recommended in the literature:
d, = e~"~"
where Z| — the coded variable that one usually tries
to connect linearly to the real variables that charac-
terize the state of the system.
The partial desirabilities (table 2) thus found are
then united into a single index of generalized desir-
ability
D — \Z~~di * da ¦ d;j • d4
The quantities D are also presented in table 2. In
the first three experiments, these quantities do not
exceed the limits that determine the state of the sub-
ject of research as "good," and only in the fourth
experiment the quantity D = 0.355 indicates that
the state of the population should be evaluated as
"poor." These evaluations are right, naturally, only
in so far as those premises are true that provide the
basis for construction of the desirability scale for
each index.
An examination of the quantities of partial desir-
ability in table 3 [sic) indicated that a reduction in
the culture "quality" in the presence of diesel fuel
and berol in concentrations of 100 mg/1 and 10 mg/
1, respectively, is related to a fairly sharp drop in
the desirability for duration of the lag phase, and in
particular, for the number on the background of
more or less constant quantities of production and
logarithmic growth rate. Thus, the introduction of
conventional scales of desirability makes it possible
not only to obtain a certain generalized character-
istic for the state of the system, but also to more
distinctly analyze the reasons for a change in this
state.
REFERENCES
Maksimov, V. N., Gidrobiologicheskiy zhurnal, 13. No. 4
(1977), 34-45.
Maksimov, V. N., and V. D. Fedorov, Primeneniye metodov
metematicheskogo planirovaniya eksperimentov pri otyskanii
optimal'nykh usloviy kul'tivirovaniya mikroorganizmov ("Use
of Methods of Mathematical Planning of Experiments in
Searching for the Optimal Conditions for Growing Micro-
organisms"), Izd. MGY, 1969, 128 p.
Sepetliyev, D., Statisticheskiye metody v nauchnykb meditsin-
skikh issledovaniyakh ["Statistical Methods in Scientific Medi-
cal Studies"), Moscow: Meditsina, 1968, p. 80-85.
Fedorov, V. D., Biologicheskiye nauki. No. 10(1974).
Fedorov, V. D., Gidrobiologicheskiy zhurnal, II, No. J (1975),
Harrington, E. C.,Industrial Quality Control, 21, No. 10(1965),
p. 494-498.
82
-------
FUNDAMENTALS OF CONTROL OVER CHEMICAL
CONTAMINATION OF THE ENVIRONMENT, PRE-EMINENCE OF
MEDICAL INDICATIONS OVER ALL OTHER APPROACHES TO
ENVIRONMENTAL PROTECTION
N. F. IZMEROV and I. V. SANOTSKIY
The growth of civilization, particularly the scien-
tific and technical developments of our century with
uncontrolled use of natural resources have led to a
drastic deterioration in the environment. Intense
contamination of the environment is primarily re-
lated to the development of the chemical industry
which creates an enormous number of new com-
pounds with unstudied biological properties.
During the 60-year existence of the USSR, the
problem of protecting workers from the effect of
harmful substances has been raised to an unprece-
dented height. On the basis of comprehensive theo-
retical studies, principles and methods have been
formulated for evaluating the toxicity and danger of
chemical compounds, including the principle of
stages: synchronism of toxicological studies with
stages of the technological production of new com-
pounds. The classification of danger and toxicity
has been substantiated, and an expert and complete
toxicological evaluation has been made of many
thousands of substances. Sanitary standards have
been formulated for harmful substances of the envi-
ronment (MPC—maximum permissible concentra-
tion), including air in the working zone—about 900
compounds (the most complete list in the world).
Using mathematical models, methods have been
developed to establish reference safety levels for
the action (RSLE) of harmful substances in working
zones according to physical and chemical proper-
ties, parameters of acute toxicity, and thresholds of
the selective action of toxins. The GOSTs (all union
state standards) "Systems of Labor Safety" have
been set up. The evaluation of toxicity and danger
of chemical compounds and the formulation of sani-
tary standards include not only criteria for direct
action, but also criteria of indirect action (genetic
action, accelerated human aging, etc.).
Control and regulation of contamination for pro-
tecting human health are important factors in the
comprehensive analysis, study and protection of
the environment.
1. The main principles for control of environ-
mental chemical contamination are embodied in So-
viet legislation which primarily emphasizes pre-
vention of harmful discharges into the atmosphere,
water and soil. The decree of the CPSU Central
Committee and the USSR Council of Ministers of
January 14, 1960 "Measures for Further Improve-
ment of Medical Service and Protection of the
Health of the USSR Population" makes these tasks
the responsibility of the republic Councils of Minis-
ters and the local Soviets of Workers Deputies. The
law "Environmental Protection" of the RSFSR,
1960 (article 12 and others) obliges the ministries
and departments to prevent contamination of the
environment by implementing the necessary tech-
nological processes or by constructing effective
treatment facilities.
A special decree of the USSR Council of Minis-
ters ordered the construction of gas-purifying and
dust collecting devices at industrial enterprises and
power plants in 1963-67 in the USSR republics.
The decree of the CPSU Central Committee and
the USSR Council of Ministers of July 5, 1968,
"Measures for Further Improvement of Public
Health and the Development of Medical Science in
the Country," provides for the creation of laborato-
ries studying problems of industrial environmental
protection in order to control the working zone and
external environment. "Fundamentals of Public
Health Legislation of the USSR and Union Repub-
lics" (1969) stipulates that environmental pro-
tection is the responsibility of all state organs and
public organizations. These organizations must pro-
vide the harmonic development of physical and
spiritual forces, health, a high level of work capac-
ity and longevity, active life of citizens, prevention
and further reduction in morbidity, invalidism, mor-
tality, and must eliminate factors and conditions
which adversely affect the health of the popula-
tion. The decree of the USSR Supreme Soviet
"Measures for Improving Environmental Protection
and the Efficient Use of Natural Resources" (1972)
states that scientific and technical progress must be
combined with a cautious approach to nature and its
resources, and must create the most favorable con-
83
-------
ditions for life and health, for work and rest of the
workers.
The decree of the CPSU Central Committee and
the USSR Council of Ministers (December, 1972),
implements this action by specific assignments to
the ministries and departments.
The 24th and 25th CPSU Congresses raised the
question of environmental protection: accelerated
scientific and technical progress must be accom-
panied by the prevention and elimination of dan-
gerous contamination of the air and water, and de-
pletion of the earth's resources. There are increased
requirements on the planning and economic organs,
and the planning organizations for the design and
construction of new enterprises. Environmental
protection by existing plants must be improved.
The decree of the 25th CPSU Congress on the
main directions for the development of the USSR
national economy for 1976-80 dictates "more active
development and introduction of technological pro-
cesses which reduce waste products and maximize
their use, as well as systems for using water in a
closed cycle.
"Specialized production of equipment, products
and materials necessary for the creation and opera-
tion of highly effective treatment facilities at indus-
trial plants and in the cities."
"Other laws and decrees on the environment
have the same basic tenets (law on land use, law on
water use, GOSTs of the System of Standards on
Labor Safety (1976) and others).
Together with intragovernmental measures, the
USSR is also currently participating in collective in-
ternational measures for environmental protection
and efficient use of its resources.
2. One practical implementation of environmental
protection is the strict observance of sanitary norms
and regulations for control of chemical con-
tamination of the environment (production, com-
munal, household, natural); correct selection of tar-
gets, type of control (continuous, periodic), sensi-
tivity and selectivity of control methods depending
on the danger of the subjects and the environmental
targets of the environment, etc.
The strategy for control of the environment must
be based on medical indices which must meet the
economic requirements. In long-term planning of
new technology, these indices must correspond to
sanitary and economic criteria.
3. Current sanitary standards necessary for envi-
ronmental control prevent contamination which en-
dangers human health. Meanwhile the long-range or
immediate indirect effects on health are evident.
For this reason, we previously have formulated
another definition of sanitary standards.
The maximum permissible concentration of a
chemical compound in the environment is that con-
centration which — directly or indirectly through
ecological systems, as well as through possible eco-
nomic damage — does not produce somatic or men-
tal diseases in humans periodically or throughout
their lifetime (including latent and temporarily neu-
tralized diseases). Neither does it cause changes in
the state of health which exceed the limits of the
adaptive physiological reactions which can be de-
tected by modern research methods immediately or
in remote periods of life of the present and next gen-
erations (I. V. Sanotskiy, 1971).
In our opinion, this formulation corresponds to
the concept of MPC currently accepted for each en-
vironment. The new (however, already proposed
previously) elements are:
1) Regard for the action (including indirect) of
the environment as a whole.
2) Regard for the natural physiological reactions
of adaptation (similar action of the substance must
not be harmful).
3) Regard for the possibility of unfavorable re-
mote changes, including changes in the progeny.
4) A method for calculating the economic losses
from environmental contamination has apparently
not yet been sufficiently developed. It is known that
such a method exists in pisciculture. The task of
creating a similar method for calculating the losses
related to morbidity and invalidism of the popu-
lation continues to be very urgent.
5) Possibility of obtaining more immediate in-
direct data on the harmful effect of chemical envi-
ronmental contaminants upon the health of the pop-
ulation: loss of health from a reduction in the food
reserve, in particular from a decrease in protein
consumption, can be determined in general. To a
certain degree this can be used to control the effect
of pesticides on the nutritional properties of plants;
however specific computations are complicated.
6) More remote indirect data (for example, ex-
penditure of capital investments necessary to main-
tain health, to prevent or compensate for damage
from water and atmospheric corrosion of construc-
tion materials) cannot presently be used to deter-
mine the harm to health reliably (computed for du-
ration of life, for labor effectiveness, for the period
of efficiency, for the reproductive period, etc.)
7) Protection of the environment from harmful
contamination must be based on the principle that
medical and biological indices have preference over
the technical needs of today and over possible eco-
nomic losses. It should be realized that ways can be
found to radically change technology or to organize
production (the main sources of contamination) so
as to completely meet the medical, technical, and
economic requirements.
One example is establishing the MPC of sUicosis-
producing dust at many mines due to underground
crushing of ore and the introduction of new efficient
84
-------
technology which reduces the number and intensity
of sources of dust emission (decrease in the silicosis
morbidity plus high savings), the replacement of
benzene-containing glues in industry with more ef-
fective ones (protection of health in combination
with high economic savings), the elimination of
dangerous stages in a number of pharmaceutical,
paint and varnish, and other chemical products
(protection of health in combination with high eco-
nomic savings), and many others.
It is evident that the correlation between sanitary
standards and technical (or economic) achieve-
ments makes it impossible to reach a contamination
level which is dangerous from a medical viewpoint.
The sanitary standards must be progressive and al-
so must favor engineering advances. Often the for-
eign literature points to the supposedly inevitable
destruction of the natural environment as a result of
scientific and technical progress. Such a viewpoint
is unacceptable to Soviet hygienists, who believe
the development of science and technology in itself
does not necessarily result in negative con-
sequences. The problem is how to use the fruits of
scientific and technical progress in the interests of
all mankind without violating the ecological balance
in nature.
85
-------
THE RELATIONSHIP OF CLINICAL AND EPIDEMILOGIC STUDIES
OF THE ENVIRONMENT TO EACH OTHER AND TO PUBLIC
POLICY DECISIONS
ROBER S. CHAPMAN
I. INTRODUCTION
Before I begin my talk, I would like to introduce
myself to you. I am an epidemiologist with the U.S.
Environmental Protection Agency's Health Effects
Research Laboratory in Research Triangle Park,
North Carolina. In the time that I have worked with
the U.S.E.P.A., I have been most concerned with
the determination of the effects of air pollution on
community health. The pollutants I have studied
most closely are emissions from large stationary
sources and automobiles, and reaction products of
these emissions. The examples I give him in this
talk all relate to these types of pollutants.
Today I wish to describe the ways in which I have
observed the disciplines of environmental epide-
miology and environmental clinical studies to relate
to each other. (For the purposes of this talk, 1 am
defining "clinical studies" as all experimental stud-
ies of the effects of the environment on human
health. Under this broad definition, it is not neces-
sary that "clinical studies" be performed in strictly
clinical settings.)
I have three goals in my talk. My first goal is to
describe the major strengths and limitations of epi-
demiologic and clinical studies relative to each oth-
er. My second goal is to describe how the findings
of epidemiologic and clinical studies can be com-
bined to enhance our knowledge of pollution's ef-
fects of health. (Let me add here that the results of
experimental studies of animals contribute greatly
to our knowledge, and often supply important hy-
potheses for testing in human studies. However,
since Dr. Cranmer will soon be discussing studies of
animals, I will minimize discussion of them in my
talk.) My third goal is to discuss briefly my concept
of the scientist's role with respect to those who
make public environmental decisions.
II. STRENGTHS AND LIMITATIONS OF
EPIDEMIOLOGIC AND CLINICAL
STUDIES RELATIVE TO EACH OTHER
One of the most important objectives of environ-
mental health researchers is to develop quantitative
relationships which accurately describe human re-
sponse to a wide range of doses of environmental
contaminants. Neither epidemiologic nor clinical
studies alone can provide all the information needed
to develop such relationships completely. The epi-
demiologic approach is primarily observational,
whereas the clinical approach is primarily experi-
mental. Both approaches must be used to gain a
complete description of pollution's effects on hu-
man health. In the following paragraphs I will out-
line some of the major strengths and limitations in-
herent in these two basic approaches as they are
used in environmental health studies.
A. Strengths of Epidemiology Relative to Clinical
Studies
The epidemiologist does not produce change; he
merely observes it. Thus he is able to study the full
spectrum of pollution effects, from subtle changes
of uncertain significance to dramatic changes of
very clear significance. In our studies at EPA we
have investigated such a range of effects. At one
end of the severity range, we have conducted stud-
ies to determine whether exposure to several dif-
ferent types of pollution affects the pulmonary func-
tion of young schoolchildren. We have observed, as
is shown in Figures 1 and 2, that children living in
New York City, which experiences elevated levels
of sulfur oxide and particulate pollution, have gen-
erally had lower age- and height-adjusted mean
three quarter second forced expiratory volume
(FEV0.75) than comparable children living in a less
polluted community about 75 miles from New
York, [1,2]. Similarly, as is shown in Figures 3 and4,
children living in Birmingham, Alabama, which has
elevated levels of particulate pollution only, had
lower age- and height-adjusted mean FEV0 75 than
children in the less polluted city of Charlotte, North
Carolina [3]. The pollution-related differences in
pulmonary function that we have observed have
been quite subtle, and as yet their long-term clinical
significance remains quite unclear.
Near the end of the severity range, we have per-
formed epidemiologic studies of a variable whose
ominous clinical significance is quite clear, the
86
-------
Figure 1. Mean age and height adjusted FEV,) 75 in white schoolchildren, Riverhead vs. New York City average, 1970-71
FEV0.75
LITERS
2.0
1.8
1.6
1.4
1.2
MALES
O = RIVERHEAD
A = NYC AVERAGE
FEMALES
9-13 YEARS
1.0
0 1—L
5 -8 YEARS
I
NOV-
DEC
JAN
FEB-
MAR
APR
prevalence of chronic bronchitis. To date, exposure
to combined sulfur oxide and particulate pollution
has been quite consistently associated with mea-
sured prevalence of this illness [4], The consistency
of this association as observed in EPA studies is
shown in Table 1. I know of no way to document
the influence of pollution on illnesses as severe as
chronic bronchitis except through epidemiologic
studies. Ethical and practical considerations quite
clearly prohibit any experimentation intended to
produce lasting and debilitating illness in human
subjects.
A second asset of the observational method is the
epidemiologist's ability to examine, to a greater de-
gree than the experimenter, the effects of pollution
exposure on those whose health is already com-
prised. In such people, the epidemiologist can even
measure the degree to which pollution exposure
may have a fatal result. Perhaps the most striking
example of this fact is the observation of some 4000
deaths more than expected immediately following
the great London Smog of 1952 [5]. Virtually all
people dying prematurely had suffered impaired
9-13 YEARS
NOV-
DEC
JAN
FEB-
MAR
APR
health before the smog episode. More recently, a
U.S.E.P.A. examination of mortality statistics has
shown about 23 deaths more than expected just af-
ter a high pollution episode in Pittsburgh, Pennsyl-
vania in November 1975 [6]. This figure represents
an excess of about 14% over expected. During this
episode in Pittsburgh, the average concentration of
total suspended particulate matter in the air was
about 900 fig/m3, and the average concentration of
sulfur dioxide was about 200 fig/m3.
I hasten to add that experimental studies of pa-
tients with compromised health have often yielded
most interesting and useful results. For example,
clinical researchers at E.P.A. have shown quite
clearly that exercising patients with angina pectoris
experience chest pain more quickly when exposed
to 50 parts per million carbon monoxide than when
breathing pure air [7]. To date, this remains one of
the most important findings available with respect
to the health effects of carbon monoxide. Thus I am
in no way suggesting that people with impaired
health do not fall within the legitimate purview of
the clinical investigator.
87
-------
Figure 2. Mean age and height adjusted FEV0 ,75 in white schoolchildren, Riverhead vs. New York City average, 1971-72
FEV075 o = RIVERHEAD
LITERS A = NYC AVERAGE
2.0
p—
MALES
FEMALES
1.9
Cy .A
•
1.8
1.7
9-13 YEARS
1.6
—
— 13 YEARS
1.5
-
-
1.4
—
—
1.3
-
1.2
— ^^5 -8 YEARS
1.1
-
- 0^5 8 YEARS
1.0
—
0 1
r i i i
' 1 1 1
FALL WINTER SPRING FALL WINTER SPRING
OCT- (FEB-MAR) (APR-MAY) OCT- (FEB-MAR) (APR-MAY)
NOV
NOV
A third strength of the epidemiologic method is
that it allows assessment of the effects of very long-
term exposures to pollution. Quite probably, the ef-
fect of pollution on chronic bronchitis that I have
described requires at least a decade to develop.
Quite conceivably, it requires two decades or more.
It seems most unlikely to me that experimental ex-
posures of such duration can ever be performed.
A fourth opportunity enjoyed by the epidemiolo-
gist is his ability to study very large samples of
people. For example, each of the four studies of
chronic bronchitis that I have mentioned had at
least 6000 subjects. In clinical studies, the number
of subjects is usually very much smaller than this.
Indeed, I have rarely read a clinical study of pollu-
tion exposure effects which employed more than
about 50 subjects. Most such studies have em-
ployed considerably fewer. Because he is able to
study large samples, the epidemiologist is often able
to assign statistical significance to differences which
would emerge from an experimental study as non-
significant.
Fifth, the environmental epidemiologist studies
the effects of ambient atmospheres in which people
actually live. He is thus able to study the effects of
very complex interactions among pollutants, even
though he is usually unsure of the exact nature of
these interactions. Because the ambient atmo-
sphere of virtually all cities contains a very complex
mixture of pollutants, it is entirely conceivable that
many if not most of the effects which have been as-
sociated with pollution exposure are due to these
complex interactions, and not to individual pollu-
tants. To date, it has not been possible to reproduce
a realistic variety of pollutants and their reaction
products in experimental situations. Because the
ambient atmosphere is so complex, the experiment-
er may never be able to reproduce it fully.
Sixth, the epidemiologist can make measure-
ments with only minimal intrusion into study sub-
88
-------
2.0
1.9
1.8
1.7
M I*®
£ 1.5
00 ^
® £ 1.4
1.3
1.2
1.1
1.0
0
1_1
FALL
MALES
WINTER
SPRING
u>
a>
[ L
FALL
FEMALES
oCHARLOTTE
~ BIRMINGHAM
-O
-a —
WINTER
SPRING
Figure 3. Average FEV».TS in Charlotte and Birmingham white children aged 5-8 years in 1971-72 school year
-------
FALL
MALES
WINTER
SPRING
2.0
1.9
1.8
1.7
§ 1.5
£ 1.5
O
Si 1,4
Lu
1.3
1.2
1.1
1.0
0
I
FALL
FEMALES
o CHARLOTTE
~ BIRMINGHAM
WINTER
SPRING
Figure 4. Average FEV0 73 in Charlotte and Birmingham black children aged 5-8 years in 1971-72 school year
-------
TABLE 1
CHRONIC BRONCHITIS PREVALENCE RATES, DISTRIBUTED BY SMOKING STATUS AND POLLUTION
EXPOSURE: UNITED STATES STUDIES, 1969-1970
Community
Exposure &
Smoking Status
Low Pollution
Nonsmokers
Smokers
High Pollution
Nonsmokers
Smokers
Utah
Mothers
4.16
15.80
5.20
22.25
Fathers
3.00
19.60
6.81
26.80
Chronic Bronchitis Prevalence (in %)
1.08
11.78
2.54
12.88
1.25
17.05
3.48
18.36
New York
Idaho-Montana
Mothers Fathers Mothers Fathers
2.00
13.90
6.04
18.08
4.60
13.89
15.87
21.71
Chicago
Blacks
Whites
9.20
8.74
9.84
13.52
4.00
15.20
5.32
18.08
jects' routines, and with only minimal removal of
subjects from familiar surroundings. In clinical re-
search, on the other hand, subjects are generally re-
moved from their daily routines for considerable pe-
riods of time, and are almost always placed in very
unfamiliar surroundings.
I cannot quantitate the degree of stress imposed
on subjects by experimentation. Nor can I predict
the exact degree to which such stress may influence
study outcomes. However, it seems quite conceiv-
able to me that the stresses attendant on the experi-
mental situation may sometimes have a subtle but
important effect on the outcome of experimental
studies.
Finally, 1 wish to digress for a moment to caution
against the popular notion that experimental studies
are fairly easy to duplicate. Let me present an ex-
ample. Over the last several years in the U.S., there
has been debate regarding the effects on time dis-
crimination of exposure to fairly low levels of car-
bon monoxide. This debate has stemmed primarily
from the disparate results obtained by two investi-
gators in experiments which were designed to be
identical. One investigator in the state of California
observed subtle but statistically significant impair-
ment of time discrimination in subjects exposed to
50 parts per million of carbon monoxide.[8] Another
investigator in the state of Wisconsin observed no
such impairment at 50 ppm carbon monoxide. [9]
Both of these investigators are experienced and re-
spected. However, even though care was taken to
duplicate experimental conditions, different results
were observed. I believe that this example under-
scores the point that, even in experimental studies,
certain factors affecting study results remain
beyond the control of investigators.
B. Strengths of Clinical Studies Relative to
Epidemiology
As I have mentioned, I am an epidemiologist. It is
thus perhaps nature that I have discussed the
strenghts of epidemiologic studies first. However,
nearly all of the strengths of epidemiologic studies
are accompanied by limitations inherent in the epi-
demiologic method. Many of these limitations, hap-
pfly, are offset by strengths of the experimental
method employed in clinical studies. I wish to dis-
cuss now the most important strengths of clinical
studies as I have seen them from an epidemiolo-
gist's perspective.
First, the experimental researcher is usually able
to make more numerous and complex physiologic
and biochemical measurements of each individual
subject than is the epidemiologist. For this reason,
clinical studies, to the present time at least, are bet-
ter suited than epidemiologic studies to the detailed
explanation of physiologic and biochemical mecha-
nisms which accompany subject's symptomatic re-
sponses to pollution exposure. For example, a large
epidemiologic study of symptoms such as chest dis-
comfort and coughing was conducted in the area of
Los Angeles, California [10]. (This area frequently
experiences elevated concentrations of photochem-
ical oxidant air pollution, which results from in-
teractions between sunlight and automobile ex-
haust. An important component of this type of pol-
lution is ozone.) In this epidemiologic study, the
frequency of symptoms among student nurses was
observed to increase with increasing levels of pho-
tochemical oxidants. However, for logistic and eco-
nomic reasons, it was impossible to include in the
study a set of detailed measurements of the sub-
jects' pulmonary function.
To learn the effect of photochemical pollutants on
pulmonary function, we have had to rely mainly on
clinical studies. To date, we have learned from clini-
cal studies of exercising subjects that exposure to
1500 ng/m3 (0.75 parts per million) of ozone pro-
duces a fall in maximum transpulmonary pressure,
suggesting an inhibition of subjects' capacity to take
a full inspiration [ 11]. Exposure to the same concen-
tration of ozone also produces an increase in the
airways' resistance to air flow. Exposure to (725 /xg/
m3) 0.37 ppm of ozone has been observed to pro-
duce a decline in rates of air flow when the lungs are
filled to one-half of vital capacity, [12] as is shown
in Figure 5. In these experimental studies, coughing
and chest discomfort were observed among sub-
jects, as they were observed in the epidemiologic
study of student nurses. These experimental studies
91
-------
Figure 5. Effect of SO2 and Oj on maximal mid-expiratory flow rate (MMFR)
100
O
cc
I-
2
O
o
u_
O
z
UJ
u
cc
UJ
a.
70 h-
60 r—
50 h-
0.37 ppm SC>2 (n = 4)
0.37 ppm O3 (n = 8)
0.37 ppm SO2 + 03(0 = 8)
EXPOSURE TIME
(HOURS)
END
OF EXPOSURE
do not prove that ozone was the only agent, or even
the principal agent, in the promotion of the symp-
toms observed in the student nurses. However, the
experimental studies provide most interesting infor-
mation as to physiologic mechanisms that may ac-
company the symptoms experienced in the commu-
nity during photochemical oxidant exposures.
I mentioned that in epidemiologic studies, the in-
vestigator can assess the effects of very long-term
exposures to air pollution. However, a closely re-
lated limitation of epidemiologic studies is that the
epidemiologist is rarely able to determine with pre-
cision the duration or level of exposure required to
promote the effects that he observes. As I have
mentioned, elevated long-term exposures to com-
bined sulfur oxide and particulate pollution have
been linked quite consistently with elevated preva-
lence rates of chronic bronchitis. However, the ex-
act duration and level of exposure required to pro-
mote this disease remain very uncertain. To date, it
has proven impossible to conduct epidemiologic
studies for periods of time long enough to observe
this disease develop in a healthy population.
Thus the second advantage of clinical studies rel-
ative to epidemiology is the experimenter's ability
to determine the time course of physiologic changes
very precisely. In one of the clinical studies I have
described, [12] the decline in subjects mid-ex-
piratory airflow at 725 fig/m3 (0.37 ppm) ozone was
observed to decrease quite steadily over the first
hour of exposure and to remain fairly constant over
the second hour. When exposure was discontinued
after two hours, recovery of full function began al-
most immediately. It is difficult to conceive of such
detailed physiologic characterization in a large-
scale epidemiologic study.
1 have mentioned that the epidemiologist ob-
serves the effects of the complex pollutant mixtures
that actually exist in the ambient air. Unfortunately,
this opportunity to observe realistic situations com-
es only at a price of great uncertainty as to just
which pollutants are promoting the observed ef-
fects. It has proven extremely difficult to find loca-
tions for epidemiologic study in which only a single
air pollutant is present. As yet, statistical methods
are not fully equal to the task of disentangling the
effects of one pollutant, or of a specific interaction
between two pollutants, from the effects of other
92
-------
pollutants or interactions. Thus the third strength
of clinical studies is their ability to determine the
effects of individual pollutants. For example, as the
studies I have discussed indicate, we are beginning
to acquire a detailed picture of the effect of ozone
upon pulmonary function. As I will discuss later,
knowledge of individual pollutants' effects can be
most useful in disentangling the complexities that
exist in ambient atmospheres.
Fourthly, the clinical investigator is often able to
administer finely graduated doses of pollutants to
subjects. The epidemiologist must rely on factors
beyond his control to provide the pollution doses
whose effects he observes. Also, it has not yet prov-
en possible for the epidemiologist to measure pollu-
tant doses for individual subjects as accurately as
can the clinical investigator. Thus, to the present
time at least, clinical studies have provided consid-
erably more certainty than epidemiologic studies as
to the exact exposure concentrations required to
produce specific effects.
Fifthly, the clinical investigator is also able to ex-
amine the effects of specific interactions between
pollutants more easily than is the epidemiologist.
For example, in one of the clinical studies I have
described, [12] exposure to 0.37 ppm ozone alone
resulted in a 12% reduction in mid-expiratory air-
flow rate. Exposure to 0.37 ppm sulfur dioxide
alone resulted in no discernible reduction of in flow
rate. However, exposure to both gases simultane-
ously resulted in a 34% reduction, suggesting an in-
teractive effect over and above the effect of each
individual pollutant. It is difficult to conceive of an
epidemiologic study yielding such a precise quan-
titation of the effects of an interaction between two
specific pollutants. Such quantitation is most valu-
able because, in this case, it generates concern as to
what the health effects of pollution may be in areas
which experience elevated levels of both sulfur
oxides and ozone.
Finally, many epidemiologic studies are cross-
sectional in nature. In such studies, the investigator
measures both pollutant concentrations and health
indices at a single point in time, and therefore can-
not be entirely certain as to whether elevations in
pollutants levels actually preceded elevations in ill-
ness rates or physiologic impairment. Thus the epi-
demiologist must generally rely on the weight of ac-
cumulated evidence, and not on the results of any
single study, to help him in his judgments of the ef-
fects of pollution on human health.
On the other hand, in clinical studies temporal
relationships between pollution exposure and the
appearance of physiologic effects are usually clear-
er than in epidemiologic studies. Thus clinical stud-
ies are often free of a factor which may quite se-
verely limit the interpretation of epidemiologic
studies. I hasten to add that, since numerous other
factors, some of which I have mentioned, can affect
the results of clinical studies, I believe that these
studies should be replicated no less than epidemio-
logic studies.
II. COMBINING EPIDEMIOLOGIC AND
CLINICAL RESULTS TO ENHANCE
KNOWLEDGE OF POLLUTION EFFECTS
I have given quite a long recital of differences be-
tween epidemiology and clinical studies. In this re-
cital, I sincerely hope I have not created the impres-
sion that these two disciplines work toward oppos-
ing purposes. To the contrary, I strongly believe
that results from both disciplines can be brought to-
gether to produce a much more complete assess-
ment of pollution's health effects than can the re-
sults from either discipline alone. (Of course, the
results of experimental studies of animals enhance
the assessment further still. However, as I have
mentioned, I am arbitrarily limiting myself to dis-
cussing studies of humans.)
I would like to present an example of the way in
which epidemiology and clinical studies can be used
together to enhance our knowledge of the effects of
pollution exposure. First, let me describe the re-
sults of an epidemiologic study conducted in the
Los Angeles area between 1959 and 1964 [13]. Over
these years, the investigators measured the times
required for high school athletes to run over a 2.2
mile course during competitive track meets. In all,
results from 21 meets were analyzed. For each of
the six years in the study, the proportion of the
track team failing to improve running time between
meets was correlated with pollutant levels during
the hour of the race and each of the three hours pre-
ceding the race. Photochemical oxidant levels dur-
ing the hour before the race bore a stronger relation-
ship to impairment of athletic performance than lev-
els of any other pollutants measured. The overall
correlation coefficient between this photochemical
oxidant level and the proportion of runners failing
to improve their times was 0.88. The investigators
also computed coefficients of correlation between
pollution levels and impaired performance separate-
ly for the years 1959-61 and 1962-64. When the
years were separated in this way, the relationship
between oxidant levels in the hour before the race
and impaired performance became even stronger.
As Figure 6 shows, for each of these two groups of
years the correlation between oxidant and impaired
performance was 0.945, which denotes a very close
association indeed.
The results of this study strongly suggest that
photochemical oxidant exposure promotes impair-
ment of athletic performance. However, these re-
sults do not indicate the physiologic mechanisms
underlying the impairment. Nor do they indicate
which specific component of the broad class of pho-
93
-------
Figure 6. Percent of runners with decreased performance,
distributed by oxidant concentration in the hour
before track meets (Reference 13)
Percent with
decreased
performance
80
-iCs 1959-1961
r= .945
O 1962-1964
r= .945
AL- 1 1 r1
0.10
0.20
0.30
tochemical oxidants was most important in promot-
ing the observed effect. To gain insight in these
areas, we must turn to two clinical studies which
have been performed over the last several years in
Canada [14,15].
In the first of these studies, different groups of
health young adults were exposed for two hours to
0.37, 0.50, and 0.75 parts per million ozone. Half of
the total of 28 subjects exercised intermittently dur-
ing exposure; the other half rested throughout ex-
posure. After the exposure period, all subjects un-
derwent a three-stage submaximal exercise test,
with work loads adjusted to 45,60 and 75 percent of
predicted maximal aerobic power. All subjects also
underwent exposures to clean air, while observing
the same experimental procedure as during ex-
posure to ozone. Shortly after cessation of effort,
physiologic measurements were made.
The m^jor physiologic effect of ozone exposure
was a change in respiratory pattern. That is, ozone
exposure produced increases in respiratory fre-
quency in most subjects and decreases in tidal vol-
ume in most of them. The severity of these effects
increased with the delivered dose of ozone, and
with the amount of exercise during the exposure pe-
riod. No changes in oxygen uptake or minute vol-
ume (the amount of air breathed per minute) were
Oxidant concentration in ppm
observed in this experiment. This result suggests
that in healthy young people exercising sub-
maximally, an increased respiratory frequency is
able to compensate for a decreased tidal volume.
In a second experiment, the same investigators
exposed 11 healthy young adults to 0.75 ppm ozone
for two hours. All subjects exercised intermittently
during exposure, and upon cessation of exposure
exercised maximally until exhausted. Following the
same procedure, all subjects were also exposed to
clean air.
In these subjects, tidal volume following ozone
exposure was lower than following clean air ex-
posure. However, unlike the subjects exercising
submaximally, the subjects exercising maximally
experienced no compensatory increase in respira-
tory frequency following ozone exposure. Thus, in
the maximally exercising subjects, minute volume
and oxygen uptake were both decreased after ozone
exposure.
The results of these experimental studies are
most consistent with the hypothesis that ozone
stimulates irritant receptors in the lungs, and con-
sequently produces a restriction of the lungs' in-
spiratory capacity. (That irritant receptors should
be stimulated comes as no surprise in light of the
frequent observation of coughing and sore throat in
94
-------
experimental subjects exposed, to ozone.)
These experimental results provide a most inter-
esting insight into the physiologic mechanism that
may underlie the epidemiologic observation of im-
paired performance in high school athletes. Please
remember that these athletes were also exercising
at maximum levels. In light of the experimental re-
sults, it is entirely conceivable that, on days of high
oxidant exposure, ozone irritated the athletes' res-
piratory systems fufficiently to produce restriction
of their lungs' inspiratory capacity. However, it al-
so appears that the runners would not have been
able to develop a compensatory increase in respira-
tory frequency, and the total amount of air that they
could have moved in and out of their lungs would
therefore have decreased. Consequently, their oxy-
gen uptake and their maximum sustainable work
load would have decreased, leading finally to im-
pairment of running time.
The Canadian experimental results are relevant
to the epidemiologic results in another important
way. As 1 mentioned, the epidemiologic results of
the athlete study give no indication of what specific
component of photochemical oxidant pollution con-
tributes most importantly to the observed effect on
running times. The experimental results do not rig-
orously prove that ozone alone produced this ef-
fect. However, they do suggest that exposure to
ozone alone, at least on days when ozone levels
were high, may have exerted an effect sufficient to
impair running times. Of course, it is also possible
that other components of photochemical pollution
contributed to the observed effect.
III. THE RELATIONSHIP OF SCIENTISTS TO
ENVIRONMENTAL DECISION MAKERS
As I mentioned early in this talk, one of our most
important goals as environmental researchers is to
develop relationships which accurately express the
effect of pollution doses upon physiologic and path-
ologic responses. Because a great many administra-
tive decisions are based on environmental research,
we are under no small pressure to develop these
relationships as quickly as possible.
However, 1 believe that we have a second re-
sponsibility, which is just as important as the con-
duct of research. This is the responsibility to com-
municate research results as completely and clearly
as possible to those who make public decisions. For
three reasons, I believe that this responsibility is at
least as difficult to discharge as is our responsibility
to perform research. First, our responsibility to
communicate and educate requires us to learn as
much as we can about the researches of others, not
only in our own discipline, but in other disciplines
as well. As I hope I have shown, the synthesis of
information from different disciplines can greatly
enhance our knowledge of pollution's effects, and
can greatly strengthen the confidence with which
public decisions can be made.
Second, our educative responsibility often re-
quires that we express ourselves in language that
people without technical backgrounds can under-
stand. For me, this has proven to be a difficult task
indeed, particularly when concepts have been com-
plicated, or when the current state of knowledge
has been ambiguous.
Third and finally, I believe that our educative
function requires us to state very clearly the uncer-
tainties still inherent in numerous areas of environ-
mental research. Indeed, for every question that we
are able to answer through our research, new ques-
tions seem to arise. At times, uncertainty arises
from conflicting research results. At other times, it
arises from limitations in available methods for con-
ducting and interpreting research. For epidemiolo-
gists, the long-term effects of pollution represent an
area of great uncertainty, because many locations
have not yet been polluted long enough for such ef-
fects to develop. One such location is Los Angeles,
which has experienced elevated photochemical oxi-
dant levels for only about the last 30 years.
The systematic study of the relationships be-
tween man and his environment is still a very young
discipline. Before it is fully mature, it must over-
come the obstacles that all young things must over-
come. I fully believe that many of the uncertainties
we face will be resolved in time. I also believe that
the best atmosphere for advancement of our efforts
will be one of vigor coupled with great patience on
the part of researcher and decision maker alike.
REFERENCES
1. Shy, C. M. et al. Ventilatory Function in School Children:
1970-1971 New York Studies. In "Health Consequences of
Sulfur Oxides: A Report from Chess, 1970-1971." U.S.
Government Document No. EPA 650/1-74-004. May 1974.
pp. 5-109 through 5-119.
2. Chapman, R. S., V. Hasselblad, R. Burton and J. Williams.
Air Pollution and Children's Ventilatory Function: I. New
York, 1971-72. II. Comparison of New York, 1971-72 to
New York, 1970-71. U.S.E.P.A. Intramural Technical Re-
port. June 30, 1976.
3. Chapman, R. S. et al. Air Pollution and Childhood Ventila-
tory Function: I. Exposure to Particulate Matter in Two
Southeastern Cities, 1971-72. U.S.E.P.A. Intramural Tech-
nical Report. June 16, 1976.
4. Chapman, R. S. et al. Chronic Respiratory Disease in Mili-
tary Inductees and Parents of Schoolchildren. Arch. Envi-
ron. Health 27:138-142.
5. U.S. Department of Health, Education and Welfare. Air
Quality Criteria for Sulfur Oxides. National Air Pollution
Control Administration Document No. AP-50. April 1970.
pp. 119-120.
6. Riggan, W. B. et al. Daily Mortality Models: Air Pollution
Episodes. Presented at the Eighth International Meeting of
the International Epidemiological Association. San Juan,
Puerto Rico. September 17-23, 1977.
7. Anderson, E. W. et al. Effect of Low-Level Carbon Mon-
oxide on Onset and Duration of Angina Pectoris. Ann. In-
95
-------
tern. Med. 79:46-50. July 1973.
8. Beard, R. R., and G. A. Weitheim. Behavioral Impairment
Associated with Small Doses of Carbon Monoxide. Am. J.
Pub. Health, 57:2012-2022, 1967.
9. Stewart, R. D., P. E. Newton, M. J. Hosko, and J. E. Pe-
terson. Effect of Carbon Monoxide on Time Perception.
Arch. Environ. Health. 27:155-160, 1973.
10. Hammer, D. I. et al. Los Angeles Student Nurse Study: Dai-
ly Symptom Reporting and Photochemical Oxidants. Arch.
Environ. Health. 28:255-260, May 1974.
11. Bates, D. V. et al. Short-term Effects of Ozone on the Lung.
J. Applied Physiol. J2(2):176-181. February 1972.
12. Hazucha, M. and D. V. Bates. Combined Effect of Ozone
and Sulphur Dioxide on Human Pulmonary Function. Na-
ture 257:50-51. September 4, 1975.
13. Wayne, W. S.,P, F. Wehrle, and R. E. Carroll. Oxidant Air
Pollution and Athletic Performance. JAMA /W( 12)901-904
March 20, 1967.
14. Folinsbee, L. J., F. Silverman, and R. J. Shephard. Exer-
cise Responses Following Ozone Exposure. J. Applied
Physiol. 46(6)996-1001. June 1975.
15. Folinsbee, J. J., F. Silverman, and R. J. Shephard. De-
crease of Maximum Work Performance Following Ozone
Exposure. J. Applied Physiol. 42(4):531-536, 1977.
96
-------
ANALYSIS OF THE PRIORITY SERIES OF CONTAMINANTS FROM
THE PHYSIOLOGIST'S VIEWPOINT
[Otsenka prioritetnogo ryada zagryaznitely s pozitsiy fiziologa]
PROFESSOR B. A. TASHMUKHAMEDOV
One of the main criteria for evaluating the priority
of a series of contaminants is their physiological ef-
fect on organism®.
There is a large amount of data on the correlation
between the inhibiting action of various chemical
agents on the functional parameters of cells in vitro
and the toxicity of the same substances for the en-
tire organism. This correlation has already made it
possible to include in experimental toxicology a
number of rapid methods for evaluating new toxic
substances. Therefore, an examination of the mech-
anisms of action of contaminants on the biochemi-
cal and biophysical processes occurring in the cell
can be a very promising method of evaluating their
priority.
The category of biospheric contaminants con-
tains an extremely wide range of substances. Nev-
ertheless, they can be divided into a number of cate-
gories depending on the nature of their action on
certain aspects of cellular metabolism.
The most widespread contaminants of industrial
wastewater are heavy metal ions. Their toxic action
is governed by inactivation of the SH-containing en-
zymes as a result of oxidation of thiolic groups. The
thiolic groups are present in the molecules of the
most diverse enzymes and are often included in the
composition of their active center.
At the same time the metals can form complexes
with the carboxyl, amide, and other polar groups
(Dixon and Webb, 1966). The toxicity of ions of var-
ious metals is related to the strength of the coordi-
nation bonds; therefore, in many cases the order of
toxicity coincides with the order of coordinating ca-
pability of Irving-Williams (Ca < Mg < Mn < Fe <
Co < Zn < Ni < Cu). By having a broad spectrum
of action, the ions of heavy metals block the most
diverse metabolic processes in the cells. By enter-
ing the cells, the polyvalent ions (Ni+!, Al+3, La+3)
interact electrostatically with the negatively
charged membrane surface, and compete with the
calcium for the specific Ca-bonding sites on the
membranes (Khodorov, 1970). As a result of the
"stabilizing" action of the ions, the normal pene-
trability of the membranes is reduced, which results
in a disruption in the functions of the membranes in
the nerve and muscle cells and the synaptic trans-
fer.
Another class of substances which significantly
affects the functional state of cells is the surface-
active compounds or detergents which produce
complete degradation of membranes, and thus, de-
compartmentalization of the cells and their result-
ant death.
It should be noted that the membranes are one of
the cell targets. Apparently, this may be why many
natural toxins selected by lengthy evolution contain
protein toxins, that is, cytotoxins which interact
with the bimolar phospholipid framework of the
membranes, and thus labilize their structure. Ex-
amples of such membrane active compounds are
certain toxins of snakes, insects, higher plants, and
mushrooms.
The mode of action of many toxic chemical
agents is a disruption of the energy metabolism of
the cells. These include respiratory poisons such as
cyanides, azides, and hydrogen sulfide. These sub-
stances, due to their ability to form stable com-
plexes with heavy metals, inactivate the cyto-
chrome oxidase system of the respiratory chain.
The respiratory poisons also include one of the
most important contaminants of the atmosphere,
that is, carbon monoxide, which, in competing with
oxygen, is joined to the restored form of cyto-
chrome oxidase.
Certain pesticides of general systemic action, in
particular, rotenone, are true respiratory poisons,
and it is precisely this property of theirs which de-
termines their toxicity. At the same time respiratory
inhibition may be not the main, but instead the side
effect of some pesticides, for example, chloropi-
riphos and dischlorophos.
A broader class of substances which affect cellu-
lar energetics is the so-called disconnectors of oxi-
dizing phosphorylation which do not disrupt the
respiratory chain of mitochondria, but repeatedly
stimulate respiration. These substances disperse
the energy of the oxidation-reduction reactions, and
thus unite the cells with adenosine triphosphate.
97
-------
They are primarily fat-soluble substances with
weak acid properties which are introduced into the
membrane of the mitochondria, and by working as
artificial transmitters of protons, create in it a pro-
ton by-pass which reduces the polarization of the
inner membrane.
Similar properties are possessed by the fat-sol-
uble ions which eliminate membrane polarization as
a result of the direct passage through the membrane
according to the electrical gradient (Skulachev,
1969). These substances inhibit not only the oxidiz-
ing phosphorylation of the mitochondria, but also
the photophosphorylation of plant chloroplasts.
The classic disconnecting agents of laboratory
studies are dinitrophenol and tetrachlorotrifluo-
romethylbenzimidazole. Different derivatives of
benzimidazole are used as effective herbicides. It is
quite evident that these substances are not strictly
selective, and to an equal degree affect the mito-
chondrial apparatus of both plant and animal orga-
nisms. Acaricides from the group of formamidines
are strong disconnectors (1.10~6M), and therefore
also cannot be viewed as selective agents.
A very convenient method for evaluating the dis-
connecting activity is to study the action of different
chemical agents on the conductivity of artificial
biomolecular phospholipid membranes which can
be viewed as models of biological membranes. It
was found that many toxic chemicals used in prac-
tice have a proton conductivity several orders of
magnitude higher than that of artificial membranes,
and therefore can be included with the typical dis-
connectors. Included in this group is the wide-
spread herbicide dichlorophenoxyacetic acid (2,4-
D), and the defoliants butyphos and butyl captax.
Besides the different toxic chemicals used in
practice, the disconnectors can also include many
plant phenol compounds (Baraboy, 1976), and in
particular, gossypol from the cotton plant, which
with increased permissible concentrations in oil
cakes and groats, produce poisoning of cattle. Fi-
nally, the classic disconnecting agents are the dif-
ferent phenol compounds found in industrial waste-
water. Their action is not limited to suppression of
the energy metabolism of cells. Besides continuous
resynthesis of ATP, the mitochondria control the
level of intracellular calcium which in turn is a regu-
lator of the most diverse cellular functions. The dis-
connectors, regardless of their origin in the orga-
nism, depolarize the inner membrane of the mito-
chondria and release the calcium accumulated in
them. This is of particular importance in the syn-
apses, since an increase in the concentration of in-
tracellular calcium within the synapses results in an
increased probability of the interaction of vesicles
filled with the mediator in the presynaptic mem-
brane, and an increase in the frequency of mediator
release into the synaptic gap. As a result, the fre-
quency of action potentials rises producing hyper-
kinesia and musculature tremor.
Apparently, the most vulnerable targets in the or-
ganism for different toxic substances are the syn-
apses of the central and peripheral nervous system.
It is possible that this is why the most toxic plant
and animal poisons contain neurotoxins which
block various links in the synaptic transmission.
The most effective of the artificially synthesized in-
secticides also block synaptic transmission.
Of the 700 different chemical compounds used in
various countries as pesticides, the most active is
the organophosphorous group of insecticides which
highly specifically inhibit enzymes with esterase ac-
tivity, and primarily, cholin-esterase of cholinergic
synapses. The reactions which are catalyzed by es-
terases are implemented in two stages, that is,
transfer of the acyl group to the enzyme and further
hydrolysis of the acylated enzyme. Under the ac-
tion of organophosphorous inhibitors on the en-
zyme, the substituted phosphoryl group is trans-
ferred forming a phosphorylated enzyme which is a
very stable compound and practically does not hy-
drolyze. For example, diisopropylfluorophosphate
is an extremely active inhibitor capable of inhibiting
cholinesterase even in a concentration of 10~10 M
(Dixon and Webb, 1966).
Due to a certain conservatism in the paths of evo-
lution, the cholinergic type of synapses are repre-
sented both in insects and in higher warm-blooded
animals. The chemism of cholinergic synapses in
various types of animals is built on the same prin-
ciple, and unfortunately, a certain minor difference
found by physiologists has no importance for organ-
ophosphorous compounds which, to an equal de-
gree, block this mediation system regardless of the
evolution of the organisms. This is precisely why
we do not have sufficiently selective chemical re-
sources, and the insecticides intended for insects
fairly effectively act on higher animals, and man as
well. During poisoning with toxic doses of anti-
cholinesterase substances, acetylcholine is accumu-
lated in the synapses, which results in poisoning of
the organism by its own mediator. Death results
through a spasm in the bronchi which stops respira-
tion, while for substances which penetrate through
the hematoencephalic barrier and block the central
cholinergic synapses, spasms occur with sub-
sequent paralysis. Chronic intoxication with organ-
ophosphorus compounds results in irreversible dis-
orders in the behavioral reactions of warm-blooded
animals.
As a consequence of the high plasticity of insect
metabolism, they can adapt to insecticides, for ex-
ample, flies resistant to DDT, however the adaptive
abilities of higher warm-blooded animals are very
limited.
Data on the selectivity of the best insecticides
98
-------
used in agriculture are based on their different paths
of metabolizing by insects and warm-blooded ani-
mals. In some cases higher organisms inactivate
them more quickly, in other cases, due to the pecul-
iarities of metabolism in insects, primordially in-
active predecessors are transformed into toxic de-
rivatives with strong anticholinesterase activity
(Park, 1973). However, it appears that all living
things were made from one clay, and therefore the
Higher warm-blooded animals are not indifferent to
the chemical synthetic insecticides.
Finally, it is necessary to discuss another group
of widespread toxic contaminants. Many toxic
chemicals can have a mutagenic effect by eliciting a
change in the structure of chromosomes and for
their number in the cellular nuclear apparatus. In
addition, according to the mutation theory of car-
cinogenesis, strong mutagens must also be strong
carcinogens (Zyuss, et al., 1977). X-ray irradiation
and numerous chemical mutagens, for example,
yperite, have both a carcinogenic and mutagenic ac-
tion. Nitrosamines are classic carcinogens which al-
so have a strong mutagenic action.
However, regardless of the possible link between
mutagenicity and carcinogenicity, there are among
the chemical contaminants substances with strong
mutagenic and carcinogenic activity. The latter, in
particular include different polycyclic hydrocarbons
of coal tar and amino-azo dyes.
The most unique carcinogen found as a food con-
taminant is the group of aphlotoxins, that is, furo-
coumarins containing a lactone and cyclopentane
ring which produce the toxic mold, Aspergillus
flavus. Aphlotoxins appear in the mold of peanuts,
potatoes, bread, flour and cotton seeds under condi-
tions of high humidity. Aphlotoxins are very poi-
sonous (lethal dose 1 mg/kg) and can be the cause of
death of domestic animals as a result of liver necro-
sis. Chronic intoxication results in the appearance
of primary liver carcinoma. In rats, tumor forma-
tion was observed in 100% of the cases within 60-80
weeks with a concentration of aphlotoxins in the
diet 0.015 mg/kg (Pokrovskiy and Tutel'yan, 1976).
It should be noted that epidemiological observa-
tions in certain countries have established a high
level of primary cancer of the liver in regions where
the population consumes a large amount of aphlo-
toxins with the food.
By examining the different types of contaminants
from the viewpoint of their physiological effect on
the cells, one can draw the conclusion that the
group of respiratory poisons and disconnectors of
oxidizing phosphorylation produces more dramatic
changes in the cells and thereby acts in lower con-
centrations than the ions of heavy metals.
The most dangerous among the contaminants, re-
gardless of their origin, are the substances with
mutagenic and carcinogenic activity. In comparison
with these threatening factors even the regulation of
nervous-paralytic toxins of synaptic action is sec-
ondary.
However, in order to evaluate the priority order
of each type of contaminant, it is apparently neces-
sary to take into account not only their physiologi-
cal action on the organism and the consequences
under laboratory conditions, but also the distribu-
tion and amount of certain contaminants within the
biosphere.
REFERENCES
1. Dixon, M., and E. Webb. 1966. Fermenty [Enzymes], Mos-
cow: Mir.
2. Khodorov, B. I. 1970. Problema vozbudimosti [Problem of
excitability], Moscow: Meditsina.
3. Skulachev, V. P. 1969. Akkumulyatsiya energii v kletke [Ac-
cumulation of energy in cell], Moscow: Nauka.
4. Baraboy, V. A. 1970 Biologicheskoye deystviye rastitel'nykh
fenol'nykh soyedineniy [Biological action of plant phenol
compounds], Kiev: Naukova Dumka.
5. Park, D. V. 1973. Biokhimiya chuzherodnykh soyedineniy
[Biochemistry of foreign compounds], Moscow: Meditsina.
6. Zyuss, R., V. Kintsel' and Dzh. D. Skribner. 1977. Rak: ek-
sperimenty i gipotezy [Cancer: experiments and hypotheses],
Moscow: Mir.
7. Pokrovskiy, A. A., and V. A. Tuzhel'yan. 1976. Lizosomy
[Lysozomes], Moscow: Nauka.
99
-------
BIOLOGICAL EFFECTS OF CERTAIN ORGANIC DERIVATIVES
OF CARBAMIC ACID
[Biologicheskoye deystviye nekotorykh organicheskikh proizvodnykh
karbaminovoy kisloty]
A. P. SHITSKOVA, R, A. RYAZANOVA and F. F. ERISMAN
In this age of intensive scientific and technical
progress, industrialization, and urbanization, a con-
stantly increasing number of chemical compounds
is appearing in the biosphere which have an ad-
verse effect on the human organism, populations,
and ecosysterns. These increasing sources of con-
tamination expand their spatial borders and pro-
duce unfavorable changes in the global environ-
ment. Thus, the same questions of environmental
protection have troubled specialists in different pro-
fessions in various countries of the world.
The advances of the scientific and technical revo-
lution under conditions of the socialist society are
being used for further improvement in the material
and cultural level of the population, and for the pro-
tection and improvement of the environment. Na-
tional economic plans provide for the maximum
limitation and elimination of potentially harmful
factors related to the development of industry,
transportation, and agriculture.
The struggle against environmental con-
tamination is based on the hygienic aspects of stan-
dardizing chemicals circulating in the environment,
and the development of preventive measures direct-
ed towards improving working conditions and the
life of the population.
Determination of the concentration limits within
which environmental factors exceed the physiologi-
cal limits and are harmful to the organism is of pri-
mary importance in the formation of hygienic stan-
dards and the solution to the problem of introducing
new chemical compounds into the human habitat.
An evaluation of the danger of chemical com-
pounds is based on the results of lengthy experi-
mental studies which use highly sensitive physio-
logical, biochemical, toxicological, and other re-
search methods. Using the knowledge of modern
biology and medicine, it is possible to define the na-
ture of action of chemical compounds on the entire
test organism or on its individual organs and sys-
tems.
Among the different chemicals capable of pollut-
ing the environment, the pesticides occupy a signifi-
cant place due to their extensive use of chemicals in
agriculture.
Of the newly synthesized and applicable pesti-
cides, special attention should be given to the deriv-
atives of thio- and dithiocarbamic acid. These com-
pounds differ in chemical structure, physicochemi-
cal properties, persistence in the environment, and
degree of toxicity to animals and man. The majority
of them are potentially dangerous or are known to
have a damaging effect on the gonads, elicit muta-
genic and blastomogenic effects, change the immu-
nological reactivity, and cause disorders in the ferti-
lizing capacity and the intrauterine development of
the fetus.
The final link in the penetration of pesticides into
the human organism is mainly through food prod-
ucts, a source of potential danger.
Pesticides, being active compounds, disrupt me-
tabolism, which results in a change in the immuno-
logical reactivity of the organism. Thus, for ex-
ample, after a three month exposure to 1/50 LD50
(50% lethal dose) of Zineb (25 mg/kg), the formation
of allergic antibodies was elicited significantly ear-
lier than other signs of toxicity appeared. Zineb in
the minimum dose of 1/5000 LDso. elicited a more
pronounced reaction of autoantibody formation
within seven months after the start of the experi-
ment than a dose of 1/50 LD50. These data indicate
that large doses suppress the allergic reaction and
that the threshold of allergic action can be lower
than the threshold of the toxic action. Reduction in
the intensity of immunity and change in the immu-
nological status of the organism is also indicated by
inhibition of the antibody forming function after
vaccination with typhoid vaccine; an increase in the
degree of microorganism spread on the skin; a re-
duction in the manufacture of antibodies after im-
munization; and a change in the content of total pro-
tein and S-reactive protein, protein fractions, and
the complementary activity of lysozyme.
According to modern ideas, the genesis of tumor
growth is based on changes in the immunobiological
reactivity of the organism, which makes it possible
100
-------
to assume a possible immunological dependence
and relationship between these two processes.
The carcinogenic effect can be viewed as a result
of the interaction of three factors: the organism, the
amount of active agent, and time. The threshold
quantity of the carcinogen thus determined has a
relative nature and depends on individual sensitivi-
ty, the number of experimental animals, and the
procedure used to evaluate the blastomogenic ef-
fect. Classic experiments to reveal the blastomo-
genic activity of chemical compounds are very
time-consuming due to their duration over the en-
tire life-span of the test animals. The characteristic
elicits the need for using rapid methods such as or-
gan cultivation with transplacental entrance of the
chemical compounds and the method of oncogenic
viruses.
On the basis of increased sensitivity of the em-
bryonal tissue in experimental animals, the trans-
placental method was used to study the blastomo-
genic activity of derivatives of dithiocarbamic
acid—Zineb, Cichos, preparation 275, and calcium
salt of dithiocarbamic acid. Their effects differed
depending on its physicochemical properties and its
affiliation to a certain group of compounds. For ex-
ample, morphological changes in the embryonal tis-
sue occurred at various stages of growth.
Thus, in a 1/10 LDS0 (130 mg/kg) exposure to the
calcium salt of dithiocarbamic acid, diffuse hyper-
plasia of the cultivated embryonal murine kidney
occurred on the fourth day of its growth, that is, at
the same time as with the well known carcinogen,
ethylnitrosourea. In the next days of cultivation,
focal hyperplasia was observed in the renal epithe-
lium and cystic expansion of the canaliculi.
The histomorphological disorders which occur
can be evaluated (according to the classification of
L. M. Shabad) as the first pretumoral changes
which are capable under specific conditions of be-
coming malignant tumors.
Zineb at the same dosage (1/10 LDM) elicited only
diffuse hyperplasia on the 18th day of growth of the
embryonal kidney.
On the basis of these and other findings, a recom-
mendation for the use of rapid methods of hygienic
evaluation can be made. These rapid methods can
be used for the preliminary evaluation of the blas-
tomogenic activity of chemical compounds, and the
selection of the most toxic compounds for further
studies on blastomogenicity using the generally ac-
cepted classical methods.
In the evaluation of the biological action of deriv-
atives of dithiocarbamic acid, special attention was
given to detection of the embryotoxic, teratogenic,
and gonadotoxic effects, and the establishment of a
quantitative correlation of the general toxic and
specific action.
Disorders in embryonal development occur in the
pregnant female both under the indirect influence of
chemical compounds on the developing embryo,
and by direct penetration through the placenta.
The embryotoxic effect of Zineb and preparation
275 in doses of 1/20 LD50 was characterized by the
death of the embryos, anomalies in the develop-
ment of the progeny (lack of extremity, shortening
or curvature of tail), retarded physical development
or delay in the growth of the body mass, and death
of the progeny in the postnatal period.
Pesticides can produce similar changes in the pro-
cesses of ovo- and spermatogenesis. Studies have
also shown that chronic exposure to Zineb and
preparation 275 in the indicated dose results in the
disruption of the estrous cycle, drop out of the es-
trus phase, and dominance of the dormant phase.
Histological analysis of the ovarian structure has re-
vealed an increase in the number of yellow bodies
and atretic follicles.
The gonadotoxic effect in male rats was charac-
terized by a change in the function and histological
structure of the testicles. The mobility time of the
spermatozoa was reduced, the acid and osmotic re-
sistance was decreased, and the number of sperma-
tozoa was diminished. Histologically, changes were
seen as an intensification in desquamation and dis-
organization of the spermatogenic epithelium, the
appearance in the lumen of the canaliculi of multi-
nuclear giant cells, a flattening of cells in the cubic
epithelium in the seminal vesicles, and a reduction
in the number of secretory granules.
Since the frequency of reproductive disorders in
a number of cases was related to the action of ge-
netic factors, one of the limiting criteria for harmful-
ness in the evaluation of the biological action of the
dithiocarbamic acid derivatives can be the index of
mutagenic activity. The cytogenetic method was
used to establish the dependence of frequency of
chromosomal aberrations in the murine bone mar-
row cells on the dosage of the pesticides Zineb,
preparation 275, and Cichos. Chromosomal dis-
orders in the form of paired fragments and meta-
centric formations, were observed at pesticide
doses of 1/5 and 1/20 LD50 most often in the first
days of the exposure. Chromosomal disorders were
not found during the acclimation period.
The detection of the isoeffective level, according
to indices of general toxicity, and the gonadotoxic
and mutagenic action of the studied pesticides,
makes it possible to predict the effect level for man
and incorporate a definite safety coefficient which
guarantees the safe use of pesticides.
The coincidence of the general toxic and cyto-
toxic effects of preparation 275 with its action in a
dose of 1/200 LDM in a chronic experiment has per-
mitted an objective substantiation of the hygienic
standard for the pesticide in reservoir water.
The evaluation of the biological action of dithio-
101
-------
carbamic acid derivatives has in recent years em-
phasized studying the condition of the cardiovascu-
lar system, since the cardiovascular system is most
often involved in the pathological process in addi-
tion to damage to other organs and systems. Based
on this, a chronic study of the action of preparation
275 on the cardiovascular system was performed
which considered the functional characteristics of
the heart and vessels, biochemical and morphologi-
cal indices, and a comparison of them with integral
changes.
Analysis of the bioelectric cardiac activity in rats
exposed chronically to preparation 275 in doses of
1/200, 1/2000 and 1/20,000 LD50 (respectively, 4,
0.4, 0.04 mg/kg) indicated that large doses produce
tachycardia, reduction in the amplitude of the R
wave, and an increase in the length of the interval
QT and PQ.
Changes in the bioelectric activity of the heart
were accompanied by morphohistochemical dis-
orders in the myocardial tissue, a reduction in the
activity of the enzymes succinic dehydrogenase and
NAD-diaphorase, and an increase in the content of
lactate dehydrogenase. Changes in the intracellular
energy exchange accompanied by the disruption of
protein and carbohydrate exchange resulted in a
change in the tissue respiration, the oxidizing
phosphorylation associated with it, and in the de-
velopment of dystrophic changes in the myo-
cardium.
Analysis of the qualitative alterations according
to the index of the condition of the cardiovascular
system proved to be the most stress-sensitive mea-
sure, and therefore permitted establishment of the
threshold of chronic action of preparation 275 at a
level of 1/2000 LD5o- This threshold was taken into
consideration in the hygienic standardization of the
pesticide in reservoir water.
Experimental studies which evaluated the biolog-
ical action of dithiocarbamic acid derivatives have
convincingly shown the urgency and importance of
studying remote consequences: timely detection of
the blastomogenic, gonado-, embryotoxic, muta-
genic, and imunobiological effect; as well as analy-
sis of their effect on the cardiovascular system.
The discussion of dithiocarbamic acid deriva-
tives, has shown that an evaluation of the biological
action of chemical compounds must be based on a
multilateral analysis of the state of the organism as a
single system, and on the need for using a compre-
hensive approach to hygienic standardization.
102
-------
USE OF ANIMAL AND LABORATORY TESTS TO SCREEN FOR
TOXIC EFFECTS
MORRIS F. CRANMER, Ph.D.
ABSTRACT
There are more agents contaminating, or with the
potential to contaminate, the environment than
there are resources to test via comprehensive tox-
icological protocols; therefore, the establishing of
priorities for compounds to be tested and the tailor-
ing of protocols is essential. Development of rapid
screening methods and an understanding of the
mechanisms of toxicity are the greatest challenges
facing toxicology. Currently great effort is being
placed on the development and validation of predic-
tive in vitro mutagenic and/or carcinogenic
bioassays. Unfortunately, from the standpoint of
time and resources, the need to validate results
from in vitro tests in mammalian species will remain
for the foreseeable future. Criteria for selection of
the most appropriate animal model includes: a com-
prehensive evaluation of all screening data, consid-
eration of structure-activity relationships, metabol-
ic fate, pharmacokinetics in candidate models,
consideration of use patterns, possible human ex-
posure, human epidemiological data and qualitative
and quantitative estimates of effects expected to be
produced. Multidisciplinary protocols include the
potential for exposure from conception to old age.
The assurance of absence of an unreasonable risk to
reproduction, behavior, birth defects, cancer and
heritable defects should be the product of animal
and laboratory tests to screen for toxic effects of
chemicals in man's environment. The advantages
and limitations of existing methodologies and an
evaluation of developing techniques are discussed
with emphasis on risk estimation.
INTRODUCTION
The proliferation of new compounds by industry
has created a need for a benefit/risk evaluation of
these valuable chemicals in order to provide a ratio-
nal policy on pesticide regulation. To provide this
policy, certain exigencies in the fields of carcino-
genesis, mutagenesis and teratogenesis must be de-
fined and solved.
Few people dispute the fact that technology has
in large measure contributed to the achievement of
our present standard of living. Accompanying the
benefits, however, are many subtle and sometime
gross effects that potentially threaten the health of
our society. The existing implications to this, and
future generations, demands the application of ra-
tional policies on utilization of existing and new
chemical compounds that will enable the highest
possible standard of living accompanied by accept-
able risk-to-benefit ratios.
Persons suffering from an incurable, fatal disease
would not wish to be deprived of treatment with a
particular drug because of some vague potential
danger of cancer in the distant future. Similarly,
persons beyond the reproductive years certainly
have less concern for exposure to chemicals that
produce birth defects or genetic change than do
young adults. In short, society accepts considerable
risks when the risks are necessary and when accept-
able alternatives do not exist, but it is predictably
unwilling to accept risks when information quan-
titating those risks is not available. The toxicologist
is faced with the dilemma of understanding the
mechanism by which toxicity may manifest itself,
developing and/or refining test methods, and pro-
viding rational assessment of the resulting data. It is
imperative, therefore, that the world-wide scientific
community provide to the governmental bodies the
means for quantitating these risks so that, in turn,
the public can, with understanding, titrate and as-
similate these data. Valid estimates of risks must be
made. It is the purpose of this presentation to dis-
cuss existing tests and to delineate the use of animal
and laboratory techniques for assessing toxic ef-
fects.
CARCINOGENICITY
Several facts which contribute to the uncertainty
of toxicological evaluations should be stated clear-
ly. There is no way to guarantee absolute safety!
Small populations of experimental subjects, either
animal or man, provide an imprecise basis for com-
parison to a large human population of variable ge-
netic/disease states, cultural backgrounds and ages.
Toxicological assessments are made singularly and
103
-------
humans are exposed to a milieu. It should be equal-
ly clearly understood that a proper experimental de-
sign will minimize noise and maximize comparisons
and that we are constantly expanding our tox-
icological armamentarium.
Is the toxicologist faced with a paradox of abso-
lutes? What are the approaches available in at-
tempts to generate reasonable policy and guides for
chemical use and control? An examination of the
involvement of the toxicologist in guaranteeing an
adequate and acceptable food supply will be illumi-
nating. There are three major control strategies
available for the regulation of toxicants, including
carcinogens: 1. the all or none approach, (for ex-
ample, the Delaney Clause of the Food, Drug, and
Cosmetic Act [FD&C Act]) ; 2. the use of safety fac-
tors, (commonly applied to noncarcinogenic le-
sions); and 3. statistical extensions beyond the ex-
perimentally observable range. Each approach has
its proponents and critics, its advantages and disad-
vantages.
All or None Approach
The Delaney Clause of the FD&C Act is an all or
none approach, and an understanding of complica-
tions in the quest of absolute safety is required. The
Delaney Anticancer Clause contains two main seg-
ments; one for human and one for animal food addi-
tives. The segment, addressing human food addi-
tives states that in evaluating the safety of such
compounds used in food-producing animals, con-
sideration must be given to the safety from possible
residues in the products of those animals which are
a source of food for man. When there is insufficient
evidence to establish that a finite or negligible resi-
due of the compound is safe in human food, or when
the anticancer clauses contained in sections
409(c)(3)(A), 512(d)(1)(H), and 706(b)(5)(B) of the
Act are applicable, a zero tolerance (no residue)
must be required. Under the provisions of the anti-
cancer clauses, no compound may be administered
to animals which are raised for food production if
such compound has been shown to induce cancer
when ingested by man or animal, unless such com-
pound will not adversely affect the animal and no
residues, as determined by methods of analysis pre-
scribed or approved by the Secretary (DHEW), are
found in the edible products of such animals under
conditions of use specified in labeling and reason-
ably certain to be followed in practice (1).
How Does One Establish the Toxicity; e.g.,
Carcinogenesis of a Compound?
A protocol advanced by the National Cancer In-
stitute for carcinogen screening calls for 50 male
and 50 female animals to be tested at or near the
maximum tolerated dose and a like number at half
that dose. A maximum tolerated dose ideally would
be that which does not kill the animal except via
tumor production in significantly less than a normal
lifespan. The choice of using high doses is a statisti-
cal expedient in order that high incidences of tu-
mors above background can be detected with small
sample sizes. There is no biological basis for use of
high doses. Such high doses may completely alter
metabolic pathways, adsorption and distribution
(1).
Positive and negative results are treated in a com-
pletely different manner. If the screening test shows
positive carcinogenic action, under the Delaney
Clause there is no alternative but to ban the com-
pound even if more than adequate information was
available to perform a risk/benefit analysis. All too
often, a negative result is interpreted as indicating a
noncarcinogenic compound. The negative cannot
be proved statistically. Thus, it is common practice
to take an arbitrary fraction, say 1/100, of the mini-
mum "no-effect" dosage as safe. Again, the mini-
mum "no-effect" dosage is ill-defined and is a ran-
dom variable depending on the number of animals
tested. Two statements will be repeated in this and
subsequent discussions. First, it is argued that such
an approach has worked over the years. Second, do
we have epidemiological evidence to show that no
small increases in cancer have resulted from such
environmental chemicals? Until recently we were
not aware of the vinyl chloride problem even
though millions of pounds are manufactured each
year. Most new chemicals have not been in the en-
vironment long enough for effects to be noted where
long latent periods may exist.
Zero Tolerance—No Residue
If one defines zero as complete absence, the di-
lemma of a no residue concept quickly takes form.
First, let us consider the ways by which one might
attain the complete absence of a residue. The first
would be to never allow contact, and the second
would be to consider a rate of removal or transfor-
mation that after a given waiting period would result
in complete removal.
Since the first approach effectively eliminates the
use of a chemical, let us proceed with the concept of
removal. The process of removal may be passive,
e.g., the removal of a persistent pesticide from a
food by rain or washing, or it may be via active ex-
cretion or metabolism. If an enzymatic process is
involved, the process will be accelerated.
If one solves for time needed to achieve the re-
moval of the last molecule, it is a very long time
indeed. Further practical complications arise via
the determination of analytical methodologies
which would be acceptable for determining zero. To
carry the discussion to the extreme, we would have
to analyze the complete extract of the complete
104
-------
sample with an analytical sensitivity of one mole-
cule.
We will probably all agree that there are certain
advantages to the use of biologically active chem-
icals and there is also the need to insure that the
human population is not exposed to hazardous resi-
due levels. One reasonable approach would be to
insure the absence of any hazardous quantity of a
residue. The advantage of this approach would first
be to fix the amount of residue which is expected to
be hazardous and at the same time determine the
sensitivity of a method required for analysis to sup-
port regulatory actions. Rephrased, requirement for
methodology would be defined by need rather than
state of the art. Use of a chemical would not be ap-
proved until acceptable methodology was devel-
oped.
As is often the case, problems are not solved,
they simply are reshaped, for we now have the
problem of determining the acceptable residue lev-
el.
The Delaney Clause, or any similar all or none
approach, is likely to be inadequate in two respects.
First, it provides a false sense of security by ignor-
ing the problem of "false negatives" which may re-
sult from inadequate testing. The FDA is charged
with the responsibility of attempting to minimize
such occurrences, but the question remains as to
how to best accomplish this formidable task. Sec-
ond, because of current toxicological ignorance, we
have little to offer as a substitute for the Delaney
Clause which requires banning of food additives
shown to be carcinogenic in animal tests. However,
with adequate data yet to be produced, the benefits
of a food additive in preventing food poisoning, for
example, might be documented to far outweigh a
carcinogenic risk which may occasionally occur on-
ly late in life.
Safety Factors
Safety evaluation at the present time is founded
on the concept of the "maximum no-effect dose."
The procedures are designed to determine the in-
take over extended periods (including a lifetime)
that will not produce the injurious effects character-
istic of the substance when given in large, that is,
toxic amounts. Also important is the exclusion of
the possibility that these "subtoxic" amounts will
produce some hitherto unsuspected reaction. A
summary of the kinds of specific studies usually un-
dertaken can be found in the paper by Friedman and
Spiher (2).
The unique difficulties inherent in safety evalua-
tion arise from the unusual goal of attempting to
prove scientifically that no deleterious effect has
taken place, i.e., to prove the negative. Experi-
ments are usually designed to establish that phe-
nomena, apparently resulting from experimental
manipulations, are real, are not artifacts or have not
occurred simply by chance. On the other hand, the
more appropriate concern would be to ensure that
the absence of positive findings (assuming adequate
protocols and procedures), is not due to chance or
to the inadequacies of sample size. Pursuing this
point supports the awareness that positive findings
may be artifacts, and therefore adequate probing of
techniques and replication of experiments to verify
findings is mandatory. Insistence on any desired de-
gree of assurance against making a wrong con-
clusion is standard operating procedure. Conven-
tionally, a statistically significant finding must have
a probability of no more than one chance in twenty
of being a chance occurrence, and often risks of on-
ly 1 in 100, or 1 in 1000, or less, are desired. Clearly
the severity of an all or none approach to avoid the
risk of a false positive reinforces the desire of a peti-
tioner for the clearance of a compound. Have we
dealt equally with false negatives?
A practical approach for dealing with these un-
certainties for noncarcinogens has been the use of
the 100-fold margin of safety. Substances to be add-
ed to food should not demonstrate an effect in ani-
mals when fed at a dose at least 100 times greater
than the likely human exposure. Our intuition tells
us that this approach has usually worked very well;
however, we should not forget the absence of an
experimental or theoretical basis. When followed
blindly, rather irrational experimental practices, in-
terpretation and rationalization can be made.
There have been attempts to apply safety factors
to carcinogens in our food supply. One of the latest
discussions was by Weil (3), where a safety factor
of 5,000 was suggested. Weil argued, as had Fried-
man (2), that it was contrary to "scientific judg-
ment" to try to extrapolate mathematically beyond
the range of experimental observation. Weil sug-
gested that it was, however, more scientific to use a
safety factor of 5,000.
The application of a safety factor established
from a "no effect level" in a toxicological evalua-
tion has a number of pitfalls which were succinctly
summarized by Weisburger and Weisburger (4):
It seems to us a "no effect dose'' for a carcinogen
is a highly relative level which applies only for the
precise experimental conditions generated. While
similar considerations hold for drugs, the risk is not
nearly so intense. More often than not, an improper
dose rate for rapidly acting drugs is detected almost
immediately and appropriate remedial action can be
taken. With chemical carcinogens and their long,
latent period, the disease condition resulting from
inappropriate selection of dose levels and alteration
of environmental conditions leading to potentiation
may become visible only years after the exposure.
At that time remedial action is obsolete and often
worthless.
105
-------
It is necessary to add to the Weisburger remarks
that a no effect dose, with the exception of thresh-
old, is sample size dependent and therefore is not
some absolute reference point.
A few comments on "threshold" are an appropri-
ate prelude to a discussion of methods for mathe-
matical extrapolation. The concept of a threshold
dose is based on the premise that a smaller dose will
not produce an effect. There are several problems
with demonstrating the reality of a threshold; re-
peated examination of the bioassay will demon-
strate variability even within the same individual,
and heterogeneity of the population will influence
the responses observed. Many toxicologists have
stated that for any compound there must be a "bio-
logically insignificant dose." There is little doubt
that this is true; however, what is our definition of
insignificant? A case in point are reports which have
been used to estimate that 3-5 percent of those
people hospitalized have drug complications severe
enough to extend their duration of care, a very omi-
nous statistic.
Mathematical Extrapolation Models
The recent papers by Mantel and Schneiderman
(5), and Hoel et al. (6) have described some of the
problems in determining risk to carcinogenesis. Due
to, at least, the toxicological uncertainties of ex-
trapolating risks from relatively high experimental
dosages in animals to low human exposure levels,
there are many people who propose complete prohi-
bition when a chemical is demonstrated to be a car-
cinogen. A modification would be to use a con-
servative method of linear extrapolation from an
upper confidence limit on the experimental result
back to a zero response at zero dosage. This proce-
dure is described by Gross, Fitzhugh and Mantel (7)
and the 1971 FDA Advisory Committee on Pro-
tocols for Safety Evaluation (8). This procedure is
based on the premise that at low dosages, many
dose-response curves are concave upward and a
straight line is a conservative upper limit to such
curves. In the simplest case with a single dosage
and no spontaneous background occurrence of tu-
mors, the extrapolation would proceed from setting
an upper confidence limit on the observed tumor
rate at the experimental dosage, and constructing a
line back to zero. Such a straight line is likely to be
above the true dose-response curves at low doses
(Figure 1). For low dosages, the one hit curve is
approximately proportional to dosage (linear).
Data in man, either dose-response or metabolic,
may suggest greater or lesser sensitivity than the
experimental animal. Human data are seldom avail-
able, but when they are available it generally is not
clear how such data should be employed in a mathe-
matical procedure for prediction of dosages produc-
ing low risks. Much more epidemiological data is
needed. A current example of this is the need to use
the human data from benzidine as a component of
setting water effluent standards by the EPA.
Petitioners should be encouraged to conduct ex-
periments in more than one species. Selecting the
lowest tolerance for extrapolation to man from the
species tested, in order to be conservative, may
tend to discourage testing in several species but ap-
pears to be the most prudent approach. Perhaps to
encourage testing in more species, the slope for ex-
trapolation could be increased as the number of spe-
cies is increased. For example, if the Mantel-Bryant
procedure is used in a single species, a slope of one
could be used unless experimental results indicated
a shallower slope. If more species were tested,
steeper slopes could be allowed for extrapolation
with each species while still employing the lowest
tolerance from among the species tested, if the ex-
periments were done with sufficient precision that
the lower confidence of the slope could be deter-
mined statistically with high confidence to be great-
er than the dose to be used. For example, an experi-
mental slope of 4 with a lower boundary of 3 might
allow for using a tolerance level of 1.5 rather than 1.
This procedure is only a suggestion which should be
investigated with existing data to determine its
workability.
Experimental Design
In testing for carcinogenicity, it is not clear that
current experimental designs and analytical meth-
ods are the best that can be developed. It is difficult
to detect and estimate the dose for even a high risk
when the spontaneous background rate is also high.
However, it may not be desirable to choose a strain
of a species of animals with a zero or near zero
spontaneous rate, as that strain may be resistant to
the chemical. It may be desirable to consider rela-
tive rather than absolute rates.
The choice of responses to analyze (e.g., propor-
tion of animals with tumors, number of tumors per
animal, or time to tumor) will dictate the experi-
mental design. Consideration must be given to the
range of dosages, number of dosages, numbers of
animals, length of feeding (total dose), and times of
sacrifice, if any.
If a procedure such as the Mantel-Bryan proce-
dure were adopted for extrapolation, it is possible
to calculate "acceptable dosages" for given risks as
a function of the proportion of the experimental ani-
mals producing tumors (which may be zero) and the
numbers of animals employed.
Considerably more research is needed in the de-
velopment of experimental protocol for predicting
carcinogenicity of chemicals, and I feel the NCTR
will impact heavily on this area.
A component of any responsive safety assess-
ment program must be the reward of excellence.
106
-------
Upper confidence limit
_ Percent
^ tumors
PO
true dose-response
observed response
background" level
Figure 1. Linear Extrapolation
-------
Too often approaches set rigidly to meet all possi-
bilities lend themselves to manipulation and testing
by rote. The ramifications of metabolism require-
ments have been discussed previously. The proper
extrapolation approach encourages good animal
husbandry by requiring that data be balanced for
competing risks. This process would penalize an ex-
periment conducted on diseased animals not living
out a normal lifespan. The procedure also encour-
ages the use of larger numbers of experimental ani-
mals and encourages the development of dose-re-
sponse data and data below that which might be ac-
cumulated at doses producing alterations of normal
metabolic mechanisms.
Let us consider hypothetical experiments con-
ducted with three dosage groups of fifty or one hun-
dred animals with low and high spontaneous tumor
rates. Data would be obtained at three doses but let
us assume that the highest dose is not to be consid-
ered because of a high percentage of early deaths.
The examples were chosen with the minimum
number of animals with tumors in the treated
groups which give a statistically significant increase
above background with 99% and 80% confidence,
using a 2 x 2 chi-square one-sided test with Yate's
correction for continuity.
Upper 99% limits on the observed proportion in-
crease in tumors were calculated using the proce-
dure of Mantel and Bryan [9] which employs Ab-
bott's Formula to adjust for background. The val-
ues of experimental dose + safe dose were
calculated using the conservative slope of one and a
risk which we are 99% confident will never exceed
one per million in a total lifetime without other com-
peting risks. For example, with 50 treated and 50
control animals with seven and zero tumors, re-
spectively, the safe dose equals the test dose di-
vided by 16,100. As we experience control animal
tumor rates greater than zero, the usual case, the
safe dose must be even lower because of increased
uncertainty in the test data. If the control tumor rate
is as high as 20% and the test animals as high as
44%, the experimental dose would have to be di-
vided by 87,700.
It can be quickly appreciated that the use of 100
rather than 50 animals results in increased statistical
confidence and a smaller safety factor.
The effort to select responsive strains but with
low background tumor rates is also a desirable ex-
perimental detail which is encouraged.
In summary, in our hypothetical experiment, the
experiment utilizing lower doses lowered the safety
factor from 87,700 to 36,000; the experiment with
larger groups lowered the safety factor to 17,600,
and the choice of proper strains lowered the safety
factor to 2,800. Rephrased, if a petitioner performed
the minimum experiment, a factor of 87,700 would
be required. If the petitioner did a dose response
experiment with only double the group size and se-
lected a sensitive strain with low background tumor
rate, the safety factor would be 2,800, a 31 fold dif-
ference.
Where a series of compounds are being tested, it
would be likely to use a larger number of controls or
to be able to pool the control animals from several
tests.
If there were no tumors in 300 control animals
and two tumors in 100 experimental animals, the
safety factor could be reduced still further to 2,300,
The public health is protected with greater cer-
tainty and useful compounds are not unnecessarily
restricted when dose response studies are con-
ducted utilizing more animals with low background
rates.
Deficiencies in Work to Date and Factors to be
Considered in Protocol Development.
Most of the deficiencies in carcinogenic testing
result mainly from the concept that this testing in-
volves only the determination as to whether or not a
compound can be made to produce a neoplastic tu-
mor. However, it is not recognized that carcinogen-
ic testing must, of necessity, consider both qualita-
tive and quantitative factors. The main deficiencies
in past studies of these factors involve primarily
two areas, i.e., experimental design and definition
of end points.
1. Experimental Design.
a. Statistics: The bulk of the technical litera-
ture reflects the lack of statistically valid
experimental design including adequate
numbers of animals at low levels of carci-
nogenic response.
b. Dose-Response: Only limited use of dose
response studies are reported in the techni-
cal literature for the purpose of determin-
ing tumor incidence and time to tumor in
terms of dose rate and total dose. The pre-
diction of risk at a given exposure level re-
quires dose response information.
c. Low-Dose Studies: Little information is
available on dose-response studies at low
levels of exposure and response. At low
levels of exposure environmental factors
may alter extensively the quantitative as-
pects of a response.
d. Mathematical Models: There is only limit-
ed mathematical definition of the dose-re-
sponse curves at low levels of exposure in
terms of variables affecting a chemical car-
cinogenic response.
e. Life-Shortening: There is limited use of ex-
perimental designs which permit proper
observation and evaluation of life-shorten-
ing effects of chemical carcinogens in rela-
tion to dosage.
108
-------
f. Age Sensitivity: The hazards involved in
exposure to a chemical carcinogen depend
not only on the nature of the chemical it-
self, the route of exposure, and the extent
of exposure in terms of amount and time,
but also on the susceptibility of the animal
at the time of exposure. There are only lim-
ited studies available on the influence of
age on the sensitivity of an animal to a
chemical carcinogen.
g. Recovery: There is a lack of evaluation of
the possible regression or progression of
pretumorous lesions such as hyperplasia in
relation to dosage.
h. Tumor Growth Rate: The technical litera-
ture shows an impressive lack of study of
the possible dependency of tumor growth
rate on dosage.
i. Reproducibility of Results: There is an ex-
tensive lack of evaluation of the quan-
titative reproducibility of chemical carcino-
genic testing.
2. Endpoints:
a. Tumorigenesis:
Tumorigenesis is as important an endpoint
as carcinogenesis. Benign tumors may
cause death in man and animals without
even undergoing malignant transformation.
There can be no doubt from a survey of the
technical literature that benign neoplasms
are often precursors of malignancies. In the
light of present knowledge, all tumorigens
must be regarded as potential carcinogens.
Hyperplasia and number, type, grade, and
individual distribution of tumors must all
be carefully used as endpoints in the evalu-
ation of chemical carcinogenesis.
b. Time to Tumors:
In some cases the only manifestation of an
effect consists of an earlier occurrence of
tumors in the treated animals than in the
controls. Time to tumor may be a very sen-
sitive endpoint permitting estimation of
"acceptable exposure levels" from dose-
time to tumor curves. This endpoint in
chemical carcinogen testing merits further
in-depth study.
c. Life Shortening:
As indicated earlier, there is limited use of
experimental design which permit proper
observation and evaluation of life shorten-
ing effects of chemical carcinogens in rela-
tion to dosage.
d. Pathology:
It is of the utmost importance that a com-
plete and accurate pathological examina-
tion be conducted on all animals used in
carcinogenic studies. All lesions, including
precancerous lesions such as hyperplasia,
must be described. Number, type, grade,
and individual distribution of tumors must
all be carefully evaluated in a chemical car-
cinogenesis study. The lack of proper path-
ological capabilities often limits this most
critical aspect of such a study,
e. Biochemistry:
The evaluation of carcinogenic hazards for
man is based on a judgment of all available
information. That is, it is based not only on
the carcinogenic bioassay, toxicity tests,
epidemiological data, and on the extent and
route of exposure of man, but also on meta-
bolic, biochemical, and pharmacokinetic
studies. Each compound must be evaluated
individually on the nature of its absorption,
distribution, metabolism, retention, and
excretion. Biochemical endpoints, as an in-
dicator of or response to carcinogen ex-
posures, are not usually included in carcin-
ogenic bioassays. Identification of reliable
indices that relate directly to tumorigenesis
would be invaluable to possibly define sus-
ceptible or non-susceptible individuals in
an animal population, or to possibly deter-
mine the time to onset or irreversibility of
lesions during precancerous induction peri-
od. This concept is highly important to and
related to the chronic low dose carcinogen-
ic bioassays. However, the current status
of this concept has not been defintely prov-
en or confirmed and, as such, must be con-
sidered as an activity peripheral to the
large bioassay study at this time.
Logically, any stimulus such as a chemical
carcinogen producing an anabolic or pre-
cancerous change in a tissue such as liver
should produce some response, such as
stimulation or inhibition of an enzyme(s)
that can be detected biochemically in the
affected tissues or possibly in the blood.
The inherent problem is to select or find the
proper biochemical endpoint. Recent evi-
dence resulting from studies on the effect
of radiation on biological systems indicates
that mammalian cells have the capability of
repairing damage to their DNA. More re-
cently it has been demonstrated that many
chemicals such as AAF form covalent
bonds with DNA, and are removed by a
process of "unscheduled DNA synthesis"
or DNA "repair synthesis". The process
appears to involve the excision of the dam-
aged segment of DNA with concomitant re-
placement by repair synthesis. The impor-
tance of this process was made evident
with the demonstration that the resistance
109
-------
of numerous tumors to chemotherapeutic
agents could be correlated with their level
of DNA repair activity. Tumors resistant to
chemotherapeutic agents were found to be
susceptible in the presence of DNA repair
inhibitors such as caffeine or chloroquine.
It is often assumed that DNA repair always
acts in a protective way by removing dam-
aged DNA segments or bound chemical
residues. It is known, however, that the
probability of an error in DNA replication
which might result in a mutation increases
with the extent of DNA synthesis. The pos-
sibility of AAF producing mutations in
DNA by stimulating extensive DNA repair
synthesis is a real one and must be consid-
ered in any study concerning the role of
DNA repair in carcinogenesis. In any
event, a more complete understanding of
how a cell repairs the damage inflected up-
on its genetic information by chemicals in
general, and carcinogens in particular, will
be necessary. An understanding of the role
of DNA repair in carcinogenesis is basic to
the question of whether small chemical in-
sults to a cell are completely repaired or ac-
cumulate over a long period of chronic ex-
posure,
f. Pharmacokinetics:
In order to provide a firmer basis for evalu-
ation of results obtained in the large chron-
ic low-dose carcinogenic bioassay, it will
be essential to develop a correlation be-
tween dietary level of the carcinogen, total
and/or daily intake of the chemical, incor-
poration of chemical into the target site and
the incidence of tumors as a function of du-
ration and level of exposure. Involved also
in this correlation is the need to evaluate
the role of blood levels (total as well as un-
bound) and urinary excretion patterns of
the chemical and/or its metabolites. The
overall concept or process described is the
basis and definition of pharmacokinetics.
Pharmacokinetics basically measures rates
of chemical absorption, distribution, tissue
binding and storage, metabolism, and elim-
ination. Elimination in this case meaning
excretion through urine, feces, and expired
air. Mathematical models are designed to
analyze results by means of computer sim-
ulation.
Approaches
It is clear that human exposure to many chemical
carcinogens is inevitable at the present time and in
the foreseeable future. It follows that a need exists
for capabilities which would permit an evaluation of
the relative hazards posed by different chemical
carcinogens. The development of methodology for
adequately evaluating carcinogenic risk involves
two major approaches. The first is the establish-
ment of a carcinogen dose-response relationship
using various endpoints such as tumor prevalence,
time to tumor, life shortening, etc. This carcinogen
dose-response relationship must permit some math-
ematical extrapolation downward on the curve so as
to facilitate determination of risk at levels of realistic
exposure. The second approach is to develop meth-
odology and concepts which will permit extrapola-
tion of results to man.
COMPARISONS OF HUMAN AND ANIMAL
DATA FOR CANCER INCIDENCE
A few examples are given which demonstrate the
wide range of sensitivity of humans and animals ex-
posed to chemical carcinogens. The examples are
strictly illustrative. No attempt has been made to
assemble all of the pertinent data. Rather, a few
representative data have been chosen. The results
cannot be used for species conversion factors or for
setting relatively "safe" dosages for the chemicals
discussed. The purpose, here, is to demonstrate the
wide and unpredictable disparities that exist in risk
estimation of cancer between human and animal
data.
The relative sensitivity of dosages for animals to
man is given by
Animal Dose
Human Dose
Human Dose
ArtimaTDose
Relative _ Animal Tumor Incidence
Sensitivity- Human Tumor Incidence
_ Animal Tumor Incidence
~ Human Tumor Incidence
As rough approximations to the relative sensitivi-
ties, only single data points have been used. Results
from different doses would yield different relative
sensitivities.
Two methods of extrapolation have been used to
illustrate the calculation of doses for maximum life-
time cancer risks of one in a million: Mantel-Bryan
[9] and linear extrapolation. The data have been
corrected for spontaneous tumor background in-
cidence. No attempt has been made to adjust the
data for competing causes of death or to fully utilize
dose-response data where available. Thus, the lim-
its only serve as comparative and not suggested tol-
erance limits.
DIETHYLSTILBESTROL
The use of diethylstilbestrol (DES) has had a sig-
nificant impact on beef production. Cattle fed ra-
tions containing DES have approximately at 15 per-
cent faster weight gain with 10 percent less feed
than non-DES fed animals; this increased efficiency
110
-------
means that the livestock produced saves about
$8.50 per animal. Without DES estimates show that
most prices at market levels could be 15-20 cents
per pound higher; nationwide, the increased costs
of maintaining beef supplies could be more than
$400 million annually.*
The total quantities of DBS and/or its diphos-
phate ester derivative now produced in this country
are not available; in 1952, the last available date,
1,800 kg were produced. In 1972, a total of 3,053 kg
of DES were reported to have been imported
through the principal U.S. Custom districts; how-
ever, this was a significant decrease from the 5,355
kg imported in 1971.
As the regulatory agency, the Food and Drug Ad-
ministration faces a severe problem. On one hand,
DES is considered unacceptable as a growth stimu-
lator because of lack of proof of safety of trace meat
residues, but as a drug in the contraceptive area, it
has been considered for approval for human usage
under special and critical circumstances. Based on
finding of 2 ppb of DES residues in 5% of liver sam-
ples, and assuming that 2% of the average diet is
beef liver, it is estimated that a person would have
to consume 5 million pounds of liver per year to
equal the intake from one treatment of day-after or-
al contraceptives or 10 million pounds a day to
equal the treatment for prostatic cancer.
The action taken by the Food and Drug Adminis-
tration to ban DES as a dietary growth stimulant for
meat production animals and the subsequent re-
versal of this action by Court decree has prompted a
vigorous debate over the adequacy of current test
models to determine potential hazards of estrogenic
chemicals (DES specifically) and other hormones
used as growth promoters. Areas openly discussed
include: 1) estimations of carcinogenic potential as
a function of estrogenic activity; 2) adequacy of ani-
mal species for carcinogenicity of residues found in
food products for human consumption; 4) and spec-
ificity and sensitivity of analytical methods.
This dilemma points out the apparent lack of re-
liable data upon which to base sound regulatory de-
cisions. Existing methodologies and protocols ap-
parently are not adequate to provide the necessary
information with compounds having hormonal, spe-
cifically estrogenic, activity. These deficiencies can
be corrected only through in-depth coordinated
studies, specifically directed to provide data bases
and protocols that will enable regulatory agencies to
make more realistic estimates of potential adverse
effects in man based on results of toxicity in experi-
mental animals.
Estimates of virtually safe levels based on Gass'
dose response data [ 10| analyzed by the Mantel-
"Interestingly, this cost to the American people is equal to the
then existing budget of the NCI.
Bryant approach using a slope of 1 probit/log dose
and an upper limit of risk of 1 : 1,000,000 resulted in
a calculated level of 0.007 ppb. These calculations
are contained in Part 2, Hearing Regulation of DES,
Subcommittee of the Committee on Government
Operations, House of Representatives, Dec. 13,
1971, p. 118, and were made by Dr. M. Adrian
Gross.
There are critical research needs which must be
addressed before substantive progress can be made:
1. define activity, affinity for receptors and fate
and effects of estrogenic chemicals in man and
experimental animals;
2. examine and evaluate normal levels and fluc-
tuations of endogenous estrogenic chemicals
in man and experimental animals; and assess
the potential effect of the combination of en-
dogenous and exogenous hormone;
3. clarify the role and interaction of internal
mechanisms regulating hormone secretion in
the carcinogenic process induced by adminis-
tration of exogenous hormone;
4. evaluate existing and where necessary devel-
op new methods for the qualitative and quan-
tative determination of estrogens and estro-
genic activity in various biological samples, to
facilitate sensitive and specific identification
of residues of chemicals in various materials
from animal experimentation; and
5. evaluate the potential utility of in vitro sys-
tems as predictors of in vivo responses.
DES INGESTION
There is no evidence that treating humans for
prostate cancer with a therapeutic dosage which is
equivalent to 10 ppm of the diet produces cancer.
Gass et al. [10] obtained a tumor incidence of ap-
proximately 20 percent in mice at 10 ppb, a dosage
100 below the human therapeutic dosage. Of
course, extrapolations from the animal data result
in extremely low "safe" dosages.
DES-TRANSPLACENTAL CANCER
It is estimated from the Boston Collaborative
Program that the risk of cancer of the vagina or cer-
vix is less than 1 in 2000 for female offspring of
mothers treated with DES during pregnancy. A
rough estimate of their average maximum dosages
is 75 mg/kg/day. Preliminary animal studies have
been shown to produce up to 80 percent metaplas-
tic-neoplastic response in male mice at 100 ng/kg/
day [11]. Thus, the mice were at least a million
times more responsive than humans. Extrapolation
to a "safe" dose from such a high response in the
animal data is extraordinarily risky. It is of interest
to note that all of these tolerance doses are well
above a maximal intake of 0.0001 mg/kg/day that
might occur from DES residues in beef liver.
Ill
-------
AFLATOXIN B,
Using the formula for relative sensitivity, it ap-
pears that the rat is at least 500 times more sensitive
to Aflatoxin Bi than humans (IARC) with a wide
discrepancy in calculated safe doses between rats
and man [12]. Wogan [12], in his paper entitled
"Carcinogenic Effects of Low Dietary Levels of Af-
latoxin B, in Rats" states:
"It is important to evaluate these and related data
in the context of their relevance to the assessment
of public health hazards associated with aflatoxin
exposure of human populations. From the present
experiment, as well as many previous publications,
it is clear that aflatoxins are very potent liver car-
cinogens for most animal species, and exposure to
man of any level must be regarded as representing
some carcinogenic risk. However, the limited data
on intake and liver-cancer incidence available from
studies in human populations suggest that direct ex-
trapolation of potency information from animals to
man is unwarranted. In populations consuming af-
latoxins at dietary levels of the same order of mag-
nitude as those shown to be effective carcinogenic
levels in rats, the incidence of liver carcinoma is far
lower than would be expected on the basis of the
animal data.
Nonetheless, every effort should be extended to
minimize exposure whenever possible. Total elimi-
nation of anatoxins from all dietary components
may prove to be difficult or impossible to achieve
owing to the circumstances under which they are
introduced into food commodities. As analytical
methodology permits detection of lower levels of
contamination, technological processes for the
complete prevention of contamination may be inef-
fective or so expensive as to impair or destroy the
utility of important foods. In this circumstance,
continued use of foods will necessarily depend on
risk-benefit analysis, taking into account the scien-
tific consideration of the other elements relevant to
the evaluation."
DDT
Assuming that DDT has produced cancer fewer
than 1 percent, if any, in the occupational^ ex-
posed group, the mouse is at least 35 times more
sensitive than humans. The tolerance levels for
maximum risk estimation of one in a million were
made by utilizing the entire dose-response curve
with the highest tolerable dose extrapolated from
the 250 ppm group, which had a 6 percent incidence
of malignant liver tumors [13].
BENZIDINE
Benzidine is of particular interest because in this
case man is more sensitive than the test animal. The
animal data comes from unpublished results of Ves-
selinovitch. The human data were obtained from
workers exposed to benzidine for 5 to 8 years [14].
Obviously, tolerance limits based on the human
data are lower than those based on animal data.
The viewpoints which I have represented here
are shared by a growing number of scientists, law-
yers, legislators, and members of the general pub-
lic. I hope that these examples stimulate thoughts
and perspective in probing the questions plaguing
the field of carcinogenicity testing. Many of these
same problems will be seen again as 1 discuss the
laboratory tests and animal systems available for
mutagenic and teratogenic testing.
Mutagenicity
The scope of the problem of environmental mu-
tagenesis is related to the extent of potential human
exposure, the length of the latency period between
mutation and detection of mutation, and irreversi-
bility of the damage. It is estimated that the human
population is exposed to 500 new commercial chem-
icals per year. Some of these chemicals, e.g. drugs,
involve limited populations; other chemicals, e.g.,
pesticides and automobile-exhaust pollutants, re-
sult in much greater human exposure. In the United
States, because of efficient distribution systems,
close to 100% of the population may be exposed to a
chemical shortly after development and production.
Further, the annual use of synthetic organic com-
pounds in the U.S. is approximately 55 billion kg of
over 1,000 different chemicals with an annual rate of
increase of almost 10% per year.
Many suspected mutagenic agents have been de-
tected in water, air, soil and food of which all serve
as a source of human exposure. Therefore, it must
be concluded that human exposure to various
classes of chemicals, including potential mutagenic
agents, is extensive. These agents may enter the en-
vironment through (a) intentional or unintentional
release during synthesis, (b) distribution of chem-
icals, (c) biological transformation of inert com-
pounds, or (d) naturally occurring components of a
biological system.
The effect of mutagenic agents on the environ-
ment and consequently society can be summarized
as follows:
1. Most mutations are deleterious, ranging from
the extreme, sterility or embryonic death, to
such harmless effects as changes hair color.
2. The effect of mutagenic compounds is statisti-
cal, i.e., they increase the incidence of muta-
tion above an already occurring background
rate.
3. Detection of an increased mutation rate in hu-
mans usually does not identify the causative
agent. Since it is not possible to treat the cause
of the phenotypic expression or prevent its
transmission to subsequent generations, pre-
vention of an increase in the mutation rate is
112
-------
the only practical protection of the human
population.
4. Since many compounds which are beneficial
to the human population are suspected muta-
gens, accurate assessment of mutagenic po-
tential is important for cost/benefit and risk/
benefit decisions.
The choice of organism and system for mutage-
nicity testing is not one for which a single test or
even a series of tests are as yet routinely accepted.
In addition, the type of specific test will in all prob-
ability be determined by the chemical class which
will be under investigation.
Although to date few compounds have been
shown to be mutagenic to a mammalian species
(probably due to inadequate data), the list of chem-
icals known to He mutagenic in available test sys-
tems is substantial [15-18]. Various compounds rep-
resenting such classes as alkylating agents, azo
dyes, epoxides, aflatoxins, phosphoric acid esters,
nitrosamines and polycyclic hydrocarbons have
been shown to be mutagenic by at least one crite-
rion. Many of these compounds are also carcino-
genic and/or teratogenic, and they are compounds
of the human environment.
In order to adequately evaluate the mutagenic po-
tential it will be necessary to: (a) develop new pre-
dictive test systems, (b) obtain quantitative data in
various test systems, (c) determine factors influenc-
ing mutation frequency, and (d) determine mutation
rates in humans as a basis for evaluation of test sys-
tems.
The importance of mutagenesis in toxicological
research is emphasized in the definition — genetic
damage transmitted to subsequent generations via
the germ cells, i.e., the lesion is not expressed in the
organism in which the causative event occurs, but
rather is first expressed in the offspring of this orga-
nism (F, generation). In subsequent generations,
the lesion is no longer confined to the original prog-
eny unless the damage results in lethality or sterility
to the F, generation before it reaches reproductive
age.
Most of these mutational events are not observed
as altered phenotypes in the first generation be-
cause of the heterozygous genotype but may be ob-
served in the F2 and subsequent generations. The
impact of this genetic damage is compounded be-
cause it now involves individuals who have never
been exposed to the mutagenic agent.
Genetic damage can be divided into two classes:
macrolesions and microlesions. Macrolesions in-
volve a segment of the chromosome of such magni-
tude that it is detectable upon microscopic examina-
tion of appropriate chromosome preparations.
These lesions are often referred to as chromosome
aberrations.
Microlesions refers to genetic damage which may
range from a change in a limited number of bases in
the DNA molecule to damage of a small segment of
the chromosome not detectable upon cytologic ex-
amination.
The assignment of a mutation to the macrolesion
or microlesion class is merely an operational defini-
tion limited by refinements of microscopic tech-
niques and focuses on the need for improved meth-
odology for identifying mutagen damage.
Reports indicate mutations may be involved in
many medical problems. An extensive compilation
of inherited disorders [19J lists 1,876 genetic de-
fects, the majority of which are associated with a
pathologic effect.
Macrolesions, because they involve major
chromosomal aberrations and gross phenotypic
changes, are often expressed in the F, generation. It
is estimated that 4% of all pregnancies are inter-
rupted due to chromosomal anomalies [19], or con-
versely, macrolesions have been detected in 20-to-
33% of all spontaneous abortions [20,21]. The in-
cidence of macrolesions in preimplantation zygotes
may be higher [22]. It is estimated that 3-to-4% of all
human zygotes implants surviving the first trimester
of pregnancy contain a macrolesions [20]. Approxi-
mately 0.5% of all human newborns exhibit malfor-
mations which are associated with macrolesions
and the actual incidence of macrolesions is approxi-
mately 1% [19].
It is more difficult to estimate the frequency of
microlesions because in the heterozygous state they
are not usually manifested clinically, may affect on-
ly a limited number of bases in the DNA molecule
and probably only a single abnormal protein is syn-
thesized. Consequently, phenotypic normality may
be maintained because alternate pathways may be
able to compensate for any deficiencies which may
occur. Secondly, it appears that most gene products
in mammalian cells are present in excess of require-
ments for phenotypically normal metabolism: there-
fore, under normal conditions no effect would be
readily observed in the heterozygous state [23].
This does not suggest that macrolesions in the het-
erozygous condition are inconsequential, but rather
that their impact may be diffuse and subtle; nor
should this imply that the mutation is not detectable
by appropriate techniques.
The incidence of microlesions in the human popu-
lation has been estimated at greater than 1%, with
the spontaneous (background) microlesions muta-
tion rate estimated at 0.14 gametes [19]. The muta-
tion rate for autosomal dominant mutations is ap-
proximately 3 x 10~4 per newborn. The mutation
rate of the X-chromosome (sex-linked trait) is esti-
mated at 2.0 - 6.8 x 10~s per generation. The hu-
man microlesion mutation rate following radiation
exposure has been estimated at 2 x 10~3 mutations
per gamete per roentgen [19].
113
-------
It was not until the 1960's that concern associated
with the human exposure to environmental chem-
icals and resulting mutagenesis was widely ex-
pressed. This concern was reflected in the 1969
MRAK Report [24], which has probably served as
the single most important impetus for the increased
awareness and scientific commitment to toxicologi-
cal methods development and research in the
United States. This concern is reflected in the de-
velopment and refinement of the following basic in
vivo and in vitro test methods available for assess-
ing potential mutagenic activity.
In order to discuss the strengths and weaknesses
of these let me briefly describe some of the major
assays available before I contrast them.
In Vitro Methodology
Ames Test. The utility of bacterial strains aug-
mented by various mammalian liver microsomes,
particularly those employing the Salmonella typh-
imurium strains and commonly referred to as the
Ames test, has gained considerable notoriety in the
past 5 years. A positive test in this system appears
to be a good indicator for potential mutagenicity in
higher organisms. Its predictability as a screening
test for carcinogenicity, based on a significant cor-
relation between compounds that cause mutagenic
effects in the Salmonella tester strains and the same
chemicals being reported as carcinogens in higher
organisms, has led some workers to predict the car-
cinogenic potential of the compound from the muta-
genic activity in the bacterial strain. There is no
doubt that this assay is an excellent indicator of ge-
netic activity. However, the site specificity encoun-
tered with this approach and the fact that this proce-
dure measures only certain specific, rather than all,
genetic alterations make it difficult to evaluate
"mutagenic activity" in relation to effects on the
total genome [25,26]. Caution, therefore, should be
exercised in our speed to classify compounds as
carcinogenic/mutagenic in man based on their muta-
genic activity in a tester strain of Salmonella.
Host Mediated Assay. The host-mediated assay
can employ microbial or mammalian cells as in-
dicator organisms. The comparison between the
mutagenic action of the compound in the organism
directly with the host-mediated assay indicates
whether the host can detoxify the compound or
whether mutagenic products can be formed as a re-
sult of host metabolism.
The host-mediated assay is not a true mammalian
test system unless mammalian cells are used as in-
dicator organisms. In general, toxicity tests in mam-
mals are considered more relevant for regulatory
purposes than similar tests in lower organisms. Of
the two mammalian cell types currently used to de-
tect point mutation in the host-mediated assay, the
mouse leukemia cells are preferable over the Chi-
nese hamster cells as they proliferate more readily
in the host and possess a near diploid chromosome
complement permitting simultaneous evaluation of
point mutations and cytogenic changes.
This assay with Salmonella has considerable flex-
ibility in that a variety of tester strains may be em-
ployed. Strains have been derived to differentiate
frameshift and missense mutagens, and to identify
nucleotide sequence-specificity of mutagens. In ad-
dition, repair-deficient strains and membrane trans-
port variants are available.
In the host-mediated assay the host is responsible
for absorption, distribution, metabolism, detoxifi-
cation and excretion of the compound. If the host
does not handle the agent in a manner similar to
man, the predictive value of the assay is decreased
[27[.
Dominant Lethal Test. Dominant lethal mutation is
a genetic event that kills the individual which car-
ries it in a heterozygous state. The dominant lethal
test is used to screen chemicals for this type of ge-
netic damage. The damage is characterized by pre-
implantation loss of non-viable blastocytes and
early embryonic death [28]. It has been shown that
these adverse effects are strongly associated with
structural or numerical chromosomal anomalies in
the germ cells. Dominant lethal mutations are self-
limiting, as death readily eliminates the mutants
from a given population. As such, these mutations
may not merit much interest. However, it is postu-
lated that a chemical agent that causes dominant le-
thality would also cause point or gene mutation. Of
the chemicals tested for dominant lethality, a few
have been tested by the specific locus mutation test,
and the response was similar by the two test meth-
ods [29,31]. Of course, more data will be needed to
study the relationship of dominant lethality and
point mutation. Thus, the dominant lethal test might
serve as a useful test for detecting gene mutations,
which being generally recessive, cannot be readily
identified by simple techniques. Evidence support-
ing the validity of the dominant lethal test for muta-
gen detection is provided by Cattanach [32], who
identified paternal sex chromosome loss, aberations
of the paternal X and also specific locus autosomal
gene mutations in those progeny surviving a domi-
nant lethal test with triethylenemelamine.
In Vivo Methodology
Transmissible Translocation Test. Chromosomal
damage may be observed in the same individual that
was treated with the test compound, or in the off-
spring of the treated animal. Logically, damage ob-
served following transmission to the progeny is
more likely to have a genetic basis than that seen in
the treated animal itself. This is the basis of the
transmissible translocation test which measures an
inherited genetic change.
114
-------
The direct translocation test entails examination
of the germ cells of the animal under treatment. In
this case translocations are scored prior to com-
pletion of germinal selection and lethal alterations
cannot be distinguished from viable translocations.
Chromosome aberrations observed in early em-
bryos can be transmitted effects or non-transmitted
effects depending upon which parent was treated
and at what time the treatment was administered. If
the mother is treated after ovulation, the embryo's
role is that of an indicator organism in a type of
host-mediated assay.
Acutely-induced cytogenic damage can be as-
sayed in any somatic tissue with a high mitotic rate.
Although spleen, thymus, liver and epithelia yield
some mitotic figures, bone marrow is the tissue of
choice from the standpoint of technical simplicity
and number of mitotic cells. Usually the animal is
sacrificed and the chromosome examined a few
hours after a single or a short series of treatments.
Techniques for bone marrow biopsy allow repeated
sampling from the same experimental animal.
Cytogenic studies in chronic toxicity testing are
usually directed towards specific suspect target tis-
sues. For instance, the effects of a suspected leu-
kemogen can be observed by repeatedly sampling
peripheral lymphocytes. Such studies are primarily
of interest for the early detection of cancer.
In vivo cytogenic procedures have been histori-
cally designed to detect:
a. damage induced in germ cells — transmitted
b. damage induced in germ cells — non-trans-
mitted
c. damage induced during early embryogenesis
d. damage induced in somatic tissues — acute
e. damage induced in somatic tissues — chronic
Recently, an important aspect of mutagenesis re-
search is the development of methodology for esti-
mating the mutagenic potential of chemicals with
emphasis of an in vivo mammalian system, which
will be useful for studying the relationship between
mutagenic exposure and microlesion ("point-muta-
tion," small deletions) induction. Toward this goal,
the techniques employed in the study of inborn er-
rors of metabolism are being developed for use in
detecting induced microlesions in experimental ani-
mals. Application of this methodology allows for
detection of genetic alteration in the offspring of
mutagenic treated animals, i.e. the heterozygous
generation.
The methods of approach are (a) detection of al-
tered enzyme activity using a Miniature Centrifugal
Fast Analyzer (MCFA) which was developed at the
Oak Ridge National Laboratory and modified at the
NCTR, and (b) detection of active but altered gene
products measured by relative electrophoretic mo-
bility. The gene products are derived from the liver,
brain, kidney, and/or red blood cells.
The MCFA has the capability of assaying large
numbers of very small samples which allows for the
assay of many different enzymes from organs of
various size. Appropriate assay conditions for 30
enzymes derived from all of the above tissues have
been optimized. Also, "normal" levels of activity
and acceptable ranges of standard deviation have
been defined. Activity measurements of mutagen-
treated animals which fall outside this defined range
indicate suspect mutants; these are further charac-
terized. Electrophoretic mobility patterns of 30 en-
zymes have also been defined in various systems
including starch gel, isoelectric focusing and cellu-
lose acetate. Relative electrophoretic mobility dif-
ferences detected in mutagen-treated animals in-
dicate suspect mutants and these are further charac-
terized [33].
Initial phases of the methods development exper-
iment have been completed. In these phases, male
mice (P0) were treated either with 300R or 600 R of
X-irradiation. The (P0) males were mated to un-
treated females and an F, population was gener-
ated. The F, population was bred inter se and to
untreated partners to generate an F2 population
which was sacrificed and analyzed for inducted al-
tered gene products. The results suggest that en-
zyme activity variants were induced at the follow-
ing loci: alcohol dehydrogenase, fatty acid synthe-
tase, cytochrome C reductase, serine dehydratase
and aldolase. In addition, an electrophoretic mobil-
ity difference was detected at the locus responsible
for glucose-6-phosphogluconate dehydrogenase; in
one litter, 6-phosphogluconate dehydrogenase was
not detected. Preliminary evidence suggests this
may be a ramification of an alteration at a regulatory
site.
The second phase of the experiment has been im-
plemented. Techniques which include sample prep-
aration and characterization have been refined and
computer programs have been expanded to assist in
data analysis. Sensitivity of the total system has
been increased approximately two-fold over values
previously reported. The F2 population is currently
being analyzed [33].
In addition to the efforts in our laboratory, refine-
ment of mutagenic screening methods is truly an in-
ternational effort. These efforts will with certainty
provide a clearer insight into the inherent mechan-
ics of environmental mutagenesis.
TERATOGENICITY
Teratology may be defined as the science which
deals with the causes, mechanisms and manifesta-
tions of developmental deviations of either a struc-
tural or functional nature [34]. A teratogen may be a
chemical, drug, virus or physical agent. (Some ter-
atologists consider gene mutation which produces
congenital deformity as a fourth type of teratogenic
115
-------
agent; however, this was discussed under Mutagen-
esis). For an agent under test in laboratory animals
to be deemed a teratogen, it must alter the structure
or function of a statistically significant number of
young (versus sham controls) to a measurable ex-
tent following administration of the agent: to the
male before mating; to the female before or during
pregnancy ; or the fetus before completion of matu-
ration.
The magnitude of the problem of congenital de-
fects is shown by the following statistics. Three per-
cent of all children have a congenital malformation
of medical significance; one-third of these condi-
tions are life-threatening. By age 7 or 8, approxi-
mately 10% of children have a medical problem re-
lating to abnormal intrauterine development. Over
50% of the children admitted to hospitals are there
because of a congenital defect [34],
In examining a potential teratogenic agent, there
are three lines of defense that should be investi-
gated. First, is standard animal testing and chemical
prediction. This was largely developed following
the thalidomide disaster and includes all forms of
tests carried out in pregnant laboratory animals.
Doubtless, this defense has contributed to pro-
tection, but there is a belief that another thalido-
mide-like epidemic could occur. Hopefully, by
chemical prediction and knowledge of pharmaco-
logic action, we should be able, through better risk
assessment, to avoid potential problems.
A second defense, although not fully developed is
in vitro testing. By in vitro testing, I include the use
of tissue culture, organ culture, whole embryo cul-
ture and emerging techniques of ova culture. Most
teratologists treat in vitro as most valuable for un-
derstanding the mechanisms of teratogenesis; how-
ever, there are many who feel that in vitro tests for
screening of mutagens and carcinogens might be
highly useful in teratology.
A third defense consists of monitoring. This is
necessary because the previously mentioned de-
fenses are still inadequate. Fetal monitoring has a
number of advantages. Since the congenital-defect
rate in the spontaneously aborted fetus is much
higher than in the newborn [341, one might assume
that the fetus would provide a more sensitive index
for observing the rate of defects. In utero mon-
itoring has an additional advantage of time and ac-
curacy in determining the possible cause of insult
i.e., the mother can be interviewed closer to the
time of possible insult rather than 7 or 8 months af-
ter the fact. Further, should another thalidomide-
like epidemic occur, a wave of abnormal fetuses
might be seen as much as 6 months before the ap-
pearance of newborns with the defects.
At present, there are a number of monitoring
techniques for noting defects in newborns. Al-
though these systems probably fail to detect minor
changes in brain function or subtle metabolic aber-
rations, generally the gross physical defects are re-
corded. The larger portion of congenital disease is
identified following infancy, since over half of the
defects are not readily diagnosable at birth. Al-
though these systems are fraught with variations
due to artifacts of collecting the data, a continuous
recording by time and place and registry of con-
genital defects could provide important warning of
teratogenic action of a new chemical, physical or
infectious agent. National as well as internation reg-
isteries devoted to this end would benefit not only
the clinician but the teratologist and toxicologist.
Another extremely important defense mechanism
is that of data retrieval and utilization. It has been
estimated by Larsson as reported by Shepard [35]
that between 1957 and 1972 the number of scientific
articles on developmental subjects doubled with
about 250,000 appearing in 1972. In addition, there
has been a rapid increase in descriptions of new
dysmorphologic syndromes in the human. Effective
use of this widely spread information for solving
problems related to human congenita! defects is
needed. A system such as the Environmental Muta-
genic Information Center at Oak Ridge National
Laboratory might be very useful (36]. Also there is
a need for tests produced from computer printouts
making the material readily available to a computer
user. Such books as McKusick's Catalog of Men-
delian Inheritance in Man [37] and Shepard's Cata-
log of Teratogenic Agents [35] are good examples.
Another important concept is that of record link-
age. With present computer systems, it is possible
(and should be given serious consideration) to link
birth records of children with malformations to
chemical exposures of the mother, or previous dis-
ease and ill health of the parent or family members.
A recent preliminary report from the Kaiser Per-
manente System [38] used record linkage to study
the association between prescriptions filled for
women and serious congenital defects in their off-
spring.
Fig. 2 illustrates a model which might be used to
prevent congenital defects. It might be conceptual-
ized as an enzyme-type reaction with our aim being
to increase the speed of the reaction from left to
right. The essential portion of this reaction resides
in our understanding of mechanisms. Fortunately,
alert clinicians have been able to circumvent mech-
anism understanding and go from the human model
to the final reaction, removal of a causative agent.
The linkage of thalidomide and maternal rubella
with specific human syndromes are examples. Un-
raveling the etiology of about 70% of congenital de-
fects, whether they are due to environmental
agents, developmental genes or a combination of
the two, will require a concerted effort. Many ani-
mal models exist, and the emerging information on
116
-------
embryo pathogenesis, genetics physiology and de-
velopmental biology of these models will be in-
valuable. The in vitro analytical methods of the de-
velopmental biologist need to be applied to our un-
derstanding of the in vitro mechanism reactions.
Out of our understanding of mechanisms, hope-
fully a reaction (Fig. 2), 3, will lead to hypothesis
testing in the human. An important co-factor for
this reaction, 3a, is a basic understanding of species
and genetic subgroup differences. The rate-limiting
steps in this reaction appear to be located at la and
at 3. Reaction 4 could be represented by regulatory
governmental bodies, national and international.
No one animal species or group of related species
is ideally suited for the evaluation of human terato-
logic effects. Accumulated experience, however,
has shown that certain groups of animals may offer
advantages for particular types of use, whereas oth-
ers have been found generally inappropriate. In the
inappropriate category must be placed all species or
test systems that lack the all-important first line of
defense afforded by mammals to their developing
tissues whether these by germinal, embryonic, or
fetal. This refers to the parental adaptative mecha-
nisms by which mammals are usually able to reduce
the dosage of most physical chemical agents before
they reach such developing tissues. A few non-
mammalian species (mainly sharks and some
snakes) like mammals do not expose their germ
cells or their embryos to the outside environment;
but for other reasons these animals are not recom-
mended for teratologic testing.
The mammalian embryo and fetus benefit from a
further and unique line of defense against chemical
agents, namely, the placenta. Most chemicals in the
environment, specifically those in the maternal
bloodstream, are subject to concentration gradients
across the placenta. The placenta is often given
more credit for serving as a barrier behind which
the embryo or fetus can hide than is actually the
case. Evidence is accumulating to indicate that vir-
tually all unbound chemicals in maternal plasma
have access to the conceptus across the placenta.
Many molecules of small size (less than 600 mol.
wt.) and low ionic charge cross by simple diffusion,
others by facilitated diffusion, active transport,
pinocytosis, or perhaps by leakage. Lipophilic
chemicals are known to cross the placenta more
readily than other compounds. The question, then,
is not whether a given compound crosses the pla-
centa, but at what rate [39]. Avian and other non-
mammalian embryos have and doubtless will con-
tinue to contribue greatly to studies on embryologic
and teratogenic mechanisms; but even the study of
basic mechanisms of normal and abnormal develop-
ment are not entirely comparable between birds and
mammals. The relatively static nutritional and ex-
cretory functions in the incubating egg are in sharp
contrast to the fluctuating interchange that occurs in
both directions across the placenta. Foreign sub-
stances introduced into incubating eggs may remain
in the slowly diminishing pool for a relatively long
time, whereas most compounds as well as their me-
tabolites have a short half-life in maternal and pre-
sumably also in embryonic bloodstreams of mam-
mals.
IN VITRO SYSTEMS
These in vitro systems, whether involving whole
embryo, organ tissue, or cell cultures, are all in-
appropriate for use in embryotoxicity evaluation.
Among the many other obstacles to achieving and
maintaining conditions that would even approxi-
mate the in vivo state, the problem of circulation of
nutrients and metabolites surely ranks as a major
one. These variables superimposed on, and possi-
bly interacting with, whatever stress is introduced
by the test substance would render results from in
vitro cultures almost meaningless as compared with
an embryo in dynamic balance with maternal
homeostatis through the placenta. In vitro systems
have been useful in basic studies on nutritional re-
quirements for growth and differentiation over short
spans of development, but to maintain normal
growth and developmental schedules for longer
than several hours or a few days has not been pos-
sible.
IN VIVO SYSTEMS
Rodents and Lagamorphs. The complications that
may arise from dependence on the highly atypical
yolk-sac placenta during early organogenesis is rais-
ing concern for the heavy dependence placed on ter-
atologic investigations derived solely from these
species. Although this unique placental structure
may not be wholly at fault, there is accumulating
evidence that animals whose early embryos are de-
pendent upon it for essential transport show em-
bryotoxicity to lower doses of more chemical
agents than do other mammals. On the other hand,
the small size, short gestation period, large litter
size, lack of seasonal breeding patterns, and ready
availability of these animals add up to a strong in-
ducement to continue their use in initial screening
procedures. Not the least of the advantages is their
modest cost which permits the use of ample num-
bers for statistical evaluation of results. Additional-
ly, the gestational developmental parameters for
these species are adequately defined.
Ungulates. Among the smaller hooved mammals,
the pig and sheep and possibly the goat, seem to be
the most promising subjects for use in the evalua-
tion of human teratic risk. The agricultural back-
ground of these animals assures a wealth of infor-
mation on the breeding performance of various
stocks. Although they have not been used as exten-
117
-------
CONGENITAL DEFECTS-MECHANISMS -
APPLICATION AND REMOVAL
MECHANISMS
2
Human Models
la
(Genetic or
environmental)
Animal Models
(environmental ^
or genetic) 1 ^
in vivo
2a
"2b"
Embryo-
pathogenesis
Genetics of
development
Physiology
Developmental
Biology
in vitro
Culture
tissue
organ
embryo
Mutagenesis
tests
Human
Application
3a (hypothesis
and tests)
p
Removal of
Causative Agent
(environmental
or genetic)
Figure 2. A model which might prevent congenital defects.
-------
sively in pharmacologic and toxicologic investiga-
tions as some of the preceding species, a significant
amount of information probably can be found. The
larger size of these animals would require large
samples of the test compound for dosing, but it is
unlikely that sheep or pigs would be used in early
stages of evaluation when only small samples of a
new compound are available.
Nonhuman Primates. The advantages and disad-
vantages of using these animals have recently been
reviewed in some detail by Wilson [40], and the
present discussion will be limited to generalities. To
date only two genera, Papio (baboons) and Macaca
(several species), have been demonstrated to be
promising for use in the evaluation of human teratic
risk. Disregarding economic considerations, does
the use of primates offer a greater degree of security
in the evaluation of safety for man than does the use
of other forms?
The answer to this question is conditioned on the
answers of two corollary questions. It is taken for
granted that primates will not be used in mutagen-
icity testing or for preliminary teratogenicity
screens, both of which uses require larger numbers
than are practical. The first and most basic ques-
tion, then, concerns whether the agent to be tested
is one that is needed for use during human preg-
nancy, for example, antiemetic, antidepressant,
anticonvulsant, or oral hypoglycemic, for which an
equally safe and effective counterpart does not al-
ready exist. (Life-saving procedures used against
neoplasms or major infections are not questioned
because their use would presuppose elective abor-
tion when unexpected pregnancy is diagnosed). Al-
though not absolutely essential, these needed drugs
when prudently used can alleviate much suffering
and discomfort in pregnant women. Before use dur-
ing human pregnancy, however, they must be sub-
jected to the most rigorous animal test that can rea-
sonably be devised.
The second question concerns the type of animal
or battery of tests that will have maximal predictive
value for man (Table 1). Briefly, an initial level
might consist of a teratogenic screen involving rela-
tively large numbers of a readily available animal
like the rat. A test substance causing no significant
embryotoxicity would then advance to the next lev-
el in which carnivore or an ungulate would be used,
depending on which showed greater similarity to
man in the metabolism of the compound in ques-
tion—information that could be acquired during pri-
or pharmacologic studies. The higher costs and
lower fecundity of this second test animal, perhaps
dogs or pigs, would dictate the use of smaller num-
bers than were used in the initial screen, but the
general range of effective dosage would already
have been defined.
The results of the second-level test would deter-
TABLE 1.
MULTILEVEL TESTS FOR TERATOGENICITY"
No. of
Order
Suitable
Pregnant
of test
Purpose Animals Species
1st
Find embryotoxic
Rat, mouse,
130-150
dose range
hamster or rabbit
2nd
Confirm or ad-
Ugulate or
40-60
just above
carnivore
3rd
Only if use in
Macaque or other
30-50
human pregnancy
non-human primate
is needed of likely
"Modified for Wilson, 1973
mine whether further animal investigations were
necessary. If overt embryotoxicity occurred at lev-
els only moderately higher than the anticipated
therapeutic level in man, the compound would
probably be disqualified for consideration for use in
women of reproductive age. On the other hand, if
the compound caused no adverse effects at many
times the expected human-use level, it might be
considered ready for clinical trial without further
animal testing. If, however, there were uncertainty
about the margin of safety between dose that
caused embryotoxicity in the second species and
the therapeutic need, then a third level of animal
tests in non-human primates would seem justified.
Thus, the use of nonhuman primates for embryo-
toxicity evaluation is recommended only for specif-
ic purposes, namely: (a) drugs clearly needed to re-
duce severe discomfort or disease during pregnancy
(e.g., anticonvulsants, hypoglycemics) or 9b) drugs
that may be taken inadvertently during the early
part of undiagnosed pregnancy (e.g., contracep-
tives, anorexants).
The foregoing has implied that embryotoxicity
tests in nonhuman primates do indeed provide an
added measure of security. This is probably true
when they are used in sequence with proposed sub-
primate tests. It remains to be proved, however,
that in absolute terms primates are more reliable
predictors of human embryotoxic risk than are oth-
er mammals. Careful scrutiny of available literature
[41] reveals that man and other primates may not
have identical teratogenic sensitivity to more than a
few types of agents, namely, thalidomide, sex hor-
mones, and probably ionizing radiations. A degree
of similarity in embryotoxic response could be
found as regards infections and some chemical
agents, but it now seems very unlikely that a precise
correlation in all aspects of teratogenic susceptibil-
ity can be expected. A preferential status for simian
primates in preclinical tests can be fully justified on-
ly when such animals can be shown to be more like
man than are other test species. The basis for com-
parison on which this judgement is made must take
into account all of the following: (a) overall phylo-
genetic relatedness, (b) similarity of maternal me-
119
-------
tabolism of chemicals, (c) comparability of placen-
tal transfer functions, and (d) likeness in embryonic
sensitivity to extrinsic agents generally.
At present our laboratory is conducting collabo-
rative studies in order to further evaluate the chim-
panzee as a human surrogate for toxicological,
metabolic and/or teratogenic investigations. Prelim-
inary data indicates that the endocrine patterns of
the chimpanzee, indeed, closely resemble those of
the human during the various stages of development
and in some aspects throughout pregnancy, in-
dicating that for hormonally active compounds this
may be the preferred animal model. When in the
evaluation process, a known "best" animal model
exists on the basis of these and other criteria, then
toxicologic investigations of compounds within the
chemical class should be required in that species at
a dosing regime at and above that which might be
postulated for human therapeutic use [33],
Regardless to the validity of the test method or
animal model, it is not sufficient to test only those
compounds of destined human exposure. All for-
eign substances which can, through any route, find
their way into the environment must be evaluated.
Today's environmental burden is as much a result
of accident and ignorance as it is from appropriate
application. Examining a few of the major disasters
which have occurred will provide an idea of the ex-
tent, and possible calamities, which might acciden-
tally result.
The use of methyl mercury as a fungicide and in-
dustrial chemical led to the widely known disasters
which occurred in Japan and in the Middle East
with Alkyl mercury. In Japan, Minimata Bay was
contaminated with industrially discharged mercury
which when methylated moved through the food
chain and resulted in a human burden through eat-
ing fish, which proved embryotoxic, neurotoxic, le-
thal, and produced delayed neurologic and immuno-
logical dysfunction [42J. In the Middle East, ex-
posure was through consumption of grain treated
with methyl mercury as a fungicide. The results
were of catastrophic proportions, and the delayed
effects which are expressed postnatally are just be-
ginning to be understood [43,44].
The wide spread use of the herbicide 2,4,5-T
which contained a small amount of the chemi-
cal impurity, 2,3,7,8-tetrachloro-dibenzo-p-dioxin
(TCDD) in Southeast Asia gave rise to a great deal
of concern. Let me discuss briefly my opinion as to
the relative risks both to man and the environment
due to 2,4,5-T with less than 0.1 ppm TCDD and a
comparison with TCDD from other routes of entry.
Researchers at NCTR and elsewhere have prov-
en, I believe, that the currently available 2,4,5-T is
teratogenic. The effect does not appear to be due to
a generalized non-specific effect on the maternal an-
imal and TCDD plays no discernible role at the cur-
rent levels found in 2,4,5-T (45-531. It seems to me
that there are two concerns from TCDD: (a) envi-
ronmental half-life; and (b) biomagnification in graz-
ing animals which I have detailed previously [54).
From this discussion [54] and using a linear extrapo-
lation model, it would take 2250 years of use of 0.1
ppm TCDD to produce the same "damage" as has
already been done. Therefore, I personally believe
our concern over damage to the environment is
overstated when we know the use of 2,4,5-T will
never approach past levels and that the TCDD con-
taminate level has been reduced from 25 ppm to 0.1
ppm.
Dioxins in the environment are important, but I
feel that the pesticide will contribute little, if un-
toward control over the quality of production is
maintained. The problem lies not with the pesticide,
but with inadvertent contamination of the environ-
ment. As an example let me draw on data from a
May 1975 article by Carters/ al. [55]. Between Feb-
ruary and October of 1971, a production plant in
Missouri sprayed waste oil residues of hexachloro-
phene amounting to about 50,000 kg contaminated
with 350 ppm TCDD in order to control dust. To
equal this, one would have to use 400,000,000 lbs of
currently available 2,4,5-T. This abuse of industrial
waste disposal is not isolated and must be stopped.
Along the same line, and emphasizing the need
for careful adherence to safety in chemical manu-
facturing, is the recent explosion in Seveso, Italy.
An explosion at a Swiss subsidiary of Hoffman La
Roche caused a 500-gallon vat of trichlorophenol to
explode releasing approximately 4.4 tons of TCDD
[56].
Indeed, the impetus of the last decade on exten-
sive safety evaluation has dwelled on the conduct of
tests, sometimes solely for the sake of testing,
rather than the understanding of mechanics of tox-
icity. Today we are aware that it is the mechanisms
of toxicity, in vitro and in vivo, which need to be
understood. Before rational evaluation of all the
data generated can be meaningful, baseline animal
model data and selection criteria are necessary.
This will require extensive resources, time and sci-
entific communication on an international scale. We
have for too long selected our animal models arbi-
trarily and with too little understanding as to how
the model correlates with in vitro assays, other ani-
mal models, and finally man. Extrapolation remains
the Achilles heel of toxicology.
SUMMARY
The tests which have been discussed, whether
they are in vitro or in vivo, whether they are de-
signed to measure carcinogenicity, mutagenicity or
teratogenicity, all are designed to determine a po-
tential toxic effect.
However, we must remember that these tests: are
120
-------
no better than the animal model; no better than the
design of the experiment; no better than the health
of the experimental animals; and no better than the
technicians or senior professionals involved in re-
cording or interpretation of the data.
Current legislative actions within the U.S. have
attempted to address these variables and to provide
a greater consistency in the quality and reliability of
experimental non-clinical data. Legislation such as
the Good Laboratory Practices specifically address
those variables mentioned above and others such as
protocol design, data collection, handling and re-
porting, and quality control.
A worldwide scientific effort to provide this same
type of consistency would be of immense value to
all governments since chemicals frequently become
environmentally ubiquitous crossing arbitrary geo-
graphic boundaries.
LITERATURE CITED
1. Cranmer, M. F. 1977. Estimation of Risks Due to Environ-
mental Carcinogenesis. Med. Ped. Oncology 3: 169-198.
2. Friedman, L. and A. T. Spiker, Jr. 1971. Proving the Safety
of Food Additives. FDA Papers 5: 13-15.
3. Weil, C. S. 1972. Statistics vs. Safety Factors and Scientific
Judgment in the Evaluation of Safety for Man. Toxicol.
Appl. Pharmacol. 21: 454-463.
4. Weisburger, J. H. and E. K. Weisburger. 1968. Food Addi-
tives and Chemical Carcinogens: On the Concept of Zero
Tolerance. Food Cosmetic Toxicol. 6: 235-242.
5. Mantel, N. and M. A. Schneiderman. 1975. Estimating
"Safe" Levels, a Hazardous Undertaking. Cancer Res. 35:
1379-1386.
6. Hoel, D. G., D. W. Gaylor, R. L. Kirchstein, U. Saffiotti
and M. A. Schneiderman. 1975. Estimation of Risks of Irre-
versible, Delayed Toxicity. J. Toxicol. Environ. Health I:
133-151.
7. Gross, M. A., O. G. Fitzhugh and N. Mantel. 1970. Evalua-
tion of Safety for Food Additives: An Illustration Involving
the Influence of Methyl Salicylate on Rat Reproduction. Bio-
metrics, 26: 181-194.
8. FDA Advisory Committee on Protocols for Safety Evalua-
tion. 1971. Panels on Carcinogenesis Report on Cancer Test-
ing in the Safety Evaluation of Food Additives and Pesti-
cides. Toxicol. Appl. Pharmacol. 20: 419-438.
9. Mantel, N. and W. R. Bryan. 1961. Safety Testing of Carcin-
ogenic Agents. J. Natl. Cancer Inst. 27: 455-470.
10. Gass, G. H., D. Coats, N. Graham. 1964. Carcinogenic
Dose Response Curve to Oral Diethylstilbestrol. J. Natl.
Cancer Inst. 33(6): 971-977.
11. McLachlan, J. A., H. C. Shah, R. R. Newbold, and B. C.
Bullock. 1975. Effect of Prenatal Exposure of Mice to Dieth-
ylstilbestrol on Reproductive Tract Function in the Off-
spring. Toxicol. Appl. Pharmacol 33: 190.
12. Wogan, G. N. 1974. Carcinogenic Effects of Low Dietary
Levels of Aflatoxin B, in Rats. Food Cosmet. Toxicol. 12:
681-685.
13. Tomatis, L., V. Turusov, and N. Day. 1972. The Effects of
Long-Term Exposure to DDT on CF-1 Mice. Int. J. Cancer
10: 489-506.
14. Zavon, R. R., U. Hoegg, E. Bingham. 1973. Benzidine Ex-
posure as a Cause of Bladder Tumors. Arch. Environ.
Health 27: 1-7.
15. Fishbein, L., W. G. Flamm, and H. L. Falk. 1970. Chemical
Mutagens. Environmental Effects on Biological Systems.
Academic Press, New York, pp. 98-306.
16. Obe, G., K. Sperling, and H. J. Belitz. 1971. Some Aspects
of Chemical Mutagenesis in Man and in Drosophila. Angew.
Chem. Intern. Ed. 10: 302-314.
17. Shaw, M. W. 1970. Human Chromosome Damage by Chem-
ical Agents. Annual Rev. Med. 2/: 409.
18. Vogel, F., and G. Rohrborn eds. 1970. In: Chemical Muta-
genesis in Mammals and Man. Springer-Verlag, Berlin, pp.
519.
19. United Nations Scientific Committee of the Effects of Atom-
ic Radiation, the Genetic Risks of Ionizing Radiation. 1966.
General Assembly Document, 21st Session.
20. Vogel, F. 1971. Why Must Drugs and Other Environmental
Agents be Tested for Mutagenicity. Archiv. Toxicol. 28: 77-
92.
21. Environmental Mutagenic Hazards, Committee 17, 1975.
Sci. 187: 503-514.
22. Rohrborn, G., I. Hansmann, and O. Kuhn. 1971. Induced
Chromosome Aberrations in Early Embryogenesis of Mice.
Human Genetik. //: 316-322.
23. Paigen, K., and R. Ganschow. 1965. Genetic Factors in En-
zyme Realization. Brookhaven Symp. Biol. 18: 99-114.
24. Report of the Secretary's Commission of Pesticides and
Their Relationship to Environmental Health. 1969. U.S. De-
partment of Health, Education, and Welfare, pp. 565-654.
25. de Serres, F. J. 1976. Mutagenicity of Chemical Carcino-
gens. Mut. Res.4/(1): 43-50.
26. de Serres, F. J. 1976. The Utility of Short-Term Tests for
Mutagenicity. Mut. Res. 35(1): 1-2.
27. The Testing of Chemicals for Carcinogenicity, Mutagenicity,
and Teratogenicity. A White Paper, Health and Welfare
Canada. 1973.
28. Bateman, A. J. 1966. Testing Chemicals for Mutagenicity in
a Mammal. Nature, 210: 205-206.
29. Cattanach, B. M. 1971. Specific Locus Mutation in Mice. In
Chemical Mutagens, Principals and Methods for Their De-
tection. A. Hollaender, Ed. Vol. 2, Plenum Press, New
York, pp. 535-539.
30. Bateman, A. J. and S. S. Epstein. 1971. Dominant Lethal
Mutations in Mammals. In Chemical Mutagens Principles
and Methods for Their Detection. A. Hollaender ed. Vol. 2,
Plenum Press, New York, pp. 541-568.
31. Ehling, U. H. and A. Neuhauser. 1972. Induction of Specific
Locus Mutations by Tumor Inhibitors in Male Mice. Sym-
posium on Laboratory Animals in Drug Testing, Hanover.
32. Cattanach, B. M. 1967. Induction of Parental Sex-Chromo-
some Losses and Deletions and of Autosomal Gene Muta-
tions by the Treatment of Mouse Post-Meiotic Germ Cells
with Triethylenenelamine. Mut. Res. 4: 73-82.
33. National Center for Toxicological Research. 1977. 5th Anni-
versary Report, Jefferson, Arkansas, 165 pp.
34. Wilson, J. G. 1973. Abnormalities of Interauterive Develop-
ment in Non-Human Primates. In "The Use of Non-Human
Primates in Research on Human Reproduction" (E. Diczfo-
lusy and C. C. Standley, eds.) pp. 261-292. WHO Res. and
Training Ctr. on Human Reprod., Stockholm.
35. Shepard, T. H. 1973. Catalog of Teratogenic Agents. John
Hopkins Press, Baltimore.
36. Environmental Mutagenic Information Center. 1973. Chem-
ical Mutagens Principals and Methods for Their Detection.
Vol. 3: 282-285, Plenum Press, New York.
37. McKusick, V. A. 1971. Mendelian Inheritance in Man: Cata-
logs of Autosomal Dominant, Autosomal Recessive and X-
Linked Phenotypes. 3rd Ed. John Hopkins Press, Baltimore.
38. Milkovich, L. and B. J. Van Den Berg. 1974. Effects of Pre-
natal Meprobamate and Chioridiaz-epoxide Hydrochloride
on Human Embryonic and Fetal Development. New Eng.
Jour. Med. 29/(24): 1268-1271.
39. Wilson, J. G. and F. C. Fraser, ed. 1977. Handbook of Ter-
atology, Plenum Press, New York.
40. Wilson, J. G. 1972. Mechanisms of Teratogenesis. Amer. J.
Anat. 136: 129-132.
121
-------
41. Wilson, J. G. 1973. Environment and Birth Defects. Aca-
demic Press, New York.
42. Kitamura, S. 1968. Minimata Disease, Study Group Mini-
mata Disease, Kumamoto University, Japan.
43. Spyker, J. M., S. B. Sparber and A. M. Goldberg. 1972.
Subtle Consequences of Methylmercury Exposure: Behav-
ioral Deviations in Offspring of Treated Mothers. Sci. /77:
621-623.
44. Spyker, J. M., M. Smithberg. 1972. Effects of Methyl-
mercury on Prenatal Development in Mice. Teratology 5:
181-190,
45. Courtney, K. D., and J. A. Moore. 1971. Teratology Studies
with 2,4,5-Trichlorophenoxyacetic Acid and 2,3,7,8-Tet-
rachlorodibenzo-p-dioxin. Toxicol. Appl. Pharmacol. 20:
396-403.
46. Moore, J. A., B. N. Gupta, J. G. Zinkl, and J. G. Vos. 1973.
Postnatal Effects of Maternal Exposure to 2,3,7,8-Tetrachlo-
rodibenzo-p-dioxin. Environ. Health Perspective, Experi-
mental Issue No. 5, September.
47. Neubert, D., and I. Dillman. 1972. Embryotoxic Effects in
Mice Treated with 2,4,5-Trichlorophenoxy Acetic Acid and
2,3,7,5-Tetrachlorodibenzo-p-dioxin. Naunyn-Schmiede-
berg Arch. Pharmachol. 272: 243-364.
48. Holson, J. 1976. Personal Communication.
49. Allred, P. M. and J. R. Strange. 1978. The Effects of 2,4,5-
Tetrachloro-dibenzo-p-dioxin on Developing Chick Km-
bryos. Accepted by Arch. Environ. Cont. Toxicol.
50. Strange, J. R., P. M. Allred and W. E. Kerr. 1976. Terato-
genic and Toxicological Effects of 2,4,5-Trichlorophenoxy-
acetic Acid in Developing Chick Embryos. Bull. Environ.
Cont. Toxicol. 15(6).
51. Strange, J. R. and W. E. Kerr. 1976. Teratogenic and Tox-
icological Examination of 2,4,5-Trichlorophenoxyacetic
Acid in Developing Chick Embryos. Toxicol, 6.
52. Kearney, P. C\, E. A. Woolson, and C. P. Ellington, Jr.
1972. Persistence and Metabolism of Chlorodioxins in Soils.
Environ. Sci. Tech. 6: I0I7-I0I9.
53. Gehring, P. J., C. G. Kramer, B. A. Schwetz, J. Q. Rose,
and V. K. Rowe. 1973. The Fate of 2,4,5-Trichlorophenoxy-
acetic Acid (2,4,5-T) Following Oral Administration to Man.
Toxicol, Appl. Pharmacol. 26: 352-361.
54. Cranmer, M. F. 1977. Hazards of Pesticide Development
and Mammalian Toxicity: Carcinogenicity, Teratogenicity,
and Mutagenicity. Proceedings 15th International Congress
of Entomology, 719-735.
55. Carter, C. D., R. D. Kimbrough, J. A. Liddle, R. B. Clinc.
M. M. Zack, W. F. Barthel. R. E. Koehler. and P. E. Phil-
lips. 1975. Tetrachlorodibenzodioxin: An Accidental Poison-
ing Episode in Horse Arenas. Sci. 188: 738-740.
56. Gilbert, S. 1976. Toxic Cloud Hits Italian Town in Unpre-
meditated Disaster. The Washington Post, p. A-18. July 28.
122
-------
CURRENT STATE OF THE PROBLEM OF AN ECONOMIC AND
EXTRA-ECONOMIC EVALUATION OF MAN'S EFFECT ON THE
ENVIRONMENT AND WAYS TO FURTHER FORMULATE IT
[Sovremennoye sostoyaniye problem} ekonomicheksoy i
vneekonomicheskoy otesenki vozdeystviya cheloveka na okruzhayushchuyu
sredu i puti eye dal'neyshey razrabotki]
I. P. GERASIMOV
Brief summaries of the report which are pub-
lished in the collection prepared for this symposium
are the final document compiled by the speaker and
adopted at the First International Scientific Sym-
posium of the Commission of the International Geo-
graphical Union on Environmental Problems. This
scientific symposium took place May 23-31, 1977 in
the Czechoslovak Socialist Republic and was dedi-
cated to the theme of this report. At its last and
closing meeting the text of this document was
adopted; it represents the common viewpoint of all
the symposium participants on the state of the prob-
lem.
The adoption of the closing document was pre-
ceded by a multilateral discussion based on regional
materials presented by the Czechoslovakian spe-
cialists. Without touching on these materials we will
characterize the most important, in a principle re-
spect, reports of the participants at the Conference
according to their subject. We believe that these re-
ports, when taken together comprehensively prove
the main conclusions of the cited document, and at
the same time, the current state of the examined
problem.
Professor T. V. Zvonkova (USSR) discussed the
general nature of research conducted in the Soviet
Union on this topic. In her report, she described
three strategic directions in this research:
1) fundamental theoretical studies on the inter-
relationships in the natural (standard) and trans-
formed anthropogenic ecosystems;
2) studies of favorable and unfavorable anthro-
pogenic effects on the environment;
3) study of the practical aspects of the problem.
The tactics of the research can be described as
follows:
— study of the changes in the functional struc-
ture of ecosystems;
— detection of the stability of natural ecosystems
and their genetic orders to anthropogenic factors of
varying nature;
— determined prediction of the effect of different
contaminating substances;
— probability prediction of environmental
change;
— correction of technical and socioeconomic
plans for development with regard to disorders in
the environment.
Extensive basic research is being conducted in
the USSR to develop the scientific bases for rational
development of natural resources and to evaluate
the positive and negative economic consequences
of man's effect on the environment.
Professor D. Borchert (United States) character-
ized the state of the problem in the United States.
He dwelt on the main difficulties in solving this
problem — the peculiarities of an unplanned, mar-
ket-oriented economy; insufficient awareness of the
population, and mainly of the industrialists, of the
magnitude of damage done to the environment; the
need for uniting industry to take measures to pro-
tect the environment; and the emergence of con-
flicts about solutions for the problem.
A matrix that includes everything affecting the
environment (farms, cities, regions with their pecul-
iarities) must, in his opinion, emerge on the national
level and take into consideration the spatial crite-
rion. The need for planning will coincide with the
need for taking environmental protection measures.
Professor Borchert dwelt, in more detail, on a sys-
tem for controlling land resources in Minnesota,
such as vegetation, soils, slope exposition, land
use, and cost. He indicated that information for the
land use network is verified and supplemented by
aerial survey methods. On the basis of these data,
maps are compiled for using and conserving the en-
vironment which can be employed to simulate pro-
cesses, to regionalize territories, and rationally ar-
123
-------
range industrial and agricultural areas. In addition,
Professor Borchert noted that, in the United States,
a number of other works are underway. To deter-
mine the effect on the environment of urbanized ter-
ritories studies on land monitoring and the problems
of elemental calamities are being conducted. An at-
las has been created of the 20 major urbanized re-
gions in the United States. It consists of 1200 maps
that show population density, water supply, trans-
portation, and sources of contamination.
L. M. Shevchenko (USSR)'s report elaborated on
Professor Borchert's. She dwelt on Soviet study of
United States experience, stressing that interest in
this study was elicited by the fact that the United
States leads the world in the magnitude and in-
tensity of man's effect on the environment. The
broad research that has been done in the United
States, including research by the corporation "Re-
sources for the Future", has already produced cer-
tain results in the formulation of a technique for ec-
onomically evaluating man's effect on the environ-
ment; however, on the whole, American technique
cannot yet be considered completely developed.
It was stressed that American work on economic
evaluation of man's effect on the environment is
based on cost analysis by a profit and loss model. In
certain cases, analysis of damage is based on the
magnitude of losses caused to the economy by ex-
acerbation of environmental conditions, and in oth-
ers, it is determined by direct expenditures that
agencies agree to (in the form of fines, for example)
in order to obtain the right to inflict a certain dam-
age on the environment. Such evaluations are being
made in the United States primarily in respect to the
consequences of a varying types of action. Despite
the use of statistical and mathematical methods of
computation, they are still to a significant degree of
a random nature. These evaluations are usually giv-
en for several variants of an optimal solution which,
as a rule, produces sharp conflicts between the in-
terested companies and departments. In American
scientific circles, bitter debates have been con-
ducted also on general concepts of economic devel-
opment in relation to environment, since even the
already adopted laws on environmental protection
are not always kept by industrialists. Essentially, in
the United States, the greatest attention until now
has been directed on evaluations of the con-
sequences of contamination, in the atmosphere and
on bodies of water, although American scientists
fairly unanimously recognize the importance of
complex evaluations, including extra-economic and
social evaluations. The formulation of a technique
for such evaluations is recognized as being a very
complex matter, and research in this area is at the
most elementary stage. It is important to stress that,
in the United States, there is an ever clearer trend
towards formulation of geographical approaches to
evaluation of man's effect on the environment. In
particular, the opinion is ever more positively ex-
pressed that measures on environmental protection
have the most effective results in cases where they
are planned regionally.
The report of A. V. Antipova (USSR) contained a
critical analysis of certain Canadian works dedi-
cated to the problem of man's effect on the environ-
ment. The interest of Soviet researchers in the Ca-
nadian experience is very natural, since both coun-
tries solve the problem of environmental protection
under similar natural conditions. The diversity of
these conditions has resulted in diverse research
procedures and has governed the width and flexibil-
ity of Canadian general approaches to understand-
ing the interaction between man and nature. In Can-
ada, more attention is given than in the Soviet
Union to the study of land use structure as one of
the most important factors in environmental preser-
vation or contamination. On the basis of a tradition-
al analysis and classification of economic land use,
the Canadian service for land inventory has con-
ducted work on the entire populated section of the
territory (2.5 million kms). They have defined the
nature of modern land use in the territory, have
evaluated land suitability for various uses, and have
formulated regional schemes of recommended use
with regard to the preservation and improvement of
environmental quality. These schemes can un-
doubtedly be viewed as a geographical basis for
analysis of the interaction between man and the en-
vironment and for an economic evaluation of this
interaction. However, such an approach reveals the
limitation inherent in classifications by economic
use only, and does not encompass current or pre-
dictable environmental conditions. However, an
important exception is the work of Professor Dan-
zero, whose approach is close to that of the Soviets.
Professor K. Monteyro (Brazil), who was con-
cerned with urbanized ecosystems, (in the ex-
ample of Sao Paulo) believes that in modeling the
processes in the "man-environment" interface, it is
necessary not only to consider economic factors,
but also man's most basic effects on the environ-
ment. He touched upon the effect of man's presence
on the environment in urbanized areas (for ex-
ample, the relationship between weather dynamics
and the type of city, thermal effects of the city,
etc.), as well as the direct and reverse relationships
that are necessary to consider in the compilation of
such a model. Professor Monteyro believes that the
main principles in the solution of the "man and en-
vironment" problem must be universal, regardless
of the social-economic structure of the country;
however, in the developing countries, it is neces-
sary to pay especial attention to dynamics of the
development of interrelationships between man and
124
-------
the environment, taking into account the progress
of this growth.
Doctor B. Batayls (Mexico) indicated that the
"man and environment" issue in Mexico has ac-
tually changed from a technical problem into an ec-
onomic, political, and social problem.
Private ownership of the land and resources of
the county, and national and international tourism
have resulted in various violations of land use and
cultural monuments in the country. The develop-
ment of industry in the cities has led to an increase
in adverse effects of solid emissions on the environ-
ment.
Professor A. Kostrovitskiy (Poland) underlined
that an increase in the socioeconomic level of life
generally leads to greater changes in the environ-
ment. Therefore, the problem of environmental pro-
tection is both humanistic and scientific. The best
form of environmental protection is spatial planning
and rational use of land. The latter requires a master
approach in which systems must be divided into
subsystems, taking into account the interacting eco-
logical, economic, and social factors.
Professor M. Raza (India) indicated that, for In-
dia, three circumstances are very essential for the
examined problem:
— nonuniform use of natural resources;
— the integrated and undivided nature of natural
and economic aspects of the problem (for example,
famine is the greatest contaminant of the environ-
ment);
— social conflicts.
In India, environmental protection acquires a di-
verse nature in needs and problems vary widely
from city to city and even within one city problems
may be solved in different ways. On the whole, the
agglomeration of industry, which results in the
growth of the economy, increases the profits of indi-
viduals only.
Professor Raza expressed the hope that special-
ists of the developed countries would formulate a
special concept for the developing countries in the
areas of environmental protection and economic
evaluation of the anthropogenic effect on it.
Professor E. Mazur (Czechoslovak Socialist Re-
public) dwelt on questions of regional planning and
prediction in the Czechoslovak Socialist Republic
with regard to environmental protection. In this
country, many charts have already been published
and some are presently being prepared dealing with
environmental protection. They include the size of
technical installations and the degree of environ-
mental contamination, given in economic and extra-
economic indices. These charts will have great na-
tional economic importance.
As already indicated, the head of the symposium
Academician I. P. Gerasimov (USSR) proposed the
formulation of a final text of the entire discussion in
the form of a common viewpoint in a document with
the title: "Current State of an Extra-Economic and
Economic Evaluation of Man's Effect on the Envi-
ronment." This text was adopted by all the sym-
posium participants, and was published as sum-
maries of this report. We will present it once more.
An evaluation of man's effect on the environment
is an important component in control and opti-
mization of the environment. The elaboration of
evaluation methods is especially important for sci-
entific prediction. In studying the problem, scien-
tists of various countries and different specialties
have accumulated considerably knowledge of the
mechanisms involved in the interaction of man and
the environment, and have formulated various mod-
els for this interaction.
To a certain extent, models make it possible to
determine the direct adverse effects of man on the
environment — at least those of an economic, social,
medical, and ecological nature — and to propose
certain methods for analyzing direct economic dam-
age. This evaluation of damage to the environment
by man is very important in many cases, and must
be taken into account in evaluating the economic
efficacy of capital investments in environmental
protection.
At the same time, it is necessary to stress that
"society-environment" interactions which are the
scientific basis for economic and extraeconomic
analysis, have not been sufficiently studied. The fol-
lowing directions of future research are the most
important:
— further comprehensive study of the inter-
actions between man's activity and the con-
sequences of this activity;
— study of the chain reactions in natural, eco-
nomic, and social systems and subsystems elicited
by the aforementioned activity, and their local man-
ifestations;
— elaboration of methods for evaluating social
and other phenomena;
— searches for methods to study the cumulative
effects by means of summing the negative con-
sequences of various types.
The above is especially important in light of the
fact that, currently, the qualitative and quantitative
indices used to determine damage are very in-
complete. This is primarily because of a deficiency
of quantitative data on a number of effects and their
consequences. Secondly, some damaging effects
cannot be quantitatively evaluated at all. Therefore,
further study of the damage must have an inter-
disciplinary nature, both to improve the evaluation
of traditional elements of the damage, and analyze
quantitatively all the elements not previously taken
into account.
125
-------
SOCIAL VALUATION FOR ENVIRONMENTAL POLLUTION
CONTROL
RALPH C. d'ARGE
I. Prologue
During the past several decades there has been
substantial research accomplished concerning the
efficiency and efficacy of various types of environ-
mental management strategies. In the United States
and Western Europe, a growing realization has oc-
curred as to the fundamental interdependence be-
tween types of environmental pollution. Some au-
thors, such as Kneese and d'Arge (1969), have ar-
gued that the widespread concern with air and
water pollution, urban congestion, landscape dete-
rioration and other environmental impacts of eco-
nomic growth are evidence that spillovers are an al-
most pervasive phenomena of most societies. Be-
cause spillovers are present in most of the
production and consumption processes of the econ-
omy, they cannot be viewed as somewhat freakish
anomalies which disturb an otherwise smoothly
functioning social system. Pervasiveness of these
spillovers appears to be generated from three funda-
mental considerations. First, the concept of materi-
als balance is suggestive that any production or con-
sumption process will generate a certain class and
amount of residuals and these residuals will be dis-
tributed among various types of receiving environ-
mental media, i.e., the airshed or water course, de-
pending on the nature of the product and ef-
fectiveness of controls across various media. A
second and more difficult problem to overcome is
the lack of basic incentives for proper management
of these residuals by planning agencies or individual
firms. By incentives, I mean the ability to induce
proper environmental concern and actions when
confronted with choices of production or consump-
tion processes and alternatives for waste treatment.
A primary example of this problem is in the Ruhr
industrial basin in West Germany with perhaps the
most highly tuned water quality management
agencies in the Western world. That is, the develop-
ment of "Genossenschaften" with a wide range of
tools for water quality management in the late nine-
teenth century. While tools for water quality man-
agement were available, air quality was not ade-
quately regulated and normally water borne residu-
als became airborne. These rather general
considerations of materials balance and incentives
have provided substantial impetus toward new de-
signs of environmental management strategies. A
third pervasive problem stems from the lack of sub-
stantial information on three key elements of de-
signing environmental management strategies.
These three key elements are: (J) The development
of precise and scientifically defensible relationships
between the amount of pollutant emitted into the
environment and harmful effects that might occur.
For example, how incidence rates of skin cancer
vary with tonnage of nitrogen fertilizers applied to
crops would be one example of such a relationship.
These relationships are often referred to as " dose-
response" curves or transfer functions. (2) Each
type of harmful effect to man would have to be eval-
uated and weighted. Some of these effects that
should be considered include: human health effects:
effects on ecosystems; changes in productivity of
workers; harmful impacts on capital equipment or
other aspects of the economy; and finally, the effect
on amenities and recreation available to individuals,
i.e., black rivers or yellowish brown horizons. For
many environmental pollutants observed today,
these three information components are just not
available. At best, dose-response relations between
chemical carcinogens and their effects on human
beings are extrapolated from animal studies and,
as one epidemiologist remarked, "Mice are not
men." (3) Another important element is to be able
to identify all control or regulatory options and with
full understanding of their costs. The control options
may involve an entire spectrum of alternatives in-
cluding "end of pipe" treatment; changes in chem-
ical inputs; substitutes within the production pro-
cess; and finally, possible avoidance by those indi-
viduals being affected. In other words, for effective
environmental management strategies to be con-
ceived, one must be able to plot logically and com-
pletely a mapping between pollutant dose, ultimate
sink and location, and an economic-social valuation
of both the short-term and long-term responses to
this emission. Unfortunately, for most of the impor-
126
-------
tant environmental contaminants such a "cause and
effect" relationship with sound scientific under-
pinnings is not available. Thus there is a fundamen-
tal information problem in environmental manage-
ment not encountered in many other management
problems.
It is the purpose of this report to examine in detail
recent trends in types of environmental manage-
ment strategies; new concepts in assessing benefits
and costs associated with regulation; and the rela-
tively unique character of uncertainly on environ-
mental effects and how this can be adjusted for in
making rational environmental decisions.
II. Types of Environmental Management Strategies
There are perhaps as many types of environmen-
tal management strategies as there are administra-
tors and their critics involved in environmental pro-
tection throughout the world. Each individual ad-
ministrator is likely to perceive a set of unique
problems associated with any given strategy and
thereby try to accomodate them in a rather unique
fashion. Given this observation, there appear to be
four underlying components to any given environ-
mental management strategy. There is the identifi-
cation or discovery phase; the monitoring phase;
assessment phase; and finally, the regulatory phase.
In the identification and/or discovery phase, the key
element is how much knowledge is available as to
the potential effects of the pollutant. Unfortunately,
in most instances in the United States, the discov-
ery of a harmful effect occurs after the pollutant has
had substantial negative effects. With a few ex-
ceptions, including the new Ames test on carcino-
gens, there appear to be no systematic ways of dis-
covering how harmful a given pollutant is until at
least some of the harm has occurred. A prime ex-
ample of such a problem confronting the United
States today is with the emission of the fluorocar-
bons into the stratosphere. While preliminary evi-
dence indicates there is likely to be a positive tem-
perature effect and a possible increase in skin can-
cer associated with fluorocarbons, scientific
evidence is so weak at this point that accurate map-
pings between pollutant emissions and social effects
is impossible; but if the United States continued to
emit fluorocarbons into the stratosphere for a long
enough period, scientific evidence would accumu-
late as to actual harm. The question then arises as to
what environmental stragegy the United States
should follow. One strategy would be to allow emis-
sions to continue until harm is established; a second
would be to ban the use of fluorocarbons as fast as
possible and take no chance as to harmful effects.
Some middle ground strategy might be devised that
would allow for very early detection of harmful ef-
fects, but not provide a total ban on emissions. It
appears to me that for these types of environmental
issues, which seem to be growing, i.e., toxic sub-
stances cannot as yet be handled in terms of rational
environmental decisionmaking. The discovery of
harmful effects will probably continue to be rela-
tively haphazard and without design. While this
may be the case on certain types of environmental
pollutants, it would seem to be worthwhile in terms
of a cooperative research program to try to design
new methods for identification of potential pollu-
tants before they actually enter the economy and
the natural environment. It has been estimated that
the United States produces more than 30,000 new
types of toxic substances each year, and therefore
any type of advance identification system that was
not economically efficient would have little chance
of implementation.
The second major ingredient in the design of envi-
ronmental management strategies is the monitoring
phase; that is, a phase where direct observations
and measurements are made on the effects of a pol-
lutant and its dispersion to ultimate sinks. In recent
studies in the United States, an attempt has been
made to estimate a materials balance between
sources and sinks for various types of pollutants.
Unfortunately, knowledge on the nature of trans-
mission of pollutants through the natural environ-
ment and via man-induced routes is only in its infan-
cy, e.g., the ultimate sink for fluorocarbons or nitro-
gen oxide for fertilizers. Again, one of the
substantial difficulties in the monitoring component
or phase of analysis is deciding how much to invest
in developing accurate monitoring systems to
sample concentrations of the pollutant and its dis-
persion. In Los Angeles, for example, there are
more than 20 fixed point air monitoring stations
within the metropolitan area. However, these mon-
itoring systems given only a vague idea of variabili-
ty by street or locale in actual concentrations of ni-
trogen oxides or carbon monoxide. The critical is-
sue, in terms of monitoring, is to decide how much
to invest such that the expected value of additional
information is just counter-balanced by the cost of
additional monitoring. Unfortunately, the expected
value of additional monitoring information cannot
be accurately predicted unless at least a partial as-
sessment is made as to the use of this information in
designing regulatory strategies. Thus, we seem to
be caught in a dilemma, at least with regard to the
monitoring phase, that a partial assessment and par-
tial environmental management strategy must be
designed simultaneously with the design of the
monitoring system.
The third phase I have identified as assessment
implies just that, in essence taking all the available
information on the environmental pollutant and ex-
amining how realistic it is and what it implies with
regard to the seriousness of the environmental
problem. In the Western World, either implicit or
127
-------
explict benefit-cost analyses have been used exten-
sively in this phase; that is, the process of summing
all the benefits of a given regulatory act or given
degree of control and comparing these with costs of
control to consumers, producers, and policy-
makers. However, only in recent years have econo-
mists been able to even suggest techniques for accu-
rately measuring the benefits of regulation for vari-
ous kinds of air or water quality parameters; and, in
fact, there is still a continuing debate in the U.S. as
to what constitutes an appropriate measure of bene-
fit. We will return to this issue later in discussing
techniques of assessment of benefits and costs;
however, it should be recognized that assessment
depends critically on the political acceptability of
valuation process for regulating a given environ-
mental pollutant. At least in the U.S., the trend has
been toward identifying an environmental standard
on the basis of health effects, excluding any other
consideration of benefits or risks. Yet, such assess-
ment strategies can be misleading. For example, the
debate on nuclear powered electrical energy devel-
opment and inherent risks, one researcher demon-
strated that air pollutant effects from coal fired
plants may be much worse in terms of expected hu-
man mortality than risks of nuclear radiation [Lave
(1975)]. However, assessment technologies have
undergone some reasonably substantial advances in
recent years. These have been primarily in the di-
rection of assessment and measurement of benefits
and costs associated with regulation in terms of
conceptual content and empirical validation. How-
ever, in certain instances there have been few, if
any, advances. Traditionally, benefit-cost analysis
attempts to value benefits and costs occurring at
vaious points of time, and yet for some of the most
serious environmental issues there are very sub-
stantial lag times between emission or dose and the
harmful effect. For example, for most suspected
carcinogens in water courses, the response time in
terms of human health effects may be as long as 20-
30 years. However, a dollar of cost in 30 years is not
comparable with a dollar of benefit or cost occur-
ring right now. This is because, in benefit-cost anal-
ysis, benefits and costs are weighted according to
the time of occurrence and benefits in the future are
valued at less than they are now. The common prac-
tice in benefit-cost analysis is to use the concept of a
discount rate or rate of interest to discount the fu-
ture in relation to the present. Such procedures may
lead to rather exotic if not absolutely absurd results.
For example, a billion lives lost in 1,000 years may
be equivalent to less than one life being lost in 1977.
Many researchers including some from the U.S.
National Academy of Sciences, have recognized
the weaknesses associated with traditional benefit-
cost analysis applied to these very long-term com-
parisons associated with chronic effects of environ-
mental pollutants. Yet, no reasonable alternative
has yet emerged to aid environmental decision-
makers.
The final phase 1 would like to refer to as the reg-
ulatory phase, which in essence amounts to deci-
sion as to how and when to regulate a given envi-
ronmental pollutant. The key issue here is policy
costs and/or values associated with regulation and
this is an area where the free market economies
have had and will continue to have the most diffi-
culty.
There appear to be two fundamental types of reg-
ulatory controls, one which I will refer to as direct
controls and the other as indirect controls. Direct
controls, by definition, are those which are applied
at or to the source of the environmental problem
and which threaten sufficient penalities as to make
avoidance extremely costly. The continuous mon-
itoring and closing down of a particular factory
which habitually violates an emission standard is an
example of such a direct control. Direct controls
leave little or no latitude for individual decision-
making. There is only a single possible link between
an infraction and agency action. The price of not
complying is established at a level high enough to
insure universal compliance.
Indirect controls, alternatively, are defined as en-
vironmental management strategies with at least
two links of expected causation between problem
source and application of control. One example of
such an indirect control is a pollution tax which es-
sentially operates on the basis of two behavioral
postulates. The first is if the polluting firm (or indi-
vidual) is taxed for waste discharge, it will attempt
within the bounds of efficiency to avoid this charge
by altering production, adopting waste control tech-
nologies, relocating, or by some other means. The
second behavioral postulate is that if the firm
reduces waste discharge by a certain amount, so-
cietal losses will also be concurrently reduced. Di-
rect and indirect controls are distinguished by the
number of their technological or behavioral links.
However, it appears that most economic controls
today, at least in the U.S., have been mixtures of
both direct and indirect controls. For example, a
law requires motorists to stop for a red light, but if it
is not obeyed and the violator is caught, he is penal-
ized. However, the penalty is usually not high
enough to command universal compliance. Zoning
laws can also combine direct and indirect control
strategies by containing both penalty and variance
provisions, the latter based on the behavioral as-
sumption that affected parties will comply and not
request variances.
These behavioral linkages, even in what seem the
most simple control stragegies, can require sub-
stantial amounts of research and background infor-
mation before the control strategy in which they are
128
-------
embodied can be satisfactorily implemented. For
example, a gasoline tax to reduce automotive emis-
sions seems relatively simple, but the assumption
that, even after one allows for differences in gas
consumption between vehicles of different weights
and makes, there is a fixed technological link be-
tween miles driven and pollutants emitted is weaker
than one might think. It has been carefully docu-
mented that, apart from mileage driven, the ways in
which different people drive, i.e., acceleration,
downshifting, stopping, have markedly different im-
pacts on exhaust emissions of reactive hydro-
carbons, carbon monoxide, oxides and nitrogen. In
analyzing the relationship between gasoline taxes
and vehicle emissions, one must therefore include
the possibility that a higher gasoline tax aimed at
encouraging vehicle operators to drive shorter dis-
tances at slower speeds to reduce gasoline con-
sumption might instead increase total exhaust emis-
sions. Such a perverse type of emissions impact
could also be anticipated for water borne emissions.
The standard which is based on mean monthly flow
rather than on daily variations might reduce infor-
mation costs but increase the probability of positive
incentives in the short run.
It is important in determining where controls
should be applied, and to what degree, that those
subject to regulatory efforts should know how many
and what kinds of alternatives are available and
how much they cost. Thus, to make the buyer pay a
purchaser excise tax on steel from a plant where
emissions from the coke and quenching processes
are damaging to the environment is a highly indirect
method of control, since it assumes the purchasers
will buy less steel and the plant's production will be
reduced, and thereby emissions will be indirectly
reduced. The steel purchasers or users obviously
have fewer methods of control, i.e., to reduce de-
mand or possibly negotiate with steel producers,
but steel producers can resort to precipitators, use
sealants on coke oven doors, relocate or redesign
coke ovens, or institute almost any conceivable
process change which is less costly than paying a
fee or charge. Controls that are more direct may be
favored over less direct controls because linkages
between cause and effect are fewer in number. How
behavior relationships influence the actions of a
public agency in relating environmental quality
measures to emissions is highly uncertain. There is
the uncertainty surrounding the transfer functions
relating emissions to ambient quality measures.
Ambient quality standards are generally established
independently of the level of damages, control
costs, or mandate and budget of the agency, at least
in the U.S. In consequence, a very indirect control
is applied with a large number of behavioral transfer
functions and assumptions included. A less indirect
control using taxes or charges contains, by defini-
tion, fewer behavioral linkages and thereby implies
fewer uncertainties. Thus, information costs may
tend to be lower for controls which are more direct.
However, enforcement and related monitoring
costs may become prohibitively high if control costs
are too direct and/or oppressive, i.e., closing down
a factory in regions highly dependent on that facto-
ry for employment and economic growth.
In Figure 1 I have depicted the hypothesized rela-
tions between the total cost of control and the de-
gree of directness of the control; that is, the number
of behavioral and/or technological linkages between
the incentive introduced and the point of control.
What this figure is suggestive of is that there is some
combination of regulatory strategies where the total
costs of control are minimized. That is, where the
offsetting effects of information and enforcement
costs are reduced as low as possible. The U.S. ex-
perience, in terms of the regulatory phase, I would
suggest, has been rather mixed. For cases where
monitoring and enforcement costs have been ex-
tremely low and the technologies appropriate for
pollution control well understood, they have
seemed to work, i.e., requiring participators on new
coal fired electric generation plants which are
owned by public utilities. Where enforcement and
information costs have been extremely costly, i.e.,
for controlling automotive emissions to meet mini-
mum health standards in various areas, there has
been extreme difficulty in finding direct controls
that would be workable. Most of the effort has been
concentrated on the rather indirect approach of
controlling for average emissions of the vehicles be-
fore they leave the automotive plant. How efficient
this indirect control has been in reducing emissions
has not yet been evaluated.
III. Assessment of Benefit and Cost Measures
Reduction or elimination of environmental pollu-
tants of various types will be beneficial to one or
more of the following categories: human health; ec-
onomic productivity; recreation; aesthetics; and
ecosystem function and productivity. For assess-
ment, benefits need to be at least qualitatively eval-
uated and, if possible, quantitatively measured. The
central issue is how to identify the probable effects
and accurately measure associated benefits. One
approach often adopted by public agencies in the
U.S. is to evaluate the benefits in terms of the posi-
tive contribution to the economy including savings
in health costs, the economic productivity of less
corrosion of equipment, and increased amounts or
types of potential recreational activities potentially
being available. Following this conceptual ap-
( proach, benefits of not perturbing an ecosystem
' could potentially be valued as to their direct and in-
direct contribution to the economy. An example of
this economic approach is the valuation of reduced
129
-------
Figure 1. Hypothesized Relation Between Total Cost of Controls and Directness of Controls**
"Along the horizontal axis are presumed to be sets of different controls
with the most direct to the left and most indirect (greater behavioral relation)
to the right.
*"Ralph C. d'Arge, "Environmental Policy Costs: Definition, Measurement,
and Conjecture," Cost Benefit Analysis and Water Pollution Policy, Peskin
and Seslcin (eds.), The Urban Institute, Washington, D.C. (1974).
130
-------
human mortality from increased earnings and re-
duced medical costs [Rice (1965), Lave and Seskin
(1970)]. It can be anticipated that for almost any
conceivable environmental problem this approach
would substantially undervalue actual benefits to
society. Actual health costs do not, except at a min-
imum, reflect accurately how individuals or collec-
tive groups value good health. While this approach
has been extensively used for benefit-cost evalua-
tions in the U.S., at best it may be misleading and,
at worst,it may lead to incorrect environmental de-
cision making. For example, individuals may place
a very high value on additional years with reason-
able health during the latter part of their lives. How-
ever, because of retirement, they may well not have
substantial future earnings capability which would
provide them with a low benefit amount relative to
highly productive younger members. Whether so-
ciety should value older people at less than younger
people is an ethical issue far beyond the scope of
my discussion. However, it does indicate the seri-
ousness of the problem with utilizing direct effects
on the economy as a measure of environmental ben-
efits.
A relatively new evaluation technique has
emerged in recent years in the U.S. and some of the
Western European countries. The valuation of ben-
efits and costs of environmental control are as-
sessed by what individuals are willing to pay to
avoid the environmental pollutant or how much
they must be compensated to be exposed to it
[Bohm (1972), Brookshire (1976)]. The fundamental
idea underlying the concept is that when individuals
are willing to pay more for avoidance of one pollu-
tant as compared to another, more benefits should
be assessed to the first pollutant's control. In Figure
2 are depicted the income related and "willingness
to pay" related measures of economic benefits to
the economy where income equals abed and "will-
ingness to pay" equals abed. Only in the extreme
case where unit value does not change with quantity
available will "willingness to pay" and income
measures be identical. This is depicted in the lower
graph of Figure 2. Some economists in the Western
World have finely tuned this valuation procedure,
which is often called the "Pareto improvement prin-
ciple," named after the Italian economist, Vilfredo
Pareto. Under this principle the regulation of a pol-
lutant is regarded as socially worth while if it would
benefit some people without making anyone else
worse off. In some instances in the Western coun-
tries this has been interpreted to mean that a partic-
ular environmental regulation could make some
people socially better off so that they could com-
pensate other individuals who are made worse off.
Following Mishan (1971) for health effects on indi-
viduals, there are essentially two mqjor measures.
One is the maximum that one would be willing to
pay to avoid added exposure to environmental risks
or to pollutants. A second measure is equal to the
minimum amount a person would accept if he were
to be exposed to the additional environmental risk
such that he would be as well off as before. Clearly
for environmental health problems, these measures
are likely to be substantially different. If the individ-
ual must pay, he is constrained by income. Alterna-
tively, where the payment flows to the affected par-
ty, the amount requested could be very large and
perhaps even infinite.
In recent years three relatively new approaches
have emerged to attempt to measure these values.
Western economists generally have been worried
about the so-called "free rider" problem where it
pays for the individual not to provide a responsible
answer because it pays for him to influence the out-
come. However, recent studies indicate that such a
problem perhaps does not exist [Bohm (1972),
d'Arge, et al, (1977)]. That is, individuals will at-
tempt to provide a reasonably honest reponse to
maximum bids or minimum compensation ques-
tions for various types of environmental attributes.
These three new approaches that I wish to review
briefly are: (1) bidding and asking game situations;
(2) observations as to hypothetical or actual sub-
stitution behavior; and (3) observations on actual
behavior associated with risks where these observa-
tions have been specialized to types of environmen-
tal hazards. The approach of bidding-asking games
is perhaps the most simple of the three in the sense
that one provides the person with a set of environ-
mentally related alternatives and asks him to bid
across them and also to indicate minimum com-
pensation across them. To report in more detail on
how such a valuation procedure is undertaken, I
will describe briefly research we have just com-
pleted on attempting to value visibility in the south-
west part of the U.S. The policy question with re-
gard to visibility is whether to build 15-20 large coal
fired electric generation plants in this region and to
what extent to control emissions of sulfur oxides
and particulates. In the past 4 to 5 years there have
been a number of efforts to measure the value of
visibility, which appears to be almost the classic
case of a true public good; that is, one individual's
enjoyment of it is not impaired by others [Randall
(1974), Brookshire, et al, (1976)]. In each case these
researchers attempted to use the concept of bidding
games developed by Davis (1963) and Bohm (1972),
among others, to infer the value of clear skies, that
is, the lack of noticeable air pollution concentra-
tions that might impinge upon visibility. In Figure 3
is depicted the basic structure of the problem of
measurement and an outline of the approach. Es-
sentially what was attempted was to develop rela-
tionships between air pollutant emissions and am-
bient concentrations of particulates, to connect this
131
-------
Figure 2. Measures of Benefits
Value
Surplus Value
Gross Income
Quantity
Value
Quantity
to scientific measures of visibility reduction, and
then to examine how the public's perception of visi-
bility change occurs. Tests were then made using
bidding games for various visibility changes. In
terms of these linkages, the first one between air
pollutant emissions and ambient concentrations re-
quires using some form of atmospheric dispersion
model. Applications of such models appear to work
for a single plume but not very well in examining the
regional dispersion of small particulates with wind
patterns continuously changing, which character-
izes much of the Southwestern U.S. This research
project did not involve the development of a dis-
persion model beyond an attempt to examine how
132
-------
Figure 3
Air Pollutant Emissions
Ambient Concentrations
Scientific Measures of
Visibility Reduction
Public's Perception of
Visibility Change
Direct Methods of
Valuation
Indirect Methods
of Valuation
Bidding or Asking
Games
Hypothetical or Actual
Substitution
useful an average concentration approach might be.
From ambient concentrations to scientific measures
of visibility reduction, we were able to find a sub-
stantial amount of research on light extinction and
light contrast. Equation 1 reflects the best estimate
at this time between scientific measures of visual
range and mass concentrations in a non-urban envi-
ronment:
Lv -
.02 + .01
+ .01
-.003
j M(£tg/m3
io-
(1)
where Lv is meteorological range in miles and M is
the mass of aerosols. The actual coefficients were
developed utilizing results by Horvath and Noll
(1969), Horvath and Charleson (1969), Charleson
and Alquist (1968), and Alquist and Charleson
(1969), along with Ettinger and Royer (1972). I will
not go into detail on this relationship except to note
the range of estimates for coefficients reflects ap-
proximately a 90 percent confidence interval. Need-
less to say, there seems to be enough evidence to
connect, even without finely tuned diffusion mod-
els, pollutant emissions to reductions in scientific
visibilities as measured by meteorological range.
The next relationship between scientific measures
of visibility reduction and the public's perception of
the visibility change is much more difficult to docu-
ment. However, some preliminary tests for the
Southwest U.S. comparing measured visibility from
cameras as contrasted with reported visibility by
U.S. Park Rangers with 20-20 vision have been un-
dertaken. The correlations between the two esti-
mates were extremely high. Thus it appears reason-
able to assume that the relationship between scien-
tific measures of visibility reduction and the
public's perception of visibility change would al-
most be a one-to-one correspondence, provided
each of the members of the public had 20-20 vision
(corrected or uncorrected).
The next step in the experiment was to construct
a set of photographs taken from the same location
for the purpose of presenting distinct differences in
visibility. A set of photographs is appended to this
paper. By taking pictures, different levels of visibil-
ity were obtained in terms of the photographs. I
should note that this may be a very expensive way
to obtain adequate representations of alternative
environmental states or characteristics through
which persons can respond to bids. For example,
there may be no method of depiction of the environ-
133
-------
mental hazard without actually harming the envi-
ronment. The starting point for the economic analy-
sis component of the study was this set of photo-
graphs with well defined scientific measures of
visibility that varied from approximately 25 to 75
miles. The next step was to construct question-
naires for obtaining maximum bids and minimum
compensation of persons where the bids were
across various combinations of the photographs. In
Table 1, are recorded monetary amounts for the
various bids and compensation. It is clear upon ex-
amination of Table 1 that substantial differences oc-
curred between minimum compensation and maxi-
mum willingness to pay, and there was a substantial
variation in individual responses. For the experi-
ment three types of potential biases were analyzed.
These were: (1) strategic bias, which means an at-
tempt by any individual to influence the outcome or
results of the bidding process; that is, through his
bid or amount of compensation he attempts to influ-
ence environmental policy; (2) information bias,
which is a potential set of biases induced by the test
instrument, the interviewer or the process and its
effects on the individual's responses. For example,
if the individual were asked to pay a higher utility
bill as opposed to a wage tax, western economic
theory suggests that the maximum amount he would
pay would be different; that is, he would have more
ways of avoiding the tax and finding substitutes if it
were a wage tax as opposed to a utility bill; and (3)
hypothetical bias, which amounts to the potential
error due to not confronting the person with a real
situation. Asking someone what they will do a pri-
ori is not the same as confronting them with a set of
well defined prices or costs on events and making
them pay for them. However, the degree of hypo-
thetical bias may be relatively small if one can in-
duce the individual to believe he will have to pay.
Then, in essence, what he is paying is for a contin-
gent claim not unlike decisions with regard to
weather, the growing of crops, etc.
In order to examine the extent of such biases, we
structured the interview process such that first the
game was explained to them and then, after the indi-
vidual had given his maximum bid, we gave the in-
dividual a hypothetical mean bid and asked him
whether or not he would alter his own. The "game"
was structured so the individual would presume
chat he would have to pay the "average" bid, not
his own. The presumption was that if his bid was
below the mean bid and he desired to increase the
magnitude of the average bid, he would bid higher.
Alternatively, if his goal was to reduce the mean
bid, he would revise his bid downward. Only in the
extreme case where the individual's maximum bid
was identical to the mean bid would there be no in-
centive for the individual to change if he were acting
strategically. In addition to this process, we also
Table I
Measures of Visibility Benches
Standard
Type of Measure Mean Bid Deviation
(1976 U.S. Dollars)
Maximum willingness to maintain $ 6.60 $ 5.51
75 mile as opposed to 25 mile
visibility
Minimum compensation to have 25 $91.36 $166.75
as opposed to 75 mile visibility
"Payment per month per household unit.
questioned the individual about his bid being too
low. We suggested his bid along with others was not
sufficient to keep power plant emissions at present
levels for sustained high quality of ambient air and
then asked if he would revise his bid. Some individ-
uals did increase their bid, which suggests their first
maximum bid was not a maximum. Whether this is
positive evidence of a lack of strategic bias by indi-
viduals is unclear. However, individuals may be
acting strategically by subjectively forming their
preferences as to reflect their maximum bid, select-
ing the bid appropriately, and then not revising.
However, it appears to be an additional indication
along with the results of Bohm (1972), that individ-
uals generally do not act strategically, at least in a
meaningful manner to bias the outcome of the re-
sults. In addition to tests on strategic bias, we also
examined various forms of information bias, essen-
tially trying to establish influences on various as-
pects of the game. We found that bids varied sub-
stantially across individuals according to whether a
high or low starting bid was suggested by the inter-
viewer. The major findings of this study were as fol-
lows. First, that reasonably accurate measures of
willingness to pay for maintaining visibility can be
obtained through application of bidding-asking
games. A high degree of replicative consistency was
found between the empirical results of this study
and previous experiments. Secondly, estimates for
minimum compensation appear to contain greater
elements of variability and arbitrariness and refusal
to provide reasonable estimates. Additional re-
search is therefore necessary before minimum com-
pensation measures can satisfactorily be used for
environmental policymaking. Third, serious bias
problems were discovered for willingness to pay
measures associated with the level of suggested (or
starting) bid by the interviewer, payment vehicle
used, and the amount of information on previous
bids that was given to the respondent. This suggests
that the amount of information available to the re-
spondent will influence the magnitude of the bid and
thereby the resulting estimate of aggregate benefits
of air pollution control. Therefore, bidding games
appear to be highly situation or event and time spe-
cific in their orientation; that is, results from one
game are not necessarily applicable to other situa-
134
-------
tions and times. The observed payment vehicle in-
fluence was consistent with what western neo-
classical economic theory would predict. However,
aggregate benefit estimates depend thereby on how
payments are made. In consequence, traditional
benefit-cost analyses must provide the mechanism
for how payments are to be collected prior to esti-
mation of benefits. Fourth, with the exception of in-
come and education, other socioeconomic differ-
ences did not consistently affect observed maxi-
mum willingness to pay. This is suggestive that
bidding game questionnaires and procedures could
be substantially shortened, omitting characteristics
other than income and education, at least in the
U.S. situation. First, little or no evidence was ob-
tained as to the presence of detectable strategic bias
or the desire by respondents to influence the out-
come of the game by offering non-truthful bids. In a
related part of the experiment, however, more than
one-third revised their bids when confronted with
an undesirable result given their previous bid.
Whether this is evidence indicating the presence of
strategic downward bias in their first bid or the ef-
fect of new information or both cannot be ascer-
tained, given our research design. Sixth, a sub-
stantial difference is observed between the measure
of average willingness to pay and minimum com-
pensation. Whether this is explained by confusion
over the concept of compensation or ownership of
the common property resource and/or there is a
large income-wealth effect cannot be determined.
Because a positive and significant income effect was
observed across both minimum compensation and
maximum willingness to pay, minimum compensa-
tion would be predicted by western neoclassical ec-
onomic theory to be larger than willingness to pay
and obviously this occurred.
The next approach that I wish to briefly discuss is
what might be called the substitution activities ap-
proach, where the individual is confronted with dif-
ferent kinds of environmental hazards or environ-
mental quality and is given the option of suggesting
what they might do. Specifically, whether they
would substitute indoor for outdoor activities or
take other evasive actions to avoid the environmen-
tal hazard. For example, a particular individual may
decide to specialize in indoor sports if photochemi-
cal smog or other concentrations of pollutants are at
high levels for most of the weekends the individual
has available for recreation. Some preliminary stud-
ies have been completed in the U.S., again in the
Southwest region [Blank, et al. (1977)]. It was
found that individuals typically do alter both their
expenditure patterns and activity profiles in re-
sponse to the various levels of air pollutants, or at
least they suggested they would undertake such
substitutions. From preliminary evidence we have
found that the typical cost of this substitution lies
somewhere between the maximum willingness to
pay and the minimum compensation of most indi-
viduals. That is, it provides a reasonably accurate
middle bound for valuing visibility on an individual
basis for environmental risks or pollutants. Sub-
stitution "costs" across the visibility range from 75
to 25 miles for typical visibility was approximately
$60.00 per month. This estimate lies between the
range of maximum willingness to pay and minimum
compensation that was derived from a bidding-ask-
ing game approach reported on earlier. Thus, how
people suggest that they allocate their time and ac-
tivities in response to changes in environmental
risks might be an appropriate future research area
for assessing benefits. While the substitution ap-
proach has demonstrated feasibility for one ex-
ample, it is highly questionable whether it will be
useful in generalized environmental applications.
The difficulty, of course, is that most persons do not
fully understand the health or environmental con-
sequences of exposures, particularly long-term
chronic exposures to various kinds of air borne or
water borne pollutants. Recent studies in the U.S.,
for example, have demonstrated that halogenated
methanes may be responsible for certain types of
cancers of the liver and kidney. One major haloge-
nated methane obviously is chlorine used to decon-
taminate drinking water. However, it cannot be an-
ticipated that the average citizen has the requisite
knowledge with respect to probabilities of disease
to make a rational tradeoff between consuming non-
chlorinated water and subjecting themselves to a
very small probability of death from various types
of cancers; and there are not many substitution pos-
sibilities for the individual to undertake. Alterna-
tively, for recreation or aesthetic effects of environ-
mental pollutants, it appears that the substitution
approach might lead to some scaler of individual
values which can then be used to assess environ-
mental benefits.
The third promising approach of estimating envi-
ronmental benefits focuses on the issue of risk and
how individuals and/or societies value being sub-
jected to various levels of environmental risks. In
recent years a number of researchers have explored
the problem of valuing human lives and its mirror
image, the disvalue associated with death when the
probability is one. It has been generally concluded
in these studies that there is no reasonable econom-
ic methodology for assessing the value of human
lives lost because of some form of environmental
contamination [Mishan (1971), Shelling (1961)].
However, Western economists in the last several
years have explored possibilities of valuing dif-
ferences in the probability of death and what this
implies about individuals' valuations of taking in-
creased chances associated with the work habitat,
recreational activities, and environmental risks
135
-------
[Thaler and Rosen (1974), National Academy of
Sciences (1977)J. Essentially, this approach in-
volves examining how individuals make tradeoffs
with respect to economic values inclusive of wages
and other benefits with risk taking associated with
obtaining an economic value of the probability of
death. The relationship is therefore calculable from
observations on individual behaviors. Such a rela-
tionship is depicted in Figure 4. The depicted rela-
tionship indicates that it can be anticipated if the
probability of death of a particular occupation or ac-
tivity increases, the economic value associated with
that activity must also increase. Thaler and Rosen
have made empirical estimates of such a relation-
ship across selected occupations in the U.S. and
have found that this implies that the calculated rela-
tionship for the value of death with probability of
one is approximately $340,000. While such an esti-
mation process leaves a great deal to be desired
with regard to valuation, it does focus on a way to
assess the benefits of reduced environmental risks.
That is, if an individual is willing to accept X units
of economic value for an increase of. 10 in the prob-
ability of death, then it might suggest that if a partic-
ular environmental decision reduced the environ-
mental risk by .10, the individuals involved would
be willing to pay at least X units of economic value
for it to be accomplished. Of course, there are a
large number of underlying assumptions in trying to
adapt observed risk taking of individuals to a partic-
ular situation concerning the environment. For ex-
ample, individuals who wash windows on high
buildings may have a peculiar attitude towards risk
which would not be representative of the popu-
lation. Also, environmental risks may fundamental-
ly be different in terms of the psyches of individuals
as contrasted to their own risk taking in recreational
or other activities. Third, risk taking appears to be
related in some definite manner to income [Fried-
man and Savage (1948), Pratt (1960)|. Finally, there
is the extreme difficulty in establishing whether risk
taking by individuals should be evaluated the same
Figure 4.
Economic
Value
Probability of Death
136
-------
as risk taking by society as a whole; that is, societal
preferences may not be adequately reflected in ob-
servations on individuals with respect to increased
risks [Arrow and Fisher (1971)). While I spent very
little time in examining the various qualifications of
these new assessment approaches for benefits, all
seem to promise at least a first attempt at valuing
the removal of environmental hazards or pollutants.
Unfortunately, they are at this time based almost
exclusively on observing individual behavior or
asking individuals as opposed to deriving a collec-
tive evaluation or preference ordering associated
with environmental quality. Continued research in
the direction of translating individual into societal
preferences most obviously needs to be accom-
plished.
IV. Uncertainty and Environmental Management
Increasingly, it is recognized that many environ-
mental issues are surrounded with uncertainty of
various types. For example, in the mapping from
environmental pollutant dosage to ultimate human
effects, there are large uncertainties with each
transfer function or dose-response relationship.
There are also extremely large uncertainties as to
the effects, both on humans and on the ecosystem.
In a recent study on effects of supersonic aircraft on
world climate, the following uncertainties were
documented [d'Arge, et at, (1974); Grobecker
(1974)]. One, there are large uncertainties in the cli-
matic effects of various levels of emissions of
oxides of nitrogen, sulfure oxides, particulates, and
water vapor and their effects on surface climate.
Some tropospheric changes may involve irrever-
sibilities in the natural environment, although no
substantive evidence of this is now available. Two,
there are extremely large uncertainties in the trans-
lation of tropospheric climatic changes into quan-
titative biological effects. Three, there are very
large uncertainties as to how social communities
and the economic system adjust to large scale cli-
matic changes, or even to small climatic per-
turbations. Four, none of these substantial uncer-
tainties are likely to become near certainties with
exacting precision within less than one or two dec-
ades. Recent climatic modelling efforts have in-
dicated that the assumed negative effect on surface
temperature may, in fact, be reversed and that flight
of supersonic aircraft will increase surface temper-
ature. Another example of these extreme uncer-
tainties can be identified with respect to toxic sub-
stances and their chronic effects on humans. For
each new toxic substance there is an extreme range
of possible chronic human effects, none of which
will be documentable until they in fact occur.
Most environmental decisionmaking in the past
has presumed a knowledge of the dose-response re-
lation without adequately documenting its exis-
tence. What I should like to suggest is that the un-
certainties associated with environmental decision-
making are much broader and more complex than
those observed in most other decision sciences; and
these very large uncertainties have the effect of
making current economic analyses almost useless.
That is, if there are order of magnitude or greater
uncertainties encompassing the sign of the effect,
then economic costs of benefits cannot easily be es-
timated. One recent study of the environmental ef-
fects of stratospheric flight found that both a posi-
tive and negative temperature change was possible
(with almost an equal subjective probability of
each), which was estimated, by querying various
scientists [National Academy of Sciences (1975)].
For many important environmental decisions, the
type and magnitude of effects will not be known at
the time the decision is made or contemplated.
Thus, it appears that environmental decisionmaking
will contain a fundamentally normative base; that
is, decisions will have to be made on the basis of
incomplete and inadequate information on effects
and causation. In addition, the number and types of
environmental pollutants that will affect individuals
outside a given country is increasing. Examples in-
clude atmospheric changes induced by C02, strato-
spheric pollution, inadvertent weather modifica-
tion, oceanic deterioration, and problems with the
electromagnetic spectrum. A concerted effort is
needed to design new methods of valuation and per-
haps new institutions to utilize them where a com-
mon ground is reached, both in evaluating the scien-
tific uncertainties and in valuing potential effects.
REFERENCES
Ahlquist, N. C. and R. J. Charleson, "Measurement of the
Wavelength Dependence of Atmospheric Extinction Due to
Scatter," Atmospheric Environment 3 (1969), pp. 551-64.
Arrow, K. J., and A. C. Fisher, "Environmental Preservation,
Uncertainty, and Irreversibly," Quarterly Journal of Econom-
ics, (1974).
Brookshire, D. S., B. C. Ives and W. D. Schulze, "The Valu-
ation of Aesthetic Preferences," Journal of Environmental Ec-
onomics and Management Vol. 3, No. 4 (1976), pp. 325-346.
Charleson, R. J. and N. C. Ahlquist, "Brown Haze: NOj or
Aerosol," Atmospheric Environment 3 (1969), pp. 653-56.
d'Arge, R. C., "Environmental Policy Costs: Definition, Mea-
surement, and Conjecture," Cost Benefit Analysis and Water
Pollution Policy, H. M. Peskin and E. P. Seskin (eds.), The
Urban Institute, Washington, D.C. (1975).
d'Arge, R. C., "Economic Valuation of Visibility: The Four
Corners Experiment," prepared for the Resource Economics
Workshop, University of Chicago (May 4, 1977).
d'Arge, R. C., et. al., "Economic and Social Measures of Bio-
logical and Climatic Change," Vol. 6, Climatic Impact Assess-
ment Program, Department of Transportation, Washington,
D.C, (1974).
d'Arge, R. C., et. al., "Valuation of Aesthetic Preferences: A
Case Study of the Economic Value of Visibility," preliminary
draft prepared for the Electric Power Research Institute, Palo
Alto, California (September 1977).
David, Robert K., "Recreation Planning as an Economic Prob-
137
-------
lem," Natural Resources Journal, Vol. 3, No. 2, (October
J 963), pp. 239-49.
Ettinger, H. J. and G. W. Royer, "Visibility and Mass Concen-
tration in a Nonurban Environment." Air Pollution Control
Association Journal 22 (1972), pp. 108-11.
Friedman, M. and L. J. Savage, "The Utility Analysis of
Choices Involving Risk," The Journal of Political Economy
(1948).
Grobecker, A. J., el. al.. The Effevts of Stratospheric Pollution
by Aircraft, Report of Findings, Climatic Impact Assessment
Program, Department of Transportation, Washington, D.C.
(1974).
Horvath, H. and K, E. Noll, "The Relationship Between Atmo-
spheric Light Scattering and Visibility," Atmospheric Envi-
ronment 3 (1969), pp. 543-52.
Horvath, H. and R. J. Charleson, Journal of the American In-
dustrial Hygiene Association (1969).
Kneese, A. V. and R. C. d'Arge, "Pervasive External Costs and
the Response of Society," in Joint Economic Committee, U.S.
Congress, The Analysis and Evaluation of Public Expenditure:
PPB System, Vol. 1,91st Congress, 1st Session (1969), pp. 47-
64.
Lave, L., "Economic Comparisons of Nuclear and Coal Fired
Electric Plants," American Economic Association Meetings,
Dallas, Texas (Dec. 27, 1975).
Lave, L. and E. P. Seskin, "Air Pollution and Human Health,"
Science. Vol. 169, No. 3947 (Aug. 21, 1970), pp. 723-733.
Mishan, K. J., Cost-Benefit Analysis. London: Allen and Unwin
(1971).
National Academy of Sciences, /.mironrnt-ndif /mjxu I n) Slrtilo-
spheric flight: Biological and ( limattc I ff11 !\ of Aircraft
Emissions in the Stratosphere, Climatic Impact Committee
(1975).
" Non-fluorinated Halomethanex in the Environment." Panel on
Low Molecular Weight Halogenated Hydrocarbons of the
Coordinating Committee for Scientific and Technical Assess-
ments of Environmental Pollutants, National Academy of Sci-
ences, Washington, D.C. (1977).
Pratt, John W., "Risk Aversion in the Small and in the Large,"
Econometricu (1964), pp. 122-136, Vol. 32:1-2.
Randall, A., B. Ives, and H. Eastman, "Bidding Games for Valu-
ation of Aesthetic Environmental Improvements," Journal of
Environmental Economics and Manuxement 1 (1974), pp. 132-
149.
Rice, D. P., "Estimating the Cost of Illness," Public Health
Service Publication No. 947-6, (1966).
Schelling, T. C., "The Life You Save May Be Your Own,"
Problems in Public Expenditures, S. B. Chase, Jr. (ed.),
Washington, D.C.: The Brookings Institution (1965).
Thaler, R. and S. Rosen, "The Value of Saving a Life: Evidence
form the Labor Market," Household Production and Con-
sumption, Studies in Income and Wealth. Vol. 40, N. E. Ter-
teckgj (ed.), New York: National Bureau of Economic Re-
search (1975).
138
-------
SOCIOECONOMIC BASIS FOR THE DEVELOPMENT OF INDUSTRY
WITH REGARD FOR ENVIRONMENTAL PROTECTION
[Sotsial'no ekonomicheskoye obosnovaniye razvitiya prozvodstva s uchyetom
okhrany okruzhayushchey sredy]
M. YA. LEMESHEV, K. G. GOFMAN, A. A. GUSEV
I. Correlation of Socioecological and Economic
Criteria in Future Planning of Environmental
Protection
When implementing environmental protection,
society must choose between satisfying the demand
for material goods created by industry and satis-
fying the demands for a clean environment. In our
opinion, one of the promising ways to solve this
problem is to implement environmental conditions
which are permissible with respect to socioecolog-
ical criteria by the end of a stipulated period. Eco-
nomically optimal values of environmental parame-
ters must subsequently be selected.
In any environmental protection plan, there must
be no contamination concentrations which would
impair human health or the development of ecologi-
cal systems. Indices must be formulated for the
maximum critical concentrations (MCC) and maxi-
mum ecological loads on the ecosystems (MPEL)
[ 1] in order to meet the socioecological environmen-
tal requirements. These indices characterize the
minimum permissible level of environmental pro-
tection, in contrast to maximum permissible con-
centrations (MPC) characterizing future optimal
levels of environmental protection.
The system of hygienic environmental standard-
ization in the USSR is based on permissible con-
tamination indices in the form of maximum per-
missible concentrations (MPC) of environmental
contamination. These indices indicate the maxi-
mum requirements for environmental protection in
terms of human health. The environmental con-
tamination MPC is that concentration which has no
unfavorable effect on human health, efficiency of
the population, or future generations which can be
detected by current research methods. Neither does
this concentration "impair" the hygienic living con-
ditions of the population [1]. The MPC of a con-
taminant in the environment is that concentration
limit at which there is still "no unfavorable effect
that can be detected by current research methods
on the state of health and efficacy of the popula-
tion or on future generations of people, and there is
no impairment of hygienic living conditions of the
population" [2].
The need for such standards to control the quality
of the environment is indisputable. If one examines
the MPC standards from the viewpoint of socioeco-
nomics, then one cannot help noticing their similar-
ity to a number of other social standards for the sat-
isfaction of demands — scientifically based dietary
norms, standards for providing the population with
living space, durable goods, public health services,
etc. These standards, just as MPC standards, are
based on the need to create the best conditions for
human vital activity, as well as to guarantee human
health. In this sense, MPC standards and the other
standards mentioned above for providing the popu-
lation with important life goods, are long-term opti-
mal standards for the quality of human life, not min-
imum permissible standards such as minimum
wage, minimum pension, etc.
It is important here to stress, however, that prior-
itizing toward the most rapid achievement of hy-
gienic norms for environmental purity (MPC) as
compared, for example, to most rapid achievement
of housing standards, would not necessarily be
more advantageous from the standpoint of public
health. Thus, the MPC should, in our opinion, be
viewed as the long-term standard for environmental
quality only for actually contaminated areas — that
is, such areas where contamination at the start of
the planned time period already exceeds the MPC
level.
It would be erroneous to underestimate the diffi-
culties of standardizing the absolutely inadmissible
levels of contamination (exceeding the indices of
MPC and MPEL). However, in our view, these dif-
ficulties do not justify identifying MPC as the gener-
al standard for long-term level of environmental pu-
rity, with MCC and MPEL indicating maximum
permissible contamination concentrations in con-
taminated regions. The formulation of standards
like MCC and MPEL, just like standards of MPC
139
-------
for the plant and animal world (currently, MPC is
generally determined from the results of research
on the action of contaminants on man) is, in our
opinion, one of the major tasks in setting up overall
environmental quality standards. All standards of
this type are by nature temporary, since hygienic
standardization of contamination is extremely com-
plicated, and it would be incorrect to assume that
the current level of our knowledge as reflected in
active MPC standards is absolute.
A socioecological limitation on long-term plan-
ning for permissible contamination levels is that
there must be no greater danger of contamination at
the end of the planned period as at the beginning.
From the USSR classification for substances ac-
cording to the probability of their adverse effect on
the human organism, it is possible to make a sum-
mary hygienic evaluation of a given level of con-
taminant by computing the index of its danger [3J.
For specially protected areas (preserves, recrea-
tional zones), it is expedient to establish stricter re-
quirements so that the concentration of hazardous
substances in the environment will not exceed that
achieved at the start of the planned period.
The planning for environmental control, then, di-
rectly depends on arriving at levels that will result
in neither impairment of the ecological balance over
the long-term nor contaminant concentrations that
are irreversible because of their medical and biolog-
ical consequences. Within these limitations, it is
possible to isolate from the numerous possible envi-
ronmental conditions during the planned period a
subset of contamination conditions permissible by
socioecological criteria. All the states of the envi-
ronment that form the subset can be viewed as per-
missible, though not as optimal, by socioecological
criteria in so far as they stay within those two limi-
tations.
To validate the strategy for transition from the
initial baseline level of contamination in the envi-
ronment to its long-term values, it is necessary lo
use an economic criterion of effective reduction in
contamination and the concept of economic opti-
mum of contamination which is related to it.
Modern scales of economy and the nature of con-
tamination effects on the efficacy of social produc-
tion make it necessary to include in production cost
projections a special component that can be called
costs of contamination [4]. We will refer to the pa-
rameters that characterize production waste vol-
umes as physicochemical properties, and we will re-
fer to the conditions for discharging wastes into the
environment as the vector of technogenic dis-
charges. Then, with each vector of technogenic dis-
charge, we can estimate a certain cost of con-
tamination — national economic outlays governed
by the given vector of technogenic discharge. These
expenses would include: 1) outlays necessary for
reducing discharges entering the environment to a
level corresponding to the assigned value of the
vector of technogenic discharges (costs of pre-
vention); 2) outlays to compensate for negative so-
cial consequences of discharges entering the en-
vironment according to the accepted value of
the vector of technogenic discharges (costs of
compensation); and 3) outlays for replacement of
losses in raw materials and products with exhaust
and waste water (raw material losses)'. These com-
ponents of contamination costs are interdependent,
and there is an inverse relationship between the
costs of prevention and the costs of compensation.
The same type of relationship also exists between
the prevention costs and the raw material losses,
but here, the decrease in raw material losses is
reached only with conversion to closed (low-waste)
technological processes. It follows that a decrease
in one of the components of contamination costs
does not always indicate a reduction in the con-
tamination costs as a whole or decrease in social
outlays for the national economy (including also the
total amount of contamination costs).
Uncontaminated environment is no longer freely
available. Environmental purity must now be
viewed as a goal that requires specific — and often
great — outlays to reach. At the same time, envi-
ronmental purity as "utility" for social consump-
tion differs from other utilities. Usually the appro-
priation by an individual (or group) of a given utility
excludes the possibility of its appropriation by other
individuals (groups). Purity of the environment is an
indivisible and collectively consumable good. In
contrast to "traditional" types of natural resources,
it can be monopolized as a subject of management
only by society as a whole, and not by individual
nature-utilizing enterprises. Therefore, only society
as a whole (in the person of its planning organs) can
guarantee the calculation and compensation of con-
tamination costs by their "offenders" — that is, the
enterprises that generate the technogenic dis-
charges.
"Economic optimum of contamination" refers to
that level of environmental contamination at which
the minimum regional contamination costs are
reached (or the minimum regional summary produc-
tion costs, including the total amount of con-
tamination costs). The economic optimum of con-
tamination occurs when the rise in costs for pre-
venting contamination with a fairly small reduction
in level becomes equal to the decrease in com-
pensation costs and raw material losses which
thereby occurs — that is, when maximum costs of
preventing contamination are equal to the maxi-
mum costs of compensating for the social con-
1 Costs of compensation and raw material losses are often
called the economic damage from contamination.
140
-------
sequences of contamination and raw material loss-
es. The reaching of an economic optimum of con-
tamination can be considered a task for the first
stage of long-range planning for environmental pu-
rity.
The economic optimum for environmental con-
tamination must be determined not for all possible
levels of contamination, but for the contamination
levels permissible by socioecoiogical criteria. Cal-
culation of the corresponding limitations in planning
and controlling environmental purity makes it pos-
sible to evaluate them economically, determine
what the pre-eminence of socioecoiogical criteria in
long-term planning of environmental purity will cost
society. Here, in many cases, the socioecoiogical
limitations can turn out to be excessive; that is, they
do not affect the economic optimum of contami-
nation. In these cases, one can speak of the con-
sistency of socioecoiogical and economic criteria
for environmental purity.
After the economic optimum level has been
reached, the next task on the agenda is second stage
long-term planning transition from the economic
optimum of contamination to its social optimum,
that level of contamination at which universal ob-
servance of MPC is guaranteed for all the standard-
ized ingredients with regard for the cumulative ef-
fect of their harmfulness.
As indicated previously, long-range standards for
environmental purity are socially and economically
close to standards that characterize the scien-
tifically substantiated (creating the best conditions
for normal vital activity) levels for providing people
with the main material necessities — food products,
housing, clothing, etc. Therefore, a natural criterion
for distribution of resources among the types of ac-
tivity necessary for reaching scientifically sub-
stantiated standards for the quality of human life
(production of various types of consumer goods and
"anticontaminating" measures) at the stage of tran-
sition from the economic to the social optimum of
contamination is, in our opinion, the minimization
of a "time gap" between the reaching of these stan-
dards. That is the most desirable situation would be
simultaneous "emergence" to the scientifically sub-
stantiated norms for feeding the population, provid-
ing housing, etc., and reaching the standards for en-
vironmental purity. In other words, after the mini-
mum requirements (dictated by socioecoiogical
criteria) for environmental purity have been met
and the possibilities have been exhausted for in-
creasing the effectiveness of social production by
reducing environmental contamination (that is, the
economic optimum of contamination has been
reached), policy for controlling contamination must
be oriented towards reaching roughly the same de-
gree of approximation to the scientifically sub-
stantiated standards for the entire spectrum of basic
indices that characterize the quality of human life.
Such is, in concise form, one of the possible ap-
proaches to an agreement of socioecoiogical and ec-
onomic criteria in the formation of a strategy for
controlling environmental purity. The practical ap-
plication of this approach requires an essential in-
crease in the scientific validity of standards for envi-
ronmental purity as well as all other standards for
the quality of human life, and therefore it is hardly
possible in a complete volume in the near future.
However, methodological requirements that
emerge from the proposed approach and that con-
cern the validation of long-term environmental pro-
tection plans with regard to their socioecoiogical
admissibility and economic attractiveness are of
practical importance.
II. Planning of Agricultural Production with Regard
for Protecting the Environment from
Contamination with Pesticides
The use of pesticides protects agricultural crops,
but it has a number of undesirable consequences for
the environment. Among the possible environmental
protection measures, the most important is replace-
ment of stable and toxic pesticides with less stable
and less toxic ones. Other measures — for example,
complete abandonment of pesticide use and a tran-
sition to biological methods of protecting plants —
are currently not justified since, despite their poten-
tial, they do not yet produce the desired effect.
The group of toxic pesticides' includes granazon,
octomethyl, phosphamide, metaphos, dichloroeth-
ane, and others. The less toxic group includes car-
baphos, carbyne, nitrophene, celtan, cotoran, pro-
metrine, and others. The use of low-toxic pesticides
as compared to toxic ones is more expensive be-
cause crops have to be treated more often, the prep-
arations cost more, and the effectiveness is lower.
At present, determinations of the economic efficacy
of replacing toxic pesticides by less toxic ones have
underestimated the effect of the latter. This is due to
the fact that economic damage from environmental
contamination by toxic pesticides is not considered.
It is not presently possible to quantitatively deter-
mine the economic environmental damage by pesti-
cides. However, even low estimates of economic
damage, considering only urban population moribi-
dity, emphasize the economic expediency of replac-
ing toxic pesticides with less toxic ones. This
proved to be the case for example at several Uzbek
SSR farms [5].
Potential economic damage from pesticide con-
tamination was considered in the optimization task
for specialization and combination of the agricul-
ture branches in the Uzbek SSR, where a selection
'For brevity, we will call the group of stable and toxic pesti-
cides toxic, and the group of less stable and less toxic pesticides,
low-toxic.
141
-------
structure for pesticides was designed.1 First, on the
basis of generalized medical studies, the dynamics
were determined for the economic damage from in-
creased moribidity of the urban population in rela-
tion to pesticides use. Analysis of economic damage
took into consideration additional expenditures for
medical services, losses of pure profit in agriculture
due to losses of working time in relation to moribi-
dity of the working population, and additional out-
lays for sick payments. Then, on the basis of eco-
nomic damage data for a number of years, the fol-
lowing regression equation was constructed:
y = a + b,X! - b2x2
where: y — economic damage, rubles; x, — volume
of use of toxic pesticides, kg; x2 — volume of use of
low-toxic pesticides, kg.
Numerically, the equation is
y = 31.859 + 4.972 x, - 3.155 x2 (1)
It should be noted that the negative sign in the
regression coefficient with x2 was obtained on the
basis of computerized statistical information that
characterizes the effectiveness of using low-toxic
chemicals from the viewpoint of reducing damage
from pesticide use.
The coefficient of the multiple correlation of
equation (1) is R = 0.824. The calculated value F —
criterion equal to 8.65 with tabular 4.67 (for 5% er-
ror probability), that is F calculated > F tabular.
The calculated values t — criterion for regression
coefficients (a, b,, b2) are the following: ta = 10.465;
tb, = 3.073; tb2 = 3.394. The tabular value of t —
criterion for the given set with 5% error probability
equals 2.16; that is, all the coefficients of regression
are statistically significant.
Thus, according to the statistical characteristics
cited above, the equation for the relationship be-
tween economic damage and increased rural popu-
lation morbidity with given volumes of pesticides
has a linear appearance. There are published de-
scriptions of regression equations for the relation-
ship of increased morbidity and concentrations of
air contaminants from industrial and motor trans-
port discharges. In reference 6, a nonlinear relation-
ship is shown between morbidity and concentra-
tions. In our case, the relationship appears to be lin-
ear; this could be the result of averaging data from
medical examinations from all over the republic.
And this, on the other hand, is very important from
the viewpoint of solving the optimization task.
Of distinct importance also is the formulation of
the function for the dependence of damage on pesti-
cide use in relative amounts. The corresponding re-
'T. I. Iskandarov (Tashkent Medical Institute) and T. D. Dos-
chanov (Institute of Economics of the USSR Academy of Sci-
ences also participated in the solution of the given problem.
gression equation has the following form:
y = -a + bx
where: y — economic damage in '7c in relation to its
quantity for the reference year; x — portion of toxic-
pesticides in the total volume of chemicals used, in
%; a, b — parameters of the regression equation.
Numerically, this equation has the following ap-
pearance:
y = -121.238 + 1.627 x (2)
The correlation coefficient in equation (2) equals
0.797; ta = -7.296; tr, = 4.375, which is higher than
the tabular value of t — criterion equal to 2.14 with
5% error probability. The parameter b character-
izes the elasticity coefficient for the replacement of
toxic pesticides by low-toxic ones; that is, it shows
by what percent the damage will be changed when
there is a 1% alteration in the fraction of toxic pesti-
cides in the total volume of pesticides employed. It
follows from (2) that, with an increase in the frac-
tion of toxic pesticides by 1%, the economic dam-
age is increased by 1.6%.
Since in equation (1) the regression coefficients
are statistically significant, it is accurate to view
them as standards in the medium-urgent planning of
agriculture in the republic. These standards reflect
economic damage due to the use of toxic pesticides.
The dependence of economic damage on increased
morbidity of the rural population due to the use of
pesticides was included in the optimization task
which is presented below. Since such a task does
not consider the arrangement of branches within the
republic, it belongs to structural tasks.
In the optimization task for each agricultural
agency, two possible variants are examined that re-
flect the varying structure of pesticide use. As will
be shown further, the optimal plan for each agency
can include either one variant or the other, or a
combination of the two. The first use variant is that
proposed by agricultural specialists, and the sec-
ond, that proposed by medical specialists. Agricul-
turists suggest a use in which toxic pesticides are of
relatively great importance as compared to the use
structure proposed by medical specialists. The pro-
posal of agricultural workers is explained by the
fact that toxic pesticides are less expensive and
somewhat more effective in plant protection — that
is, they are more suitable from a purely economic
standpoint without regard for economic damage.
The medical proposal was based on moribidity in-
dices in which economic parameters were not con-
sidered.
Because the task before the republic involves not
only crop raising, but also animal husbandry, two
possible variants were adopted for a herd of cattle
as technological production variations.
142
-------
Now we will examine the optimization model.
We will introduce the following notation and
classifications: i and 1 — number and quantity of
indices for the agricultural plan considered in the
economic and mathematical model; j and J — num-
ber and quantity of possible variants for fulfilling
the plan according to the i-th index (in our case J| =
2); k — number of production resources; K, — sub-
set of types of resources without pesticides (land,
water, labor, etc.); K2 — subset of types of re-
sources of pesticides (K2 = 2 — groups of toxic and
low-toxic pesticides); 1 and L, — number and quan-
tity of aggregated types of products; xu — unknown
numerical values of i-th indices in the plan with j-th
variants of their implementation (dimensions of i-th
branches of plant growing and animal husbandry
measured in ha, head of cattle, etc.; i-th indices for
supplementing production resources, for example,
purchased feed, etc.); y — unknown numerical val-
ue of economic damage; ct — costs for realization
for i-th products (for commercial products c( > 0,
for the other i indices, including for non-commercial
(intermediate) products c( = 0); Su — outlays per
unit of numerical value of i-th index with j-th meth-
od; Uij— productivity of i-th branch with j-th meth-
od (crop yield, annual milk yield, etc.); ak — re-
sources of k-th production factors; aku—outlays of
k-th factor per unit of numerical value of i-th index
with j-th method; bku — output of k-th resource
from unit for the numerical value of i-th index with
j-th method (for example, the output of organic fer-
tilizer for plant growing from unit of branch of ani-
mal husbandry); xk — unknown numerical values of
k-th type of pesticides; dmU — coefficient of con-
version of i-th production produced by the j-th
method into aggregated m-th product; fL and hL —
lower and upper borders for 1-th additional limita-
tions; giU — coefficient of 1-th additional limitations
(coefficients of correlations between groups of
crops or individual crops, between mature groups in
animal husbandry, if they are isolated in the model,
etc.); bk — regression coefficients which character-
ize the effect of each group of pesticides on the eco-
nomic damage from increased morbidity of the ur-
ban population: — free term of regression equation
characterizing the effect of unconsidered factors on
morbidity; bm — assignment of the state plan for
purchase of the m-th product; A — set of assigned
classifications.
The mathematical notation for the examined
model is as follows (model 1):
1) the maximum of pure profit in agriculture mi-
nus the economic damage from increased morbidity
of the urban population due to the use of pesticides
S = max S(tt)
0 and \a > 0). On the basis of the optimal
solution in this case, additional calculations were
made to determine the weighted average of crop
yield, the percentage structure of pesticides, and
the specific expenditures for the appropriate
agencies. The weighted average of crop yield and
the expenditures were determined as follows:
- _ «llXll + «12Xi2
«1 1
Xu + Xl2
where: au, oi2 — yield or specific outlays, respec-
tively, for first and second variants for each exam-
ined agency; Xu, xw — dimensions of i-th branches;
di — corresponding weighted average indices.
In so far as the percentage structure of the pesti-
cides according to variants was designated as the
only one according to crop growing agencies, then
in the given case, the weighted average percentage
of low-toxic pesticides in the total structure of their
use is:
143
-------
* PiXil + &X12
x„ + x«
where: j3j, /32 — percentage of low-toxic pesticides
in their total structure for the first and second vari-
ants, respectively; /3j — weighted average percent-
age of low-toxic pesticides for each examined agen-
cy.
In the optimal plan there was a total of 50.19% of
sowing area from the total resource in the republic,
for which Xu > 0, Xi2 > 0. For those agricultural
agencies in which xn > 0 and X|2 = 0, the sowing
areas were 22.34%, and for those cases where Xi, =
0 and Xb > 0, they were 27.47%.
The findings indicate the importance of calcu-
lating economic damage as planning solutions are
adopted for the development of the agriculture in
the republic.1 Thus, for 27.47% of the sowing areas
treated with pesticides, the structure of their use
corresponds to the "medical" variant, while for
50.19%, the structure corresponds to the "weighted
average" variants for individual crops. This is be-
cause, with the limited resources and conditions of
mandatory fulfillment of the state assignments for
purchases of agricultural products, there is no pos-
sibility to apply across the board the "medical"
variant for employment of pesticides. And only for
22.34% of the sowing areas, the structure of pesti-
cides use corresponds to the "agricultural" variant.
Table 1 presents the results of calculation for the
optimal structure of sowing areas according to mod-
el 1, solved according to socioeconomic criteria. If
all the area treated by pesticides is distributed for
crops, where xu > 0 and xu > 0 separately accord-
ing to the "medical" and "agricultural" variants,
then, as is apparent from the data of table 1, 57.22%
of the sowing areas are treated with pesticides ac-
cording to the "medical" variant, and 42.78%, ac-
cording to the "agricultural."
To determine the economic effectiveness of envi-
ronmental protection measures, solutions are
adopted for the development of agriculture in the
republic which consider economic damage from in-
creased morbidity of the rural population due to the
use of pesticides. It is necessary to re-solve the
problem without considering the index for econom-
ic damage in the maximizing function (3). In other
words, with the same limitations the problem is
solved for the maximum of pure profit, that is ac-
cording to the production criterion:
S*(rr) = £ (CiUu - Su)Xij -* max (5)
i.)
From the viewpoint of the socioeconomic crite-
'In the future we will call the first variant "agricultural" and
the second — "medical."
rion (3), the solution of the new problem for produc-
tion criterion (5) with the same system of limitations
is a permissible solution for the optimal socioeco-
nomic problem of model I.
We will designate the permissible solution to the
problem according to criterion (5| by xtJ, y, and i.k.
Since [51, does not include the parameter of damage
and does not affect the solution, then in the per-
missible solution, there is a simple calculation of the
quantity y according to condition [4|, depending on
the values of xk obtained in this solution.
In order to compare the socioeconomic indices in
the optimal and permissible solutions, we designate
S* = max S*(w) and subtract from S* the computed
value y, this is, the pure profit minus damage S' in
the permissible solution is: S' = S* - y. On the
basis of the possible differences in the optimal and
permissible solutions of the problem according to
the socioeconomic criterion, in general the correla-
tion S^S' must be observed.
Table 2 presents certain results of the permissible
solution of the socioeconomic problem which differ
from the optimal (see table 1). Thus, in the per-
missible solution, 65.94% of the areas treated with
pesticides belong to the "agricultural" variant and
only 34.06% to the "medical," while in the optimal
solution, the structure of areas treated with pesti-
cides is 42.78 and 57.22% respectively. This oc-
curred because the production criterion lacks the
parameter of economic damage that affects the per-
missible solution, due to which a "shift" occurred
in it to the side of the "agricultural" variant.
It is apparent from table 3 that the use of a so-
cioeconomic criterion for optimization makes it
possible to obtain a solution for the development of
the agriculture of the republic with the optimal com-
bination of the aforementioned opposing factors. If,
in the permissible solution for production criterion,
the pure profit is higher (S* > S"), then the econom-
ic damage also is greater (£ > y). At the same time,
the value of the socioeconomic index (pure profit
minus economic damage) which considers both the
factor of growth in pure profit and the factor of pro-
tecting the rural population from adverse con-
sequences of the use of pesticides was higher in the
problem solved by the socioeconomic criterion of
optimization (S > S').
'Strictly speaking, in model I both the favorable and adverse
effects of the use of pesticides in agriculture are considered. On
the one hand, the "agricultural" variant for the use of pesticide*
affects to a greater degree the drop in specific outlays for produc-
tion, and increase in output of agricultural products necessary to
the society, and a rise in the pure profit than the "medical." On
the other hand, the "agricultural" variant to a greater degree
affects the increase in economic damage. Thus, the optimal solu-
tion for socioeconomic criterion (maximum of pure profit minus
economic damage) is formed with regard for these opposing fac-
tors.
144
-------
TABLE 1
OPTIMAL STRUCTURE OF SOWING AREAS OF AGRICULTURAL CROPS IN THE REPUBLIC ACCORDING TO
THE RESULTS OF THE SOLUTION TO THE PROBLEM WITH REGARD FOR ECONOMIC DAMAGE FROM THE
USE OF PESTICIDES (IN %)
including
Sowing areas
I variant
II variant
Agricultural crops
under cultivation
("agricultural")
("medical")
Grains, total
28.56
16.81
11.75
including wheat
10.25
10.25
—
barley
10.03
—
10.03
rice
3.94
2.22
1.72
corn
4.34
4.34
—
Industrial, total
47.21
18.21
29.00
including cotton
46.25
18.21
28.04
ambary
0.55
—
0.55
tobacco
0.41
—
0.41
Potatoes, vegetable-
cucurbitaceous, total
6.68
—
6.68
including vegetables
2.89
—
2.89
cucurbits
2.42
—
2.42
potatoes
2.37
—
1.37
Fodder, total
17.55
7.76
9.79
including lucerne for hay
8.31
—
8.31
lucerne for grain
3.52
3.52
—
fodder
corn for silage
2.71
2.71
—
corn for grain
1.53
1.53
—
fodder
joughara
0.65
—
0.65
sugar beets
0.83
—
0.83
Total sowing area
100.00
42.78
57.22
TABLE 2
STRUCTURE OF SOWING AREAS OF AGRICULTURAL CROPS IN THE REPUBLIC AS A RESULT OF SOLVING
THE PROBLEM ON THE PRODUCTION CRITERION (IN %)
including
Sowing areas
I variant
II variant
Agricultural crops
under cultivation
("agricultural")
("medical")
Grains, total
25.48
6.46
19.02
barley
8.83
—
8.83
rice
3.94
3.94
—
corn
2.52
2.52
0.41
Industrial, total
48.26
17.85
including cotton
47.31
47.31
—
ambary
0.55
0.55
—
tobacco
0.41
—
0.41
Potatoes, vegetable-
6.24
cucurbitaceous, total
7.61
1.37
including vegetables
3.85
—
3.85
cucurbits
2.39
—
2.39
potatoes
1.37
1.37
—
Fodder, total
18.65
10.25
8.40
including lucerne for
hay
5.87
5.87
—
lucerne for
grain fodder
3.29
3.29
—
corn for
silage
4.09
—
4.09
corn for
grain fodder
4.31
—
4.31
joughara
0.23
0.23
—
sugar beets
0.86
0.86
—
Total sowing area
100.00
65.94
34.06
145
-------
TABLE 3
COMPARATIVE SOCIOECONOMIC AND ECONOMIC RESULTS OF OPTIMAL AND PERMISSIBLK SOLUTIONS
FOR THE SELECTION OF THE STRUCTURE OF USE OF PESTICIDES IN THE AGRICULTURE OF THE UZBEK
SSR (IN GENERAL VIEW)
No. in
order
Indices
Economic damage
from increased mor-
bidity of rural popu-
lation due to use of
pesticides
Pure profit in
agriculture
Pure profit minus
economic damage
Optimal
solution
S"
S
Permissible
solution
S*
S'
Correlations of results
in optimal and per-
missible solutions
S" < S'
S > S'
REFERENCES
1. Izrayel', Yu. A., "Permissible Anthropogenic Load on the
Environment," in Vsestoronnyy analiz okruzhayushchey pri~
rodnoy sredy. Trudy II sovetsko-amerikanskogo simpozium
["Comprehensive Analysis of Environment. Proceedings of
Second Soviet-American Symposium"), Leningrad: Gidro-
meteoizdat, 1976.
2. Sidorenko, G. I., Ye. I, Korenevskaya, M. A. Pinigin, G. N.
Krasovskiy, "Hygienic Bases for Environmental Pro-
tection," in Vsestoroniy analiz okruzhayushchey prirodnoy
sredy. Trudy sovetsko-amerikanskogo simpozium, Lenin-
grad: Gidrometeoizdat, 1975.
3. Pinigin, M. A., R. M. Barkhudarov, I. K. Dibobes, "Quan-
titative Criterion for Summary Evaluation of Contamination
of Atmospheric Air," in Trudov institute! obshchey i kommun-
al'noy gigiyeny im. A. N. Sysina AMN SSSR ["Proceedings
of A. N. Sysin Institute of General and Municipal Hygiene of
the USSR Academy of Medical Sciences"), ed. 3, 1976.
4. Gofman, K. G., "Costs of Contamination of the Environment
as an Element of Socially Necessary Expenditures," in
Ekonomiko-matematicheskiye issledovaniya zatrat i re-
zul'tatov ["Economic and Mathematical Studies of Outlays
and Results"), Moscow: Nauka, 1976.
5. Gusev, A. A. and T. D. Doschanov. "Determination of Com-
parative Effectiveness of the Use of Toxic Chemicals with
Regard for the Consequences of Contamination of the Envi-
ronment," in Pervaya shkola "Melody planirovaniya i up-
ravleniya prirodnymi resursami" ["First School 'Methods of
Planning and Controlling Natural Resources' "), Moscow:
TsEMI AN SSSR, 1975.
6. Mel'nik, L. G., "Questions of Predicting Damage to the Na-
tional Economy from Contamination of the Atmosphere for a
Long Period," in Ekonomicheskaya otsenka i ratsionainoye
ispol'zovaniye prirodnykh resursov (tezisy dokladovl ["Eco-
nomic Evaluation and Rational Use of Natural Resources
(Summaries of Reports)"), Moscow: TsEMI AN SSSR, 1973.
146
-------
BASES FOR MATHEMATICAL MODELING OF THE MOISTURE,
HEAT, AND SALT TRANSPORT PROCESSES IN SOILS TO
DETERMINE POLLUTION
[Osnovy matematicheskogo modelirovaniya protesessov perenosa vlagi, tepla i
soley v pochvo-gruntakh dlya otsenki zagryazneniya]
I. G. GRINGOF and YU. M. DENISOV
Problems of protecting soils from pollution and
salinization arise in irrigated fanning using ferti-
lizers and insecticides. These problems are also re-
lated to the contamination of underground waters
and sewer-drainage discharge. The problems may
be solved by developing mathematical models for
the transport of moisture, heat, and salts in soils.
As is known, soils consist of a skeleton, the roots
of plants, ice, water, and moist air. We shall give
the characteristics of these phases the phase index
i, which has the following values: for the soil skele-
ton — 1; plant roots — 2; ice — 3; water — 4, and
moist air — 5.
Within the soil we shall distinguish the elemen-
tary volume A V and designate the volume of the i-th
phase included in AV by AV,. Then:
AV = £ AVi
(1)
We shall divide equation (1) by AV, and we shall
designate (AVi)/(AV) by at — the relative volume of
the 8-th phase:
ai = 1 0 s a, s 1
(2)
The interaction of the phases with each other is
accomplished through their interfaces, which there-
fore are fundamental characteristics of a multi-
phase medium.
Let us use ASu to designate the area of the mutual
contact surface between the i-th and j-th phase in a
separate volume AV. The ratio of ASU to AV is the
specific surface of mutual contact /3U. Let us set:
and
fl =
P" ay
A, = Pu
(3)
The total specific surface of the i-th phase /3,
equals:
fit = 2 P»
(4)
Let us also introduce the important characteristic of
a multiphase system which represents the charac-
teristic linear dimension of the phase 8t — the ana-
log of its thickness (or the hydraulic radius). We
may express this dimension by the relative volume
of the phase and its total specific surface:
8, = K
£1.
A
(5)
where K is a dimensionless parameter which de-
pends on the geometric form of the phase. K ranges
from 2 to 3, and the low value of K refers to the case
when the pores have a shape which is almost cy-
lindrical and the high value pertains to the case
when the pores are almost spherical [3],
It may be shown that there are comparatively
simple expressions for the specific surfaces of ice,
water, and air, depending on their relative volumes
for given specific surfaces of the skeleton and the
roots, the specific surface of their mutual contact,
and their relative volumes:
Pt ~ (Pi + Pt P12)
!-(<*,+ at)
k - 1
K
~ Pi
If there is water and ice in the soil, then:
P43 ~ (fit + P2 P12)
The sp
equals
1
as
K - I
K
(6)
(7)
(8)
1 - (a, + aiii).
The specific surface of the skeleton /SK by definition
Pv. = (Pi + Pt ~ Pu)
(9)
147
-------
The total specific surface of water (j34) is:'
fi* = (fll fit ~ fiu) 1 ~~
«3
K - 1
K
1 - («j + a2)
1 +
(10)
The specific surface of ice /33 is:
1 -
«4
1 — (a1 + a2 + aa)
K - I
/^3 ~ (fii + @2 Pit)
1 +
1 -
"3
K - t)
K
1 - (a, + a2)
(11)
Formula (10) is valid when ay > 0, and formula (11)
is valid when a3 > 0. At ay = 0 we must set /3y = 0,
and at a3 = 0 we must set /33 = 0. This is true be-
cause these phases have, for two interfaces (double
layer) and at —~ 0, interfaces which do not vanish as
they approach each other.
We shall introduce into this discussion the den-
sity of the phase which we shall designate by pt (the
mass of the phase per its unit volume). The density
of the soil p in this case may be expressed by the
following equation:
P - £ «iPi
(12)
Let us use U, to designate the velocity of the i-th
phase. Then the equation of the conservation of
mass of the i-th phase will have the following form:
+ divP,«,o, = 2 <»
dt j
(13)
where is the mass of the j-th phase which changes
into the i-th phase per unit time per unit soil vol-
ume.
The following equation for the heat transport of
the i-th phase can be used:
CWi(f + U,grad T,j + £ Lueu =
= div <*iX| grad T, + X riTTTT ~ Tl) <14)
Here, C*i, T|, K, are the specific heat capacity, tem-
perature, and thermal conductivity of the i-th
phase, respectively. The quantity Lu is the specific
heat for the transfer of the i-th phase into the j-th
phase; XT is a dimensionless proportionality coeffi-
cient, which approximately equals 6.
In this equation, the quantities (Ljj, eu) are given a
"plus" sign if the phase transitions are endo-
thermic, and they are given a "minus" sign if they
are exothermic. In addition,
L„ = L2J = 0 u T3 < Tm, T< s Tm
where Tm is the temperature of the phase equilibri-
um of water and ice.
We shall use e45 to designate the mass of water
transformed into vapor per unit volume of soil per
unit time. We may also designate this quantity by
€4w, that is:
Then eM = ew4 = - c45 is the condensation of vapor
into water. In a similar manner, e35 = «3vb is the vol-
atilization of ice and eM = ew3 = -e35 is the sub-
limation of ice.
It can be shown that:
— c4«
C3s ~ €3w —
4Dw
a.
4Dw
«»
045(pHi ~ Pw)
Pasipm ~ Pw
(15)
(16)
Here, Dw is the diffusion coefficient for vapor in air,
pw is the density of vapor in air entrained in the
pores, and pw is the density of vapor saturating the
space above the i-th phase.
Phi
Phi(T„) exp
1 I
Liw
Rw
1
\_
T,
284
p4RwTi85
(17)
where T0 = 273° K, Rw is the gas constant of vapor,
845 is the specific surface energy of the air-water
boundary, and pHi(t(>) is the density of saturated va-
por over a flat surface at a temperature of Tu =
273° K.
The equation for conservation of mass of vapor
is:
dpw«5
at
+ diva5(pwU5 - Dw grad pw -
- pwae grad T5) = £ «iw
(18)
Here, as is the heat diffusion coefficient. The ab-
sorption e42 of the water in the soil by the plant roots
is, as a rule, a physiologically active process, and
takes place with consumption of free plant energy.
For the aeration zone, we have:
€« = p4 *£[PW ~ («P + P2 - P2C)] (19)
For the region where the pores are completely filled
with water:
= P* ~-[P,c + (P4C + Pz - P2C)]
(20)
In the formulas given above, /3% is the active (ab-
148
-------
sorbing) specific surface of the roots, R,, is the resis-
tance of root pocket, P,, is the absorbing pressure of
root pocket, a4 is the specific weight of water, P4 is
the absorbing potential of soil moisture (expressed
in a layer of water), P2 is the inter-root pressure, P2t
is the inter-root osmotic pressure, P4 is the pressure
of soil moisture, and P4, is the osmotic pressure of
soil moisture.
The force £2 expended by the roots located in a
unit volume of soil on absorption equals:
*42
Pi
P.<
(21)
The quantities /3£and P,«. are the physiological con-
trol parameters of the plants.
In the aeration zone, that is, when a4 < 1 - (al +
a2 + «;i), the following forces act upon water lo-
cated in the AV volume of soil:
1. The force of gravity
-<*474 grad Z • AV
2. The skeleton force caused by attraction of wa-
ter to the soil skeleton
a4 grad ^ AV
3. The meniscus force caused by the curvature of
the air-water surface of contact
-a, grad (P5 + 845 div a,5) AV
4. The force caused by osmotic pressure of the
soil moisture
n4
a4KsT4 grad £ n4S
6 - 1
5. The force of friction [ ]
AV
_ y xu
-------
The formula obtained (25) for the filtration coeffi-
cient coincides with the formula obtained in the
monograph by Perpin and Chudnovskiy [7|.
The second factor:
1 -
1 - (a, + a2)
(26)
takes into account the influence of the ice present in
the soil upon the filtration coefficient.
Finally, the third factor
relationships shows that the theory describes quite
accurately the given phenomenon. The components
under the gradient sign represent the gravitational,
skeleton (matrix), meniscus, and osmotic potential.
We shall designate the sum of the last three com-
ponents of op4 fsee formula (19)), and we shall des-
ignate the sum of the second and third components
as (for an isotonic solution). This quantity may
be written as follows:
4
Here, M, N, m are parameters characterizing the
given soil and determined on the basis of observa-
tions using optimization methods.
Figure 2 shows the theoretical (solid line) and em-
pirical curves taken from the work of Philip and Ku-
lik [14,15] (dashed line) for the dependence of lgi|<4
0.8
0.6
0.4
0.2
//
//
//
//
//
/
/
//
/
0.2
0.4
0.6
0.8
Figure 1. Dependence of the ratio of the moisture conductivity coefficient KB on the filtration coefficient K^. The solid line represents
the theoretical curve; the dashed line represents the empirical curve.
150
-------
on y,. The parameters M, N, m are located along
the empirical curves using the optimization method
developed for this purpose by Islamov. The results
of the comparison indicate a fairly good agreement
between theory and experiment.
For the region in the soils where the pores are
completely filled with water, that is, when a4 = 1 -
(a, + at + a:t), the flow of water per unit of porous
surface q4 equals:
Q4
K.
I -
1 - (a, + a2)
grad
1
H
mi'i
a48-
1
1 - (a, + a2)
Z + — P4
a 4
1 N„
"4 S a 4
Let us assume:
G(x, y, Z, t) = 0
is the interface of the aeration zone and the region
where the pores are filled with water, and NG is the
unit vector of the normal to this surface directed
toward the aeration zone. Thus on this surface, the
following boundary condition must be satisfied:
Ng = q4lG • Ng (29)
^4a'G
The interface G(x, y, z, t) = 0 is movable. The rate
of its movement along the normal to the surface NG
can be determined from the following equation:
UNG =
2( divq4a + — ^ e4J
1 P* j
Nc grad a5lc
(30)
We may show that the flow of the mass of the s-th
component fls in the i-th phase per unit of surface of
porous medium is:
fls (i + Aim
{Fis + Als[fis x H,] +
+ Ai(Fls • H,)H|} (31)
where Cis is the concentration of the s-th com-
ponent in the i-th phase. The concentration of the
s-th component in the soil Cs is:
0.2 0.4 0.6 0.8 f*
0.2 0.4 0.6 0.8 *»~
Figure 2. Theoretical (solid line) arid empirical (dashed line) mixtions of the absorbing potential of the toil moisture (dependence of
If on v>4). (a) according to J. R. Philip (b) according to Kulik.
151
-------
Cs = y C,(a,
AiS
Ki,C
(32)
where e* is a proton charge, Z*s is the valence of the
s-th particle, is the magnetic permeability of the
i-th phase, c is the speed of light, K(s is the resistance
coefficient of the s-th particle in the i-th phase, and
H4J/ days
0 t-flTn rt-n f Iff
T1 n r
j
t Irmlmi
H( is the strength of the magnetic field in the i-th
phase.
, r Ms . K:,Tt I
F|S = U, + — q - • —¦ grad C,s -
>--k
e*Z*s
Kta
grad U(
(33)
Here, ms is the mass of the s-th particle and U, is the
a)
~h i
UBk
[mi
In.
n
lelcL
On "*/<
Figure 3. Chronological behavior of residues (a), water supply (b), level of soil waters (c), and the flow of returning waters
(d), on the massif of the old irrigation of the Golodnaya Steppe (1-measured, 2-calculated, 3-soil component of
returning waters).
152
-------
40
30
20
10
0
8.0
6.0
4,0
2.0
0.0
2) ms
a
\jMf\
/ ^/days
/
/wr
b)
/Vy/
0,210
Qa> <*/days
Figure 4. Chronological behavior of water regime elements of the old irrigated zone of the Golodnaya Steppe in 1965-
1966 (a-air humidity deficit, b-totai evaporation, c-soil moisture in the aeration zone, d-moisture exchange
with the soil waters in the aeration zone.
potential of the electric field in the i-th phase.
For neutral particles, Z\ - 0 and ai( = 0, thus
greatly simplifying the equation for transport of this
component.
We should note that the quantities KsTt/Ku and
e*ZVKt, are the diffusion coefficient Du and the mo-
bility Ci, of the s-th component in the i-th phase.
The equation for conservation of mass of the S-th
component in the i-th phase is:
+ div f. = 2 ^(C„„ - CJ +
<9t T P*
(34)
On the right side of equation (34), the first term
153
-------
Figure 5. Change in the flow of returning waters (a) and level of the soil waters in model sections during the irrigation pro-
cess. (1-section I, 2-section II, 3-section III).
-------
represents the transition of the s-th component from
the j-th phase into the i-th phase, by dissolution; the
second term represents the transition of the h-th
component into the s-th component within the i-th
phase due to chemical reactions. The third term
takes into account the transition of s-th component
from the j-th phase into the i-th phase in the case of
phase conversion.
In this equation, ps is the density of the s-th com-
ponent (this should not be confused with the con-
centration of the component in the phase C|S), C*js is
the concentration of the s-th component in the j-th
phase changed to the i-th phase, C*js s Cjs, rijs —the
mass of the s-th component adsorbed from the i-th
phase per unit of the interface of the i-th phase with
the j-th phase.
The system of equations obtained is quite com-
plex and must be generalized for practical use.
Figures 3 and 4 give the results of calculating the
water regime of the old irrigated zone of the Go-
lodnaya Steppe, and Figure 5 gives the calculated
data for the dynamics of the water regime of three
regions (model problem).
REFERENCES
1. Budagovskii, A. I. 1955. Vpityvaniye vody V pochvu. (Ab-
sorption of water into soil). Publisher Academy of Sciences
USSR, Moscow, p. 137.
2. Budagovskii, A. I. 1964. Ispareniye pochvennoi vlagi.
(Evaporation of soil moisture). Publisher Nauka, Moscow,
p. 243.
3. Denisov, Yu. M. 1968. Perenos vlagi I tepla V pochve.
(Transfer of moisture and heat in the soil). SARNIGMI. 39
(54): pp. 3-19.
4. Denisov, Yu. M. and Ye. B. Trofimova. 1974. Matemati-
cheskoye opisaniye nekotorykh fizicheskikh kharakteristik
snezhnogo pokrova. (Mathematical description of several
physical characteristics of snow cover). SARNIGMI. 15
(96):69-7l.
5. Denisov, Yu. M. and V. P. Ovcharenko. 1975. Mnogo-
faznaya magnitogidrodinamicheskaya model atmosfery.
(Multi-phase Magneto-hydrodynamic Model of the Atmo-
sphere) SARNIGMI, 1975, Issue 21 (102), pp. 3-29.
6. Denisov, Yu. M. 1974. Nekotorye biologicheskiye printsipy
I ikh ispolzovaniye V izuchenii pronitsayemosti kletochnoi
membrany. (Several biological principles and their use in the
study of the permeability of cell membrane). Institute of Cy-
bernetics, Izd. Uzbek Academy of Sciences. 73 Tash-
kent.: 129-140.
7. Nerpin, S. V. and A. F. Chudnovskii. 1967. Fizika pochvy.
(Soil physics). Nauka, Moscow, p. 583.
8. Nerpin, S. V. and A. F. Chudnovskii. 1975. Energo I mas-
soobmen V sisteme rasteniye — pochva — vozdukh. (Ener-
gy and mass exchange in a system — plant — soil — air).
Gidrometeoizdat, Leningrad, p. 358.
9. Sou, S. 1971. Gidrodinamika mnogofaznykh sistem. (Hydro-
dynamics of multiphased systems). Mir, Moscow, p. 536.
10. Rakhmatullin, Kh. A. 1956. Osnovy gazodinamiki vzaimo-
pronikayushchikh dvizhenii szhimayemykh sred. (Gas dy-
namics of interpenetrating movements of compressible envi-
ronments). PMM. 20 (2): 184-195.
11. Chudnovskii, A. F. 1976. Teplofizika pochv. (Thermophys-
ics of soils). Moscow, p. 352.
12. Chailds, E. 1973. Fizicheskiye osnovy gidrologii pochv.
(Physical bases of soil hydrology). Gidrometeoizdat, Lenin-
grad. p. 427.
13. Wallis, G. 1972. Odnomernye dvukhfaznye techeniya. (Uni-
dimensional two-phased flows). Mir, Moscow, p. 440.
14. Filip, Dzh. R. 1972. Teoriya infiltratsii. Izotermicheskoye
peredvizheniye vlagi V zone aeratsii. (Infiltration theory.
Isothermal transformation of moisture in the aeration zone).
Gidrometeoizdat, Leningrad, pp. 6-81.
15. 1974. Metody rascheta vlagoperenosa V zone aeratsii. (Me-
todicheskiye ukazaniva). (Methods for calculating moisture
transfer in the aeration zone). Minsk, p. 82.
16. Denisov, Yu. M. and A. I. Sergeyev. Matematicheskaya
model oroshayemogo massiva, sostoyashchego iz otdelnykh
uchastkov. (Mathematical model of an irrigatable area, com-
posed of several sections). SARNIGMI. 39 (120): 15-25.
155
-------
ABSTRACT
Models of Economic Growth and the Environment
RALPH C. d'ARGE
In recent years there has been substantial re-
search activity by economists in the Western World
to reconcile concepts of traditional economic
growth and general equilibrium with physical laws
and natural processes occurring in the environment.
In this presentation three models were developed,
examing first what role conservation of matter and
energy plays in economic theory and how it may
alter traditional economic models of planning; sec-
ond, describing how finite natural resources and fi-
nite assimilative capacities of the environment re-
strict traditional models of economic growth; and
third, examining how entropy might enter as a con-
straint in economic planning models. It was shown
that traditional microeconomic models are not en-
tirely consistent with either the first or second laws
of thermodynamics. It was also shown that the
mapping between atomistic competition and effi-
ciency is broken entirely when the first law of con-
versation of matter and energy is taken into ac-
count. Further, it was demonstrated under rather
general assumptions that consumption per capita
may increase or decrease but is unlikely to remain
constant for any optimal strategy of development. It
was also demonstrated that the pure theory of ex-
haustion of resources is complemental to a pure
theory of environmental policy. Finally, it was con-
jectured that a symmetry exists as to the effects of
both the first and second laws on economic models
and that both are a form of constraint that elicit
well-defined shadow prices.
156
-------
RECORD
of the Third US-USSR Symposium
on Comprehensive Analysis of the Environment
Tashkent, USSR, October 10-14, 1977
In accordance with the principles laid down in the
records of the Working Group Meeting held in the
USSR in October 1976 and the visit of Soviet spe-
cialists to the US in August 1977, under the aus-
pices of Project 02.07-21; and in accordance with
the Memorandum of Implementation from the Fifth
Session of the US-USSR Joint Committee on Coop-
eration in the Field of Environmental Protection,
the Third US-USSR Symposium on Comprehensive
Analysis of the Environment was held in Tashkent,
USSR from October 10-14, 1977.
The Symposium was co-chaired by Mr. C. R.
Gerber, Associate Assistant Administrator for Re-
search and Development of the US Environmental
Protection Agency and Academician Yu. A. Izreal,
Chief of the Main Administration of the USSR Hy-
drometeorological Services.
The participants in the Symposium presented pa-
pers on general approaches to pollution control, and
on health, ecological and economic effects of pollu-
tion. The following problems were discussed: com-
plex and comprehensive assessment of pollution ef-
fects on the environment as the basis for developing
programs for environmental quality control and
monitoring, studies of pollutant effects on human
health by means of clinical and epidemiologic inves-
tigations, pollutant effects on biota using both field
and laboratory tests and mathematical modelling,
and economic analysis of pollutant impact on the
environment. Attention was also given to determin-
ing permissible loading on ecosystems, to setting
ecological standards and to ecological monitoring
principles. General discussion was also held.
Both sides agreed to publish the proceedings of
the Symposium on Comprehensive Analysis of the
Environment during the second quarter, 1978; the
American side in English and the Soviet side in Rus-
sian. The co-chairmen will prepare and exchange an
introduction at the Joint Committee Meeting to be
held during November in Washington D.C. The
American side will translate the Soviet papers from
English into Russian. At this time, both sides will
also exchange final texts of the papers and related
illustrations.
During the meetings both sides discussed future
cooperative activities under Project 02.07-21, Com-
prehensive Analysis of the Environment, and
agreed that 1) a Fourth US-USSR Symposium on
Comprehensive Analysis of the Environment would
take place in the US in 1979; 2) this Symposium
would focus on ecological, health and economic as-
pects of analyzing environmental impact in one or
more specific geographic areas either developed or
under development and on assessing permissible
loading on these areas; 3) to this end, one or two
specialists from each side would be exchanged in
1978 to study environmental impact in specific de-
veloping regions and to discuss plans for the Fourth
Symposium; 4) both sides would exchange pro-
posals for the visits by March 1, 1978.
The Symposium and meetings were held in an at-
mosphere of friendly cooperation and were of mu-
tual benefit to both sides. The US specialists wish to
express their gratitude to the Soviet delegation for
organizing the symposium and for the gracious hos-
pitality shown them during their visit. They also
want to thank the interpreters for their excellent
services.
This protocol was signed in Tashkent October 14,
1971 in two copies — English and Russian — both
copies being equally valid.
157
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1 . REPORT NO. 2.
EPA-600/9-80-024
3, RECIPIENT'S ACCESSIOf*NO.
4. TITLE AND SUBTITLE
Third Joint U.S./USSR Symposium on the Comprehensive
Analysis of the Environment
6. REPORT DATE
May I 980
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Compiled by Elaine Fitzback
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
SAME AS BELOW
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO. "
I n-house
12. SPONSORING AGENCY NAME AND ADDRESS
Office of Research and Development
U.S. Environmental Protection Agency
Washington, DC 20460
13. TYPE OF REPORT AND PERIOD COVERED
14. SPONSORING AGENCY CODE
EPA/0RD
15. supplementary notes ^gg a ) so First, Joint Symposium ( EPA-600/9-75-004 ) and Second
Joint Symposium (EPA-600/9-76-024),
16. ABSTRACT
These proceedings include papers presented at the Third Joint U.S./USSR
Symposium on the Comprehensive Analysis of the Environment. The papers
were presented in English or Russian at Tashkent USSR from October 10—14,
I977. Two additional presentations, both on mathematical models (Section V)
are also included in this publication.
The publication of these proceedings is in accordance with Memorandum from
the Sixth Session of the U.S.-USSR Joint Committee on Cooperation in the
Field of EnvironmentaI Protection, signed in Washington, D.C. on November 18,
I977. This document called for independent publication of the Proceedings
in both the United States and the Soviet Union.
17. KEY WORDS AND DOCUMENT ANALYSIS
a. descriptors
b. IDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Group
18. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
19. security CLASS (ThisReport)
21. NO. OF PAGES
158
20. SECURITY CLASS (Thispage)
22. PRICE
EPA Form 2220-1 (9-73) I en
I JO ft US GOVERNMENT PRINTING OFFICE 1980.657-146/5662
------- |