oEPA
United States
Environmental Protection
Agency
Environmental Research
Laboratory
Gulf Breeze FL 32561
EPA-600/9-78-Od7
May 1978 ,_.-- —
Research and Development
First American-Soviet
Symposium on the
Biological Effects of
Pollution on Marine
Organisms
This document has not been
submitted to NTIS, therefore it
should be retained.
-------
EPA-600/9-78-007
FIRST AMERICAN-SOVIET SYMPOSIUM
ON
THE BIOLOGICAL EFFECTS OF POLLUTION
ON MARINE ORGANISMS
SYMPOSIUM SPONSORED AS PART OF THE U,S,-U,S,S,R,
AGREEMENT ON PROTECTION OF THE ENVIRONMENT
THOMAS W, DUKE ANATOLIY I, SIMONOV
U.S. Leader of Project VI-2.1 U.S.S.R. Leader of Project VI-2.1
25SSS3S1-"-
U.S. ENVIRONMENTAL PROTECTION AGENCY
OFFICE OF RESEARCH AND DEVELOPMENT
ENVIRONMENTAL RESEARCH LABORATORY
GULF BREEZE, FLORIDA 32561
-------
DISCLAIMER
This report has been reviewed by the Environmental Research Laboratory,
Gulf Breeze, U. S. Environmental Protection Agency, and approved for publica-
tion. Mention of trade names or commercial products does not constitute
endorsement or recommendation for use.
11
-------
FOREWORD
Chemical pollution and the resulting impact on the World Ocean is par-
ticularly important among the numerous problems of environmental protection
since the enormous area covered by the open waters of the World Ocean is not
under the jurisdiction of any individual government. In many instances, man
has turned the oceans which formerly separated countries and people into com-
mon reservoirs containing the discharges of wastes which accompany the produc-
tion of materials. The accumulation and spread of pollution in the waters of
the World Ocean can produce unforeseen and unwanted changes in the structure
of functioning of natural communities. This, in turn, causes a large part of
society to be concerned.
A similar situation is the incentive for coordinating the efforts of
scientists from different countries to jointly develop questions for evaluat-
ing the quality of natural waters and the scientific basis for preventative
measures to protect their living population.
The US-USSR Agreement on Cooperation in the Field of Environmental
Protection, signed in May 1972, is an important step toward international sci-
entific cooperation in solving these problems. One part of this agreement,
Project 02.06-21, provides for a study of the "Influence of Pollutants on
Marine Organisms." The creation of a program for global monitoring of pollu-
tion in the World Ocean is a necessary step for solving marine pollution prob-
lems. A system for observing the biotic component of the earth's surface
water must hold an important place in such a system. Therefore, in the first
stages, specialists should develop parameters characterizing the structural
and functional properties of natural ecosystems, and likewise develop the pos-
sibility of studying them in laboratory conditions.
In this particular direction, joint Soviet-American studies under
Project 02.06-21, "Influence of Pollutants on Marine Organisms," have devel-
oped within the framework of the US-USSR Environmental Agreement. The work
performed laid the foundation for conducting the First Soviet-American Sympo-
sium on Hydrobiological Methods for Analyzing Marine Pollution, which took
place September 20-24, 1976, at the Gulf Breeze (Florida) Environmental Re-
search Laboratory. Soviet and American specialists presented more than 20
papers which gave a multiple exposure to the contemporary state of methods
for hydrobiological analysis of basic structural components of marine ecosys-
tems and the influence of various pollutants on them.
Problems which were discussed at the successfully conducted symposium
(methods for modeling the influence of pollutants on the marine environment,
long-term forecasting and determination of permissible loads of pollutants,
unification and intercalibration of methods for determining production of
iii
-------
microorganisms of ocean bacterioplankton and phytoplankton, methods for study-
ing pollutants of varied nature and their influence on the environment in
field and laboratory conditions, results of laboratory research on the influ-
ence of pollution on the marine environment) are undoubtedly of great interest
to a large group of specialists who dedicate their activity to various scien-
tific aspects of protecting the World Ocean from pollution.
Consequently, the co-chairpersons of Project 02.06-21 have agreed to
publish the materials from the symposium, in Russian in the USSR and English
in the United States, as stated in the Memorandum from the 4th Session of the
Joint US-USSR Committee on Cooperation in the Field of Environmental Protec-
tion.
Thomas W. Duke
Director
Environmental Research Laboratory
Gulf Breeze, Florida
IV
-------
ABSTRACT
This symposium was conducted under a US-USSR Environmental Agreement,
Project 02.06-21 titled "Influence of Pollutants on Marine Organisms."
American and Soviet specialists discuss state-of-the-art for hydrobiological
analysis of basic structural components of marine ecosystems and the influ-
ence of various pollutants on these components. Participants define problems
related to methods for modeling the influence of pollutants on the marine
environment, long-term forecasting and determination of permissible loads of
pollutants, and the unification and intercalibration of methods for deter-
mining production of microorganisms of ocean bacterioplankton and phyto-
plankton. Results of laboratory research on the influence of pollution on
the marine environment are presented. Proceedings were published in English
and Russian in compliance with the Memorandum from the 4th Session of the
Joint US-USSR Committee on Cooperation in the Field of Environmental Research.
v
-------
CONTENTS
Foreword
Abstract v
Protocol vii
Participants in the Symposium x
Design of Simulation Models to Determine Biological Impact of Pollutants
on the Marine Environment 1
Problems in Modeling the Effect of Pollution on Biological Systems of
the Ocean 10
American Methods for Measuring Phytoplankton Production in the Open
Ocean 22
Relative Abundance of Sympatric Species and Model of Exponentially Broken
Rod (EBR) 37
Impact of Radioactivity on the Marine Environment 63
The Effect of Radioactive Pollution of Reservoirs on Fish 73
A Non-Standard Approach to Heterotrophy 77
The Development of Standard Methods of Measuring Microbiological
Production 87
Use of Biological Indicators for Monitoring Effect of Pollutants on
the Marine Environment 95
Microorganisms as Biological Indicators of Commercial-Domestic and
Petroleum Pollution 107
Persistence Limits in Ecological Systems 112
The Development of Standard Laboratory and Field Ecosystem Research to
Determine the Effects of Pollutants on the Marine Environment .... 118
Impact of Pesticides on the Marine Environment 126
Oil and Marine Organisms 138
Impact of Metals on the Marine Environment 144
Appendices 155
vi
-------
PROTOCOL
OF THE FIRST AMERICAN-SOVIET SYMPOSIUM
ON
THE BIOLOGICAL EFFECTS OF POLLUTION ON MARINE ORGANISMS
In accordance with the principles laid down in the Protocol from the
Working Group Meeting on Project VI-2.1, held in the USSR in July 1976, and
in the Memorandum of Implementation from the Fourth Meeting of the US-USSR
Joint Committee on Environmental Protection, a joint US-USSR Symposium on the
Biological Effects of Pollution was held in Gulf Breeze, Florida, September
20-24, 1976.
The Symposium was co-chaired by Dr. T. W. Duke, Director, Gulf Breeze
Environmental Research Laboratory, and Dr. A. I. Simonov, Office Chief, State
Oceanographic Institute Moscow. The support from the Gulf Breeze Laboratory
was most helpful and services of interpreters were excellent. A list of par-
ticipants is attached.
The participants presented prepared papers on biological effects of
pollution on marine organisms from radioactivity, pesticides and metals and
the design of methods of phytoplankton, and microbiological production, as
well as the development and use of models to determine the movement of pollu-
tants and biological impact of pollutants on the marine environment.
Stimulating discussions were held after each presentation and impor-
tant aspects of these discussions will be included in the published proceed-
ings of the Symposium. In addition, the following proposals were discussed
concerning future joint research projects:
A. MODELING
Specialists discussed models now available, their strengths and weak-
nesses. It was proposed that this group jointly undertake the development
and validation of models for some specific areas of the World Ocean. For ex-
ample, a model for the calculation and prediction of primary production in
the Gulf Stream including the system of the North Atlantic Current and models
for microcosms were suggested.
B. PRIMARY PRODUCTION
Specialists having discussed existing methods for determining primary
production came to the conclusion that there is a significant need for improv-
ing the methods—indeed, a need to create a new method. It was agreed that a
VII
-------
joint effort to develop methods for determining primary productivity and ef-
fects of pollutants on primary productivity would be worthwhile.
C. MICROBIOLOGY
US and USSR specialists came to the conclusion that it is worthwhile
to conduct joint microbiological studies in laboratory conditions and sea-
waters.
D. RADIOACTIVITY
Experts expressed an interest in obtaining more detailed information
on the uptake—accumulation and effect of transuranic radionuclides on marine
organisms. Also, they expressed the desire to conduct joint work on inter-
calibration of methods of analysis and with subsequent discussion at the next
Symposium.
Both sides discussed and gave a high evaluation of the intercalibra-
tion results of hydro-biological methods of analysis made on the joint US-USSR
cruise R/V Moskovskiy Universityet in July and August of 1975 and prepared a
joint report for publication. The sides expressed desire that the American
specialist, Dr. Iverson, a participant in the joint intercalibration, meet at
Moscow State University (Moscow) in December of 1976 with his Soviet col-
leagues to complete the discussion of results on the intercalibration of hydro-
biological analyses.
Both sides also agreed to publish the Symposium papers and highlights
of the discussion during 1977.
The co-chairmen agreed to write the text of the Introduction to the
Proceedings of the First Hydrobiological Symposium via correspondence by De-
cember 1, 1976. They also agreed to supply each side with final text of
papers in both languages by April 1, 1977. Both sides agreed to publish the
planned but not read papers as abstracts.
Members of this Symposium recommended that there be discussion of the
results of the above mentioned studies and other joint studies at the Second
Hydrobiological Symposium. It is desirable to hold this Symposium in the
Soviet Union within the next one and one-half to two years.
Both sides expressed the desire to conduct expeditionary observations
in 1977 or 1978 in the North Atlantic on the above mentioned questions. Co-
chairmen will discuss the details of the expeditionary observations during
1977.
Vlll
-------
After the meetings, the Soviet scientists toured the research facili-
ties at Florida State University in Tallahassee.
The Symposium and meetings were held in an atmosphere of friendly co-
operation and have been of mutual benefit to both sides. The Symposium ex-
pressed its thanks to Dr. Duke, organizer of the joint meeting.
A. I. S imonov
U.S. Leader of U.S.S.R. Leader of
Project VI-2.1 Project VI-2.1
IX
-------
PARTICIPANTS IN THE SYMPOSIUM
AMERICAN SIDE
Dr. Thomas W. Duke, Director
Gulf Breeze Environmental Research Laboratory
U.S. Environmental Protection Agency
Gulf Breeze, Florida 32561
Dr. Ford A. Cross, Director, Division of Ecology
National Oceanic and Atmospheric Administration
National Marine Fisheries Service
Beaufort, North Carolina 28516
Mr. David J. Hansen, Supervisory Research Aquatic Biologist
Gulf Breeze Environmental Research Laboratory
U.S. Environmental Protection Agency
Gulf Breeze, Florida 32561
Dr. Richard L. Iverson, Assistant Professor of Oceanography
Florida State University
Tallahassee, Florida 32306
Dr. John H. Martin, Associate Professor of Biology
Moss Landing Marine Laboratory
P.O. Box 223
Moss Landing, California
Dr. John McN. Sieburth, Professor of Oceanography
Graduate School of Oceanography
University of Rhode Island
Kingston, Rhode Island 02881
Dr. Samuel C. Snedaker, Associate Professor of Biological Research
University of Miami
4600 Rickenbacker Causeway
Miami, Florida 33149
Dr. Frank G. Wilkes, Chief, Processes and Effects Branch
Gulf Breeze Environmental Research Laboratory
U.S. Environmental Protection Agency
Gulf Breeze, Florida 32561
-------
SOVIET SIDE
Prof. Anatoliy I. Simonov, Doctor of Geographical Sciences
Head of the Section of Hydrodynamics and Scientific Base for
the Protection of the Marine Environment of the State
Oceanographic Institute of the Hydrometeorological Service
Moscow, USSR
Prof. Vadim D. Fedorov, Doctor of Biological Sciences
Head of the Chair of Hydrobiology
Moscow State University
Prof. Mikhail V. Gusev, Doctor of Biological Sciences
Dean, Department of Biology, and Head of the Department Chair
of Plant Physiology and of the Laboratory of Physiology
and Biochemistry
Moscow State University
Prof. Mikhail E. Vinogradov, Doctor of Biological Sciences
Deputy Director and Head, Plankton Laboratory of the Institute
of Oceanography, Academy of Sciences USSR
Professor of the Chair of Hydrobiology
Moscow State University
Ivan B. Tokin, Doctor of Biological Sciences
Director, Institute of Marine Biology of the Academy of Sciences
USSR
Murmansk, USSR
Irina A. Shekhanova, Candidate of Biological Sciences
Head of the Radiation Biology Laboratory
Ail-Union Scientific-Research Institute of Marine Fisheries and
Oceanography
Moscow, USSR
Yury A. Anokhin, Candidate of Physico-Mathematical Sciences
Deputy Head, Laboratory of the Institute of Applied Geophysics
of the Hydrometeorological Service USSR
Moscow, USSR
Konstantin S. Burdin, Docent, Candidate of Biological Sciences
Chair of Hydrobiology, Department of Biology
Moscow State University
XI
-------
DESIGN OF SIMULATION MODELS TO DETERMINE
BIOLOGICAL IMPACT OF POLLUTANTS
ON THE MARINE ENVIRONMENT
by
Samuel C. Snedaker
Rosenstiel School of Marine and Atmospheric Science
University of Miami
Miami, Florida 33149
INTRODUCTION
Models for simulating the behavior of ecological systems in their natu-
ral state or after an acute or chronic perturbation must be rigorous both in
concept and in adherence to physical laws. The primary objective of this
paper is to review the bases for constructing models incorporating systems'
responses to pollution and to explore the conceptual bases for such models.
THE CONCEPTUAL BASES
POLLUTION DEFINED
Pollution may be defined as the inducement of a stress that can poten-
tially drain energy from the normal functioning of the system. In response,
affected organisms become less capable of performing their life-supporting
functions and ultimately less competitive. The organisms are therefore sub-
ject to replacement by others better adapted to the prevailing circumstances.
In severely polluted environments no organism may possess the necessary sur-
vival adaptations. By defining pollution in terms of energy, empirical mea-
surement is facilitated and the effect of a pollutant is more easily incorpo-
rated in a multicomponent model.
POLLUTION AS A SOURCE OF POTENTIAL ENERGY
Essentially, the two main classes of pollutants are distinguishable on
the basis of kind or type, and quantity or intensity. Some pollutants (for
instance, atomic radiation, the heavy metals and certain organic biocides)
provoke only negative responses from living organisms. The severity of re-
sponse is generally proportional to the quantity or intensity of the pollu-
tant. Other commonly cited pollutants (e.g., temperature, mineral nutrients,
salinity, etc.) provoke negative responses only when present in quantities
-------
which the living organisms cannot tolerate. However, the same conditions may
represent the optimum survival requirements for organisms in other systems
where the same pollutant is present in natural quantity. This is consistent
with the concept of the environmental continuum in which various species find
optimum conditions as well as conditions under which they cannot exist. Thus,
a stress and a drain of potential energy in one species may in fact be an
auxiliary source of potential energy for another species (cf. Odum, 1971).
THE SPECIES VERSUS THE SYSTEM
A common inadequacy of many biological models, particularly those asso-
ciated with a pollution problem, is the tendency to focus on parameters de-
scribing the component species and population assemblages, as opposed to para-
meters descriptive of the larger system. A species population may be com-
pletely lost from the system as a result of a pollutant. A population model
may fail to indicate that a better-adapted species would be able to colonize
the system to the point that its appropriate system function could be main-
tained. Likewise, a system's model emphasizing system parameters may faith-
fully simulate a stable system function (such as community metabolism) without
indicating that a particular component species of interest was lost. A model
usually focuses on the problem at hand and the question(s) under study, while
the empirical data must be obtained through a research design sensitive to
both system and species parameters.
KINDS OF MODELS
Quinlan (1975) has determined that all models may be subsumed under one
of three model categories, i.e., biodemographic, bioenergetic or biogeochemi-
cal. Each has specific attributes and disadvantages. Biodemographic models
incorporate parameters descriptive of the dynamics of the species populations,
but occasionally may indicate a population whose material content exceeds that
available in the parent ecosystem. Bioenergetic models, like biogeochemical
models, are conservative in that energy or matter may be fully accounted for
in the model. Energy and mass balances in this context can be used to assure
quality in a simulation exercise. Bioenergetic models, however, often assume
that the participants in any process behave in proportion to their specific
energy content. These models ignore, for instance, the synergistic effect of
an interaction or the amplification effect of a feedback. Biogeochemical
models provide the most complete representations. They are inherently con-
servative, and their processes are not allowed to continue if one or more re-
actants are depleted. Modeling methodology promulgated by Odum (1971) basi-
cally satisfied the principles of conservation and causality by assigning to
model participants energy values proportionate to the quantity of the work the
system can perform.
Many contemporary models, in addition to supporting research, propose
to model energy and energy flow when, in fact, participants in the model are
commonly stocks and flows of organic matter. (Energy is the capacity for
doing work. The term flow in this context is a misnomer, for it associates a
flow with a thermodynamic state.) To resolve this particular dilemma, one
may either model carbon as a biogeochemical model or assign work-generating
energy values of Odum's bioenergetic models.
-------
ARE SIMULATION MODELS PREDICTIVE?
The prediction of a future state can be made only from a knowledge of
past and present states and the experience gained by observing analogous,
time-dependent situations. Once programmed, a model may be visualized as a
hard-wired network of storages and flows. The pattern is not subject to
change during simulation. If an altered real-world system results in a dif-
ferent pattern in the arrangement and linkage of components, this occurrence
would not be observed during a simulation exercise. Thus erroneous conclu-
sions could be inferred. One common method that partially allows for circuit
changes during simulation is to program the alternative arrangements, using a
logic program as a switching mechanism in some predetermined manner. This
method, however, requires that the optional future states should be deter-
mined in advance, as well as the conditions for any given state. Obviously,
one purpose of modeling is to gain an insight into future states and to sug-
gest that answers used in formulating the questions are circular and contra-
dictory. The preferred mode of model design sufficiently aggregates the par-
ticipants in the model so that minor changes in the network pattern do not in-
validate the simulation. After the behavior of the aggregated model is under-
stood through simulation under varied conditions and is validated, the model
then can be redesigned to include more detail about the participants. The art
of knowing how to aggregate a biological model and how to select the level of
necessary detail may transcend the importance of operational computer/mathe-
matical techniques in a successful modeling exercise.
A MODELING LANGUAGE
The need to bridge the gap between real-world observations and the
mathematical formulations of the operational models has evoked the use of sym-
bolical languages that explicitly portray relationships (Forrester, 1961;
Odum, 1971). These symbol languages permit stocks, flows, and various forms
of interactions and processes to be visualized as a whole and subsequently to
be parameterized for simulation exercises. The energy language of Odum (1971)
is used here to review the construction of models that would convey the behav-
ior of ecosystems and their forms of response to pollution.
DESIGN OF THE BIOLOGICAL MODEL
The simplest model that can be constructed to simulate the gross behav-
ior of an ecological system aggregates the producers and the consumers as
separate participants. It also links them with two flows representing the
trophic relationship and the regenerative feedback of the elemental constitu-
ents of organic matter (Fig. 1). In this representation of the system, the
sun, as the primary forcing function, powers the photosynthetic machinery that
builds new organic materials out of the inorganic constituents. Both pro-
ducers and consumers exhibit auto-catalytic properties that induce a growth
response as some function of the available energy and matter. In the produc-
tion-respiration model, two passive storages represent the system's total
stocks of organic and inorganic matter. Because all real-world systems are
open to flow of energy and matter, each passive storage has external inputs
-------
and ..outputs. This provision essentially couples the system of interest with
the larger total system. Two heat sinks account for energy lost when work is
done in accordance with the second law of thermodynamics, the bioenergetic
mode. As models for biogeochemical cycling in a mass balance simulation, the
heat sinks may not be necessary. Since this basic model is parameterized with
empirical values and simulated, the basic patterns of productivity and respira-
tion (i.e., community metabolism) can be evaluated in relation to the inten-
sity of the driving force and the inputs and outputs of organic and inorganic
materials. In its most complex form, this model is highly predictive but sac-
rifices detail.
Figure 1. A basic conceptual model for an ecosystem emphasizing primary pro-
duction and respiration with storages for organic and inorganic
materials coupled to the larger system. Producers and consumers
are auto-catalytic. Photosynthesis is the product of the inter-
action between sunlight and inorganic materials. Heat sinks ac-
count for the heat loss as a result of work. (Symbols from Odum,
1971)
I -
o
1
2
3
4
J. -
sunlight
producers
organic materials
consumers
inorganic materials
inputs from source external
to system of interest
1,2
2,3
3,4
4,1
output exported beyond system
of interest
net organic production
consumption by consumers
products of respiration
uptake of inorganic materials
photosynthesis (product of sun-
light and limiting nutrient)
-------
The basic model in Figure 1 can be progressively expanded by adding
components to describe and detail specific functions associated with the sys-
tem, provided the conservative characteristics are retained.^ In Figure 2, for
example, gross primary productivity is modified by the circulation of water in
such a manner that the uptake of the essential nutrients (and gases) becomes,
over a certain range, a function of the delivery mechanism, irrespective of
the total quantity available in the system. The interaction symbol is an ex-
pression of the limiting factor concept. Over the range in which one factor
is limiting, the response is linear and operationally incorporated in the
final simulation program as simple multiplication of the two interacting
flows. (Whereas simple interactions are linear over a response range, the be-
havior of the system as a whole assumes nonlinear characteristics.) A second
example of the expansion of the basic model is illustrated in Figure 3. Here,
a stock of organic matter is detailed to include some component organic frac-
tions of interest with respect to two distinct kinds of consumers. One con-
sumer can draw nonselectively from either storage in proportion to its quan-
tity. The second consumer is substrate specific. The total mass of consumers
in this latter category varies with the quantity of material in this single
storage.
Figure 2. The basic production-respiration model illustrating the role of
currents and circulation in juxtaposing inorganic materials and
producers. (Symbols from Odum, 1971)
I - circulation and currents
X - stirring and transport of inorganic materials.
This mode of construction permits the resulting model to reflect the
real-world system as seen through the eyes of the investigator. And, depend-
ing upon the question(s) being asked, no two models will be necessarily iden-
tical. In contrast, other modeling efforts utilize a fixed compartment-flow
model in which the real-world system is superimposed by varying the size of
stocks and by transferring coefficients linking them. Such models are
-------
generally preferred by investigators more interested in studying models per se
than as tools in understanding the structure and function of a real-world sys-
tem or specific problem.
J6.1
Figure 3. The basic production-respiration model with the storage of organic
materials divided into two compartments based on composition and
the relative feeding strategies of two groups of consumers.
(Symbols from Odum, 1971)
Q and Q,. - organic matter of two different compositions
Q, and Q, - consumers with two different feeding strategies
Jj. ,.; J,. ,.; and J - consumption by respective consumer group
J / Q J^-j £ f 3
MODELING POLLUTANT IMPACT
The impact of pollutants on a biological system via its living compo-
nents can be incorporated in a model in the same manner specific functions
were included in the foregoing examples. Two examples of pollution-induced
stress are provided in Figures i and 5, which respectively illustrate the
species-selecting characteristics of temperature through the differential ac-
celeration of metabolism and the variable response to an input of a biocide.
Temperature (Fig. 4) is illustrated as having a similar overall response rela-
tive to each of the living components through the acceleration of metabolism
such as might occur in the thermal plume of a coastal power plant. The two
consumer participants in this example can be parameterized to show a differen-
tial response in the drain of metabolic energy for a given temperature. A
similar acceleration of potential energy loss is shown for the producers, but
in this instance temperature also is shown to have a positive amplification
effect on primary production through enzymatic stimulation (Jorgensen and
Nielsen, 1965). The result of this "push-pull" effect may be manifested dur-
ing simulation as an increasing turnover time without changing the standing
stock of producers (McKellar, 1975) . Many other effects of elevated or
-------
otherwise altered temperature regimes can be similarly included in a model
when specific relationships are known.
Figure 4. The impact of elevated temperature in accelerating the metabolism
of living organisms illustrated as drains of potential energy. Dif-
ferential temperature responses by the component species would be
represented by the appropriate coefficients on the energy drains.
Producers may be stimulated by small temperature rises offsetting
the accelerated metabolic losses (Symbols from Odum, 1971)
I - source of heat raising ambient water temperatures
X - temperature stimulation of gross primary production
X , Xr and X - temperature-induced acceleration of metabolism
Provisions are included in the model in Figure 5 for the input and ef-
fect of biocide, which for discussion here is illustrated as a lipid-soluble
chlorinated hydrocarbon. It is diagrammed as being introduced into the pool
of materials available for uptake by the consumers. However, this kind of
compound is basically insoluble in water and is absorbed onto the surfaces of
solid substrates. Now consider the two stocks of organic material as con-
trasting-sized classes of organic particulates. Because of surface-area-to-
volume ratio differences, the two-particle classes differentially sorb the
biocide. The result is that one fraction carries a more concentrated biocide
burden, thus affecting the two consumer groups in different degree, depending
on their time-varying feeding strategies. The drain of potential energy on
the consuming organisms can be incorporated in a simulation model as a meta-
bolic acceleration (similar to that diagrammed for temperature), with the rate
of potential energy loss proportional to the body burden.
-------
Figure 5. The introduction of a chlorinated hydrocarbon entering two organic
storages differing in surface-area-to-volume ratios. The water-
insoluble biocide sorbed onto the surfaces of the organic materials
would result in different concentrations and consequently different
doses to the consumer based on their time-dependent feeding strate-
gies. (Symbols from Odum, 1971)
I - source of chlorinated hydrocarbon
Q and Q - organic particulates of two different size classes
^- O
J and J - differential rates of absorption onto particulate
•3 , b 3 f 2.
surfaces
The model in Figure 5 represents a summation of Figure 1 plus the addi-
tion of four components to include specific circumstances. Other components
could be added in like manner for greater detail or for incorporating other
specific circumstances. At any level of increasing complexity, given the per-
tinent data, simulations could be made for such purposes as understanding the
behavior of systems with these characteristics or for developing hypotheses
for subsequent testing.
SUMMARY
Simulation models to determine the biological impact of pollutants can
assume any one of three forms (bio-demographic, bioenergetic or biogeochemical)
if they are conservative and mimic real-world, cause-effect processes. The
term pollution is sometimes difficult to define precisely. It can be modeled
relatively easily, however, when one considers a pollutant as an inducement of
8
-------
a stress that alters the potential energy allocation process of a system.
(More energy goes into stress-compensating processes and less into competi-
tion.) Depending on the kind and quantity of a pollutant, the impacted system
may exhibit many different types of compensatory strategies. The possibility
that a pollution-induced response may be manifested in a significant altera-
tion of the basic network pattern of the system argues for highly aggregated
models that are insensitive to internal network changes. The modeling method-
ology of Odum (1971) is particularly useful in the design of simulation models
to determine the biological impact of pollutants, because rigorous models can
easily be constructed to include any observable and measurable process. Simu-
lation exercises are viewed here as a tool for understanding system processes,
rather than as the end result of a research task.
LITERATURE CITED
Forrester, J. W. 1961. Industrial Dynamics. Wright-Allen, Cambridge, Mass.
464 pp.
Jorgensen, E. G., and E. S. Nielsen. 1965. Adaptation in plankton algae.
Pages 38-46 in C. R. Goldman, ed., Primary Productivity in Aquatic Envi-
ronments. Mem. Inst. Ital. Idrobiol., 18 suppl. Univ. California Press,
Berkeley.
McKellar, H. N. 1975. Metabolism and models of estuarine bay ecosystems af-
fected by a coastal power plant. Ph.D. Dissertation, Univ. Florida,
Gainesville.
Odum, H. T. 1971. Environment, Power and Society. Wiley-Interscience, New
York. 331 pp.
Quinlan, A. V. 1975. Design and analysis of mass conservative models of eco-
dynamic systems. Ph.D. Dissertation, Massachusetts Inst. of Technology,
Cambridge, Mass. 450 pp.
-------
PROBLEMS IN MODELING THE EFFECT OF POLLUTION
ON BIOLOGICAL SYSTEMS OF THE OCEAN
by
Mikhail Ye. Vinogradov
Laboratory of the Institute of Oceanography, USSR
The increasing tempo of industrialization, the intensification of ag-
riculture, the rapid growth of population have given rise to rapidly progress-
ing pollution of the biosphere, the extent and rate of self-purification of
which are limited. To prevent an ecological crisis, special investigations
are necessary both in the control of the processes of pollution of ecosystems
as well as in the change in their functioning under conditions of pollution.
MODELING OCEANIC ECOSYSTEMS
From the point of view of anthropogenic pollution of the ocean, it is
significant that it first affects the surface layers, i.e., the most produc-
tive ecosystems of euphotic waters, which are the energy provision for the
existence of the entire population of the ocean. It is natural that the in-
vestigation of the influence of pollution on the surface associations is of
greatest importance.
Up to the present, extensive materials have already been collected
about the biology of marine organisms, their interrelations and relations with
the abiotic medium, including anthropogenic effects. There has arisen the
practical possibility of beginning to analyze the functioning of associations
under various external conditions. The complexity and variability of ecosys-
tems and the difficulties of investigations arising in connection with this
have forced particular attention to be directed to modeling the processes
taking place in them. It makes it possible to describe the set of inter-
actions in the ecosystem and not only to predict with sufficient validity the
behavior of the system with a change in one or another of its parameters or
one or another of the external effects, but also to evaluate certain situa-
tions arising in real systems and amenable with difficulty to direct measure-
ment.
Works in producing mathematical models of various processes occurring
in marine, primarily pelagic, ecosystems and models of the general pattern of
the functioning of these ecosystems have been conducted during recent years
in the Soviet Union. Some of the results obtained indicate that the models
are quite adequate (Vinogradov et al., 1972, 1973). This has made it possible
to carry out computer experiments with these models and, in particular, to
10
-------
evaluate the effect of the various factors of anthropogenic pollution on the
functioning of natural ^sterns (Vinogradov et al. , 1975) .
Another approach. which at present is still only beginning to be de-
veloped, is the production of special models of the influence of pollutants on
ecosystems, where the pollution is considered as some stochastic process
(Brusilovskiy, 1975).
I will discuss below some of the results and proposals resulting from
the works of collaborators of our laboratory and my colleagues, primarily
B. S. Fleyshman, P. M. Brusilovskiy, O. G. Mironov, E. A. Shushkina, V. F.
Krapivin and V. V. Menshutkin.
DETERMINISTIC MODELS OF PELAGIC ECOSYSTEMS
Investigations must, of course, begin with the simplest and most ac-
cessible for studying systems. Pelagic ecosystems are such systems. The sig-
nificant homogeneity of the biotope causes the decisive situation in pelagic
associations of having trophic relations, which are comparatively easy to
evaluate quantitatively. The abiotic conditions, including the effect of pol-
lutants, which have a direct influence on the functioning of pelagic associa-
tions can be determined and quantitatively evaluated comparatively easily. At
later stages, when methods for investigating pelagic associations are avail-
able, one will be able to turn to the study and prediction of the behavior of
systems including the population of both the water strata and the bottom of
the sea.
The structure of models must be based on dynamic representations of
the succession of pelagic associations (Vinogradov, Gitelzon and Sorokin,
1970), i.e., with the development of the association with time, its structu-
ral and functional properties vary and consequently their response to external
unfavorable effects.
In considering pelagic systems of the open ocean, it is necessary to
take into account that the system developing in time together with the current
of "aging" water moves in space. Thus, one must consider as the elementary
unit being modeled the succession of the association on the entire path of
water from the time of ascent into the euphotic layer to the time of descent.
In other words, one must model the temporal change of an association moving in
space. Of course, there cannot be equilibrium (energy balance) of the system
at each moment of time. The accumulation of energy takes place in the system
close to the zone of formation, while its dissipation occurs "downstream." It
is in the "accumulation period that the system reacts particularly strongly to
inhibiting factors" (Federov, 1975). Thus, pollution is most harmful in the
zones of system formation—upwellings and divergences of flow in the tropics
or during the spring development of plankton in temperate latitudes.
According to the approach developed by Lyapunov (1963) and Odum (1972),
the ecosystem can be represented in the form of a set of individual elements
functioning relatively autonomously, between which are communications channels.
The role of signals passing along these channels can be performed by certain
portions of matter (energy) or information. One can accordingly distinguish
11
-------
the material and information communications between the elements. One can
also distinguish two methods of transmitting matter and energy from one ele-
ment of the ecological system to another. The first (flow) is necessarily re-
lated to some transformation of matter or energy. The second (transfer) is
related only to active or passive displacements of the elements in the water
stratum. The entj.^3 ecological system can be divided into cells such that
within each cell the elements communicate only by flow, while transmission of
matter and energy between cells is accomplished by transfer. The cell of a
pelagic ecosystem of the tropic ocean is presented in Figure 1 as an example.
It is evident from Figure 1, where the effect of pollution was not evaluated,
that such effects can be taken into account.
However, it is obvious that to construct an adequate model it is neces-
sary to have a sufficiently complete representation about how matter and energy
are propagated between the corresponding elements, what regularities and in-
fluences determine the intensity of flow between the elements, what enters the
system, what leaves or is extracted from it and in what amount. In particular,
it is necessary above all to investigate experimentally the directionality and
degree of effect of various pollutants on the basic living elements of the
system. Without such estimates, any model cannot rise from the level of
purely qualitative speculations and loses its predictive properties.
In having a representation of the flow of matter within a cell and the
regularities of its transfer from cell to cell, one can describe the function-
ing of an ecosystem (association) with a system of balance equations, which
have the form:
fl
dz
=-aT
- - h P
od + n R. +
L i
i=p, b,
3
R - Up - £ C . + k
P L PI
3 b
3p
3z
'I ••' 4
3X.
i- =uTD..-R. -y.x.
L 31 i i
i=p,b,d,f-L. . .f4,
c. .
i:
j=f ...f^, s ...s
i=fr..f4, sr..
12
-------
Figure 1. Block diagram of cell in the model of the functioning of a pelagic
ecosystem of the tropical regions of the ocean. 1—living elements
of the ecosystem; 2—inert elements of the ecosystem; 3—groups of
elements; 4—flow of matter; 5—trophic communications; 6—transfer
of matter; 7—solar radiation energy; 8—information communications;
9—transfer between cells; 10—ecosystem cell. k—turbulent dif-
fusion; oj—gravitational precipitation of phytoplankton and detri-
tus; T—solar radiation; C—biogenic elements; d—detritus concen-
tration; DOM—dissolved organic material; p and P—biomass and
production of phytoplankton; q—biomass of bacteria; f±—biomass of
protozoa; f.^—biomass of microzooplankton; £3—biomass of small
plant-eating animals; f4—biomass of large plant-eating animals;
S]_—biomass of cyclopoida; 82—biomass of predatory copepoda;
33—biomass of other predators.
13
-------
where n is the concentration of biogenic elements; p and b are the biomass of
phytoplankton and bacteria, respectively; fj-f4 are the various groups of
plant-eating zooplankton; 8^-33 are various groups of predatory zooplankton;
d is the concentration of detritus; t is the time; z is the depth; T is the
light intensity; a is the light absorption coefficient; h is the biogen con-
sumption coefficient in photosynthesis; Pp is the primary production*; D is
the coefficient of biogen evolution as a result of detritus (d) decomposition;
H is the evolution coefficient of biogenic elements in the metabolism process;
R-L is the nutritive ration of the i-th element; k is the vertical diffusion
coefficient; y is the natural mortality coefficient; wi and (03 are the rate of
gravitational precipitation of phytoplankton and detritus; Xj_ is the biomass
of the i-th element of zooplankton; Cjj is the partial ration of the i-th food
consumer due to the j-th food supply; H-^ is the unassimilated food of the i-th
element of zooplankton.
It is obvious that terms taking pollution into account can be intro-
duced into all the equations for phytoplankton and zooplankton. However, the
corresponding coefficients must be ex^_ ,aei tally evaluated for computer
checking of the model.
The proposed model makes it possible to predict the behavior of the
system with time: the change in biomass of the evolved elements with the de-
velopment of the system '"•'gure 2) and the change in its vertical distribu-
tion. Comparison of the calculations made according to this model with the
actual pattern observed in the ocean gives acceptable agreement (Vinogradov
et al., 1973), which permitted us to use it for experimental purposes, ex-
plaining how a change in one or another of the parameters and communications
affects the development of the system (Vinogradov et al., 1975). With the
introduction of the effect of pollution, one can obviously predict its in-
fluence on the developing system.
It is likely that models of the development of associations over the
reservoir area must have great value in evaluating the effect of pollutants.
One can easily convert to such models from that just considered by introduc-
ing transfer of the elements by currents into the scheme. We performed such
an experiment with the example of the Sea of Japan (Menshutkin, Vinogradov
and Shushkina, 1974). The cell presented in Figure 3 was taken and the sur-
face flow diagram of Figure 4, by which the transfer of plankton was accom-
plished. The motion of nekton (calmar and fish) was assumed active. After
assuming certain other initial and boundary conditions, we successfully ob-
tained the pattern of tb quantitative distribution of the studied elements
over the entire water area of the sea and its annual dynamics (in increments
of five days) (Figure 5).
opt opt
14
-------
4200
2400 -
1800 -
1200 -
600 -
60
70
80
90
100
Figure 2. Modeled changes in biomass of the various elements of the associa-
tion of the tropical regions of the ocean with their development.
Notation as in Figure 1.
15
-------
JW^SSSSSSS^^^SS^\^>^^
Figure 3. Block diagram of cell in the model of the functioning of the pelagic
ecosystem of the Sea of Japan: f^—biomass of boreal plant-eating
epizooplankton; £2—biomass of warm-water plant-eating epizooplank-
ton; f3~-biomass of interzonal plant-eating zooplankton; S]_--biomass
of the eurifag and predators of the boreal complex; 82—biomass of
the eurifag and predators of the warm-water complex,- nj_—biomass of
fish; n2—biomass of calmar; 6—water temperature; w—curr. it trans-
port; Y—effect of fishing; OJ2—seasonal vertical migrations of
interzonal plankton; 003—active migrations of fish and cephalopods.
Remaining notation as in Figure 1.
16
-------
Figure 4. Diagram of the distribution of ecosystem cells and the direction of
currents in the Sea of Japan.
17
-------
JANUARY MARCH JULY SEPTEMBER
p
Figure 5. Modeled results of the distribution of certain elements of the eco-
system over the water area of the Sea of Japan (Menshutkin, Vinogra-
dov and Shushkina, 1974): p—phytoplankton; f2~-warm-water plant-
eating zooplankton; f3—interzonal plant-eating zooplankton;
n—nekton (fish and calmar); 1—biomass greater than 10 kcal/m2;
2—10-5 kcal/m2; 3—5-1 kcal/m2; 4—1-0.1 kcal/m2. Points denote
cell centers.
18
-------
Again, it is obvious that an analogous pattern can be obtained with
consideration of the effect of pollutants. Thus, there is now the practical
possibility of including the influence of pollution into adequate determinis-
tic models describing the functioning of biosystems. It is necessary "for this
to obtain practical estimates of the suppression or stimulation of the ac-
tivity of mass hydrobionts by the basic pollutants.
THE STOCHASTIC NATURE OF POLLUTION
Let us now consider approaches to evaluating the penetration of pollu-
tants into the system and its response.
The incidence of the pollutant into the ecosystem occurs at random
moments of time and in random amounts. The sequence of incidence of the pol-
lutant is a sequence of similar events, which begin at random intervals of
time, i.e., they are a random flow. This idea is the basis of stochastic
models of the pollution of ecosystems. Another random quantity X^—the inten-
sity of the incidence—corresponds to each random value of t^—the time of the
i-th incidence of the pollutant. The set of pairs (t^, X^) is the pollution
flow.
The pollution flow gives a comprehensive characteristic of the situa-
tion, but does not carry any information about the reaction of the system to
the pollution. This reaction is composed of a biotic and an abiotic compo-
nent. For example, an oil film on the sea is dissipated as a result of the
evaporation of the oil, drifting into the atmosphere along with spray from
waves, coagulating and settling, but also as the result of the consumption by
various oil-oxidizing microorganisms and the accumulation of the oil hydro-
carbons in the body of other hydrobionts.
In this case, the pollution level is determined by the expression:
n (t) = I fi (t-tj, XH)
LJ -t- j_ j_
tilt
One can consider the reaction of an ecosystem not only to one but to
several pollutants with an additive or synergetic effect (Brusilovskiy, 1975).
Thus, the macroscopic pattern of the pollution of the ecosystem and the evolu-
tion of the pollution are described. These problems go beyond the scope of
this report, and I will not consider them in detail.
In the first part of the report, we spoke of the possibility and ne-
cessity of taking into account the effect of pollution on the functioning of
an ecosystem, about the microscopic pattern of pollution. The method of this
accounting can be demonstrated with an example of the clearly simplified model
of the Ferchulst-Pirl population:
|| = X[E(6) - Y(9)x] ,
n
where x(t) is the population at time t , (t) is some parameter symbolizing
the state of the medium at time t, E(9) is the natural (inherent) rate of
19
-------
increase of the population, J(Q) is the coefficient of self-suppression of the
population growth. If the pollution process forms a pollution level of the
ecosystem n (t), then the model of the population, the component of the ecosys-
tem subjected to pollution, can be described as:
n(t) = I f(t-t ^
ti
-------
"graphical" models of ecosystems and predicting their development.
2. to develop mathematical models taking into account the stochastic
nature of pollution and the possibility of controlling its extent. Some fun-
damental aspects of the development of this direction are discussed in the re-
port of V. D. Federov.
Note: The illustrations to this report are taken from: Vinogradov,
M. E., and V. V. Menshutkin. 1976. The Modelling of Open Sea Ecosystems.
Vol. 6, The Sea.
REFERENCES
Brusilovskiy, P. M. 1975. Model of Pollution in an Ecosystem.
Federov, V. D. 1975. Problems of maximum permissible effects of the anthro-
pogenic factor from an ecologist's point of view. Report.
Lyapunov, A. A., and S. V. Jablonsky. 1963. Theoretical problems of cyber-
netics. Problems of Cybernetics, No. 9.
Menshutkin, V. V., M. Ye. Vinogradov, and E. A. Shushkina. 1974. A mathe-
matical model of a pelagic ecosystem in the Japan Sea. Okeanologiya,
Vol. 14.
Odum, H. T. 1972. Environment, Power and Society. Wiley-Interscience, New
York.
Vinogradov, M. Ye., J. J. Gitelzon, and Ju. J. Sorokin. 1970. The vertical
structure of a pelagic community in the tropical ocean. Marine Biol.
6(3).
Vinogradov, M. Ye., V. F. Krapivin, B. S. Fleishman, and E. A. Shushkina.
1975. Utilization of a mathematical model for analyzing the behavior of
an ocean pelagic ecosystem. Okeanologiya 15(2).
Vinogradov, M. Ye., V. F. Krapivin, V. V. Menshutkin, Ye. S. Fleishman, and
E. A. Shushkina. 1973. A mathematical model of the functioning of a
pelagic ecosystem in tropical areas of the ocean. Okeanologiya 13(5).
Vinogradov, M. Ye., V. V. Menshutkin, and E. A. Shushkina. 1972. On mathe-
matical simulation of a pelagic ecosystem in tropical waters of the ocean.
Marine Biol. 16(4).
21
-------
AMERICAN METHODS FOR MEASURING PHYTOPLANKTON
PRODUCTION IN THE OPEN OCEAN1
by
Edward J. Carpenter and Wayne E. Esaias
Marine Sciences Research Center
State University of New York
Stony Brook, New York 11794
INTRODUCTION
This paper summarizes the basic methodology for measuring primary pro-
duction in the open ocean. There are other excellent publications which de-
scribe production measurement, most notably that of Strickland and Parsons
(1972). Our publication supplements these already existing manuals and makes
note of most recently discovered inadequacies of the 1'*C technique. Problems
include those of 1>fC isotope quality, high dark-bottle counts, scintillation
fluors, loss of C when drying filters, etc. We also present, where possible,
ways of overcoming these problems on the basis of our experience and that of
others. The methods described here are designed for primary production mea-
surement in the open ocean where there is a low concentration of particulate
matter, high light penetration, and a relatively constant concentration of
dissolved inorganic carbon. With modification, as noted by Strickland and
Parsons (1972), the methodology presented here can be adapted to use in rela-
tively low salinity water of high detrital content.
METHODOLOGY
A. SAMPLING DEPTHS AND LIGHT TRANSMISSION
IN THE EUPHOTIC ZONE
Samples for production measurements should be collected from at least
five depths based on irradiance values relative to that at the sea surface.
The exact irradiance levels are chosen to agree with the transmittance of
neutral-density screens used for shipboard incubation, but should be approxi-
mately 100, 50, 25, 10, and 3-1% of surface (0 meters) irradiance. Strong
chlorophyll maxima should also be sampled.
Supported by NSF Grant DES76-01405. Contribution No. 146 from the
Marine Sciences Research Center, State University of New York at Stony Brook.
22
-------
Irradiance values should be measured with a cosine-corrected detector
n
which is equally sensitive to either energy flux (ywatts cm" ) or quantum flux
(einsteins cm" sec" ) over the entire photosynthetically active region of the
spectrum (350-700 nm). The LAMBDA Instruments Corp. (Lincoln, Nebraska 68504)
markets a suitable underwater quantum sensor and meter that is calibrated in
terms of yeinsteins m~2sec~ . Regardless, the manufacturer's type and the
spectral response of the underwater detector should be clearly stated.
We recommend that the underwater measurements be performed in conjunc-
tion with a similar unit mounted on gimbels, in a position free from shadows
and strong reflections, and preferably during the middle of the day. If light
attenuation measurements are made in the early morning or late evening, then
the attenuating coefficients may be overestimated by as much as 25%, due to
sun angle effects. The underwater unit should be suspended as far outboard as
possible on the sunny side of the ship and should be maintained in a vertical
position.
Measurements taken over a series of depths are normalized to a constant
value of the deck unit and are plotted on 3-cycle semi-log paper to obtain a
good estimate of the diffuse attenuation coefficient K (m~ ):
K^=^^ " • lzi
IT7 _ "7 -*-v*3 -r
; \ -70 1 7
^ 1 Z2 2 1 Z2
Where Z = depth in meters, I = irradiance, and y = specific relative irradi-
ance. Depths of specific relative irradiances are then calculated:
4.605 - £n Y%
Y% K
or determined graphically.
In the event that a submarine photometer is not available, a Secchi
disc may be used to provide a very rough estimate of the diffuse attenuation
coefficient. The general formula for determining the attenuation coefficient
(K) from Secchi disc depth (D) in meters is:
*-¥
Values of K estimated from Secchi disc depths may easily be in error
by 25%, depending on a variety of factors including the amount of backscatter-
ing of light by surface water and roughness. We suggest that it be on hand
but used only as an emergency measure if the submarine photometer fails.
B. WATER SAMPLER
It is important that seawater be collected with a non-toxic sampler.
In no case should metal come in contact with water to be used in production
measurements. A sampler such as the opaque-plastic Niskin bottle is ideal
since it has no exposed metal. Also, it is darkened and will prevent exposure
of phytoplankton to high light intensities as it is brought aboard ship.
23
-------
Niskin bottles should be secured in a shaded area on ship. When water is
drawn to fill glass productivity bottles, larger zooplankton should be removed
by filtration through a plastic funnel that contains 200 ym mesh Nitex netting.
Glass bottles should be covered with a black cloth bag to shade phytoplankton
before they are placed in incubators.
C. INCUBATORS
To optimize expensive ship time, it is desirable that incubations be
carried out on ship and not in situ. Actually, little bias appears to be in-
duced by incubating samples under neutral density screens as opposed to an in-
situ incubation. Holmes (1968) compared the in-situ (in seawater) method with
productivity samples held under fluorescent light or natural sunlight with
shading by neutral density filters. He concluded that the latter two methods
gave unbiased estimates of actual in-situ production.
We recommend the simultaneous use of two types of incubation. One set
of productivity bottles should be held in an artificial light box with shading
by neutral density screens. This system allows the incubation of samples
under a known, and constant, light source. A second set of bottles is incu-
bated on deck in sunlight. Incubation in sunlight allows a measure of the de-
gree of inhibition of photosynthesis from ultraviolet and high light intensity
which is not possible under fluorescent light. It also more closely approxi-
mates natural conditions.
As mentioned previously, in short-term experiments there seems to be
no bias induced by using neutral filters rather than colored filters which
would duplicate the light found at depth. However, it should be noted that in
long-term incubations the quality of light can affect the nature of the photo-
synthate (Wallen and Geen, 1971). Recently Shimura and Ichimura (1973) mea-
sured photosynthesis in the northwestern North Pacific phytoplankton under
blue, white (fluorescent), red and green light for 3-hour incubations. They
noted that the photosynthetic efficiency in green light was 80 to 90%, and
that of blue light was 105 to 115% of white (fluorescent) light, respectively.
Gray-plastic window screening, which is available at most hardware
stores, can be used to attenuate light. We have observed the following rela-
tionship between layers of screen and light attenuation:
Layers
0
1
2
4
6
9
% Light
100
60
40
16
6
1.5
Due to variations in manufacture of screening, investigators must
check the light reduction characteristics of their own screens. To diffuse
light entering the bottles, one layer of a thin-white, translucent-plastic
sheet can be placed between the screening and the Plexiglas tubes.
24
-------
Artificial Light Incubator
Primary productivity estimates frequently must be made during overcast
conditions. In order to make meaningful inferences about the average produc-
tivity of an area, variations brought about by changes in daily irradiance
must be taken into account. The extrapolation of productivity estimates made
at low irradiance levels (as on cloudy days) to high irradiance levels (as on
sunny days) is difficult because the productivity-irradiance relationship is
curvilinear and because acclimation can result in varying slopes and satura-
tion intensities within populations. However, it is possible to make compara-
tive measurements using artificially illuminated incubation boxes which expose
the samples to a constant, saturating irradiance. This is the major reason
fdr incubating samples in an artificial light box.
The use of "daylight" fluorescent bulbs gives a closer approximation
to the spectral quality of natural daylight than incandescent bulbs. The lat-
ter emit predominantly red light, but usually produce a lower overall irradi-
ance level. It is very important to include a description of the light source
as well as the irradiance experienced by the samples. This irradiance should
be measured with a detector of the type described above. The entire artifi-
cial illumination incubator should be screened from unwanted sun and labora-
tory light.
Artificial illumination incubators also permit one to measure the po-
tential productivity of samples collected at all hours of the day. However,
care must be taken to account for diel rhythms in light arid dark COa uptake
capacity if these results are to be used to estimate the productivity of a
given area. For example, at one station occupied over a 24-hour period from
noon to noon, we found that nighttime light bottle uptake, as measured in an
artificial illumination incubator, was twice that of daytime values. Dark
bottle uptake increased by an order of magnitude during the dark period. The
variation in chlorophyll a concentration was less than 20% during the 24-hour
period.
Probably the simplest type of incubator suitable for artificial light
incubation is that based on a design by John Ryther (personal communication).
This consists of Plexiglas tubes covered with various neutral density screens
to attain the desired light attenuation. The glass productivity bottles are
slipped into the tubes, and ends of tubes are sealed with rubber stoppers.
Each rubber stopper contains a piece of hollow plastic pipe for the passage of
seawater from the ship's deck-wash system. The whole assembly is mounted at a
45° angle to a bank of fluorescent lights. Cooling seawater flows in from the
bottom and out the top and is distributed to the tubes from a manifold arrange-
ment. Temperature of the flowing seawater should checked regularly to assure
that it is the same as that of surface water.
Other incubator designs can be used to obtain good production data.
For comparative purposes, we illustrate the incubator designed by Fee (1973)
in Figure 1. Fee successfully used this incubator in a production study on
Lake Michigan. Samples were incubated for 4 hours and agitated by rotation.
In the open ocean if the incubation times are short (ca <4 hr) the bottles
need not be rotated.
25
-------
MOTOR BED
MOTOR
SHAFT SUPPORT
TRANSMISSION
TRANSMISSION
SHAFT
WATER BATH
*S
WATER BATH
»2
ROTATING WHEEL
BOTTLES WITH
-N WATER SAMPLES
ROTATING WHEEL
SHAFT
WATER BATH
'LIGHT THROUGH WINDOW
ILLUMINATING
UNIT
Figure 1. Incubator used by Fee
in Lake Michigan
(1973) for measuring phytoplankton production
The light source for this incubator was a bank of six 500-w T3Q tung-
sten iodine vapor bulbs. Each bulb was backed with a paraboloid reflector of
brightly buffed aluminum to direct maximum amount of light toward the experi-
mental changers. Tungsten-iodine bulbs emit very high irradiances and have a
very long life with little change in the quality of light emitted. The heat
released by the bulbs was largely removed by placing a rectangular Pyrex box
directly in front of the first chamber. The box was 5 cm wide and was filled
with rapidly circulated cold water.
Ambient Light Incubator
The outdoor incubator which is exposed to natural sunlight can be con-
siderably simpler than the artificial light incubator. It is possible to
simply place bottles in a rectangular box with five compartments in line.
26
-------
There are holes in partitions between compartments so that seawater from the
ship's deck-wash can flow in one end of the box and out the other. Each com-
partment is covered with a different number of neutral-density screens to at-
tain the desired light attenuation. The box is shallow so that there is little
attenuation of UV light as a result of passage through seawater. Compartments
receiving 100% of surface light contain plastic bottles which permit passage
of UV light. Those with 60% light and less contain glass bottles which absorb
UV. This system approximates the depthwise penetration of UV light in the sea.
Bottles with phytoplankton exposed to UV light could be made of quartz
glass, but these would be expensive. Instead of quartz, bottles could be made
of acrylic plastic. Ordinary Plexiglas will not transmit UV light, but those
made of polymethyl-methacrylate will allow passage. This is available as 3 mm
thick PMMA Shinkolite A acrylic plastic by Mitsubishi Rayon, Japan. In using
this plastic with shallow bottles (90 mm dia. by 25 mm deep), Ilmavirta and
Hakala (1972) observed that production in near-surface waters was reduced by
50% as compared with measurements made in Jena glass which excluded UV light.
Tests were carried out in a lake in Finland.
The optical attenuation of the neutral density screening should be
checked by placing the underwater detector within each chamber. The inside of
the chamber should be painted with flat-black paint to prevent reflections
from the walls and bottoms, or flat-dark blue to simulate the small contribu-
tion of upwelled light in situ. The chambers should be sufficiently wide to
prevent shadows from the wall from falling on the bottles. Alternatively,
screening can be placed around individual bottles that are placed within a
lucite box of flowing seawater. The bottoms of this box should be darkened
for reasons cited above. The deck units should be situated in a position free
from shadows and strong reflections.
Continuous irradiance measurements should be made during the incuba-
tion period to express production as a function of light exposure (irradiance
x time). A relatively inexpensive, self-contained solar recorder is made by
the Belfort Instrument Company of Baltimore, Maryland. Output from the solar
recorder can be compared with the submarine photometer, which has an output of
einsteins cm" sec"1, to obtain very rough calibrations.
D. DURATION OF INCUBATION
In the ambient and artificial light incubator, bottles should be ex-
posed no more than 3 hours. In the sea, phytoplankton move from one light in-
tensity to another as they drift with currents. Thus it would be unnatural to
incubate them for relatively long periods at one light intensity. In clear,
open ocean water, light penetration is quite deep and a 3-hour incubation of
these phytoplankters at one light intensity is probably not unnatural. How-
ever, in some estuaries phytoplankters may regularly move from the 1% to 100%
light intensity in a matter of minutes. Prolonged incubation at one intensity
might yield a biased result.
There are other reasons for avoiding long incubations. For example,
Vollenweider and Nauwerck (1961) have observed that the production sum of
27
-------
three or four short incubations exceeds the total of one long exposure. Also,
Harris and Lott (1973) have shown that long exposures of phytoplankton to high
light intensities leads to an increase in photorespiration. Finally, another
reason for avoiding long exposures is that glass ampules, the usual method for
storing 14C, bicarbonate, can add sufficient silica to stimulate production
over long incubation periods (i.e., 24 to 28 hours). Gieskes and van Bennekom
(1973) noted up to 1 mg-at liter"1 of dissolved silica in lkC stocks stored in
distilled water in glass ampules. They also noted that if incubation times
are relatively short there is no simulation in productivity as a result of
silica enrichment. All of these considerations indicate that duration of in-
cubation should be as short as possible.
E. 14C BICARBONATE STOCKS
We and others have experienced problems with certain manufacturers
concerning the quality of ampulated C stocks. Some have contained toxic ma-
terial, others particulate matter (Morris et al., 1971), and all are likely to
contain high silica concentrations (Gieskes and van Bennekom, 1973) . We sug-
gest that 14C bicarbonate stocks should be purchased as a crystalline solid
and ampulated by the investigator. An alternative, as suggested by Morris et
al. (1971), is to pool the lkC from all the ampules. The pooled 14C should
then be UV-irradiated to remove organic matter that is associated with lkC.
Additionally, a "zero time blank" should be made by adding the normal concen-
tration of "*C to a 125 ml glass incubation bottle containing unfiltered sea-
water (Morris et al., 1971). The bottle contents are immediately filtered and
the activity of this filter is substituted for dark-bottle uptake. Morris et
al. (1971) recommended that this be used as the standard dark uptake correc-
tion. We agree, but only if the uptake in normally incubated dark bottles ex-
ceeds 10% of the uptake in light bottles.
If the investigator chooses to dilute and ampulate his own crystalline
14C, then care should be taken that the samples are clean and free of particu-
late matter. Sodium chloride should be added to distilled water to give a sa-
linity of 35 ppt and the pH should be adjusted to 9.5 with dilute NaOH solu-
tion before adding crystalline 14C to the solution. Aliquots of the radio-
active solution are added to samples which should be immediately sealed and
autoclaved. To test for improper sealing, the ampules can be placed top down
in a solution of dye while being autoclaved. If not sealed, ampules will sink
after autoclaving and dye will be present within them.
One convenient way to prepare the 14C solution is to anticipate how
many samples will be required per station and then place the total concentra-
ted amount as, say, 5 ml, in one ampule. At the station, this volume can be
diluted with filtered seawater, and one ml of the diluted stock solution can
be added to each bottle with an automatic pipet. This system saves time and
money since glass ampules are expensive, and insures that a uniform amount of
14C is delivered to each bottle.
Generally, 10 to 20 uCi of 11+C bicarbonate should be added to each
bottle in oceanic waters and from one to 5 yCi in estuarine regions.
28
-------
14
F. DETERMINATION OF C UPTAKE
Many difficulties regarding self-absorption and standardization of ll*C
stocks arise when the thin-window gas-flow beta is used. The scintillation
counter method allows relatively easy standardization of stocks and efficient
counting of samples (Schindler, 1966).
After incubation, the entire contents of a 125 ml bottle should be im-
mediately filtered. Caution should be observed if formaldehyde is added to
the bottle to terminate carbon uptake. Ilmavirta (1974), working in Finnish
lakes, observed an average 21% decrease in the activity of samples only five
minutes after the addition of formaldehyde. However, he did not state whether
the formaldehyde was buffered or how much was added.
Arthur and Rigler (1967) have observed that, as the volume of sample
filtered ;ncreases, phytoplankton are damaged and cell contents are lost
through ?. filter. The authors used 25 mm diameter Millipore® HA filters
with a me tium pressure differential of 300 mm Hg. However, MaMahon (1973)
noted tha the retention capacity for 1'*C by the filter, expressed as radio-
activity per ml filtered, is maximum for small samples (£ 1 ml) and decreases
to a constant value when samples are larger than 100 ml. This is attributed
to absorption in the filter and probable retention of 1*C bound to unknown sub-
stances in the water or on the filter, which are eluted or exchanged by pas-
sage of volumes of water of 100 ml or more. The error can simply be corrected
by washing the filter with a few ml of nonradioactive filtered seawater before
and after the bottle contents are filtered. Allen (1971) recommends against
washing with dilute HC1 since acidification removes a significant fraction of
previously incorporated carbon.
Another problem with the Geiger-Mueller counting method is that phyto-
plankton penetrate the filters, thus increasing problems of self-absorption
(Theodorsson, 1975). This problem is also eliminated if the scintillation
counting method is used.
Yet another apparent error was demonstrated by Wallen and Geen (1968)
who noted that desiccation of algae in filters before addition of filters to
vials containing scintillator fluor could lead to losses of up to 50% of ac-
tivity. In spite of this observation, Lind and Campbell (1969) recommended
drying the filters since good water-accepting scintillation fluors were not
available. Ward and Nakanishi (1971, 1973) found that the loss of 1'*C during
desiccation increases with decreasing incubation light intensity. They com-
pared Geiger-Mueller with liquid scintillation methods for counting filters
and found that, when production was calculated on an aerial basis, results of
the latter method generally were 25-40% greater. A good water-accepting
naphthalene-dioxane fluor was developed by Schindler and Holmgren (1971) for
use with samples from freshwater lakes. However, its use might be limited in
saltwater, since any variation in the salt retained by the filters can cause
counting to be erratic (Pugh, 1973) . The xylene-based fluor, Aquasol®, can be
used, but it is costly. If Aquasol is used to count bicarbonate standards,
phenethylamine should be added to 10-20% v/v final concentration to eliminate
loss of lkC due to the low pH of the cocktail (Iverson et al., 1976). This
precaution can prevent errors up to 40% in knowledge of the initial activity.
29
-------
In addition, the high chemiluminescence of the Aquasol® system must be consid-
ered when counting plankton samples. This potential error^can be eliminated
by counting samples only after they have remained in the fluor for about a
week. The *4C counting methods of Pugh (1970, 1973) seem to offer the most
reliable results in assaying 1I+C retention by marine phytoplankton. Pugh
(1973) suggests the use of cellulose nitrate (Sartorius Membrane-filter MF 125)
filters or PVC filters since they decompose more completely in a toluene fluor
than do cellulose ester filters. Pugh (1973) obtained best results when he
used 2-methoxyethanol-toluene fluor. This consists of 1:2 v/v 2-methoxyethanol:
0.5% Butyl PBD in toluene. Using the cellulose nitrate filters and the toluene-
based fluor, Pugh (1973) did not observe any apparent loss of radioactive ma-
terial as the filtered volume was increased. Further, there was no loss of ac-
tivity when the filters were dried and stored overnight at 70°C. Also, wet
filters could be added to the fluor if necessary and counts could be made after
several weeks with no loss of activity. The scintillation cocktail Aquasol®
was not tested by Pugh (1973).
In using the scintillation counter, Pugh (1970, 1973) recommends the
use of the filter standardization method for counting intact filters. Cells
and membrane filters have a self-absorptive effect that is not corrected by ex-
ternal standardization. It is especially important to standardize by this
method since phytoplankton are known to penetrate membrane filters rather than
lie on the surface as a thin film (Theodorsson, 1975). The filter standardiza-
tion method, however, is only accurate when the weight of algae on the filters
is small (<1 mg dry wt). It would be highly unusual to have more than one mg
of algae material in 125 ml of oceanic water. The techniques involves con-
struction of a quench curve relating channels ratios to counting efficiency.
Scintillation counts are recorded in two channels. One channel counts only
weak 3 activity, while the other channel records all the activity in the C 3
spectrum. The ratio of counts in the first channel to those in the second is
called the channels ratio. As described by Pugh (1973) , the investigator first
filters natural phytoplankton and assembles several filters, each with a dif-
ferent amount of plankton. The volume filtered may vary from one ml to one
liter. Next, approximately 0.1 ml of a 14C sucrose solution is micropipetted
onto the filters and then allowed to dry. Sucrose is used because it is vir-
tually insoluble in the toluene fluor and will remain on the filters. Filters
are placed in scintillation vials, completely covered with fluor solution (10
ml of 0.5% Butyl PBD in AR Toluene), and counted. As described, counts were
recorded on two channels set for optimum of a homogeneous solution. From
these counts the channels ratios were calculated. The specific activity of
the original llfC sucrose solution is determined by counting 0.1 ml samples in
homogeneous solution in a toluenerethanol (3:1) fluor; from the computed chan-
nels ratio, the specific activity can be calculated from the standard curve
for the scintillation counter. The result is a quench curve (Fig. 2) that
takes into account the self-absorptive properties of phytoplankton and filter.
An alternative to the filtration technique of removing phytoplankton
from unassimilated llfC is bubbling (Schindler et al. , 1972). In this tech-
nique, about 20 ml from the incubation bottle is pipetted into a 30 ml tube
with fine-fritted glass at the base (Fig. 3). The pH is lowered to between 3
and 3.5, and air is bubbled through the column for about 20 minutes. This re-
moves inorganic 14C from solution, and the remaining lkC is that contained in
30
-------
the cells or which has been excreted and is organically bound. Next the in-
vestigator removes 2.5 ml from each filter tube and places it in a scintilla-
tion vial with fluor for counting. Since there is no filter, there is no ab-
sorption problem and the external standardization channels ratio can be used
for calibration. The beauty of this technique is that there is no problem
with cell rupturing when filtering. At times, Schindler et al. (1972) noted
that lkC uptake with the bubble method was twice measured by filtration.
Figure 2.
(b)
CHANNELS RATIO
Channels ratio calibration curves. Filter standardization curve
(a) calculated as a second order polynomial from 220 points
(r=0.970). Channels ratio curve (b) is the standard curve for
counting in a homogeneous solution. From Pugh (1973).
However, several workers have indicated that this method yields highly
variable results. IT: is possible that some fraction of labeled excreted or-
ganic material is hydrolyzed and converted to '*CO2, which may be purged from
the solution under these conditions and is therefore net counted (Iverson, per-
sonal communication), leading to underestimates of phytoplankton carbon fixa-
tion if excretion is significant.
G. DARK UPTAKE
The problem of high uptake of lltc in dark bottles has been encountered
by several workers. Generally, dark uptake of 1'*C should not exceed 10% of
uptake in the light. However, Holmes (1968) reported that in the open Pacific
dark uptake occasionally equalled uptake in the light. Holmes had no adequate
explanation for high dark-bottle uptake but thought that perhaps an occasional
ampule containing particulate radioactive material may have been the cause.
High dark-bottle uptake was also encountered by Steven (1971) while working in
the western tropical Atlantic. High dark-bottle counts occurred through the
year, often exceeded light-bottle uptake, and were present at all depths sam-
pled. The problem was solved by making all collections at sunrise and com-
pleting the experiments with minimum delay. When this procedure was adopted,
dark-bottle counts dropped to less than 6% of counts in light bottles.
31
-------
Figure 3. Apparatus used to drive off inorganic
ments. From Schindler et al. (1972).
C in productivity measure-
Morris et al. (1971) observed in the field and in laboratory cultures
that high "*€ uptake in dark bottles was associated with low phytoplankton
cell densities. For example, Figure 4 shows that the ratio of light carbon
dioxide fixation to that in the dark is high in relatively dense phytoplankton
culture. At lower cell densities, dark-bottle fixation increases in relation
to light-bottle uptake until uptake in both light and dark are equal. Morris
et al. (1971) had no adequate explanation for this phenomenon. Until more is
known concerning the mechanism of dark fixation, they recommend that it be ig-
nored. In this case the blank would be the time-zero control as described
under the heading "*^C bicarbonate." At present, there seems to be no adequate
explanation for high dark-bottle counts. Perhaps this actually represents car-
bon incorporation by bacterioplankton (Sorokin, 1971) , although this theory
has been criticized (Banse, 1974) . At present we must agree with Yentsch
(1974) who, in speaking of high-dark uptake, stated, "This is another one of
those indefinable mysteries which pervade the discipline."
32
-------
LU
£t
o
z
z
o
X
u.
UJ
a
X
o
0
z
0
CD
tr
o
i-
X
o
_J
L
z
0
X
"-
UJ
a
x
o
a
z
o
ID
cc
o
/
•/,
•
.
.—
^*-r-«
•T^ i i i i i
in1 in2 in3 in4 in5 me
<
IT
CELL NUMBER PER ML
Figure 4. Relationship between cell number per milliliter (log scale) and the
ratio of light COz to dark COz fixation in cultures of Dunaliella
tertiolecta and Phaeodactylum tricornutum. From Morris et al.
(1971).
H. CALCULATION OF PRODUCTION
The basic formula for calculating marine primary production has been
presented by Strickland and Parsons (1972). This is:
mg C/liter/hr =
R - Rx W x 1.05
R x N
In the formula, 1.05 is a factor used to correct for the fact that the heavier
1'*C isotope behaves differently from the 12C found in nature. N equals the
incubation time in hours. W is the weight of inorganic carbon available to
the algae. Total lUC added per bottle is represented by R. Carbon-14 retained
by the light- and dark-bottle filters are RS and R^, respectively. The for-
mula should be modified if the acidification-bubble technique is used or if
only part of an incubation bottle is filtered. It is assumed that there is
about 90 mg COz liter" , and in oceanic waters this would range between 85 and
103 mg liter"1. For measurement of available COz in low salinity waters, one
can use the techniques described by Strickland and Parsons (1972) or Rao
(1965) .
As Fee (1969, 1973) has pointed out, the interpretation and expression
is difficult. Because of problems of extracellular excretion and respiration,
it is necessary to confine incubation times to short intervals. Yet, daily
33
-------
rates are needed for a reliable estimate of annual or seasonal production.
There are problems in extrapolating rates measured in natural sunlight in sunny
weather to cloudy days. A computer model has been constructed by Fee (1969,
1973) to allow the calculation of total daily photosynthesis. The model is
available in Fortran IV language (Fee, 1971). As inputs to the model, the fol-
lowing are required: (1) shape of the curve relating photosynthesis to irra-
diance (P vs I curve), (2) highest photosynthetic rate, (3) extinction coeffi-
cient of seawater at sample site, and (4) time variation of photosynthetically
active illumination entering through the water surface over the interval of
interest, usually a day. The model does have some limitations. For example,
it does not take into account the "afternoon depression" of photosynthesis.
However, it has the advantage of allowing a large number of stations to be
processed rapidly, and it appears to closely approximate in-situ production.
The quantum yield approach developed by Bannister (1974) for nutrient-
saturated layers may represent a significant advancement but has not yet been
applied to oceanic situations and again does not adequately treat photo inhi-
bition or the afternoon depression.
In summary, it is apparent that there have been numerous technical
problems associated with the 1!*C light-dark bottle method. However, recent
advances, mentioned in the text, can alleviate many of these biases. Most of
the errors noted in the literature lead to underestimates of oceanic produc-
tion. As a guess, it would appear that, due to these biases, actual oceanic
production)may be underestimated by as much as a factor of 2 or 3. Further
research and improvements in the lf*C light-dark bottle method by both Soviet
and American oceanographers will undoubtedly yield a more accurate estimate of
primary production in the open sea.
REFERENCES
Allen, H. L. 1971. Primary productivity, chemo-organotrophy, and nutritional
interactions of epiphytic algae and bacteria on macrophytes in the littoral
of a lake. Ecol. Monogr. 41:97-127.
Arthur, C. R. , and F. H. Rigler. 1967. A possible source of error in the lf|C
method of measuring primary productivity. Limnol. Oceanogr. 12:121-124.
Bannister, T. T. 1974. Production equations in terms of chlorophyll concen-
tration, quantum yield, and upper limit to productions. Limnol. Oceanogr.
19:1-12.
Banse, K. 1974. On the role of bacterioplankton in the tropical ocean. Mar.
Biol. 24:1-5.
Fee, E. J. 1969. A numerical model for the estimation of photosynthetic pro-
duction integrated over time and depth in natural waters. Limnol.
Oceanogr. 14:906-911.
Fee, E. J. 1971. Digital computer programs for estimating primary production,
integrated over time and depth. Univ. Wis.-Milwaukee Cent. Great Lakes
Stud. Spec. Rept. No. 14. 42 pp.
34
-------
Fee, E. J. 1973. A numerical model for determining integral primary produc-
tion and its application to Lake Michigan. J. Fish. Res. Bd. Canada
30:1447-1468.
Gieskes, W. W. C., and A. J. van Bennekom. 1973. Unreliability of the ll*C
method for estimating primary productivity in eutrophic Dutch coastal
waters. Limnol. Oceanogr. 18:494-495.
Harris, G. P., and J. N. A. Lott. 1973. Light intensity and photosynthetic
rates in phytoplankton. J. Fish. Res. Bd. Canada 30:1717-1778.
Holmes, R. W. 1968. Description and evaluation of methods for determining
incident solar radiation, submarine daylight, chlorophyll a, and primary
production. Spec. Sci. Rept.-Fisheries No. 564, U.S. Dept. Interior.
31 pp.
Ilmavirta, V., and I. Hakala. 1972. Acrylic plastic and Jena glass bottles
used in measuring phytoplanktonic primary production by the l^C method.
Ann. Bot. Fennici 9:77-84.
Ilmavirta, V. 1974. Diel periodicity in the phytoplankton community of the
oligotrophic lake Paajarvi, southern Finland. I. Phytoplanktonic primary
production and related factors. Ann. Bot Fennici 11:136-177.
Iverson, R. L., H. R. B. Haker, and V. B. Myers. 1976. Loss of carbon-14
activity in direct use of Aquasol for standardization of solutions con-
taining MC-NaHCOa. Limnol. Oceanogr. (in press).
Lind, 0. T., and R. S. Campbell. 1969. Comments on the use of liquid scin-
tillation for routine determination of 14C activity in production studies.
Limnol. Oceanogr. 14:787-789.
McMahon, J. W. 1973. Membrane filter retention—a source of error in the ^C
method of measuring primary production. Limnol. Oceanogr. 18:319-323.
Morris, I., C. M. Yentsch, and C. S. Yentsch. 1971. Relationship between
light carbon dioxide fixation and dark carbon dioxide fixation by marine
algae. Limnol. Oceanogr. 16:854-858.
Pugh, P. R. 1970. Liquid scintillation counting of 14C-diatom material on
filter papers for use in productivity studies. Limnol. Oceanogr. 15:
652-655.
Pugh, P. R. 1973. An evaluation of liquid scintillation counting techniques
for use in aquatic primary production studies. Limnol. Oceanogr. 18:
310-318.
Rao, D. B. S. 1965. The measurement of total carbon dioxide in dilute tropi-
cal waters. Aust. J. Mar. Freshwater Res. 16:272-280.
Schindler, D. W. 1966. A liquid scintillation method for measuring carbon-14
uptake in photosynthesis. Nature (London) 211:844-845.
35
-------
Schindler, D. W., and S. L. Holmgren. 1971. Primary production and photo-
plankton in the experimental lakes area, Northwest Ontario, and other low-
carbonate waters, and a liquid scintillation method for determining ll*C
activity in photosynthesis. J. Fish. Res. Bd. Canada 28:189-201.
Schindler, D. W., R. V. Schmidt, and R. A. Reid. 1972. Acidification and
bubbling as an alternative to filtration in determining phytoplankton pro-
duction by the 1I+C method. J. Fish. Res. Bd. Canada 29:1627-1631.
Shimura, S., and S. Ichimura. 1973. Selective transmission of light in the
ocean waters and its relation to phytoplankton photosynthesis. J. Oceanogr.
„ Soc. Japan 29:31-40.
Sorokin, Y. I. 1971. Bacterial populations as components of oceanic ecosys-
tems. Mar. Biol. 11:101-105.
Steven, D. M. 1971. Primary productivity of the tropical western Atlantic
Ocean near Barbados. Mar. Biol. 10:261-264.
Strickland, J. D. M., and T. R. Parsons. 1972. A practical handbook of sea-
water analysis. Bull. 167 (2d ed.) Fisheries Research Board of Canada,
Ottawa. 310 pp.
Theodorsson, p. 1975. The study of 1!*C penetration into filters in primary
production measurements using double side counting. Limnol. Oceanogr.
20:288-291.
Vollenweider, R. A., and A. Nauwerck. 1961. Some observations on the lkC
method for measuring primary production. Verh. Int. Verein. Limnol.
14:134-139.
Wallen, D. G., and G. H. Geen. 1968. Loss of radioactivity during storage of
llfC labeled phytoplankton on membrane filters. J. Fish. Res. Bd. Canada
25:2219-2224.
Wallen, D. G. , and G. H. Geen. 1971. Light quality in relation to growth,
photosynthetic rates and carbon metabolism in two species of marine plank-
ton algae. Mar. Biol. 10:34-43.
Ward, F. J., and M. Nakanishi. 1971. A comparison of Geiger-Mueller and
liquid scintillation counting methods in estimating primary productivity.
Limnol. Oceanogr. 16:560-562.
Ward, F. J., and M. Nakanishi. 1973. A comparison of liquid scintillation
and Geiger-Mueller estimates of primary productivity in an in situ experi-
ment. J. Fish. Res. Bd. Canada 30:708-711.
Yentsch, C. S. 1974. Some aspects of the environmental physiology of marine
phytoplankton: a second look. Oceanogr. Mar. Biol. Ann. Rev. 12:41-75.
36
-------
RELATIVE ABUNDANCE OF SYMPATRIC SPECIES AND MODEL
OF EXPONENTIALLY BROKEN ROD (EBR)
by
V. D. Fedorov
Head, hydrobiology, Moscow State University, USSR
In 1957 Robert MacArthur (MacArthur, 1957) considered three alternate
versions of the possible distribution of N individuals among w species, and
demonstrated good agreement between the distribution of trophic birds and one
of these versions. This case corresponds to the initial assumption that in a
relatively homogeneous biotope, groups of closely related species which are
comparable by size and physiology retain a stable population ratio, i.e.,
their relative abundance turns out to be constant and can be predicted by
some model.
Formally, this case is analogous to a segment of a rod of unit length
(corresponding to the biotope) which is partitioned into w parts by w-1 points
falling randomly on it. At the points of impact, the rod is broken into parts
whose lengths are proportional to the population sizes of the individual spe-
cies, so that the mean sizes of the parts form the same ratios as the numbers
+ +-_ + -, + ... +
w w-1 w w-1 w-2
, ,
w w w-1 w w-1 w-2 w w-1
The expected abundance of the r-th species among N individuals and w
species can be predicted using the formula
r
/s N v 1 ... .
nr - w
where nr is the length of the segment having the number r in the sequence of
populations ranked by size from 1 to w. Biologically, this model, named the
"broken rod model" by biologists (henceforth referred to in this article as
the "B.R. " model), corresponds to the initial assumption that in a system in
equilibrium the niches of sympatric species are contiguous without consider-
able overlap.
In the same year, G. Hutchinson (Hutchinson, 1957) , using the concepts
of set theory, determined elegantly the fundamental niche. He also demon-
strated that the non-overlapping segments predicted by the B.R. model have a
great deal in common with the concept of a fundamental niche reflecting the
total ecological requirements of the species.
37
-------
The non-overlapping of niches in higher dimensional space discovered
by MacArthur presupposes unequivocally the absence of interspecies competition,
since otherwise, according to the Gauze-Volterra axiom, the relative abundance
of species could not be constant, i.e., they could not form population size
ratios which are stable over time in a homotypic association (what is meant is
similarity in regard to type of feed). Moreover, species in the same trophic
grouping are forced under similar conditions to a similar type of existence,
which in turn inevitably leads either to a local crowding out of species, or a
modification of the ecological requirements of one or perhaps even all species
in ways which will tend to reduce competition among them. Feed selectivity,
different resistance to variations of different variables in the biotope, and
finally, rivalry for shelter refine the relationship between partners and thus
turn out to be the causes responsible for the creation and support of stable
associations of sympatric species.
I deliberately avoid mentioning the most trivial mechanism supporting
this organization—the competition for food, which depends on the density of
the individuals N. The competition for food forces the partners to make con-
cessions, which force the "opening up" of various sections of the genome
"line," so that the diversity of requirements and phenotype adaptation can
avoid an overlapping of fundamental niches with respect to a scarce component,
and consequently the competitive crowding out of species. As a result, in a
system in equilibrium (for the case stipulated by the B.R. model), individual
species do not vanish and no species attains a population size which entails a
destruction of the ecological niches of other species. Therefore, not only
the fact that the population size ratios among the species making up the sys-
tem remain constant, but also the absence of fluctuations in the total density
of individuals in the association can be justifiably considered as a charac-
teristic of a system in equilibrium.
It was by no means coincidental that MacArthur considered those spe-
cies to be in equilibrium for which the integral in the equation
logN.(t) = logN.(0) + j r.(t) dt (2)
0
was smaller than the logarithm of the original number of individuals, where rj_
is the population growth rate of the i-th species (MacArthur, 1960). Somewhat
later, Preston (1968) considered a species to be stable if the variations in
its population size (both in time and space) over many years obeyed a log-
normal distribution.
Thus, for the case stipulated by the B.R. model, the co-existing sym-
patric species must be in equilibrium. The mutual interdependent partition-
ing of the segment by means of randomly falling points reflects only the fact
that the spacing of the niches is not discrete, which in no way implies that
one species becomes more abundant at the expense of crowding out another spe-
cies. Rather, the situation is such that several related species located in
an unoccupied but confined biotope "partition" the latter in such a way that
no free gaps are left between their niches. This final result is achieved as
a consequence of the action of mechanisms regulating the ecosystem, whose
38
-------
purpose is to compensate the activities of each individual population by the
collective activities of all other populations from various trophic groups
utilizing jointly the available biotype resources.
While the competitive crowding out mechanism can only occur as an epi-
sodic event, the functional equilibrium of the ecosystem is disturbed whenever
the total density of individuals in the sympatric species increases (or due to
any other cause leading to a functional imbalance of the ecosystem).
Returning to the B.R. model, we note immediately that it describes a
special case. The initial restrictions are so strict (sympatric equilibrium
species of approximately the same size, with a constant total density, con-
fined to a relatively homogeneous biotope) that it is actually surprising that
in certain cases the agreement between the prediction made by MacArthur's
model and reality is satisfactory. Meanwhile, the great popularity enjoyed by
MacArthur's "broken rod" model among ecologists is attributable precisely to
the good agreement detected between its forecasts and the observed relative
abundance of sympatric species in certain associations of small animals
(ciliary Ophiuroidea, Gastropoda, hermit crabs, snakes and certain fish). The
most exhaustive "biological" analysis of the broken rod model is due to Ch.
King (1964) to which we refer the reader interested in the details of the
problem. The range of w and N values for which data showing good agreement
with the forecasts made by the model were reported varied from 5 to 30 species
and from 20 to 200 individuals. At the same time, the study of phytoplankton
undertaken by Hutchinson (1958, 1961) showed poor agreement with forecasts
made by the B.R. model. I demonstrated during a study of seasonal variations
in offshore'marine plankton that the B.R. model predicts the relative abun-
dance of species in a comparatively large proportion of cases when the total
number of individuals is not large (Fedorov, 1970). On the other hand, during
the spring and fall phytoplankton bloom period, the deviations from forecasts
made by the model are not only substantial, but they also obey a particular
law, namely, the most common species are more abundant and the rarest species
less abundant than the values predicted by the model. Ultimately, this obser-
vation served as the main impetus for a revision of the original assumptions
and hypotheses discussed by MacArthur (1957). To begin with, a fact attract-
ing attention is that in the B.R. model the ratios of the lengths of the parts
depend only on the number of points falling on the rod (i.e., the number of
species w) without being dependent on N. This means that MacArthur makes no
allowances for the effect of the total density of individuals on the popula-
tion sizes. This can only occur in the absence of competition for food and
the competitive crowding out of individuals. At the same time, a more general
model of the relative abundance of sympatric species including MacArthur's
B.R. model as a special case (corresponding to the absence of competition for
food), must certainly also include an additional variable which is related to
the total number of individuals in potentially competing species, i.e., to N.
While retaining the formal analogue of the model (partitioning of a rod into
parts which are proportional to the expected average population sizes) , the
main difficulty of the problem is carried over to the plane. The problem
which arises is what function must be used to specify the non-uniform distri-
bution of the points on the rod when N varies.
39
-------
A GENERALIZATION OF THE MODEL OF
THE RELATIVE ABUNDANCE OF SPECIES
Among all possible prerequisites, we single out three which are of
prime importance in the construction of a more general model.
1. In terms of the concepts of D. Hutchinson's formal theory, the
intersection subset (Hi-Hj) formed by niches overlapping with respect to a
number of identical parameters is always smaller than the potential niche size
(Hj_,Hj) of any species, since species with niches that are completely identi-
cal cannot exist in a group of sympatric species. In other words, the inter-
section subset formed is related to the size of the fundamental niche in the
same way as the phenotype is related to the genotype. Consequently, the com-
petitive crowding out from the biotype region (corresponding to the intersec-
tion subset of one species) to the region which corresponds to the "refuge" of
the crowded species (H-L-H-pHj) , leads to phenotypic changes in the latter not
allowing one species to crowd out another completely (due to the formation of
new "cells" in the genome line). With respect to specific structural charac-
teristics of their niche, the crowded species which form an ecological reserve
of the association, are most closely related to the crowding species. A study
of diatomaceous plankton in the White Sea has shown that the most common and
rarest species react similarly to conditions in experiments conducted in situ,
whereas "neighboring" species in the sequence of populations ranked by size
react in the most dissimilar manner.
2. The intersection subset Hj'Hj turns out to be a region in which
the Volterra-Gauze axiom admits no exception. Consequently in such a system,
the population sizes can only stabilize after one species is crowded out from
the locus corresponding to the intersection subset, i.e., after the biotope is
partitioned between the crowded and crowding species according to the condi-
tion Hj_-Hi-Hj and Hi+Hi-Hj respectively.
3. The region of intersecting subsets increases with an increase in
the total density of individuals N, since the total demands on the environ-
ment made by the entire association as a whole exceed the capabilities of the
biotope at every instant of time. Therefore, competitive crowding processes
shift the population size ratio in favor of the crowding species (see pre-
requisites 1 and 2), and due to this mechanism the most common species become
more abundant and the rare species which are crowded into shelters, less abun-
dant than predicted by MacArthur's model.
A rod of finite length (0,1) can again be used as a formal analogue of
the more general case described above. Suppose that the potential size (not
the population size!) of each species is determined by a pair of points fall-
ing randomly on the rod from a height which is proportional to N. Clearly, on
the average, the increase in the distance between the falling points will be
proportional to N if the surface of the rod is elastic (see Fig. 1). Conse-
quently, the probability that the niches will overlap, i.e., that a region of
intersecting subsets will be formed, will also increase. As a result of com-
petitive crowding, one species is crowded out by another species from the
zone Hi-Hj, so that the segments (H^+Hi-Hj) and (Hj_-Hj_-Hj) obtained are propor-
tional to the population sizes of the sympatric species nj_ and nj respectively
40
-------
(not their potential sizes). Moreover, it is evident from Figure 1 that the
length of the overlapping part corresponding to the intersecting zone must be
subtracted from the segment corresponding to the requirements of the species
according to some rule (for example, the common part can always be subtracted
from the right segment, or to take another example, the smaller segment can
always be subtracted from the larger segment). Due to this mechanism which
changes the lengths of the segments with increasing N (and consequently also
the intersecting zones (Hj_-Hj)), all segments except one will become ~ncreas-
ingly smaller and the largest segment will approach un_ty. Since the first
prerequisite Nj_ ^ N-; in principle precludes the elimination of shelters ever, by
the most competitive species, when the number of species remains approxi-
mately constant (W~ const), with increasing N the decrease in the abundance of
individuals in the sequence of populations ranked by size will follow an ex-
ponentially decreasing curve.
Key: a. i.e., for N^>Nj and n^>n^
where ni = Hj_ + n^'Hj and n.; = H^ -
Figure 1
Mathematically, the observed pattern corresponds to assuming in the
broken rod model that when w-1 points fall on -i-he rod, the probability of a
point making impact at different ends of the rod is nee. the same, and that it
differs by a value (which is measurable along the rod) which varies from "-y"
to "+y" at its ends.
In view of the exponential character ot the curve describing the de-
crease in the relative abundance of species during competitive crowding, the
logical and simplest assumption is that the (w-1) points falling on the rod
41
-------
UX
follow some non-uniform distribution which is proportional to e / where y is
an unknown parameter. Hence "u" is an important parameter reflecting the
change in the ratio of population sizes of species, which depends on competi-
tion for feed (or niches?). Competition becomes more important as the total
density of individuals N increases. Moreover, when y = 0 (i.e., in the ab-
sence of competitive crowding out of species), any possible model must degen-
erate to the special case specified by MacArthur's B.R. model.
EXPONENTIALLY BROKEN ROD MODEL (E.B.R.)
Ya. I. Gol ' f and performed by no means trivial mathematical calculations
which made it possible to select the parameter y. A full justification of
these calculations is given in a supplement to the study dedicated to the
"Ecology of Phytoplankton" (to be published) . Using the former formal ana-
logue of a broken rod, essentially the generalized model reduces to the fol-
lowing:
A rod of unit length is broken into w parts. It is assumed that the
(w-1) "breaking" points are independent random variables which are identically
distributed on the interval (0,1) , with the density
p(a) = U/sinhy exp {y(2a - 1)} (3)
where y is a parameter and lVsinhy ^s a normalizating factor. We obtain the
ranked sequence by ordering the parts after the rod is broken in a decreasing
sequence. This sequence is random, since it is associated with the correspon-
ding random breaking of the rod. The problem reduces to finding the expected
values of random variables in ranked sequences. These quantities are func-
tions of the parameter y, so that they can be written as
1 > nx(y) > n2(y) > n3(y) ... > nw(y) > 0 (4)
nx(y) + n2(y) + n3 (y) ... + n^y) = 1 (5)
When y = 0, the breaking points obey a uniform distribution (p(a) = 1) , re-
flecting a special case of the model, i.e., MacArthur's B.R. model.
/s 1 r 1 ,,,
nr = — ) — (6)
r w , L k
k=r
In this formula, the order of the numbering sequence was changed (compared to
formula 1) , and the most abundant species has rank 1.
In the case where y ^ 0, the exact formula for nr(y) is considerably
more complicated
*
where r = 1,2, . . . , w
42
-------
and the functions K have the form
K
. .
r(T)
Here
w
(-1)
k-1
r-1
k-r
k-1
r-1
(k-1) !
Sinn
(8)
(r-1) ! (k-r) !
is the binomial coefficient and
(9)
w,
Sinh
i=l
Sinh (
•w-k+i
T)
Sinh g-'T)
(10)
Calculations based on formulas 7-10, made using an electronic computer, al-
lowed us to calculate "the relative lengths of the segments (which were ex-
pressed in normalized units N/w) in the range of w values from 5 to 30 and the
range of y values from 0 to 60 (with step size 0.1 in the interval 0.
-------
TABLE 1
u
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
10.0
11.0
12.0
13.0
14.0
15.0
16.0
17.0
18.0
19.0
20.0
25.0
30.0
35.0
40.0
45.0
50.0
55.0
60.0
1
2.28
2.51
2.98
3.41
3.74
3.97
4.14
4.26
4.35
4.42
4.48
4.53
4.57
4.60
4.63
4.65
4.67
4.69
4.71
4.73
4.74
4.79
4.83
4.86
4.88
4.89
4.91
4.92
4.92
2
1.28
1.23
1.08
0.90
0.73
0.61
0.51
0.44
0.39
0.34
0.31
0.28
0.26
0.24
0.22
0.21
0.19
0.18
0.17
0.16
0.15
0.12
0.10
0.09
0.07
0.06
0.06
0.05
0.05
/s
nr
3
0.78
0.70
0.54
0.41
0.31
0.25
0.21
0.18
0.16
0.14
0.13
0.11
0.10
0.10
0.09
C.08
0.08
0.07
0.07
0.07
0.06
0.05
0.04
0.03
0.03
0.03
0.02
0.02
0.02
4
0.45
0.39
0.28
0.20
0.15
0.12
0.10
0.09
0.08
0.07
0.06
0.06
0.05
0.05
0.04
0.04
0.04
0.04
0.03
0.03
0.03
0.02
0.02
0.02
0.01
0.01
0.01
0.01
0.01
5
0.20
0.17
0.12
0.08
0.06
0.05
0.04
0.04
0.03
0.03
0.03
0.02
0.02
0.02
0.02
0.02
0.02
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.00
0.00
0.00
44
-------
TABLE 2
y
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
10.0
11.0
12.0
13.0
14.0
15.0
16.0
17.0
18.0
19.0
20.0
25.0
30.0
35.0
40.0
45.0
50.0
55.0
60.0
s\
nr
1
2.93
3.46
4.61
5.72
6.59
7.20
7.65
7.98
8.23
8.43
8.59
8.72
8.83
8.92
9.00
9.08
9.14
9.20
9.25
9.29
9.33
9.49
9.60
9.67
9.73
9.77
9.80
9.83
9.85
2
1.93
1.97
1.93
1.72
1.45
1.22
1.03
0.89
0.78
0.70
0.63
0.57
0.52
0.48
0.44
0.41
0.38
0.36
0.34
0.32
0.30
0.24
0.19
0.16
0.14
0.12
0.11
0.10
0.09
3
1.43
1.35
1.14
0.89
0.69
0.56
0.47
0.40
0.35
0.31
0.28
0.25
0.23
0.21
0.20
0.18
0.17
0.16
0.15
0.14
0.13
0.10
0.08
0.07
0.06
0.05
0.04
0.04
0.03
4
1.10
0.99
0.76
0.56
0.43
0.35
0.29
0.25
0.22
0.19
0.17
0.16
0.14
0.13
0.12
0.11
0.10
0.10
0.09
0.09
0.08
0.06
0.05
0.04
0.03
0.03
0.02
0.02
0.02
5
0.85
0.74
0.54
0.39
0.30
0.24
0.20
0.17
0.15
0.13
0.12
0.11
0.10
0.09
0.08
0.08
0.07
0.07
0.06
0.06
0.05
0.04
0.03
0.02
0.02
0.02
0.01
0.01
0.01
6
0.65
0.55
0.39
0.28
0.21
0.17
0.14
0.12
0.11
0.09
0.08
0.08
0.07
0.06
0.06
0.05
0.05
0.05
0.04
0.04
0.04
0.03
0.02
0.02
0.01
0.01
0.01
0.01
0.00
7
0.48
0.40
0.28
0.19
0.15
0.12
0.10
0.08
0.07
0.07
0.06
0.05
0.05
0.05
0.04
0.04
0.04
0.03
0.03
0.03
0.03
0.02
0.01
0.01
0.01
0.01
0.00
0.00
0.00
8
0.34
0.28
0.19
0.13
0.10
0.08
0.07
0.06
0.05
0.04
0.04
0.04
0.03
0.03
0.03
0.03
0.02
0.02
0.02
0.02
0.02
0.01
0.01
0.01
0.01
0.00
0.00
0.00
0.00
9
0.21
0.17
0.11
0.08
0.06
0.05
0.04
0.03
0.03
0.03
0.02
0.02
0.02
0.02
0.02
0.02
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.00
0.00
0.00
0.00
0.00
0.00
10
0.10
0.08
0.05
0.04
0.03
0.02
0.02
0.02
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
45
-------
H
hi
i£>LnrocN] p_| f-.. f**) ^^ ^Q p«, (^ j_f^ <5ji ^« (^r) fV) (~r^ prj (~vj f\j f\j f\j f\J i~H i—^ i-H C^ C^ (~) C3 ^^
CNCNCNiHi-HiHOOOOOOOOOOOOOOOOOOOOOOO
CT^ ^D LO CO LO
*-H r^^. m ir\ s~\
ro
CNCNCNCNfNiHiHrHrHOOOOOOOOOOOOOOOOOOOO
CN
CNrOrJ-^^jrorOcNCNCNrHrHnHrHrHrHrHtHOOOOOOOOOOO
CT^ fO ^0 "^J* O^ LO LO ^5 CNJ i~H CO ^* O1^ fO vO CT^ i—I ro LO r^* O^ ^* t**^ ^D rH CO ^i* ^J1 LO
co LO co O1] LO co ^D oj co ^* ^* LO LO ^o ^o vo f^1* r*** p^* r*" r*1* co co cr^ CT^ cr^ cr^ cr^ cr^
rHrHtHCNCNCNCNCSCNCNCNCNCNCNCNCNCNCNCNCNCNCNCNCNCNCN
ooooooooooooooooooooooooooooo
3.
46
-------
rorocNrHi-trHrHOOOOOOOOOOOOOOOOOOOOOO
O OOOOOOOOOOOOOOOOOOOOOOOOOOOOO
oodoooooooodooooooooooodooooo
r-roncNCNrHiHrHrHiHOOOOOOOOOOOOOOOOOOO
CTi OOOOOOOOOOOOOOOOOOOOOOOOOOOOO
CN
ooooooooooooooooooooooooooooo
omm^rocNCNi-ir-ir-ti-HrHOOOoooooooooooooo
CO rHOOOOOOOOOOOOOOOOOOOOOOOOOOOO
CN
ooooooooooooooooooooooooooooo
f- rHOOOOOOOOOOOOOOOOOOOOOOOOOOOO
CN
OOOOOOOOOOOOOOOOOOOOOOOOOOOOO
rHi-HOOOOOOOOOOOOOOOOOOOOOOOOOOO
fN
ooooooooooooooooooooooooooooo
in CNrHrHOOOOOOOOOOOOOOOOOOOOOOOOOO
CN
ooooooooooooooooooooooooooooo
0)
-------
individuals in the natural phytoplankton by species by means of an appropriate
selection of the parameter y. To do this, it is necessary to relate the ex-
perimental N values to tabular y values. In the first place, to make a suffi-
ciently accurate selection of y, we must know the accuracy with which w and N
are being determined in samples of the natural plankton.
The following relationship was determined between the number of spe-
cies detected and the number of surveyed individuals on the basis of data ob-
tained from a quantitative treatement of two series of samples (each 'series
consisted of 50 samples taken simultaneously) of diatomaceous plankton of dif-
ferent density (thousands and tens of thousands of cells per liter) (Kol'tsova
et al., 1971):
w = a log — (11)
Using these data, I established that by counting approximately 3000 indivi-
duals, w can be determined with an error that is less than 20%. Fixing the
sample size (by counting the number of plankton counting cells), and counting
approximately 3000 cells, also allows the determination of the number of in-
dividuals per liter with an error which is less than 20%. This problem will
be discussed in greater detail in a paper dedicated to methods used in the
study of phytoplankton (Fedorov, 1977). If for some reason(s) the researcher
is forced to restrict himself to a sample consisting of a smaller number of
individuals, the number of species can be determined by means of calculation.
To do this, it suffices to determine the value of the slope a in equation (11)
from the data in this nonrepresentative sample, plot the graph of w as a func-
tion of log N/10* and then calculate w for 3000 individuals. In this case
agreement must be sought between the prediction made by the model for the num-
ber of individuals N that was found and the calculated value w.
Therefore, it is clear that for the accuracy with which N and y were
determined, while seeking relations between N and y, it makes no sense to
focus the search on segments whose sizes correspond to species represented by
a small number of individuals. It suffices entirely to determine the corre-
spondence between the abundance of the first 5 most prevalent species in the
sequence of populations ranked by size (n]_,n2/•••,ns), and the corresponding
tabulated values of the lengths of segments. In this case, in Tables 2 and 3,
for 10 and 30 species respectively, only 5 columns are left in each table.
This simplifies the compilation of a general table for natural plankton. The
magnitude of the error with which w is determined allows, in turn, a reduction
(in a valid manner) of the number of tables by at least a factor of 2 through
the inclusion of only even or odd w values in the general table (for example,
w = 6,8,...,30). In the case when the number of species w found or calculated
"does not have" its own table, the table for the value w+1 can be used, since
we can assume that in all probability some species was not detected (the case
that one species too many was detected is excluded), so that we can correct
the error by increasing the sample size. The known population size
*Usually the slope varies considerably. Thus, for White Sea diatoma-
ceous plankton, its value varies in the range 8-10. For Arctic plankton (Kara
Sea), "a" varies in the range 20-30.
48
-------
measurement error allows ascertaining the error with which the parameter y is
determined, which in turn permits us to increase the "step size" used in the
compiled summary table. Thus, in the range of y values from 6 to 20, 1.0 is
an acceptable step size, while for y>20, this step size can be increased to
5.0. A summary table for phytoplankton is presented below. Using this table,
one can find the relationship between the parameter y and the quantities w
and N.
To illustrate how this relationship is found, I present as an example
the data used in a quantitative treatment of samples collected in 38 White Sea
water area stations (June 1972). The samples were taken at 4 levels (0.5,
2.5, 5.0 and 10.0 m). Samples from the surface level were used to find the
relationship, and from the remaining levels only those samples for which log N
was less than 4 and greater than 6. The spatial heterogeneity in the distri-
bution of phytoplankton is related to the presence of patches forming micro-
scopic algae concentrations in the photic zone. According to Platt et al.
(Platt, Dickie, and Trites, 1970), the diameter of such patches varies in the
range from 1.3 to 3.9 km. Using correlation analysis, it was possible to de-
termine for White Sea diatomaceous plankton a relationship between y and log
N and also between y and (log N, w) [sic]:
y = 2.1 (log N)2 - 9.6 log N - 0.025
y = 11.1 (In N)2 - 22.1 In N - 0.025
t
y = 5.9 (log N)2 - 26.6 (log N - log w) + 44.3 (log w)2
- 15.4 log N + 0.9 log w + 33.8 (13)
where w and N are the number of species and the total number of individuals in
one liter of the sample, respectively. Of course, no "mysterious" signifi-
cance must be attributed to equations (12) and (13). By validating theoreti-
cally the existence of a relationship between the associated parameters, they
simply enable us to rewrite equation (7) in terms of the quantities whose
values were recorded. In practice, they make it possible to find y on the
basis of data about w and N, and to predict the distribution of population
sizes for the first five most prevalent species in the sequence of populations
ranked by size on the basis of Table 4 for the number of species w (or w+1).
The following statement can be made with 100% confidence. The actual con-
crete form of the relationship between the associated parameters will vary
from basin to basin and possibly from season to season (for each basin).
Moreover, serious reasons exist which in principle preclude this relationship
from being good, i.e., more or less rigorous, definite and invariant. The
only statement that can be made with confidence is that the new model will
predict the distribution of N individuals among w species better than Mac-
Arthur's B.R. model. Therefore, it is hoped that the new model represents
progress in the quest for some hypothetically ideal predictive model. It must
be assumed that at the present time we only began this quest. This is borne
out convincingly by a number of common and special reasons (pertaining in par-
ticular to phytoplankton) which are responsible for the discrepancy between
the forecast made by the model and reality.
1. Notwithstanding the fact that the introduction of the parameter y
49
-------
^"
3
^ l^vD'»mrNj(Ni-lHi-li-|i-IOOOOOOOOOOOOOOOOOO
OOOOOOOOOOOOOOOOOOOOOOOOOOOOO
OCOVDinrorOCNCNHrHrHHr-lrHrHiHOOOOOOOOOOOOO
•HOOOOOOOOOOOOOOOOOOOOOOOOOOOO
w m
HiHi-IOOOOOOOOOOOOOOOOOOOOOOOOOO
rrt i-*^ r««^
CM
HHrHi-IHi-IOOOOOOOOOOOOOOOOOOOOOOO
m I~H oj ^D LO co I**** LO r^* ^* r** cr\ oo ^^ ro o^ LO ^3 ^* co PH LO ^* I~H t-o o^ 0*1 ^-3* ^D
co co f^o ro ^^ LO o^ c^j ^* ^o !*"*• oo o^ ^p I~H I~H oj ro oo ro ^* LO vo t""^ r*1* r^** oo co co
f\j PO ^* '^ vo vo vo r^ ^* r*^ c*** '^^ t1^ co oo oo co co co co co co co co co co co co co
^ inn
-------
T)
C
•H
4->
a
o
u
CNi-HCO^D^nncNCNtNiHrHrHrHiHrHrHrHOOOOOOOOOOO
iHr-ldoddddodoooooooooooooooodoo
co ^D LO I"*** p^1 ^* LO co co o^ vO co rH o^ co v^ LO ^* co CN rH oo P** LO ^* "^ co co CN
^* *^f i~H CO VD LO ^* CO CO CN C^ 0"! OJ rH i~H rH i~H rH rH rH rH ^D ^D ^D CD ^D G) ^D ^D
r-iiHr-ioddodoooooooooooooodoooooo
i-li-lrHi-li-HOOOOOOOOOOOOOOOOOOOOOOOO
oj oj ro r*1* LO c^ LO co r*^ ro co ^* VD ^D LO ^^ ^o oj o^ vo ^* ^* co ^* I~H co vo ^* co
CO ^* VO *^ rH CO LO CO r*H ^3 0~^ CO f^^ t"^* kO VO LO LO *^J* ^* ^* CO C^J OJ 0^1 rH r~H rH rH
CSCNCNJ(NCN)i-lr-HrHrHrHOOOOOOOOOOOOOOOOOOO
iHOOOOOOOOOOOOOOOOOOOOOOOOOOOO
rHHiHOOOOOOOOOOOOOOOOOOOOOOOOOO
CO ^* ^* LO CD CO i~H 0s) LO ^D v^ CO ^D t*1^ LO CO r~H CD CT^ CO LO \Q CD CO P^* ^O LO LO ^*
kO ^O ^f i~H G^ r^~ \^ LO ^^ *sf CO CO CO OJ O4 C^J C^ C^J i—I rH t~H rH i—| ^D ^D ^D ^D ^D ^D
^Hi-lrHi-IOOOOOOOOOOOOOOOOOOOOOOOOO
tNCNCNCNiHt-I^Hi-ltHOOOOOOOOOOOOOOOOOOo'o
I—i QQ ^i cr^ i-H ^3 vo ^* ^* r** ^D csj co LO v^ r*** co o^ c^ C5 pH co *^ LO LO ["""• r^* r^* co
CO CO LO V^ CO CT^ O^ ^D ^D ^D i~H i~H i~H nH r~H rH i-H pH t~H OJ OJ C^ OJ C^J OJ CN OJ OJ O>J
CnCD^'J'rOfMtNi-lrHiHrHi-li-lrHOOOOOOOOOOOOOOO
doooo'dooooo'odoodoo'oooodododdd
CT^ CO LO rO CO O^ O^ CO ^* i~H CT^ t*^* VO ^* CO CN r~H ^H ^D 0^ 0^ f^* LO "^3* CO CO 0^ CN O^
rH ^D CO ^-D ^J1 CO CO O4 O^ O^ i—| r~H r~H nH p~H rH rH rH rH ^D C? ^D ^D ^D CD ^5 ^D ^D CD
i-iiHooooooooo"odoooooooooo'ooooo*o
OJ LO ^sf 00 ^O C*J rH ^* CT^ ^* rH CO LO CO rH ^D CO P^* VO LO ^* rH O^ r^* V^ LO ^* 'sj1 CO
LO *^t* CNJ O^ f"^ ^O LO ^* CO CO CO C^J C^ OJ OJ CNI rH rH rH i—I rH i~H ^D ^D ^D ^D CD ^D ^D
i-HrHiHOoooooodoo'ooodoododdo'ooodo
oj co co r** o^ ^* ^* oo v^ vD o^ oj r*1* CN co LO CN cr^ r*^ LO co v^ rH co LO co 0^1 I-H ^D
^D ^D ^D CO LO CO rH CT^ CO Is"* ^Q VO LO LO ^* ^f ^* CO CO CO CO C^ c^j rH rH rH rH i-H i—I
(NCNCNiHr-HrHiHOOOOOOOOOOOOOOOOOOOOOO
^D VO CO rH rH 00 CO r** 0^ f^l CO LO VO P"* 00 O^ CD rH rH CNJ O>J ^J1 LO ^O P^" P^ P^" CO CO
CO CO ^3* vO P^* P^* CO CO 00 O"^ 0^ CTi CT^ O^ CT^ O^ ^D ^D ^D ^D ^D ^D ^D ^D ^D ^D CD ^D CD
i-lrHrH^Hi-lnH^HrHrH^liHiHiH
OOOOOOOOOOOOOOOOOOOOOOOOOOOOO
51
-------
C
•H
•P
C
O
O
w
r-HiHiHOOOOOOOOOOOOOOOOOOOOOOOOOO
rH oo LO CM ^* oo CM cr^ ^D CD I-O CM co LO ^* 0*4 ^D 0*1 P1*" vo LO rH 0*1 r^ co LO ^p *^f oo
co P"* ^-D CM CT*I P"^ ^0 LO *^ *"3* oo co CM CM CM CM CM rH r™H rH i—t I-H CD CD CD CD ^D ^D CD
rHrHi-li-HOOOOOOOOOOOOOOOOOOOOOOOOO
CNCNCMiHMi-IOOOOOOOOOOOOOOOOOOOOOOO
CN
(NCNroncNCNCNrHiHrHrHiHf-IOOOOOOOOOOOOOOOO
LO CD rH ^* LO P** CD CM ^* rH P"^ LO vO oO ^O ^D f*1) CO r~H C^ ro r^ LO i^ ro r~l CO CO CO
^O t^** i"H I**1* C^ LO CO P^- ^* CO *N}* 00 r~i ^* ^D CO ^D r~H 00 ^J1 LO 0^ r~\ fO *^* LO LO VO VO
oo "sf t^* o^ I~H LO ^f LO ^o r^~ r^~ p^ oo co oo co 0*1 o^ o^ o^ o*^ cr^ ^^ ^D ^D CD c^ ^5 ^3
vo r*^ CT^ CN c^i CD I~H LO c^ LO oo r~H cr^ c^* LO ^* oo c^ I~H ^D CD r~" LO ^j* oo oo 0^1 04 c*j
^* OO ^5 CO ^«D LO ^I1 OO oO OsJ 0^ CN i~H rH i~H rH rH r~4 rH i^ r~H ^D ^D CD CD CD CD ^D ^D
•HiHi-IOOOOOOOOOOOOOOOOOOOOOOOOOO
I~H *^ ^D LO co r*** co CN t*^ o^i CT^ vo ^* o^j ^D co r*"* LO LO ^* CD oo r^* LO LO ^3* oo oo
r*** ViJ ^J1 t~{ CO vO LO ^* *^ oO oo O4 OJ Osj CN C^ rH r~*1 rH rH rH rH ^D CD CD CD CD CD CD
i-ii-irHiHocJooooooodoooooodooo'ooooo
o>
CJCNrHT-lr-lrMOOOOOOOOOOOOOOOOOOOOOOO
LO CD OO LO 0"^ O^ LO CO ^O ^D ^O LO ^O CO rH LO CD CO OJ CO LO *^ VO CD ^0 oo ^D CO f"^
LO CO rH CD VO O4 OJ^ ^O ^* OO rH CD O^ CO CO f**- I*"*1 VO LO LO LO *^ 00 OO C^] ONJ O4 rH rH
CNCsinncNCNlrHtHrHiHrHi-IOOOOOOOOOOOOOOOOO
LO o] CN oo rH vo 0s] oo LO *3* LO oo v^ CA CT^ r*** 04 LO r*"* r^~ r*^ rH oo oo oo v£) o^ r^* rH
LO LO P""* CD CD ^* LO OO 0*1 ^* 00 rH ^* v^ CO ^D C^ OO ^* LO VO CO C^3 OO ^* LO vO ^O P*~
00 *^ vO CT^ rH CN OO ^^ ^* LO LO vO VD ^O VO 1^* P"* P**" P"*1 P** P*" CO CO 00 CO CO CO 00 00
\O LO CO OO LO ^* P** rH P** ^* rH 0^ P1* LO ^* OO CM rH ^D CT^ CT^ O^ LO ^* 00 0^4 C'J ON) rH
OO 0^ CTV P** LO ^* OO 00 O^J O4 OJ rH rH rH rH rH rH rH rH ^D CD ^D ^D ^D CD C5 CD CD CD
i-HrHododoooddoooodooooooodooooo
rH OO O^ O^ yD f~H rH 00 CO OO 0^ VO ^* C*J CD 00 P"* LO ^* OO OO CT^ P*" \& LO ^* ^* OO OO
vO LO 0^ CJ^ P** vO LO ^3* OO oo CNJ fNJ O4 OJ CN rH rH rH rH rH rH ^D CD CD CD CD CD CD CD
r-Hi-H,-! oooodooooooooodoooooodoooo
^f LO rH V^ CO vD ^D CO O^ 0s) ^D 0^4 00 LO 0s] O^ P^ LO ^* C^] »~H ^O OO rH O^ CO P"* VD LO
O^ O^ 00 t1 i~H O^ CO VO LO LO ^J1 ^3* OO OO 00 C^J CN Osj OJ C^ C^] rH rH rH ^D ^D CD CD CD
i-HrHrMi-liHOodooOOOOOOOOOOOOOOOOOOO
^l1 LO O^ ^O CM LO LO rH CM P""" ^f *3* VO 0^ OO 00 OO O^ VO CM C^ O^ CM P^ OO rH CO V^ LO
^J1 kO CO P"- ^* CD P^1 LO OO r-H CD O^ CO P** P"* ^^ VO LO LO LO UO OO OO CM CM CM i—1 rH rH
CNtNCNfSCNtNrHrHiHr-liHOOOOOOOOOOOOOOOOOO
,-1
/
ooooooooooooooooooooooooooooo
52
-------
•H
4J
O
O
W
r-HrHiHiHOOOOOOOOOOOOOOOOOOOOOOOOO
CNCNrHrHr4cJOOOOOOOOOOOOOOOOOOOOOOO
CM ~
CMCMCNCNi—liHrHi-HOOOOOOOOOOOOOOOOOOOOO
CN
Cvjro^^c^rocNCNCNrHrHHHrHrHOrHOOOOOOOOOOOO
00 [""*•• i~H 00 O>] LO ^^ LO 0*^ ^J1 CD vO ^J* rH 0^ CO vO LO ^* OO CN O^ I"**" LO ^* *sF 00 OJ C^
I*"-* vO "^t1 (O CO vO LO ^j1 00 00 PO CN C^J OJ nH i~H r~H rH i~H rH i~H CD CD C^ CD CD CD CD CD
rHMi-lr-IOOOOOOOOOOOOOOOOOOOOOOOOO
co o^ I™H ^* f^ o^ f*^ 04 f*o r^* i~~t r** oo CD r^* LO co oj CD cr^ ^o f^o I™H o^ r^ to LO ^* ^*
O^ O^ CO ^* rH CN r^* vO LO ^J* ^* fO PO 00 O4 O*I OJ f^J C^ i~H rH i™H r~H CD CD CD CD CD CD
iHrHrHtHr-loOOOOCJOOOOOOOOOOCJOOOOOOO
in
i • I
CNCNJfNfMrHiHi-IOOOOOOOOOOOOOOOOOOOOOO
fsj i—i (—i ^Q r*^ OQ ^j^ OQ [—) @) ,—( [**•• ]_f) LO VD o^ oj vo I~H r^1* vo oo r*1* CD LO CD r^* ^* oj
CO OJ CO CO ^J* O^ LO t~~\ O*^ ^0 LO fO OJ rH ^D CT^ 0^ CO CO I*1^ f1^" LO ^* ^f PO f11*) OJ C^l OJ
CNronnncM(NfSi-ttHr-|r-)rHi-HiHOOOOOOOOOOOOOO
co ^D co o^ r**1 co ro LO ^* rH r*1* CN vo o^ CN ^* ^JD co o^ rH oj r*~- o^ rH ro ^J1 LO LO vo
ro LO r^- CD LO LO r~* co o^ CD CD I~H I~H I~H 04 c^ oj c^i c^j ro n f^o LO ^t* ^-J1 ^j* ^* ^* ^*
j_Q pTJ p—j Q^ \^ ^^ ^^ ^\J ^Q |^ QQ ^( f*^ £^ QQ ^^ jj^ ^1 pT) j%J |_^ (Vj ^Q ^O ^4( (Y^ f¥^ fV^ ^J
nHiHrHOOOOOOOCJOOOOOOOOOOOOOOOOOO
CD oo co oo ro c^j o~^ r^* o^ ro co *sf ^H oo v^5 PO oj CD o^ r1^ vo c^i CD co r^ LO LO LO ^*
O^ OJ *sO 00 ^D 00 V^ LO ^* ^* oO 00 00 C^ O] CN OJ O4 ^H i™H t~H t~t t~H CD ^D CD CD CD CD
HrHrHHiHOOOOOOOOOOOOOOOOOOOOOOOO
ro
i-1
CNCNtNtHrHHr-IOOOOOOOOOOOOOOOOOOOOOO
I"*** CO ^d* oo CO 00 CD i—I O^ LO f~H LO CO i~H ^* VO 00 CD i~H c^J OO OC* CD 0s! oo ^J* LO VO VO
OO ^* r**" ^D 0s) ^* vO f^1^ ^" CO O^ C^ 0^ ^D CD ^D CD f~H *~H i~H i~l i~~\ C^ CN C*^J fN C^ OJ C^
rH f"H r™H rH i~H rH i—! f~H i*H rH CN O4 C^4 CN] C^l C^J OJ CN O3 OJ CN O«J C^J 0s! CS Oj
ooooooooooooooooooooooooooooo
53
-------
T3
0)
c
•H
4J
o
o
w
I
in
rHiHrHHOOOOOOOOOOOOOOOOOOOOOOOOO
NCNrHrHrHHOOOOOodooOOOOOOOOOOOOOO
fM m
CNCNtNCNrHHrHrHOOOOOOOOOOOOOOOOOOOOO
I
CN
(NfO^r^J'nrOCNlCMCNrHiHt-HrHiHi-li-HrHrHOOOOOOOOOOO
I
'
-------
makes the model more sophisticated, it is still too simple to reflect the out-
come of events in such a dynamic system as phytoplankton in which the presence
of homeostasis is determined by a possible rearrangement of the structure
(i.e., due to a change in the relative abundance of populations!), and by a
suppression of sustained relatively stable functional characteristics (Fedorov,
1974).
2. The application of static models to dynamic systems inevitably en-
counters difficulties related to the probabilistic description of the vari-
ables. MacArthur (1960), who tried to overcome these difficulties, approxi-
mated the variations in the population sizes of sympatric species by a straight
line and separated the species into species in equilibrium and opportunistic
species. He excluded the latter from his model on the (rather unconvincing)
grounds that among opportunistic species, "their relative abundance is of
little biological interest since it is controlled by variations in the climate
and other external factors exerting an effect on the rank r" (ibid.). By this
discriminatory act, he eliminated the dependence of the model on N, i.e., he
reduced the model to consideration of a special case. In reality even the
population sizes of so-called equilibrium species fluctuate, which is an ob-
jective reason for the errors made in predictions. Returning to the descrip-
tion of events in a dynamic association, using a static model reflecting the
result of the action of regulatory mechanisms operating in the ecosystem,
"poor agreement" between the forecasts made by the model and nature should be
anticipated due to this reason alone: a number of "wild" results (since a
probabilistic model utilizes average values, thus allowing for a certain de-
gree of scatter of the data about the mean).
Thus, the model is static because it reflects the result of the action
of regulatory mechanisms but does not reflect the rates of the processes regu-
lating these mechanisms. The observed result, which manifests itself in a
change of population sizes, reflects the effect of external conditions which
had a determining effect in the basin somewhat earlier. More accurately, this
"earlier" is measured by the lag in "response" (i.e. , recorded changes in the
structure of the association) to disturbances in the biotype (for phytoplank-
ton usually 2-4 days).
3. The introduction of the parameter y in the model in implicit form
postulates a direct relationship between N and the mechanism of competitive
crowding out of species (in more precise terminology the shoving of the rare
species into the reserve or "far end" of the ranked sequence). This is un-
doubtedly a crude assumption, since the fluctuations may be related both to
keener competition for food observed in phytoplankton in the summer period
(Fedorov, Traskin and Dauda, 1973), and to the beginning of seasonal succes-
sion observed in the spring phytoplankton bloom period. In the latter case,
the end of phytoplankton "flourishing" is not necessarily due to having ex-
hausted biogenic elements and to keener competition for food; it may also be
due to biotic causes, i.e., a sharp increase in the number and activity of
phagocytes.
4. An important misgiving which I have not been able to overcome so
far is that the longest segment predicted by the exponentially broken rod
model is in fact not proportional to the individuals in a single most abundant
species, but rather to the individuals in the crowding species. The latter
55
-------
can be isolated using a method that was described earlier (Fedorov, 1969).
Although in the overwhelming number of cases there turns out to be only one
crowding species, sometimes there are two, and very rarely three such species.
This misgiving applies above all to phytoplankton representing an association
of individuals which are nearly indistinguishable. If this misgiving turns
out to be correct, after the population sizes of the sympatric species are
ordered in decreasing sequence, the dominant species must be isolated statis-
tically, and the predictions made by the E.B.R. model must be compared with
the ranked sequence of values nj in which the first term is
K
HI = I n-j.
i=l
and K is the number of isolated dominant species. I cannot state conclusively
that the original prerequisites of the model were violated, however, I do at-
test to the fact that the values of u calculated on the basis of equations
(12) and (13) refer us in the generalized table to a row whose forecasts are
in better agreement with n^ than n^.
5. For phytoplankton, the anticipated good agreement with the predic-
tions made by the E.B.R. model should occur if at least two conditions are
satisfied: 1) the individuals in the association are inadequately supplied
with mineral feed (when Liebig's law comes into force), and 2) presence of
more or less homogeneous physiological state of cells (when the proportion of
dead individuals is not excessively high). When the phytoplankton densities
are high, the proportion of inactive individuals (with yellow and green lumi-
nescence in ultraviolet light) can be substantial. Thus, for 17 species of
diatomaceous White Sea plankton, in 7 samples taken from the surface level in
one day, the individuals with red luminescence in ultraviolet light (i.e.,
individuals which were undoubtedly alive) represented approximately 67% (for
N on the order of several tens of thousands of cells/liter). Therefore it
would be more correct to compare the ratios of live cells with the predictions
made by the model. Since this cannot always be realized technically, the per-
centage of inactive cells or cells showing little activity can be considered
as the main source of errors for high values of N, because under these condi-
tions, it is very unlikely that the individuals have a surplus supply of min-
eral feed components. For relatively low phytoplankton densities, the propor-
tion of "dead" individuals is relatively low. Therefore, it appears that in
this case the relationship between y and N is not detected because of an ade-
quate supply of mineral feed to individuals. However, low biomasses in the
presence of surplus food can be observed only either during sufficiently ac-
tive consumption of microscopic algae by zooplankton organisms, or in the
presence of a definitely unfavorable effect of certain abiotic factors such
as light (too much or too little) and low temperature. In the second variant
it may happen that some species (for example, a sciophilous and relatively
psychrophilic species) which is better adapted to these conditions than the
other species can sharply increase its population size, i.e., it displays op-
portunism. As a result, the observed high value of N and a high value of y
will not reflect the results of the crowding out of partners by this species,
and it is not completely clear what kind of effect this case will have on the
quantitative aspect of the relation between the parameter y and the values of
N and w.
56
-------
ADEQUACY OF E.B.R. MODEL
To evaluate objectively the advantages of the MacArthur model and the
E.B.R. model, we compared the agreement between the data and their forecasts
using an estimate of the variance which was calculated as the mean square dif-
ference between the observed population size n^ and the population size n^
calculated on the basis of each model:
w
Y t "
LJ 3- 1
s2 = — (14)
w
We assumed that a smaller variance indicates a better forecast made by the
model. Diatomaceous plankton samples collected near the Karelski coast of
the White Sea during the vegetative season at the level below the surface were
used for comparison purposes.
To obtain a generalized seasonal index, we used the mean estimated
variance averaged over all stations
K
Is2
i2 = — (15)
K
where K is the number of stations. A comparison of the variances showed that
they differ significantly. Therefore one can assert with a high degree of
confidence that better agreement between the observed and calculated distri-
bution of individuals among species is undoubtedly obtained using the E.B.R.
model.
Fedorov and Kol'tsova (1972) demonstrated earlier a close relationship
between the pattern indicating the decrease in the sequence of populations
ranked by size and the density of the phytoplankton. They found that good
agreement with the forecasts made by MacArthur1s R.C. model was only observed
when the density was relatively low (1.10^-1.10^ cells/1). To analyze pat-
terns where agreement was poor, we isolated the so-called empirical type
(1.106-10.107 cells/1), which we named after A. Hutchinson (1960), and a cer-
tain mixed case for intermediate population sizes. The E.B.R. model greatly
simplifies the situation and allows further development of the idea of comple-
mentarity as the key principle in the formation of the structure of a phyto-
plankton association. This principle consists of the following: due to suc-
cession and competition for food, which is keenest among ecologically close
species, when equilibrium is attained in the ecosystem, ecologically remote
species thrive at the same time. This is because ecologically remote species,
whose needs appear to be mutually complementary, make "complementary contact,"
which in turn permits the most efficient utilization of the biotype resources.
Ultimately, this leads to a mosaic structure of sympatric species associations
as the main characteristic of an organization which reproduces sufficiently
faithfully MacArthur's case when the fundamental "niches do not overlap much
but there are no free gaps between them" (MacArthur, 1957).
57
-------
DISCUSSION
Curiously, attempts were made to explain the poor agreement between
the ecology and the forecasts made by MacArthur's model by practically any-
thing except the restrictive original assumptions. The most exhaustive "bio-
logical analysis" of the broken rod model was made by Ch. King (1964) to whose
study we refer the reader interested in the details of the problem. Somewhat
later the model was improved by Pilow and Arnoson (1966) and Vandermeyer and
MacArthur (1966); however, this did not result in an appreciable improvement
of the forecasts made by the model. Apparently the first person to realize
the weakness of the original prerequisites in the model was Robert MacArthur
himself. At any rate, in 1966, MacArthur (1966) agreed with the remarks made
by Pilow and retracted his model in very elegant and restrained words express-
ing the hope that "it is used only as a rough approximation of the ecology of
associations which should be allowed to die a natural death."
However, MacArthur's "renunciation" did not undermine the popularity
of the model among ecologists. Kon (1968) , Inger (1965) , MacDonald (1969),
Fedorov (1971), and Fedorov and Kol'tsova (1972) continued to use the model
even after Hairston (1969) demonstrated that good agreement between its pre-
dictions and the observed abundance could be obtained by an appropriate selec-
tion of the sample size. From this Hairston concluded that agreement with the
forecasts made by the model does not depend on the ecological characteristics
of the system considered and that consequently MacArthur's model is devoid of
ecological meaning. In a study titled "Another Look at the Relative Abundance
of Species," I analyzed Hairston's arguments, and in my opinion, presented ar-
guments which can cast some doubt on Hairston's discouraging conclusion
(Fedorov, to be published). The secret for the tenacity of MacArthur's model
is simple. MacArthur constructed a model for which 1) the mathematical calcu-
lations are far from trivial, but can be carried out to the end; 2) a probable
biological prerequisite is elegantly reduced to the problem of randomly parti-
tioning a segment into parts; and 3) in some cases good agreement is observed
between the forecasts made by the model and the facts.
The great elegance and genuine beauty of the model consist of these
three items which prevent one from attributing the success of its forecasts
to pure chance. Beauty and elegance must be a necessary attribute determining
the properties of a model. The deep intrinsic soundness of this criterion was
realized in practice by Migdal (1976), who wrote: "The concept of beauty
plays an important part in checking the validity of results and in finding new
laws. It reflects in our consciousness the harmony which exists in nature."
A reflection of this harmony reigning in nature was the conviction that the
pattern in which the abundance increases in the sequence of populations ranked
by size will always be a kind of "counterpart" of the relations between part-
ners in one ecological grouping which unifies species with a similar speciali-
zation in the association. Therefore, the concept of mechanisms responsible
for the prearrangement of the structure of an ecosystem includes a priori a
natural element "which is hidden in the harmony of parts grasped only by the
mind" (Henri Poincare, cited by Migdal, 1976). I would like to illustrate the
importance of the natural element in decision making. During a discussion of
the specific properties of the E.B.R. model, Valeriy Nikolayevich Tutuballin
did not like the effect of the ends of the rod which is essential in the form
58
-------
of the generalized MacArthur distribution proposed by me. Tutuballin was in-
clined toward the idea that "it was more natural to let the points fall on a
circle." In the case of a uniform distribution, when w + 1 points fall inde-
pendently, in the end it makes no difference whether they fall on a rod or
circle. For the case reflecting "the struggle for existence," the position of
individual points must be made dependent "so that they can repel each other."
Trying to select a formal analogue which simulates reality to which appro-
priate mathematical tools can be applied, Tutuballin proposes the use of Day-
son's model (generally applied to the energy levels of atomic nuclei). Essen-
tially this model seems to simulate successfully the conditions in natural
associations. It is presented below as discussed by Tutuballin:
"Let us consider a circle lying in some plane. Suppose that infinitely
long and infinitely thin charged filaments stand out perpendicularly to the
plant at the points Qi,...,Qn °n this circle. Then, as is well known, they
repel each other and the energy of the whole system is
w
= I
- e
(16)
Let us assume a Gibbs distribution for the points Q^,...,Qn, i.e., a distribu-
tion with the density
Cn£ exp (-Bw) (17)
where 3 = 1/T and T is the absolute temperature. Then we obtain the problem
of the equilibrium distribution of the points Q^,...,Qn which are in thermal
motion at the temperature T, and repel each other according to the laws of
electrostatics. As the temperature approaches infinity, this model becomes a
uniform distribution of independent points, i.e., it gives MacArthur's law.
At a finite temperature, the parts into which the circle is partitioned are
more nearly equal in magnitude than is assumed by MacArthur's law (at zero
temperature the parts are equal). "Consequently," Tutuballin concludes,
"either everything discussed makes no biological sense, or biological systems
must exist, in which due to the struggle for existence, the resulting distri-
bution of the species by population size is more nearly uniform than that
obeying MacArthur's law." The last conclusion drawn by Tutuballin is a suf-
ficiently strong argument for asking the question as to which model (among
various models) simulates more realistically the situation in an association.
It is possible that for some groups of sympatric species in a very mature as-
sociation (an example of which should be sought in tropical forest associa-
tions) , the forecasts made by the Dayson-Tutuballin model will come closer to
reality than the predictions made by the MacArthur and E.B. R. model. In re-
gard to phytoplankton, it appears that the overwhelming number of observations
is indicative of the opposite pattern, a fact which incidentally can be inter-
preted as an argument favoring the concept of the "immaturity" of plankton as-
sociations existing under conditions in which the biotype is disturbed con-
tinuously. Besides, forecasts made by models with rather rough approximation
hardly force us to accept an alternative in the sense that one model is un-
equivocally acknowledged as either definitely valid or invalid. If the E.B.R.
model appears to be inadequate because of the absence of a direct relationship
between the density of the individuals and the limiting factors, at least there
is the possibility that the model can be improved by introducing a new variable
59
-------
in it. If ultimately even this cannot be done, I will be consoled by the
knowledge that, nonetheless, the E.B.R. model makes predictions which agree
better with the facts (at least as far as phytoplankton is concerned) than
those made by MacArthur's B.R. model.
CONCLUDING REMARKS
It appears that my interest which focused on finding a more general
solution to the problem inspired several enthusiasts interested in general
problems of biology. Every one of them contributed some of his work toward
carrying out the appropriate calculations, and undoubtedly enhanced the over-
all progress made in the solution of the more general version of the problem.
As was to be expected, the approaches as well as the degree of their compre-
hension varied considerably. Tutuballin's model is interesting. It includes
at least two items out of the three which make the MacArthur model elegant.
Should it turn out that it conforms to certain ecological facts, its longevity
in ecology can be predicted almost with certainty.
V. A. Svetlosanov1s study, dedicated to a solution of the generalized
problem, is unfortunately mathematically incorrect. In his study formula (1)
is invalid for any function w(a) ^a, and formulas (2) and (3) are invalid no
matter what assumptions are made about w(a). However, even if the invalid
formulas are simply ignored, things are even more complicated in the ensuing
calculations. Thus, Svetlosanov obtains the relative distribution of the spe-
cies from expressions (6) and (7), defining some decreasing sequence of num-
bers. At the same time it is not clear how these numbers must be related to
the average lengths of the parts obtained or to some probabilities (the latter
was done by author in the simplest case when the points were uniformly dis-
tributed and the number of points was greater than one). Unfortunately the
shortcomings pointed out above make this study less valuable. In fact, its
applicability turned out to be too limited and it will therefore gradually
"die out."
In conclusion I would like to acknowledge my deep personal gratitude
to my colleagues and friends. Among these I thank above all V. N. Tutuballin,
M. Ye. Vinogradov, V. V. Nalimov, T. I. Kol'tsova and Yu. Barabasheva for
their valuable help during the discussion of my study and also their useful
critical comments made during the writing of the manuscript and its prepara-
tion for publication.
REFERENCES
Fedorov, V. D. 1969. Dominiruyushchiye formy fitoplanktona beloga morya
[Dominating species of phytoplankton in the White Sea]. Acad. Sci. USSR
Report 188(4):913-196.
Fedorov, V. D. 1970. Soobshchestva fitoplanktonnyky organizmov i sezonnye
ismeneniya yego struktury [A community of phytoplankton organisms and
seasonal changes in its structure]. Botanicheskiy Zhurnal 55(5):626-637.
60
-------
Fedorov, V. D. 1974. Ustoichivost ekologicheskiky sistem i yeyo izmereniye
[Stability in ecosystems and its measurements]. Acad. Sci. USSR Pub.,
Biol. Coll. #3, pp. 402-441.
Fedorov, V. D. Ekologiya fitoplanktona [Phytoplankton ecology]. (In press.)
Fedorov, V. D., and T. I. Koltsova. 1973. Eksperimentalnoye issledovaniye
otnositelnogo obiliye vidov fitoplanktona [Experimental research on the
abundance of phytoplankton species]. Vol. 3, Issue I, pp. 85-93.
Fedorov, V. D., V. Yu. Traskin, and T. A. Dauda. 1974. 0 pishchevoi konku-
rentsii u morskogo fitoplanktona [Food rivalry in marine phytoplankton].
(Regarding E. Khalbort's article "Marine Phytoplankton Rivalry in the Open
Ocean, Shore Zones and Estuaries".) Obshchaya Biologiya, No. 4,
pp. 483-493.
Hairston, H. G. 1969. On the relative abundance of species. Biology
50(6):1091-1094.
Hutchinson, G. E. 1957. Concluding remarks. Cold Spring Harbor Symp. Quant.
Biol. 22:415-427.
Hutchinson, G. E. 1961. The paradox of the plankton. Am. Nat. 95(882):
137-145.
Inger, R. F. 1969. Organization of communities of frogs along small rain
forest streams in Sarawak. J. Animal Ecol. 38:123-148.
King, G. E. 1964. Relative abundance of species and MacArthur model.
Ecology 45:716-727.
Kohn, A. I. 1968. Microhabitats, abundance and Chagos Islands. Ecology
49:1046-1061.
Kol'tsova, T. I., L. A. Konoplya, V. N. Maksimov, and V. D. Fedorov. 1971.
K voprosu o reprezentativnosti vyborok pri kolichestvennoi obrabotke fito-
planktonnyky prob [How representative are the choices given a quantitative
treatment of phytoplankton samples]. Gidrobiologicheskiy Zhurnal, Vol. 7,
No. 3.
MacArthur, R. H. 1957. On the relative abundance of bird species. Proc.
Nat. Acad. Sci. U.S. 43:293-295.
MacArthur, R. H. 1960. On the relative abundance of species. Amer.
Naturalist 94:25-36.
MacArthur, R. H. 1960. Note on Mrs. Pielou's comments. Ecology 47:1074.
MacDonald, K. B. 1969. Quantitative studies of salt marsh mollusc faunas
from the North American Pacific coast. Ecol. Monogr. 39:33-60.
61
-------
Margalef, R. 1958. Temporal succession and spatial heterogeneity in phyto-
plankton. In Perspectives in Marine Biology, symposium held at Scripps
Inst. of Oceanography, March 24-April 2, 1956, pp. 323-335.
Migdal, A. B. 1971. 0 psikhologii nauchnogo tvorchestva [The psychology of
scientific creativity]. Nauka i Zhizn, No. 2, pp. 100-107.
Pielou, E. C., and A. N. Arnason. 1966. Correction to one of MacArthur's
species-abundance formulas. Science 151:592.
Platt, T., L. Dickie, and R. W. Trites. 1970. Spatial heterogeneity of phy-
toplankton in a near-shore environment. J. Fish. Res. Bd. Canada
27(8) -.1453-1473.
Preston, F. W. 1948. The commonness and rarity of species. Ecology
29:254-283.
Svetlosanov, V. A. 1974. Ob odnoi matematicheskoi modeli raspredeleniya
osobei po vidam [One mathematical model of the distribution of individuals
by species]. Obshchaya Biologiya, 35(l):58-63.
Vandermeer, J. H., and R. H. MacArthur. 1966. A reformulation of alternative
(b) of the broken stick model of species abundance. Ecology 47:139-140.
62
-------
IMPACT OF RADIOACTIVITY ON THE MARINE ENVIRONMENT
by
Ford A. Cross
National Marine Fisheries Service
Southeast Fisheries Center
Beaufort Laboratory
Beaufort, North Carolina 28516
ABSTRACT
For the past three decades, man has released artificial radionuclides
into the marine environment predominantly by global fallout from nuclear wea-
pons testing and by controlled releases from nuclear power and reprocessing
plants. A number of ecological and experimental studies have been conducted
to determine the effect of these releases on marine organisms. Results of
these studies indicate that, within our present state of knowledge, measurable
effects on populations or ecosystems have not been observed either on a global
or regional scale.
INTRODUCTION
Contamination of the marine environment by radionuclides has caused
much concern regarding the possible effects of irradiation on marine species.
As a result, considerable research has been conducted over the past two de-
cades in an attempt to determine if the presence of artificial radionuclides
in the oceans has affected marine organisms deleteriously. As new information
relative to the effects of irradiation on marine organisms becomes available,
the possible effect of anthropogenic additions of radionuclides in the marine
environment must be constantly re-evaluated.
One problem in accomplishing this task is to define the term effect in
relation to radiation damage. For purposes of this discussion, we define ef-
fect as "the result of an environmental stress or perturbation which causes an
alteration in the functioning or form of a biological system at any level of
complexity (cellular, organ, individual, population, ecosystem)." Templeton
et al. (1976) argue that in the marine environment, populations and ecosys-
tems, not individuals, should be our ultimate concern. If we accept this ar-
gument, then radiation damage to cells, organs and individuals is significant
only when these effects are manifest at the population or ecosystem level.
63
-------
Three basic methods are used to assess the importance of anthropogenic
additions of radionuclides to natural environments, their resultant dose and
their ultimate irradiation effects on marine organisms. One method compares
natural background dose rates to aquatic organisms with dose rates received
from radionuclides introduced by man. In the second, the dose rate from an-
thropogenic radionuclides can be compared with results of laboratory irradia-
tion experiments with marine species. Lastly, field studies can be conducted
in the particular marine environment receiving artificial radionuclides. The
latter attempts to distinguish changes in ecological structure or function
(i.e., standing crop, species diversity, productivity, etc.) that can be re-
lated to increased levels of radiation. In the following discussion, each
method will be examined in an effort to evaluate the impact that man has had
on the marine environment by the introduction of artificial radionuclides.
COMPARISON OF NATURAL BACKGROUND
DOSE RATES TO MARINE SPECIES WITH
THOSE FROM ARTIFICIAL RADIONUCLIDES
The total concentration of natural radionuclides in water, sediments
and biota, and their subsequent dose to marine organisms is difficult to as-
sess because of the existence of over 60 radionuclides of natural origin and
their variable distribution in the ocean (Folsom and Harley, 1957; Koczy and
Rosholt, 1962; Cherry, 1964; Mauchlin and Templeton, 1964; Lai and Peters,
1967; and Joseph et al., 1971). In addition, dose rates to marine organisms
from both natural and artificial radionuclides will vary as a function of such
variables as pAsition in the water column, proximity to and geological compo-
sition of sediments and submerged strata, specificity for adsorption and accu-
mulation, food intake, etc.
In spite of the variables listed above, Woodhead (1973) and Woodhead
et al. (1976) compared background dose rates with dose rates from global fall-
out and from radioactive discharge in two marine environments. These two lo-
cations were the portion of the Irish Sea near the nuclear fuel reprocessing
plant at Windscale and the Blackwater estuary, England, which receives radio-
active effluent from the Bradwell nuclear power station. They summarized
available data on environmental concentrations of natural and artificial radio-
nuclides and applied this information to dosimetry models for phytoplankton,
zooplankton, mollusks, crustaceans and fish.
Estimates of dose rates from fallout radionuclides are quite variable,
but are on the same order as background dose rates (Table 1). Most of the
background dose rates for phytoplankton, zooplankton and pelagic fish resulted
from incorporated 210Po and 1*°K. For mollusks, crustaceans and demersal fish,
sediments delivered a dose rate similar to that received from incorporated ac-
tivity. For fallout, the organisms' body burden of artificial radionuclides,
particularly 137Cs, contributed the most significant fraction of the dose, al-
though 239Pu in phytoplankton and 90Sr/90Y in zooplankton and Crustacea were
of some importance.
The releases of radionuclides from Windscale produced significantly
higher doses to marine organisms in the immediate vicinity of the discharge
64
-------
than from either fallout or background (Table 1). In addition, 970 fish tagged
and recaptured near the point of discharge gave a mean dosimeter reading of
350 yrad hr"1 (Pentreath et al., 1973).
In the Blackwater estuary the dose rate from radionuclides released by
the power plant could be estimated only for oysters and other benthic organisms
and were significantly below background estimates. The total number of curies
of radioactivity (excluding tritium) released annually by this nuclear power
station exceeds the amount released by both boiling water and pressurized nu-
clear reactors in the United States by a factor of two to ten (Joseph et al.,
1971; Woodhead, 1973).
COMPARISON OF LABORATORY IRRADIATION EXPERIMENTS
WITH ENVIRONMENTAL DOSE RATES
In recent years, several review articles have summarized laboratory ir-
radiation experiments on both the lethal and nonlethal responses of marine or-
ganisms to radiation (Polikarpov, 1966; Rice and Wolfe, 1971; Templeton et al.,
1971; Chipman, 1972; Rice and Baptist, 1974). These reviews indicate that
fish, particularly their eggs and larvae, are more sensitive to radiation than
other marine species, although the actual dose rate required to produce obser-
vable effects on individuals from radionuclides dissolved in seawater are in
dispute (Polikarpov, 1966; Templeton et al., 1971). Ophel et al. (1976) gen-
eralize from these published laboratory data that a minimum acute dose of about
100 rads is required to produce some mortality in a population of marine orga-
nisms. In addition, they estimate that a chronic dose of at least one rad
day"1 (40 mrad hr"1) is required to produce observable physiological effects
in marine organisms. At field or experimental dose rates below this, they in-
dicate that observable effects are masked by inherent biological variation.
Shekhanova (this proceedings), however, states that effects on reproductive
capacity of fish have been observed at doses ranging from 0.1-0.3 rad day"1
Comparison of these estimates of minimum dose rates required to pro-
duce observable effects on individual organisms in laboratory experiments to
maximum estimates of environmental dose rates listed in Table 1 indicates that
present levels in the ocean are significantly below most levels shown to af-
fect marine organisms adversely. Maximum dose rates from fallout range from
4.6 x 10" mrad hr"1 in Crustacea to 0.147 mrad hr"1 in zooplankton. These
rates are 270 to 87,000 times less than the estimated minimum dose rate of 40
mrad hr~ discussed above. In the North Irish Sea, however, estimated maximum
dose rates from radionuclides released from the nuclear fuel reprocessing
plant at Windscale ranged from 3.9 x 10"3 mrad hr"1 for pelagic fish to 6.9
mrad hr~ for zooplankton. This maximum estimate is within a factor of six of
the one rad day" value and similar to the values obtained by Shekhanova.
Caution must be exercised, however, in extrapolating results from la-
boratory experiments to natural conditions. As indicated by Perez (this pro-
ceedings) , duplication of natural conditions in the laboratory is extremely
difficult and results obtained in artificial systems may be misleading.
65
-------
>H
EH
H
f^
U
O
H
f^
ril
Qu
t-N
i-}
^-C
EH
S
1
1 i
f.
(^
2
W
g
fa
w
a
I.
H
z
^
rh
05
O
H
&
S^
»±4
O
£-1
^
r-t
1
•**
•d
rd
^~i
^^
w ^~.
W n
EH r-
K iH
W -
W Q
P-i Q
o o
>H |S
/v
11
D fa
cn —
*
i-H
w
•i]
F-!
43
to
•H
id
0)
CJ
id
4-1
CO
2
id
o
CO
3
rH
rH
S
C
O
4-1
*S.
G
cd
i-t
0,
§
N
C
o
4-1
cd
rH
a
O
p-t
0)
o
VJ
o
M
*»
43
4-1
a.
01
T3
B
o
CN
•*
£
4-J
a
cu
•a
B
0
CM
^r?
4J
a
01
13
^
CN
•*
4-1
a
01
13
el
CM
*
43
4-1
a
cu
•o
B
0
CN
43*
4J
a.
01
•o
B
0
CN
0
CU
4J
O
Jj
C
o
c
o
cu
4J
o
§
0)
4-1
1
13
0)
cd
0)
ca
0)
4-1
13
CU
43
cd
CU
CO
IW
13
0)
43
cd
CU
ca
cu
cC
4J
1
id
ca
.-i
4-
4-1
•a
cu
43
id
CO
B
tj
M-l
"S
42
cd
cu
co
14-1
CN
1
in r-
O CM
CM
^
in '
. .
O CM
-3-
.
rH
CM
1
in o\
o r-x
OS
•
~3-
rH
|
m st
O 1^
p-^
»
lO
i-t
1
lO vO
• •
0 CN
co
1
m os
O rH
§ >s
5 n w
O 0 iH
O 4M iH
$4 cd 4J
G -H 0
^J ^3 ^
co id
rl 1-H
cd
U C
< B o)
PS n u
5 o C
H O rH
<
Z
^
O
fh^
.
o
rH
O
rH
O
CN
.
O
O
4J
•H
>
iH
4J
CJ
cd
V4
01
4J
5
0 0
SO rH
rH CN
1 1
rH rH
1 1
0 0
SO rH
rH CM
in so
rH rH
0 0
SO rH
rH CN
1 1
in so
rH rH
1 1
1 1
^- CQ
>-,
•H
•rl
4-1
O
4-1
a
<0
•H
CU
C/3
CO
O
CN
oo
rH
CO
.
CO
CM
00
oo
CN
i-l
0
H
O
o
1
CN
O
0
SO
O
»
0
1
CN
O
t
0
rx
OS
O
.
0
1
sf
O
0
O
CN
CO
.
0
1
^D
o
0
•^~
*
ff}
rH
1
O
CN
•
o
00
00
•
o
rH
o
0
^
-3-
rH
X
CO
^s 3
V fc
>CO
•HCS
4J .
cj ca
cd u
fx,
i-HCO
CdrH
13
H rl
b o) tj
0 4J W
iJ CO
P-J r-l OS
<
fa
1
o
rH
r^. ^-^
• CN
rH CO
1 1
CN CN
rH >
• CD
1
O
X
• m
rH VO
I |
CN CO
rH •
• O
O
rH
^^
CM
en
1
SO CN
O-*-'
I
O
rH
X
. CM
r^ en
1 i
<3- CN
0 •
O v
rH
rH
O
•
O
•
rH iH
CJ 4-1
3 O
C id
rl M
CU 01
4-1 Cd
0 3
00
rH
1
rH
0
oo
rH
Sf
i-H
.
O
vD
s^~
.
o
1
SO
CO
CD
O
•
00
1
o
rH
CD
r-^
•3-
rH
1
-*
•
»H
m
•
CN
1
so
CN
o
rH
cd
o
H
o -
CM -3 O
in . co oo
co co
I m
I
in
I
m o
o o co CM
CN
I
in o
CD o i i
o
OS CM -? O
. • CO 00
r^ rH co co
so i i in
I m <• i
os o • r-»
. . so o
so O CO CN
OS O
. CN st O
00 • CO 00
in rH CO CO
I I i m
co m -3 i
. CD • r*x
in • so CD
rH O CO CN
o
o o
OS .
SO CO
I I
O CM
en •
in o i
o
o co
rH .
CN CO
I I
O CM
O •
CN O I
>- 01
3
en
O
PH
CO
M
Q
B
3
4J
O
cd
i-H
cu
rH
id
CJ
n
t)
(3
•rH
cd
c
cu
4-1
c
M
iH
>
-r(
4-1
CJ
CO
M
01
4->
It)
3
4-1
CJ
cd
4-1
C
cu
B
•H
13
01
CO
CO
CO
OS
CO
CO
-3
o
o
-3
CO
I
00
o
o
Os
en
in
o
o
o
CN
OS OS
so co
1 1 1
OS CN
vO CO
rH
oo
r^ os CM
CO SO CO
CO
I
III
1 1 1
4J 4J
o a
id id
(H
cd
4-1
O
H
cd C
rH
rH
CU
>j
•o
cd
pq
C 0)
rl B
0) iH
4-1 13
c3 0)
M C/3
£
66
-------
DETECTION OF ECOLOGICAL EFFECTS RESULTING
FROM RADIOACTIVE CONTAMINATION
The detection of ecological changes caused in a marine ecosystem by
the presence of a contaminant such as radioactivity is perhaps the most diffi-
cult type of assessment discussed in this paper. Our knowledge of ecological
systems is not precise enough to predict what processes or structural charac-
teristics would be most likely affected by radiation and which therefore
should be monitored.
In the marine environment, three areas have received relatively large
quantities of radioactivity as a result of man's activiries. These are the
Irish Sea, Columbia River estuary and adjacent coastal waters, and Central Pa-
cific proving grounds. These areas have been the subject of relatively in-
tense environmental studies (Lowman, 1960; Osterberg et al., 1974; Pruter and
Alverson, 1972; Pentreath et al., 1973; Cross et al., 1975; Templeton et al.,
1971). No gross ecological change that could be related to the presence of
artificial radionuclides has been observed in the environment. Gorbman and
James (1963), however, were able to detect radiation damage in thyroids of
fish collected at Eniwetok Atoll at one-month and eight-month intervals after
a nuclear explosion. No attempt was made to predict what effect this damage
had on the population of fish living near the atoll.
Failure to detect radiation-induced changes at either population or
ecosystem levels in these environments cannot be used as an argument that no
effects have occurred. Instead it may reflect the superficiality of our under-
standing of how the system operates. We can only state that no catastrophic
mortalities have occurred and more subtle changes which we might measure could
be due to long-term fluctuations in the ecosystem.
We can infer, however, the possible effects of radiation at the popu-
lation level using knowledge of the population dynamics of commercial fish
species. This subject was examined by a recent International Atomic Energy
Agency Panel studying "Effects of Ionizing Radiation on Aquatic Organisms and
Ecosystems" (Templeton et al., 1976). The panel considered the role of den-
sity-dependent mortality in the stock-recruitment relationship in marine popu-
lations of both high and low fecundity (Fig. 1). Using commercially-exploited
fish stocks as an example, the authors concluded that "if mortality of eggs is
being enhanced by the low levels of irradiation presently existing in the ma-
rine environment, then recruitment to the stocks of highly fecund marine spe-
cies of fish is unlikely to be adversely affected unless those stocks are al-
ready at risk because of severe over-exploitation." In other words, the re-
sult of mortalities of eggs caused by irradiation would decrease larval compe-
tition for food and space and, therefore, would increase the probability of
survival for the remaining individuals.
Survival rates for highly fecund density-dependent stocks increase
dramatically at low stock sizes (Fig. 2). The curve demonstrates the relation-
ship between spawning stock size and survival for Atlantic menhaden for the
1955-1970 year class. Each dot on the graph shows the estimated egg produc-
tion and percent survival to age one by year. In this case, the decrease in
egg production of the spawning stock was caused by severe exploitation by the
67
-------
fishing industry and adverse environmental conditions. We would expect the
same population response, however, if some perturbation such as radiation was
causing mortalities of eggs and larvae. Obviously we would not expect the
density-dependent relationship to compensate for radiation-induced mortalities
in severely exploited fish stocks.
O£
t-
111
at
IU
DC
STOCK SIZE (S)
Figure 1. Density-dependent relationship between spawning stock size and re-
cruitment in highly fecund species (figure from Beverton and Holt,
1957)
In addition, evidence is presented which indicates that increased
pressure on fish stocks by over-exploitation or other stresses is compensated
for by an increase in fecundity of surviving adults. In this manner, fish
stocks have survived total mortality rates of 60-70% a year. Blaylock (1969),
for example, reported that the mosquito fish, GamJbusia affinis, increased fe-
cundity relative to controls in the presence of chronic exposure to radio-
activity in a freshwater environment.
The inherent dynamics of highly fecund marine populations, therefore,
would compensate for additional mortalities of young caused by contaminants
such as radionuclides. This compensating mechanism is limited in its capacity
to "protect" a species which experiences high mortalities in its early stages.
The actual number of mortalities that must occur to affect a population sig-
nificantly would be highly variable and dependent on a number of additional
factors such as predation, food supply, exploitation, etc.
68
-------
100(-
z
o
UJ
o
z
UJ
s
ac
O
010
a:
UJ
a.
001
58
63
62
.61
20 40 60 80 100120 140 160 180
EGG PRODUCTION OF SPAWNING STOCK X 1012
Figure 2. Relationship between spawning stock size and survival for Atlantic
menhaden (Brevoortia tyrannus) for 1955-1970 year classes.
(W. Nelson, personal communication)
CONCLUSIONS
The foregoing discussion indicates that man's introduction of artifi-
cial radionuclides into oceans has not caused significant adverse effects on
marine populations. Within the scope of our knowledge, measurable changes in
marine populations or ecosystems have not been observed in regions where arti-
ficial radionuclides have been released. In addition, most estimates of dose
rates to aquatic organisms from releases of radionuclides are comparable to
natural background dose rates and are significantly below dose rates most fre-
quently shown to cause measurable damage to organisms in the laboratory. Al-
though these discussions have centered on somatic effects, available evidence
also indicates that no adverse genetic effects have occurred (Purdom, 1966;
Templeton et al., 1976).
69
-------
If nuclear weapons testing in the atmosphere is not resumed on a major
scale, the oceans as a whole probably experienced their highest concentrations
of artificial radionuclides in the early 1960's. Since the Nuclear Test Ban
Treaty was signed, concentrations of fallout radioactivity have decreased sig-
nificantly and no widespread contamination has resulted from nuclear power
plants or nuclear-powered ships (Joseph et al., 1971) . As long as existing
levels of radioactivity in the oceans are kept at levels tolerable for man,
harm to marine organisms is expected to be minimal or non-existent.
I have two major concerns for the future: catastrophic accidents and
plutonium. The rupturing of a nuclear reactor in a coastal power plant or on
a ship could release large quantities of radioactivity over a relatively
large geographical area and expose marine organisms in this area to dose rates
significantly higher than those known to cause mortalities. It is hoped that
the present concern for safety in construction and operation of nuclear reac-
tors will continue and such an event will never occur.
Because of its emission of alpha particles, long residence time in
tissues and lack of a stable isotope, plutonium is the most toxic radionuclide
known to man. Its usefulness to industrial and military establishments is
causing the world-wide inventory of plutonium isotopes to increase signifi-
cantly (National Academy of Sciences, 1975). Increased use of this material
enhances the probability that additional uncontrolled releases to the marine
environment will occur. A study panel for the U.S. National Academy of Sci-
ences (1975) recommended: "Releases of transuranic elements to marine environ-
ment should be kept to a minimum. Any releases should be monitored, and plans
for significant increases above current levels should be carefully scrutinized
and regulated."
Although man has not released concentrations of radionuclides that
have adversely affected marine populations, constant vigil must be maintained
to assure that future uses of nuclear power do not cause widespread contamina-
tion. As our understanding of ecological processes improves and as experimen-
tal techniques are refined, research on the effect of irradiation from low
chronic dose rates may disclose effects that are presently undetectable.
ACKNOWLEDGMENTS
I would like to acknowledge Dr. David Engel and Dr. Walter Nelson of
this laboratory for their helpful suggestions in the preparation of this
manuscript.
REFERENCES
Beverton, R. J. H., and S. J. Holt. 1957. On the dynamics of exploited fish
populations. Fish. Invest. Ser. II, Mar. Fish. Gr. Brit. Minist. Agric.
Fish. Food 19. 533 pp.
Blaylock, B. G. 1969. The fecundity of a Gambusia affinis affinis population
exposed to chronic environmental radiation. Radiat. Res. 37:108-117.
70
-------
Cherry, R. D. 1964. Alpha-radioactivity of plankton. Nature 203:139-143.
Chipman, W. A. 1972. Ionizing radiation. Pages 1579-16S7 in 0. Kinne, ed.,
Marine Ecology, vol. 1, pt. 3. Wiley-Interscience, New York.
Cross, F. A., W. C. Renfro, and E. Gilat. 1975. A review of methodology for
studying the transfer of radionuclides in marine food chains. Pages 185-
210 in Design of Radiotracer Experiments in Marine Biological Systems.
Int. At. Energy Agency Tech. Rep. Ser. 167. IAEA, Vienna.
Folsom, T. R., and J. H. Harley. 1957. Comparison of some natural radiators
received by selected organisms. Pages 28-33 in The Effects of Atomic
Radiation on Oceanography. National Academy of Sciences-Nationa' Research
Council Publ. 551.
Gorbman, A., and M. S. James. 1963. An exploratory study of radiation damage
in the thyroids of coral reef fishes from the Eniwetok Atoll. Pages 385-
399 in V. Schultz and A. W. Klement, eds., Radioecology. Reinhold and
American Institute of Biological Sciences, Washington, D.C.
Joseph, A. B., P. F. Gustafson, I. R. Russell, E. A. Schuert, H. L. Volchok,
and A. Tamplin. 1971. Sources of radioactivity and their characteris-
tics. Pages 6-41 in Radioactivity in the Marine Environment. National
Academy of Sciences, Washington, D.C.
Koczy, F. F., and J. N. Rosholt. 1962. Radioactivity in oceanography.
Pages 19-46 in Nuclear Radiation in Geophysics. Springer-Verlag, Berlin.
Lai, D., and B. Peters. 1967. Cosmic ray produced radioactivity on the
earth. Pages 600-601 in Encyclopedia of Physics. Springer-Verlag, New
York.
Lowman, F. G. 1960. Marine biological investigations at the Eniwetok test
site. Pages 105-138 in Disposal of Radioactive Wastes. Int. At. Energy
Agency, Vienna.
Mauchline, J., and W. L. Templeton. 1964. Artificial and natural radioiso-
topes in the marine environment. Pages 229-279 in H. Barnes, ed., Oceanog-
raphy and Marine Biology, vol. 2. Allen and Unwin, London; Hafner, New
York.
National Academy of Sciences. 1975. Assessing potential ocean pollutants, a
report of the Study Panel on Assessing Potential Ocean Pollutants to the
Ocean Affairs Board, Commission on Natural Resources, National Research
Council. National Academy of Sciences, Washington, D.C. 438 pp.
Ophel, I. L., M. Hoppenhett, R. Ichikawa, A. G. Klimov, S. Kobayashi, Y.
Nishiwaki, and M. Saiki. 1976. Effects of ionizing radiation on aquatic
organisms. Pages 57-86 in Effects of Ionizing Radiation on Aquatic Orga-
nisms and Ecosystems. Int. At. Energy Agency Tech. Rep. Ser. 172.
Osterberg, C., W. G. Pearcy, and W. G. Curl, Jr. 1964. Radioactivity and its
relationship to oceanic food chains. J. Mar. Res. 22:2-12.
71
-------
Pentreath, R. J., D. S. Woodhead, and D. F. Jefferies. 1973. The radio-
ecology of plaice (Pleuronectes platessa L.) in the northeast Irish Sea.
Pages 731-736 in D. J. Nelson, ed., Radionuclides in Ecosystems, Proceed-
ings of the 3rd National Symposium on Radioecology, vol. 2. U.S. AEG, Oak
Ridge, Tenn. (CONF-710501-P2)
Perez, K. Persistent limits in aquatic ecosystems (this proceedings).
Polikarpov, G. G. 1966. Radioecology of Aquatic Organisms (translated from
the Russian). Reinhold, New York. 314 pp.
Pruter, A. T., and D. L. Alverson, eds. 1972. The Columbia River Estuary and
Adjacent Ocean Waters. Univ. of Washington Press, Seattle. 868 pp.
Purdom, C. 1966. Radiation and mutation in fish. Pages 861-867 in Disposal
of Radioactive Wastes into Seas, Oceans and Surface Waters. Int. At.
Energy Agency, Vienna.
Rice, T. R., and J. P. Baptist. 1974. Ecologic effects of radioactive emis-
sions from nuclear power plants. Pages 373-439 in L. A. Sagan, ed., Human
and Ecological Effects of Nuclear Power Plants. C. C. Thomas, Spring-
field, 111.
Rice, T. R., and D. A. Wolfe. 1971. Radioactivity—chemical and biological
aspects. Pages 324-379 in Donald Hood, ed., Impingement of Man upon the
Oceans. Wiley, New York.
Shekhanova, I. A. Influence of radioactive pollution of reservoirs on fish
(this proceedings).
Templeton, W. L., B. G. Blaylock, M. J. Holden, D. S. Woodhead, R. N. Mukher-
jee, 0. Ravera, M. Bernhard, and L. S'ztanyik. 1976. Effects of ionizing
radiations on aquatic populations and ecosystems. Pages 89-102 in Effects
of Ionizing Radiation on Aquatic Organisms and Ecosystems. Int. At.
Energy Agency Tech. Rep. Ser. 172.
Templeton, W. L., R. E. Nakatani, and E. E. Held. 1971. Radiation effects.
Pages 223-239 in Radioactivity in the Marine Environment. National
Academy of Sciences, Washington, D.C.
Woodhead, D. S. 1973. Levels of radioactivity in the marine environment and
the dose commitment to marine organisms. Pages 499-523 in Radioactive
Contamination of the Marine Environment. Int. At. Energy Agency, Vienna.
Woodhead, D. S., A. Frantz, M. Ijuin, G. P. Olivier, J. Sas Hubicki, E. Wan-
derer, S. W. Fowler, and M. Bezzegh-Galantai. 1976. Concentrations of
radionuclides in aquatic environments and the resultant radiation dose
rates received by aquatic organisms. Pages 5-54 in Effects of Ionizing
Radiation on Aquatic Organisms and Ecosystems. Int. At. Energy Agency
Tech. Rep. Ser. 172.
72
-------
THE EFFECT OF RADIOACTIVE POLLUTION
OF RESERVOIRS ON FISH
by
I. A. Shekhanova
Institute of Marine Fisheries and Oceanography, USSR
The distinguishing feature of contemporary pollution of the water with
artificial radioactive materials consists in the predominant localization of
these materials in the coastal and shelf zones of the ocean, in inland seas,
in rivers and lakes, I.e., in regions of increased biomass and productivity.
The possibility of increasing the concentration of artificial radionuclides in
these regions is growing in connection with the existing practice of open dump-
ing of low activity waste products in coastal regions of seas, bays and inland
reservoirs. The results of multiple measurements of levels of the content of
radioactive products in reservoirs of different kinds conducted in a number of
studies confirm this thesis.
Today, the concentration of strontium-90 in coastal waters of the
World Ocean comprises 0.1-0.2 pCe/1, in the inland and marginal seas 0.5-1.5
pCe/1, in rivers 1.5-5 pCe/1, and in lakes 3.5-12pCe/l. In a number of reser-
voirs in which the radioactive wastes of nuclear industries are discharged,
cumulative beta-activity fluctuates from 25 to 2000 pCe/1. A high concentra-
tion of certain radionuclides, including cesium-137, has been noted in bottom
sediments.
Expansion of the scales of utilizing nuclear energy entails a propor-
tional increase in the amount of radioactive wastes, part of which will un-
avoidably enter the reservoirs. The developing situation is stimulating an
attentive analysis of the biological consequences that may result from the in-
creased content of artificial radionuclides in reservoirs and determine the
threshold of permissible irradiation of fishes at different stages of ontoge-
nesis.
The biological effect during prolonged habitation of fish in the radio-
actively polluted environment is due to the power of absorbed dose and to the
cumulative dose of radiation. At all stages of ontogenesis the power of the
dose of radiation of the fish with the presence of artificial radionuclides in
reservoirs consists of the external and internal sources. Sources of internal
radiation are the radionuclides that are incorporated in given organs and tis-
sues. Sources of external radiation are the radionuclides contained in the
water, accumulated by aquatic plants, and absorbed in the bottom deposits.
Additionally, radiation resulting from the radionuclides in neighboring organs
73
-------
or tissues that are characterized by a high degree of accumulation can be ex-
ternal with respect to some particular organ.
Dose power of internal radiation is determined by the intensity of ac-
cumulation of radionuclides in the organs of the fish and by their energy of
radiation. The dose power of external radiation is closely dependent on the
ecology of fish.
Calculations have shown that the power of the dose of radiation of
ocean and sea pelagic fishes resulting from artificial radionuclides today is
2-3 orders smaller than the dose power of radiation from natural sources of
ionizing radiation. In rivers the doses of radiation of fish from natural and
artificial sources are comparable. In lakes, especially if cesium-137 is pres-
ent in them, the dose of radiation of bottom fishes from artificial sources is
5-10 times higher than that from natural sources. In reservoirs in which low
activity wastes are discharged, dose power of radiation of fish resulting from
artificial radionuclides fluctuates within limits of 0.1-1 rad/day and some-
times reaches even higher levels.
Natural radiation loads pertain to the constantly acting factors of
the environment, to which all living organisms have developed corresponding
radioresistance over the course of prolonged evolution. Consequently, there
is no basis to assume the possibility of the harmless effect of artificial
radionuclides on ocean and marine pelagic fishes with the doses of radiation
that presently exist because of them. Certain changes in the biological con-
dition of fish can exist in fresh water fish and fish that inhabit stretches
of seas regularly polluted by discharges of low activity wastes.
In coastal areas of the seas a large number of valuable commercial
species of fish spawn: the herring, the Baltic herring, the whitefish, the
sole and many others. The fry of these and other species of fish also inhabit
these areas for a long period of time. By living in the aquatoria polluted by
radioactive wastes, they are exposed to radiation at the most radiosensitive
period of ontogenesis—in the period of the establishment and development of
all organs and systems.
The problem of the biological effect of small doses of ionizing radia-
tion on living organisms is extremely complex. The scale of danger created by
chronic irradiation with low dose power cannot be predicted based on experi-
ments with acute irradiation. Conclusions regarding the harmful consequences
of radioactive pollution of the environment should come from experiments with
chronic continuous irradiation simulating the conditions that are created for
fish in reservoirs. This tenet was taken as the basis for determining the ef-
fect of radioactive pollution of the water on fish. Functional changes in dif-
ferent organs and systems and their structural changes were used as indicators.
Of the large number of investigations that have been devoted to iden-
tifying the effect of ionizing radiation on the fish, from the viewpoint of
predicting the biological consequences of pollution of reservoirs with radio-
active products, those are of interest that characterize deviations from the
normal in a range of doses that actually exist or are possible with considera-
tion of the scale of development of nuclear energy. In regard to this we
74
-------
shall analyze materials obtained during the chronic irradiation of fish with a
dose power of 0.1-1 rad/day.
The capacity for reproduction, fertility and the quality of the spawned
generation are the most significant indicators of the well-being of any living
organism. They determine in fact the numbers of the population and the re-
plenishment of the school of fish. It has been established that at all stages
of ontogenesis prolonged irradiation of the sexual organs with a dose power of
0.1-0.3 rad/day causes disruption of the process of formation and function of
reproduction. The morphological structure of the primary germ cells and their
time of appearance are similar in the irradiated and intact specimens. The
prolongation of the miotic cycle of the primary germ cells that leads to a de-
lay in sexual differentiation is noticeable. At the later stages of gameto-
genesis, the changes caused by irradiation appear more clearly.
Chronic irradiation of the testicles leads to biochemical changes that
are expressed in a reduced content of glycogen and an increase in the content
of fat. In the generative tissues cells appear with pycnoticlysis of the
nucleus.
The ovaries of fish, like those of the warm-blooded animals, are more
radioresistant than the testicles. In the process of differentiation and de-
velopment of the oocytes, irradiation of the ovaries with a dose power of
0.1-0.3 rad/day entails a change in the volume of the nucleus and protoplasm
and the nuclear-plasma ratio. There is a reduction in the number of young
oocytes, and as a result relative fertility of the irradiated females becomes
1.5 times less than normal. The generation obtained from fish irradiated in
the indicated range of doses is characterized by reduced viability.
An increase in the dose power of radiation of the gonads to 1-3 rad/
day leads to more profound changes. Under these conditions the males lose
their capacity for reproduction after irradiation of the testicles in a dose
of 550-660 rad, and the females following irradiation of the ovaries with a
dose of 1000-1200 rad.
The functional condition of the endocrine system, specifically, gonado-
tropic activity of the hypophysis, is intimately related to the reproductive
capacity of the fish. A histological investigation has shown that during
chronic irradiation of the hypophysis with a dose power of 0.1-0.3 rad/day,
growth of the hypophysary parenchyma is observed and the process of secretion
of the hormone into the blood stream is disrupted. With an increase in dose
power the effect of the action increases.
A reduction in the resistance of irradiated fish to parasitic and in-
fectious diseases has been established. This reaction is due to changes in
the hematopoietic process, the morphological composition of the blood and the
immunological reactiveness. In embryos of fish during chronic irradiation of
the axial cellular mass, responsible for hematopoiesis, it is more weakly pro-
nounced than in the intact ones. The process of differentiation of the mesen-
chyme cells into hemocytoblasts is inhibited, and the intensity of hemato-
poiesis decreases on the periblast as well. Under the effect of irradiation
there is an increase in the number of erythrocytes with morphological disrup-
75
-------
tions and there is an increase in the number of pathological mitoses. Retarda-
tion is noted in the formation of the pronephric glomeruli. The number of
leucocytes decreases, which leads to inhibition of phagocytic activity.
Changes in the process of hematopoiesis have a phase character and depend on
the power and total magnitude of the absorbed dose. A decrease in the immuno-
logical reactiveness in fry is noted upon irradiating the kidneys with a dose
of 4-5 rad with a dose power of 0.05-0.1 rad/day.
Hence, a clearly unfavorable situation for fish is developing in the
aquatoria polluted with radioactive wastes. The doses of radiation that form
there entail a reduction in the viability of fish and disruption of reproduc-
tive function. It does not seem possible to block off contaminated stretches
o€ the sea from the entry of fish into them from the purer regions. Propor-
tional to expansion of the scale of utilizing nuclear energy and the increase
in the amount of radioactive wastes discharged into reservoirs, the area of
polluted stretches will grow proportionally. This should be considered when
developing forecast monitoring.
76
-------
A NON-STANDARD APPROACH TO HETEROTROPHY
ATP ESTIMATION OF NATURAL POPULATIONS
OF SELECTIVELY-FILTERED BACTERIOPLANKTON
AND THEIR GROWTH RATES ON IN-SITU WATER
IN DIFFUSION-CULTURE!
by
John McN. Sieburth and Dennis M. Lavoie
Graduate School of Oceanography
University of Rhode Island
Kingston, Rhode Island 02881
ABSTRACT
The sieve-like properties of the Nuclepore membrane filter are uti-
lized to separate the bacterioplankton from the larger plankton, and its bio-
mass is determined by ATP assay. Diurnal growth of the bacterioplankton is
observed in a chamber with a Nuclepore membrane wall designed to cage the mi-
croorganisms while allowing naturally occurring nutrients to diffuse freely
into the culture. The results obtained at a blue-water station east of the
Azores in August 1975 are presented to show the kind of information that can
be obtained. The bacterioplankton biomass varied from 1.6 to 8.4 mg C/m3 over
the uppermost 250 m, and its distribution was strongly correlated to that of
phytoplankton and dissolved carbohydrates. Bacterioplankton exposed in growth
chambers to water continuously pumped from 50 m (oxygen maximum) and 80 m
(chlorophyll maximum) exhibited diurnal growth patterns from which daily pro-
duction was estimated to be 20 to 67 mg C/m3. A calculated mean annual pro-
duction of 13 g C/m3 is some three times the mean obtained by other methods.
INTRODUCTION
A first approximation of the functional role of a class of organisms
within an ecosystem requires a knowledge of both its biomass and rate of pro-
duction. The development of procedures for such studies of bacterioplankton
would permit observations on the distribution of biomass, factors that control
its rate of production, and evaluation of its role in the in-situ transforma-
tion of organic carbon. Obtaining such data for any group of organisms is
This study was supported by Grant DBS 74-01537 from the Biological
Oceanography Program of the National Science Foundation.
77
-------
difficult, but especially so for the bacterioplankton due to their small size,
small biomass, and rapid adaptation to changing conditions. However, these
very attributes can be used to advantage in designing new methods such as
those developed in this laboratory over the past several years. They include
a procedure for the biochemical estimation of the biomass of bacterioplankton
as well as a direct method for estimating growth rates and growth patterns in
situ. Although these technig_ues are still under development, we feel that
they show promise as a method for studying bacterioplankton in both natural
and pollution-stressed planktonic ecosystems. They offer alternatives to the
direct microscopic counting procedures for biomass and the 1'*C indirect method
of growth rate measurement used by Sorokin (1971) for characterizing bacterio-
plankton in the Pacific Ocean. During R/V Trident Cruise 170 in the North At-
lantic from Rhode Island along the shelf to the Grand Banks and across the At-
lantic via the Azores to Spain, a spectrum of water masses was examined by
these procedures. To illustrate the type of information that can be obtained,
we report here the results of one station. Station 13 was a 4,450 m deep blue-
water station east of the Azores at 36°59'N, 21°22'W that was occupied for 36
hours commencing at 0800 local time 14 August 1975.
We gratefully acknowledge the assistance of all members of the scien-
tific party of TR-170, especially Kenneth R. Hinga and Paula J. Willis for ATP
determination, Fred W. French III for pigment determinations, Kenneth M. John-
son and Curtis M. Burney for carbohydrate determinations, and James Hannon for
marine technical services.
METHODS
BIOMASS ESTIMATION BY ATP
Samples of seawater are drawn through Nuclepore filter membranes
(Nuclepore Corp., Pleasanton, Calif.) by a 100 mg Hg vacuum. The ATP of micro-
organisms retained on the filter is extracted and assayed with a procedure
similar to that of Holm-Hansen and Booth (1966) . The filter is immersed in
5 ml of boiling Tris (hydroxymethylaminomethane) buffer (pH 9.0 at 20°C) and
extracted for two minutes. The extract is then immediately frozen until as-
sayed aboard ship at the close of the station, using a DuPont Luminescence
Biometer (E. I. DuPont de Nemours Co., Wilmington, Del.) to measure the
luciferin-luciferase light reaction (Allen, 1972).
Differentiation between the ATP of bacterioplankton and that of larger
particles is accomplished by selectively filtering the 1,000 ym prescreened
sample first through a 3 ym Nuclepore filter and then through a 0.2 ym pore
size Nuclepore. The filters are then extracted for ATP. Since the Nuclepore
filter functions as a microscopic sieve (Sheldon, 1972), the two fractions
represent the ATP of particles between 3 and 1,000 ym and that between 0.2 and
3 ym. Replicate water samples of 145 ml were taken from 5 and 30 liter Niskin
bottles of PVC plastic which were washed with 0.1 N HC1 just prior to being
lowered to depth (Sorokin, 1971).
78
-------
ESTIMATION OF GROWTH RATES
By applying the bioengineering principles for the dialysis culture of
microorganisms described by Schultz and Gerhardt (1969), we developed a cul-
ture chamber (Fig. 1) in which a study population of microorganisms can be
held captive and yet be exposed to dissolved natural substrates at natural
concentrations (Lavoie, 1975). A 0.1 ym pore size Nuclepore membrane is used
as a barrier to retain the bacteria while soluble substrates are allowed to
diffuse into the culture from a continuously replenished volume of seawater
and metabolic by-products are able to diffuse out. Theoretical considerations
indicate that substrate concentration and rate of biomass increase within this
culture device should closely simulate the natural situation (Lavoie, 1975).
This is achieved by the chamber configuration and a vigorous agitation of the
liquid adjacent to both the membrane surfaces which promotes a rapid transfer
of solutes across the membrane. The population therefore responds rapidly to
changes in the quality and quantity of substrate in the seawater, thus enab-
ling the observation of natural growth patterns.
To inoculate the chambers, seawater from the 50 and 80 m depths (the
oxygen and chlorophyll maxima respectively) were aseptically filtered through
a 3 ym pore size Nuclepore filter, and 180 ml portions were put into each pre-
viously autoclaved culture device. Seawater was drawn up continuously from
50 m and 80 m depths through two polyethylene tubes, 0.6 mm inside diameter.
A high speed peristaltic pump provided water at approximately 500 ml/min to a
continuously overflowing 2-liter reservoir bottle for each depth while tubing
from each bottfle supplied water at a rate of about 100 ml/min to the medium
chambers of each of three replicate culture devices. The chambers were im-
mersed in a flowing water bath in the dimly lit wet lab, but temperature and
light conditions were less than ideal. At four-hour intervals duplicate sam-
ples were drawn from each chamber and assayed for ATP concentration.
OTHER PARAMETERS
Several other measurements were made for supportive data. Of those
reported here, chlorophyll a and phaeopigments were measured by fluorometry
(Yentsch and Menzel, 1963; Holm-Hansen et al., 1965); dissolved oxygen was
measured by the standard Winkler titration; carbohydrates were measured as
hexose equivalents by a technique recently developed in our laboratory (John-
son and Sieburth, 1976; Burney and Sieburth, 1976). Water from the upper 150
ym of the air-sea interface was obtained with a nylon screen sampler (Sieburth
et al., 1976).
RESULTS
In Figure 2 the vertical distribution of the two fractions of particu-
late ATP is plotted on the left. Pigments, dissolved carbohydrates, and dis-
solved oxygen are superimposed on the same depth scale to show their relative
distributions. For the smaller than 3 ym fraction, three ATP peaks are imme-
diately obvious: at the sea-air interface, at 50 m (oxygen maximum) aid at
80 m (chlorophyll maximum). All points except those at 30 and 40 m are
79
-------
"*.*,>. *
Figure 1. Diffusion culture apparatus: view into the 180 ml growth chamber.
Filtered air is pumped into the culture for aeration and mixing via
the aerator tube (at) and to provide turbulence at the membrane sur-
face to promote diffusion. A Nuclepore membrane (np) of 0.1 ym
porosity allows diffusion of seawater solutes into the culture
while keeping the bacterioplankton population captive. Hidden by
the membrane in the photograph is the 100 ml medium chamber, through
which untreated seawater is continuously pumped by means of the two
ports seen on the left side. The medium chamber is agitated by a
teflon-coated magnetic bar to provide turbulence at the membrane
surface. Approximate dimensions of the assembled chamber are 10 cm
on each axis.
80
-------
o
Or
O
O-
to
o
o •
o
o
o
o
f
o
X
o
§
10
o
8
Ul
o
•U O
C -H
0) M
'O Q)
•H 4J
M O
Es id
O £
+J ro
g -P
fd rd
ft
Q) rH C
si o
4-> 4-> -H
O W 4->
0 U
rf3 rj^ rd
C ^ S~i
(d < 4-1
p_l ^J p^l
El rH E-I
a) a) g
rd -H
rH -P n
O rH C
•H fd rd
4-> U ,C
S-l O 4-J
fd -H
ft M
U) Q)
H I ti I
O £ rH
rd
dog
O O U)
•rH O
4J rH 0)
A 4J 4->
•H rd
t-i CO
4-> O -H
W [-^
•H rH C
T3 O
0) 4J
Q)
D U
CN
0)
Cn
-H
rd
81
-------
statistically different from each other at the 1% confidence level. The sig-
nificance of this distribution of bacteria-sized particles becomes clearer
when the other parameters are examined. The larger particulate ATP fraction
shows a more extreme distribution as might be expected. The organisms consti-
tuting this fraction evidently were stratified principally at 30 m and 80 m.
The large peak at 30 m, with a minimum chlorophyll a concentration, implies a
preponderance of micro-zooplankton dominated by protozoa, while the maximum
for chlorophyll a at 80 m indicates the major zone of phytoplankton. At all
sampling depths the ATP-biomass of this fraction was greater than the less
than 3 jam fraction. It is perhaps more interesting to view this fact from the
opposite perspective, that is, that the less than 3 ym fraction at times makes
up more than 30% of the total biomass of microplankton less than 1000 ym in
,size. Figure 3 plots the mean growth obtained in diffusion cultures of the
smaller than 3 ym fraction exposed to water drawn from 50 m (oxygen maximum)
and 80 m (chlorophyll maximum). The diurnal growth pattern is consistent with
the thesis that these microorganisms utilize dissolved substrates that origi-
nate with the primary producers. These substrates may be released either di-
rectly, through exudation during photosynthesis, or indirectly, through graz-
ing by zooplankton, with attendant cell damage and zooplankton excretion. The
depth distribution of the bacterioplankton is given in Table 1 while the amount
of bacterioplankton produced in diffusion culture at the maxima at 50 and 80 m
is presented in Table 2. The apparent daily production for these microzones
of intense activity ranged from 21 to 67 mg C/m3 which on an annual basis would
yield 7.6 to 24.5 g C/m3 with a mean of 13 g C/m3. This compares with a range
of 1 to 6 g C/m3 estimated for the Pacific and the North Sea by Sorokin (1971)
and J. Meyer-Reil (personal communication), respectively.
TABLE 1. DEPTH DISTRIBUTION OF SMALLER THAN 3 ym PARTICULATE ATP AND COMPUTED
CELLULAR CARBON AT STATION 13, R/V TRIDENT CRUISE 170, 1000 HRS,
14 AUGUST 1975, 36°59'N, 21°22'W. MEAN VALUES ±95% CONFIDENCE
LIMITS.
_ .. rAmTM / 3 (cellular carbon)
Depth • [ATP] yg/nr 3
mg/m *
0
5
20
30
40
50
60
70
80
100
250
33.7
16.8
18.4
16.7
16.7
25.8
15.3
21.6
24.4
15.6
6.4
±0.49
± 0.09
± 0.23
± 0.25
±0.18
± 0.55
± 0.23
±0.10
±0.16
± 0.21
± 0.16
8.4 ±
4.2 ±
4.6 ±
4.2 ±
4.2 ±
6.5 ±
3.8 +
5.4 ±
6.1 ±
3.9 ±
1.6 ±
0.12
0.03
0.06
0.06
0.05
0.14
0.06
0.03
0.04
0.05
0.04
*Computed as follows: [ATP] yg/m3 x 0.25 = (cellular carbon) mg/m3
(from Hamilton and Holm-Harisen, 1967) .
82
-------
II 1)
T3
c
Itf
o
m
H
X
a
j
j
>-i
a:
ft
9
DH
O
hH
K
U §
Q £>
S EH
<: in-
D
2 0
W
O 2
>-i O
X H
o to
W Cn
ffi fc
EH H
Q
EH
tf^ ^
f~i^ f^1
CQ
0
2 Q
M W
rt 2
ry* [__(
u S
8w
EH
W
2 Q
EH CO
V itr1
f^M R^
§ ^
F-^ *•
t-1 0
ft r-
O H
H 1
CM «
W EH
EH
U -
< CO
« H
2 2
M O
M
W EH
W <
rtl EH
§ W
U EH
2 <:
H
. ESTIMATED
80 meters)
CM
w
j
PQ
fC
EH
Sg
in en
rH ^ CO
r-^
CN O 0
en CN in
rH rH
ro r^ \£>
CN CO CN
H O 0
Ti
Ti O
§0 -rH
P -H M
g rH CD
•H CD a
| *
6 5 ^
c s o
CD O CO rH CO W
Cn SH M Cn JH >H
£ ** ^^^
O -P CN C VD O
tO CN O O rH
^ rH 1 O||
g -rH CO CD CN &
tH rH CO O O
o
in
CN O
CO
CN
m
-p
o
O
-H
H
0)
a
in
rH
(0
-p
o
-p
o
•H
a
in cri
IT) IT)
cn
en oo
03 m
O O
g
-H
X
•a
o
U
o
•H
O CO W
^ M rH
Cn X! XI
•P 00 CN
CO rH CN
--nil
g -H ^ CO
LO O
VD CN
i-l CN
CN CN
<* (Ti
O O
Ti
O
•H
X!
-P
Se
T3
G VD O
O O rH
O I I
(U CN
r-
U)
cn
un
CNJ
^
Cn
3.
ft"
EH
sC
*
O
CO
co o o
83
-------
H
u.
z
o
3*
in
en
H
X
o
GO in
H H
10
o
CM
O
CM
CM
.CO
01
p -p
-P C
id *H
M O
ffl ft
id u
id
QJ W
P
-P •
-H g
U O
CO
T)
<1) *d
tr> C
id (d
U
O
in
C
•H
O
id
01 w
-H 0)
-P
Q) id
i-l O
TJ -H
U T)
CO C
•H
C
O
U
^
(U
4-)
10
id
(U
Ul
•d
(U
4->
id
a)
4J
O
4->
•d
0)
01
o
ft
X
Q)
Q)
•H
_c~|
^
0)
4-)
id
u
•H
ft
0)
^1
n
•H
C
Id
C
id
ki.
^^^
Cn
Cm
EH
<:
Q)
4->
id
rH
^
U
•H
4-1
TJ
a
U-t
o
c
id
g
0)
,c
4J
(/I
4->
C
0)
Ul
(1)
a
0)
!H
-H
1
O
H
||
^
4-1
,— |
*—^
|
CL,
*^3
D^
M-)
CsJ
D^
o
i-H
01
id
'd
c
•H
r-l
•§
O
0)
00
.
-------
DISCUSSION
The results for estimating biomass and productivity presented here are
preliminary in nature. Some of the assumptions and the procedures described
require further testing before the validity, limitations, and thus the useful-
ness of this methodology can be determined. The 3 ym particle size cutoff was
chosen arbitrarily but appears to be a fortunate choice (Hoppe, 1976).
Laboratory studies using the diffusion culture device (Lavoie, 1975)
show that when a fraction smaller than 3 ym is removed from the natural light
regime, its growth is closely coupled to the quality and quantity of nutrients
in the diffusion medium. By eliminating predation and other removal mecha-
nisms, any growth in the population is cumulative and is directly proportional
to the amount of substrate in the diffusion medium. The production observed
in the open ocean experiment implies that a substantial amount of the primary
production is converted to bacterial biomass.
The distribution and rates of growth of the bacterioplankton seem to
agree with the data on dissolved organic matter and the occurrence of protist
biomass. The 50 ATP peak occurs at the oxygen maximum, where presumably most
of the primary production is occurring (9% light level). Phaeopigments also
peak at this depth, indicating some phytoplankton cell decomposition, due
either to autolysis or to grazing by zooplankton (Lorenzen, 1967) . Either
process would provide soluble substrates for bacterial growth, as would exuda-
tion by intact cells (e.g., Thomas, 1971; Smith and Wiebe, 1976). The actual
presence of dissolved substrates is substantiated by the observed carbohydrate
concentration which has a maximum at this depth.
The 80 m peak in the smaller than 3 ym particulate ATP distribution is
also accompanied by a higher peak in the phaeopigments and another maxima for
dissolved carbohydrates. This depth is characterized by chlorophyll a maximum
(at the 1% light level) indicating the maximum for phytoplankton biomass.
Before annual productivity values per m2 can be estimated for specific
oceanic areas, a finer profiling must determine the rates of productivity out-
side the patches of plankton accumulation observed in this study where maximum
rates presumably occur.
REFERENCES
Allen, P. D., III. 1972. Development of the luminescence biometer for micro-
bial detection. Dev. Industr. Microbiol. 14:67-73.
Burney, C. M., and J. McN. Sieburth. 1976. Dissolved carbohydrates in sea-
water. II. A spectrophotometric procedure for total carbohydrate analy-
sis and polysaccharide estimation. Mar. Chem. 4 (in press).
Hamilton, R. D., and 0. Holm-Hansen. 1967. Adenosine triphosphate content of
marine bacteria. Limnol. Oceanogr. 12:319-324.
85
-------
Holm-Hansen, 0., C. J. Lorenzen, R. W. Holmes, and J. D. H. Strickland. 1965.
Fluorometric determination of chlorophyll. J. Cons. Perm. Int. Explor.
Her 30:3-15.
Holm-Hansen, O., and C. R. Booth. 1966. The measurement of adenosine tri-
phosphate in the ocean and its ecological significance. Limnol. Oceanogr.
11:510-519.
Hoppe, H. G. 1976. Determination and properties of actively metabolizing
heterotrophic bacteria in the sea, investigated by means of microauto-
radiography. Mar. Biol. 36:291-302.
Johnson, K. M., and J. McN. Sieburth. 1976. Dissolved carbohydrates in sea-
water. I. A precise spectrophotometric analysis for monosaccharides.
Mar. Chem. 4 (in press).
Lavoie, D. M. 1975. Application of diffusion culture to ecological observa-
tions on marine microorganisms. M.S. Thesis, Univ. Rhode Island. 91 pp.
Lorenzen, C. J. 1967. Vertical distribution of chlorophyll and phaeo-
pigments: Baja California. Deep-Sea Res. 14:735-745.
Schultz, J. S., and P. Gerhardt. 1969. Dialysis culture of microorganisms:
design, theory, and results. Bacteriol. Rev. 33:1-47.
Sheldon, R. W. 1972. Size separation of marine seston by membrane and glass-
fiber filters. Limnol. Oceanogr. 17:494-498.
Sieburth, J. McN., P.-J. Willis, K. M. Johnson, C. M. Burney, D. M. Lavoie,
K. R. Hinga, D. A. Caron, F. W. French III, P. W. Johnson, and P. G. Davis,
1976. Dissolved organic matter and heterotrophic microneuston in the sur-
face microlayers of the North Atlantic. Science (in press).
Smith, D. F., and W. J. Wiebe. 1976. Constant release of photosynthate from
marine phytoplankton. Appl. Environ. Microbiol. 32:75-79.
Sorokin, Y. I. 1971. On the role of bacteria in the productivity of tropical
oceanic waters. Int. Rev. Ges. Hydrobiol. 56:1-48.
Thomas, Y. P. 1971. Release of dissolved organic matter from natural popula-
tions of marine phytoplankton. Mar. Biol. 11:311-323.
Yentsch, C. S., and D. W. Menzel. 1963. A method for the determination of
phytoplankton chlorophyll and phaeophytin by fluorescence. Deep-Sea Res.
10:221-231.
86
-------
THE DEVELOPMENT OF STANDARD METHODS
OF MEASURING MICROBIOLOGICAL PRODUCTION
by
M. V. Gusev
Dean, Department of Biology, Moscow State University, USSR
The task of this report is to present suggestions concerning a uni-
fied plan of research involving the determination of microbe production in
aquatic ecosystems.
It is unnecessary to repeat that the role of microorganisms as great
utilizers and converters in the life of any natural community is vast and ir-
replaceable, in connection with which it is urgently necessary, first, to have
a knowledge of, and secondly, to utilize the activity of microorganisms in
natural biotopes. How can one estimate the character and scale of activity of
microorganisms in the habitations that interest us? This is one of the basic
problems of ecological microbiology of the present day. The approach to its
solution was developed by Vinogradskiy, and is the basis of research in this
region today.
Vinogradskiy developed the idea that, although certainly in any situa-
tion it is necessary to study the behavior of individual types of microorga-
nisms, one cannot draw conclusions regarding their role in nature by basing
one's self solely on data obtained during work with pure cultures. The inves-
tigation of the actual processes realized by microorganisms in nature should
be included in the study of behavior of microbe societies in combination with
all other (physicochemical, biological) components of the ecosystems.
The basic principles of microbiological ecology were formulated by
Vinogradskiy:
1. One should determine the physical and chemical conditions in which the
studied phenomenon occurs, related to the activity of microbes and de-
termine which groups of organisms participate in it.
2. One should identify the representatives of groups of microorganisms that
are important for the given phenomenon, and study their physiology and
interrelationships with other organisms, i.e., determine their biologi-
cal niche.
3. One should attempt to explain the true significance of their activity in
nature (i.e., in a society with other species), approaching this from
the quantitative aspect.
87
-------
All of this signifies the need for combining laboratory investiga-
tions of pure and mixed cultures under conditions that simulate various natu-
ral situations with field research that includes chemical, physical and bio-
logical characterizations of the biotope, the characterization of the state of
microorganisms and their interaction with each other and with organisms of the
other levels of the ecosystem. Works of recent years have been developing in
this very direction and this approach is recognized as a wise one by modern
investigators.
The most important aspect that integrates the demands made upon re-
search in the realm of ecological microbiology is the determination of the so-
called microbe production.
THE CONCEPT OF "MICROBE PRODUCTION"
In a general sense, the concept of "microbe production" means the dy-
namics of microbial manifestations (the biomasses, population and functions)
in the ecosystem. Functioning of the ecosystem is the cycle of substance and
transport of energy whose significant aspect is a certain equilibrium between
the synthesis of organic material and its destruction. The synthesis of ma-
terial occurs at all levels of the ecosystem; if its dynamics can on the whole
be termed production of the ecosystem, then dynamics of synthesis of organic
material (as the result of which growth, reproduction and the liberation of
organic materials into the environment occur), each component part of the eco-
system can be called production of the given member or level of the ecosystem.
In each functioning community of organisms there is activity of microorganisms,
and consequently, microbial production, i.e. , the synthesis of material by
microorganisms expressed in their accumulation of biomass or maintenance of
biomass at a single level for a certain time under conditions in which attri-
tion and consumption occur.
The activity of microorganisms occurs at all levels of the ecosystem;
one should proceed from this when determining microbial production. If the
subject concerns aquatic ecosystems, i.e,, the so-called free-living micro-
flora: epiphyte microflora of the algae type, the microbe-decomposers, the
heterotrophic bacteria that feed on organic wastes of the algae and that serve
as food for the zooplankton, the chemosynthesizing and photosynthesizing bac-
teria, as well as the pathogenic and symbiotic microflora of all the higher
and lower aquatic organisms. The pathogenic microorganisms can obviously play
an important role in populations of fishes, etc. It is obvious that methods
of determining the free-living and pathogenic microflora should be different.
The destruction of dead organisms is an important element of action of the
microorganisms and should also be investigated for all levels and components
of the ecosystem.
It is initially important to bear in mind the following variations
that are possible during the investigation of microbial production:
1. The system is supplied with biogenic elements and has suitable
physico-chemical indicators for the development of microorganisms but does not
have in general an adequate number of cells of the corresponding microorganisms
that are capable of utilizing the given biogens. This is possible for ocean
88
-------
water as the result of dilution, stirring or the absence of stirring. In this
situation there is no production, which makes it necessary attentively to deal
with the problem of the size of the inoculate of microorganisms. It is solved
by determining the number of microorganisms of different species in a given
place of habitation in combination with laboratory investigations of the ini-
tial values of the inoculate required for the beginning of growth under simi-
lar conditions.
2. Another possible situation is when one cannot measure the actual
microbe production—microorganisms are present in the system, but there is an
insufficient concentration of biogenic elements or unsuitable physico-chemical
conditions for the beginning of their multiplication. In this case one can
determine the potential activity of the group of microbes present in the water
by adding different substrates to a removed sample and following their con-
sumption and the increment in biomass (population), or the functioning of the
microbes. In such systems, the microorganisms are in a state of the quiescent
forms or in a state of low activity and readiness for reproduction.
3. With the presence in the system of a sufficient amount of bio-
genic elements and substrates for growth, the microorganisms reproduce, i.e.,
one can determine actual production. Its scales depend on the concentration
of nutrients and consequently, low production can exist which is equal to the
rate of mortality and consumption, as the result of which the total number of
cells remains constant. This is the usual situation for unpolluted water.
Upon enrichment of the medium, a flare of development, the so-called "flower-
ing" of microorganisms is possible when a significant biomass of them accumu-
lates in the system.
Apparently, there is a direct relationship between microproduction
and the dynamics of functioning of the ecosystem on the whole. In a certain
sense, microbe production can serve as an indicator of functioning of the eco-
systems. Bearing all of the above in mind, one can state that during the de-
termination of microbe production, in order to judge correctly the microbio-
logical processes that are occurring in the aqueous ecosystem, one must know
the following:
1. The chemical composition of the water (the content of P, N, and organic
substances), the physico-chemical conditions, and what primary produc-
tion was (in order to know the level of productivity of the water).
2. The composition and population of the living population of the given
biotope. It is necessary to include all these in investigating the ac-
tivity of microorganisms.
3. The general population and the population of specific groups of micro-
organisms .
4. Actual and potential microbe production in accordance with what sources
of material and energy exist in the medium.
By knowing all of these data, one can on the one hand determine the
scale of microbe production in the ecosystem, and, on the other hand, its ac-
tual value for the functioning of the ecosystem.
89
-------
METHODS OF DETERMINING MICROBE PRODUCTION
Methods of determining microbe production, particularly in fresh
water, are being intensively developed in the USSR. We shall list the most
suitable methods of determining microbe production with brief commentaries.
1. Determining the increase in the number of bacteria in isolated water
samples (Gak, 1975) .
Samples of water taken with a sterile bathometer are poured into
sterile glass dark flasks with a volume of 250 ml, with ground-glass stoppers.
Half of the samples are filtered through a preliminary filter to separate the
zoo- and phytoplankton. The total number of bacteria and the number of bac-
teria of the groups of interest are determined in the flasks containing fil-
tered and unfiltered water. The flasks are lowered into a reservoir to the
level from which the sample was taken or are incubated in aquaria in which
conditions that approximate the natural ones have been created.
Exposure time is selected experimentally for reservoirs of different
latitudes and different trophic types. It should be equal to the average time
of bacterial generation. In the mesotrophic and eutrophic reservoirs of the
middle latitutudes, G. W. Gak recommends daily exposure of the samples. Upon
terminating exposure, the number of bacteria is once again determined. The
constant of the rate of production in the filtered sample of water is deter-
mined according to the following formula (Gak, 1975) .
In b - In b 2. 303 (log b - log b )
t o t o
Kl
log ^ - Ic
o
Bt
Dg B~
o
where bo and b-t are the initial and terminal concentration of bacteria in the
filtered water.
The change in the biomass of bacteria is described by the equation
H = KB - K.B (2)
dt 1
where K^ is the constant of the rate of consumption of bacteria and B is the
concentration of bacteria in the unfiltered water.
(3)
Yt = KB (4)
where B is the average biomass of bacteria over a time t, Yt is the consump-
tion of bacteria. Then, from (3) and (4) , the production of bacteria Pfc is
determined as
bt Bt
log — - log —
P^ = B^ - B + B~ - - - - (5)
t t ° 0.4343 t
90
-------
or a simplified method of calculation
P = 1" Kt [the same as (5) ]
2. The determination of microbe production by the radiocarbon method according
to the rate of assimilation of carbon dioxide (Romanenko and Kuznetsov,
1974).
The rate of absorption of carbon dioxide by microorganisms in dark-
ness in a reservoir is a cumulative value that forms as the result of chemo-
synthesis and heterotrophic assimilation of carbon dioxide. According to the
calculations, about 6% of the biomass of heterotrophic bacteria forms at the
expense of CC>2. Perceptible production of organic material as the result of
chemosynthesis occurs only in limited cases, when one observes copious libera-
tion of hydrogen sulfide as the result of anaerobic lysis in the eutrophic or
meromictic reservoirs. In reservoirs with a normal oxygen regime in the layer
of water, bacterial chemosythesis reaches significant values (0.05-0.2 g/m3 of
biomass per day), only in the bottommost layer of water and in the upper layer
of the bottom sediments. In the meromictic reservoirs, where an anaerobic
zone is constantly forming, the role of production of chemosynthesis grows.
If the anaerobic zone is within limits of the illuminated zone, the intensive
development of the photosynthesizing bacteria is observed in it.
Methods of separate determination of heterotrophic and chemosynthetic
fixation of carbon dioxide have been described by Sorokin (1964, 1970). Only
heterotrophic fixation occurs in the surface layer, and upon determining the
vertical distribution of the heterotrophic bacteria it is considered that
heterotrophic assimilation of carbon dioxide is proportional to this distribu-
tion. The value of assimilation of carbon dioxide in darkness in the layers
adjacent to the anaerobic zone is determined. It usually exceeds assimilation
in the surface layers by a value that corresponds to the chemosynthetic fixa-
tion of CO2- One can define the values of chemosynthetic and heterotrophic
fixation of CO2 according to the increment in the biomass of bacteria by de-
termining their population or production, taking into account that the biomass
of the chemosynthesizing bacteria is wholly formed from C02, while only 6% of
the biomass forms from CO2 in the heterotrophic bacteria. Another method of
measurement is made according to oxygen consumption, since the 02/C02 ratio
differs 10-20 times for these two processes.
3. In order to identify localization of active microbial populations and the
effectiveness of bacterial biosynthesis in reservoirs, a method of deter-
mining so-called potential microbe production is employed.
In this case, a trace amount (1-100 ug/1) C-labeled dissolved organic
substance is added to the samples of water from the reservoir and the relative
activity of the bacteria for samples from different biotopes is judged accord-
ing to inclusion of the label in the biomass, that is, j-ht intens: '-j of bac-
terial biosynthesis is indirectly characterized. The method has made it pos-
sible to identify the principles of localization of active microflora in a
layer of water and in bottom sediments (Sorokin, 1970). The method was sug-
gested by Parsons and Strickland (1962). One can also name certain other
methods of determining microproduction, for example, the method of micro-
colonies, a method using overgrown glass, a method of determining production
91
-------
according to the rate of (>> consumption in the reservoir or in an isolated
sample of water (only suitable for waters rich in microflora).
A variety of methods using labeled substrates makes it possible to
judge on the microbiological destruction of organic material, the activity of
microflora that utilize methane, hydrogen, the sulfides, ammonia, etc.; the
method of radioautography is used to count colonies of specific bacteria and
to calculate the activity of individual cells. Attempts to use the principles
of continuous cultivation of microbes for determining microbial production are
interesting.
Brok and Brok (1968) studied the rate of growth of bottom algae, using
a small stream as the container of a continuous culture. They covered a stretch
bf the river with dark foil, thereby stopping the access of energy. By measur-
ing the rate of attrition of algae cells from the shaded stretch, they calcu-
lated the initial rate of growth. Similar investigations were also conducted
under laboratory conditions. A chemostat was filled with sterilized or un-
sterilized water in which the inoculates of bacteria subject to testing were
placed. Temperature and other factors were maintained at the level character-
istic for the natural environment. By knowing the difference between the rate
of dilution of the system and the rate of attrition of the organisms, one
could calculate their rate of growth with high accuracy in the absence or
presence of competing microflora in the unenriched natural water.
The most widespread and used methods of determining microbe produc-
tion developed by Soviet scientists involve determining the number or activity
of microorganisms in isolated water samples. Work with isolated samples of
water, whose results are then extrapolated to the reservoir, forces the inves-
tigator to accept a number of assumptions. Taking the samples means moving
from an open system, which the natural habitation is, to a closed system of
the isolated sample, which is equal to sharply changing the environment. An-
other possible influence of the isolated sample on the rate of reproduction of
bacteria is the effect of the walls of the vessel, since the solid surface is
a point of concentration of organic material and of fixation and enhanced re-
production of the bacteria.
However, the investigations of Gak (1975) showed that the number of
bacteria in unfiltered samples of water remains equal to their initial popula-
tion in the reservoir for one day. This indicates that normal functioning is
preserved in flasks containing water, over the course of several days at low
temperature and a slow rate of processes and over the course of one day at
high temperatures and a high rate of the process in the plankton society as
are the natural interrelationships of the organisms, i.e., one observes nei-
ther the effect of the vessel walls nor the effect of the closed sample on the
population of bacteria.
Determination of the production of bacteria in the water samples from
the reservoir predicates purifying the sample of zooplankton so that their de-
vouring the microorganisms does not blur the picture of the absolute increment
of their biomass. For this purpose, filtration through gas or a filter paper
is employed. Filtration disrupts the natural food relationships in the water.
It has been shown that an increase in the number of bacteria is observed in
92
-------
the filtered samples, and in the non-filtered samples the number of bacteria
remains constant for a day. However, this change in the population of bacte-
ria in the sample as the result of filtration is a slight one and cannot in-
fluence microbial production. Obviously, however, filtration removes the epi-
phyte microflora attached to the algae from the sample as well as the micro-
organisms attached to the detritus.
Phytoplankton influences the microflora in two ways: it liberates
(in life or posthumously) the organic substances dissolved in the water. Fur-
thermore, the intensive development of microorganisms occurs on the surfaces
of the old and dying algae cells; instances of a toxic or antibiotic effect of
plankton algae are known.
All of the flask methods of determining microbe production include
the use of dark flasks. For some reason, it is felt that this, i.e., the ex-
clusion of photosynthesis, holds differences between the filtered and unfil-
tered water to a minimum. Excluding photosynthesis is vital for the method
of determining production according to C02 fixation, so as to separate the
heterotrophic and chemosynthesizing fixation from the photosynthetic variety.
It seems to us that removal of the phytoplankton during fixation has
a great drawback for the goals of determining microproduction since it dis-
rupts the process of algae liberation of organic material, and, perhaps, of
toxins into the medium, the consumption by algae of the biogenic elements, the
cells of algae themselves are removed, some of which serve as a direct sub-
strate for the bacteria. All of this certainly strongly disrupts the natural
process of vital activity and reproduction of the bacteria for the reservoir.
Probably, this explains the unequal character of growth curves of bacteria in
the isolated samples.
Hence, we conclude the limited value of any of the methods of deter-
mining microbe production. Since this determination cannot be an end in it-
self in the ecological research, but is only a method of identifying function-
ing (or potential possibilities) of the ecosystems, then certain of the listed
methods can be recommended as standard ones, but under the condition of ob-
serving the combined approach based on the principles of Vinogradskiy.
In this regard we submit that it is necessary simultaneously to make
the following analyses in order to judge production of microorganisms in eco-
systems (with the goal of using these judgments in any ecological forecasts):
1. For representing the situation in the ecosystem:
a. Chemical analysis for the composition of the water (the content of in-
organic ions and organic material), consideration of the physico-
chemical conditions (temperature, illumination, pH, C>2 concentration,
etc.), the determination of the makeup and number of microorganisms.
b. Counting the total number of microorganisms by the aid of filter stain-
ing; counting the number of living cells by the aid of luminescent
microscopy.
c. Counting the groups of microorganisms that one can expect, based on the
composition of the water and other conditions, quantitatively, using
the method of sowing on dense elective media.
93
-------
2. For determining true microbe production:
d. Determine the total production in isolated water samples according to
the method of Romanenko-Kuznetsova-Sorokin using C according to the
rate of heterotrophic, chemosynthetic and photosynthetic fixation of
C02.
e. Determine the potential capacity of the water to produce microorganisms
by the aid of sowing various cultures in a chemostat (necessary for in-
vestigating situations with slight natural inoculation).
f. Determine the potential activity of microorganisms in ecosystems with a
natural limitation for one factor or another by means of adding and
varying conditions.
94
-------
USE OF BIOLOGICAL INDICATORS FOR MONITORING EFFECT
OF POLLUTANTS ON THE MARINE ENVIRONMENT
by
Richard L. Iverson
Department of Oceanography
Florida State University
Tallahassee, Florida 32306
ABSTRACT
Marine organisms which exhibit sensitive physiological responses to
pollutants have been used in laboratory biological assays to predict the ef-
fects of pollutants on the marine environment. The methodology for this ap-
proach is developed for single species bioassays and is under development for
laboratory experimental ecosystem bioassays. The bioassay approach will serve
as the primary means for predicting effects of pollutants before the pollu-
tants are introduced to the marine environment and will be important in moni-
toring sublethal effects of pollutants on marine organisms.
Indicator species, opportunistic species which colonize disturbed envi-
ronments, have been used for many years as indicators of pollution effects in
aquatic environments. While the classical saprobic indicator system is no
longer used in the United States, some groups of organisms such as Capitellid
worms remain useful indicators of chronic pollution. Quantitative ecological
techniques which have been developed for use in characterizing assemblages of
organisms show promise for monitoring effects of pollutants. While no single
measure appears to be a consistent indicator of disturbance of natural assem-
blages by pollutants, a combination of changes in dominance, species diversity,
and evenness can be used together with ordination methods to quantitatively
assess changes in assemblages of marine organisms caused by pollutant stress
or by reduction of pollutant stress.
INTRODUCTION
Marine organisms which exhibit sensitive physiological responses to
particular pollutants or classes of pollutants have been used in biological
assays to establish permissible environmental concentrations of pollutants.
This approach is the standard means by which potentially singificant pollu-
tants are evaluated to predict their effects on marine organisms (Goldberg,
1975). Mortality of some part of the test population is commonly used as the
primary means for defining pollutant effect. Additional criteria such as
95
-------
changes in respiration rate, inhibition of enzyme systems, and modification of
blood component characteristics have been suggested for use in defining pollu-
tant effects (Duke, 1974). Biological assay methods using single species of
organisms will continue to be the primary means for predicting pollutant ef-
fects on the marine environment prior to introduction of a pollutant. Labora-
tory experimental ecosystem assay methods for predicting pollutant effects on
marine food webs are under development and show promise. The success of both
the organism bioassay methods and the laboratory experimental ecosystem assay
methods for predicting potential effects of pollutants and for monitoring
chronic effects of pollutants in the marine environment depends on the suc-
cessful culturing of marine organisms. Many neritic phytoplankton species and
some neritic invertebrate and vertebrate species presently can be maintained
in the laboratory. Some neritic species can be raised through the various
stages of their life cycles in the laboratory; however, it is very difficult
to maintain populations of oceanic organisms in culture. Progress is being
made in culturing oceanic phytoplankton (Booth, 1975). Knowledge of pollutant
effects on marine organisms or marine communities is best developed for neri-
tic environments. The investigation of pollutant effects on oceanic species
is very important since oceanic species may be more susceptible to pollutants
than coastal species. Diatoms isolated from the Sargasso Sea were more sensi-
tive to polychlorinated biphenyl (PCS) compounds than were clones obtained
from estuaries and from continental shelf waters (Fisher et al., 1973) .
The classical saprobic system for diagnosing water pollution was de-
vised by Kolkwitz and Marsson (1908). The system is based on zonal distribu-
tion of indicator organisms which respond to pollutants emanating from a point
source. Zones of organic pollution effects and zones of recovery from pollu-
tion effects were identified based on classes or organisms present within the
different zones. The saprobic system has been attacked on several bases
(Sladecek, 1965) and is no longer used in the United States (Bartsch and In-
gram, 1966). Marine biogeographical investigations attempt to elucidate the
geographical distribution of organisms on the basis of their physiological re-
sponses to varying environmental conditions (Ekman, 1953; Hedgpeth, 1957). A
central feature of marine biogeographical investigations is the identification
of characteristic fauna which occupy particular water masses (McGowan, 1972)
or which inhabit particular combinations of depth ranges and sediment types
(Menzies et al., 1973). The combination of the biogeographical approach, with
emphasis on identification of taxa inhabiting particular regions of the marine
environment and with the tools of modern quantitative ecology, shows promise
for assessing the effect of pollutant stresses on the biota of the marine en-
vironment. This approach may be used for evaluating the quality of predic-
tions based on biological assay methods and for quantitatively monitoring
changes in marine food web structure caused by impact of pollutants or by the
reduction of pollutant impact.
BIOLOGICAL ASSAY EXPERIMENTS FOR ASSESSING
POLLUTANT EFFECTS
Many potential marine pollution problems can be assessed by a detailed
consideration of rates of release into the environment, residence time in the
environment, concentration by marine organisms, and levels of toxicity of the
96
-------
compounds in question (Goldberg, 1975). The determination of toxicity levels
at present is achieved primarily through biological assay experiments utiliz-
ing one or a few species of organisms maintained in the laboratory. The
choice of organisms for use in bioassay experiments is constrained by the
limited number of species that can be maintained in the laboratory. Experi-
mental organisms are chosen because they are easy to handle in experimental
designs, because they give particularly sensitive and reasonably reproducible
responses to particular pollutants, or because they are of direct economic
value. An additional criterion for choice of experimental organisms is their
significance in the food webs of which they are a part. Keystone species play
significant roles in controlling food web structure. A starfish, Pisaster
ochraceus, controlled food web structure of a Washington rocky intertidal bot-
tom through predation (Paine, 1966) . Introduction of a carnivorous fish,
Cichla ocellaris, to Lake Gatun, Panama, markedly affected the pelagic food
web of the lake. Carnivorous bird numbers decreased near parts of the lake
dominated by Cichla (Zaret and Paine, 1973). Sea otter, Enhydra lutis, ap-
peared to be a keystone species in Aleutian Island and California nearshore
food webs (Estes and Palmisano, 1974). Squid are hypothesized to be a pelagic
oceanic keystone species which defines pelagic food web structure through pre-
dation (Costlow, 1975). To understand and to monitor effects of pollutants on
oceanic food webs, research efforts must be directed toward identification of
keystone species and toward elucidation of pollutant effects on these species.
Bioassay methodology has been reviewed by Sprague (1969) who recom-
mended that the lethal concentration for 50 percent of individuals on exposure
to a compound (incipient, LC50) is the single most useful criterion of toxicity.
The 4-day LC50 is another common measure of acute toxicity. Bioassays of syn-
thetic pesticides indicate that toxicities of compounds cannot be predicted
and that each compound must be evaluated to assess its toxicity. The effects
of a pesticide on estuarine organisms cannot be predicted on the basis of its
observed affinity to other known pesticides or its known action on other ani-
mal species (Butler, 1971). In an attempt to account for synergistic or an-
tagonistic interaction, the toxicities of mixtures of two or more pollutants
have been represented by a single number composed of the sum of toxicities of
individual pollutants represented as fractions of the incipient LC50 (Sprague,
1970). This approach is questionable since joint effects of pollutants are
not always linear (Livingston et al., 1974). Temperature, pH, bicarbonate
alkalinity, salt content, totally dissolved solid content, and dissolved oxy-
gen concentration are capable of modifying toxicities of pollutants (Sprague,
1971; Livingston et al., 1974).
It is very important that bioassay results are correctly interpreted.
Most phytoplankton bioassay experiments which reported effects of pollutants
on photosynthetic carbon-14 fixation appear to have estimated indirectly the
effects of pollutants on phytoplankton cell division rates rather than their
effects on photosynthetic rates. Photosynthetic carbon-14 uptake rates for
phytoplankton exposed to PCB compounds and to DDT were not different from con-
trol values when normalized to a per cell basis (Fisher, 1975).
Pollutant effects can be more pronounced at particular stages in the
life cycle of organisms. A concentration of 1.0 yg/1 of the PCB Aroclor® 1254
killed juvenile pink shrimp (Penaeus duorarum) within 15 days in laboratory
97
-------
bioassay. Adult shrimp were not as susceptible to Aroclor® 1254 at low con-
centrations, but were killed within 17 to 53 days at 2.4 to 4.3 ppb of the
Aroclor® (Nimmo et al., 1971) . Crassostrea gigas and C. angulata spermatozoa
survival was more sensitive to various weights of oil suspended in water than
were eggs, embryos, or larvae, therefore the most pronounced effects of oil on
oyster populations may occur during fertilization (Renzoni, 1973). Sheepshead
minnow fry (Cyprinodon variegatus) were more susceptible to Aroclor® 1254 than
were embryos, juveniles, and adults (Schimmel et al., 1974). Pesticide resi-
dues in water samples obtained from selected United States estuaries were low
in fall and winter and rose in late spring to a peak in midsummer. This sea-
sonal pattern probably reflected periods of pesticide application for agricul-
tural purposes and time of maximum runoff from watersheds. Unfortunately, the
seasonal maximum of pesticide residues coincides with the time when maximum
numbers of larval fish and shellfish are present in estuaries (Butler, 1971).
There is growing recognition of the significance of chronic low-level
pollution which does not result in spectacular point-source kills but which
can modify the structure of marine food webs on a long-term basis. Sublethal
effects of pesticides may be particularly important in the marine environment
(Butler, 1971; Duke and Dumas, 1974). All organisms sampled from pelagic Sar-
gassum weed communities in the Atlantic Ocean were contaminated with petroleum
hydrocarbons (Burns and Teal, 1973). There was no relation between the hydro-
carbon content and the animals' supposed positions in the food chain. There
were extensive mortalities of the marine macrophytes Fucus spiralis, Mya are-
naria, and Spartina alterniflora as a consequence of a large spill of Bunker C
oil in Chedabucto Bay, Nova Scotia. Zooplankton and benthic animals appar-
ently were not acutely affected by the hydrocarbons (Conover, 1971; Scarratt
and Zitko, 1972; Thomas, 1973). Chronic effects of hydrocarbon pollution on
marine communities remain unknown.
Pollutants can modify assemblages of marine organisms through nontoxic
effects. DDT affected salinity selection by mosquito fish (Hansen, 1972).
The grass shrimp (Palaemonetes pugio) avoided water containing the herbicide
2,4-D but did not avoid water containing several insecticides in laboratory ex-
periments (Hansen et al., 1973). Fishes avoided Kraft-mill effluents in labo-
ratory experiments (Lewis, 1974). Grass shrimp, pinfish (Lagodon rhomboides)
and mosquito fish (Gambusia affinis) avoided at least one concentration of Aro-
clor® 1254 in water, but pink shrimp (Penaeus duorarum) and sheepshead minnows
did not avoid water containing any of the experimental concentrations (Hansen
et al., 1974).
There are some new approaches to pollutant assessment which utilize
cellular or subcellular components of organisms in biological assays. Tissue
culture methods can be used to monitor environmental levels of mercuric chlo-
ride which inhibited multiplication of L-cells at concentrations of 10 yg mer-
curic chloride per liter or less (Li and Troxlen, 1972). Enzyme assays may
provide a sensitive and reproducible means for assessing effects of pollutants
on marine organisms. Acetylcholinesterase activity was inhibited by organo-
phosphate and carbamate pesticides when the fish Leiostomus xanthus was ex-
posed to the compounds (Coppage, 1972). Standardized assays of enzymes of the
tricarboxylic acid cycle or of the electron transport system may provide a re-
producible means for estimating effects of pollutants on marine organisms
(John Calder, personal communication). The use of standardized enzyme assays
98
-------
could minimize data interpretation problems which arise in pollutant bioassays
when different populations of the same species of experimental organisms are
grown under different environmental conditions. Identification of changes in
tissue of marine organisms subjected to pollution stress shows promise for de-
termining whether sublethal effects of pollutants have occurred in the marine
environment. Yevich and Berry (1969) identified tumors in ovarian tissue from
clams, Mercenaria mercenaria, obtained from Narragansett Bay, Rhode Island.
Effects of various pollutants on tissue of invertebrates has been discussed by
Sparks (1972).
POLLUTION INDICATOR SPECIES
The central concept which underlies the use of indicator organisms for
monitoring effects of pollutants on the marine environment is that pollutant
stress results in a change in species composition and in numbers of organisms.
The change frequently appears to be in the direction of a simplification in
community structure with equilibrium species replaced by opportunistic species
(Woodwell, 1970). Since the assemblage of organisms present in a polluted
area reflects the effects of pollutants over some length of time, sessile or-
ganisms or organisms "with low mobility are best choices for indicator orga-
nisms. Worms, mollusks, arthropods, foraminifera, epiphytic algae and at-
tached microalgae have been used as pollution indicators in marine environ-
ments (Wass, 1967).
Worms of the genus Capitella have been used to indicate the presence of
pollutants in marine and estuarine water. Capitellid worms are found in areas
polluted by sewage outfalls (Filice, 1959; Halcrow et al., 1973) and in areas
polluted by industrial effluents (Wade et al., 1972; Pearson, 1972; Rosenberg,
1972). A single genotype of Capitella capita was selected on a short-term
basis following an oil spill in a Massachusetts bay (Grassle and Grassle,
1974). In some cases, Capitellid worms have been observed to colonize areas
which were physically disturbed but which were not polluted (Eagle and Rees,
1973). Apparently the cosmopolitan species C. capita, one of the most widely
used indicator species, is actually a composite of six species which differ
very little morphologically but which differ widely in life histories and in
reproductive modes (Grassle and Grassle, 1976).
Highly polluted areas near a sewage outfall in Biscayne Bay, Florida,
were characterized by benthic algae, Gracilaria blodgettii and Agardhiella
tenera, plus a tubiculous polychaete, Diopatra cuprea. Less polluted areas
were characterized by seagrasses, Haladule wrightii and Halophila baillonis,
together with an ophiuran, Amphioplus abditus (McNulty, 1961). Littoral algal
communities of Yugoslavian coastal waters were sensitive to organic pollution.
Absence of a dominant alga, Cystosiera barbata, and presence of Ulva lactuca
and Codium tomentosum indicated pollution effects (Golubic, 1973).
99
-------
USE OF BIO-STATISTICAL MEASURES AS INDICATORS
OF POLLUTANT EFFECTS
The strategy of ecological succession appears to be toward homeostasis
with the physical environment in the sense of achieving maximum possible pro-
tection from environmental perturbations (Odum, 1969). The accumulation of
pollutants in the biosphere has led to changes, which operate on successional
time scales, in the structure and function of natural ecosystems. Patterns in
the changes seem to be broadly similar for different pollutants in different
ecosystems. Under pollutant stress, ecosystem structure shifts from a complex
arrangement of specialized species toward species which are generalists. Spe-
cies diversity, stability of nutrient cycles, and stability of population num-
bers appear to be reduced (Woodwell, 1970). The identification and characteri-
zation of changes that occur under pollutant stress require detailed sampling
and identification of organisms in areas under pollutant stress. The large
quantity of data requires condensation into biologically meaningful statistics
that can be used to interpret changes in ecosystem structure due to introduc-
tion or removal of pollutant stresses. It has been apparent for some time
that some waste products are subjecting marine environments, particularly
nearshore ecosystems, to an increasing degree of pollutant stress. A report
on "Waste Management Concepts for the Coastal Zone" (NAS-NES Committee on
Oceanography, 1970) outlined some particularly important problem areas and re-
search requirements regarding assessment of the biological status of coastal
marine environments. The report emphasized the need for evaluation of various
statistical indices, together with evaluation of sampling variability as it
affects the indices, to determine their usefulness for assessing pollutant ef-
fects on the, marine environment. The recent literature contains several exam-
ples of the use of statistical indices to evaluate pollutant effects on marine
biota.
Bechtel and Copeland (1970) found lowest information theory diversity
indices in the most highly polluted areas of Galveston Bay, Texas. Fish popu-
lations could be divided into separate communities, each structured as a re-
sponse to environmental and pollution stress. In those areas receiving the
greatest stress, the bay anchovy, Anchoa mitchilli, was the dominant species;
tne most highly stressed areas also supported the fewest number of large indi-
viduals. A four-year investigation of fish populations of the middle Patuxent
Estuary in Maryland revealed strong seasonal cycles in number of species, num-
ber of individuals, and in several diversity indices (McErlean et al., 1973).
Trend analysis revealed downward trends in the number of species, in species
richness, in the information theory species diversity index, and in evenness.
The changes were explained by loss of species and by dominance shifts, both
of which could result in simpler community structure. No simple causal rela-
tionship could be established for the changes hypothesized to arise from gen-
eral environmental degradation. Effects of Kraft pulp-mill effluents on the
fishers of a shallow bay system along the north Florida coast were investi-
gated for two years (Livingston, 1975). Estuarine and marsh fish assemblages
in areas of acute pollutant impact contained reduced numbers of individuals
and species. A survey of a broad offshore area showed reductions in number of
individuals and number of species collected on a monthly basis. The cumula-
tive number of species collected annually was the same for polluted and unpol-
luted areas, however. Opportunistic species, relatively rare in unpolluted
100
-------
areas, appeared to be recruited to areas of pollutant impact. The polluted
areas showed decreased dominance and qualitative species differences compared
to control areas. Species richness and species diversity (H1 and the Shannon-
Weaver Index, H) were lower at highly stressed stations but were similar to
control values at stations where pollutant impact resulted in moderate reduc-
tions in the number of individuals and in the number of species. Species di-
versity by itself did not appear to be an indicator of pollutant effects for
the estuarine and marsh fish assemblages. Equitability indices were the same
for polluted and unpolluted areas of the shallow bay system. A general pat-
tern of reduced numbers of individuals and species, decreased dominance, and
variable species diversity and species richness were observed at most heavily
impacted stations.
Macrobenthic species assemblages at stations in the Elizabeth River,
Virginia, differed in terms of species content, dominant species, and species
diversity when compared to stations with similar bottom types in the Hampton
Roads, Virginia, area (Boesch, 1973). Species diversity, estimated with the
information theory index, H1, was reduced in the Elizabeth River macrobenthic
assemblage. The changes in the biological indices appeared to be related to
pollutant stress in the form of primary-treated domestic sewage entering the
Elizabeth River. Biota around a low-volume domestic sewage outfall near San
Clemente Island, California, were less diverse (in 5 diversity indices) than
were controls. Several marine macrophytes were replaced by a low turf of
blue-green algae, which exhibited higher net productivity, smaller growth
forms, simpler and shorter life histories and which were components of earlier
successional stages in the littoral zone (Littler, 1975).
Zooplankton were collected from Timbalier Bay, Louisiana, and from con-
tiguous coastal waters to assess effects of oil extraction on the zooplankton
community. No statistically significant differences in mean number of single
species populations, in species diversity, or in biomass were observed for
samples collected near oil platforms compared to control areas. Long-term
effects of oil drilling and production on zooplankton communities appeared
negligible (Marum, 1974). An investigation was conducted in the New York
Bight to evaluate small-scale variations of single species populations and
coastal zooplankton communities as they related to the disposal of acid
waters. The spatial distribution of the majority of the species was highly
aggregated, but no trends were observed to suggest that the acid wastes were
an important factor in controlling the distributions (Wiebe et al., 1973). No
significant trends were evident in species diversity estimated with the infor-
mation theory H1. Differences of less than a factor of 5 to 10 in the abun-
dance of a population between stations in coastal waters, based on single ob-
servations, could have been caused by sampling error. The effects of Aroclor®
1254 on survival of marine organisms was evaluated in a four-month laboratory
ecosystem investigation (Hansen, 1974). Amphipods were dominant in control
aquaria and in aquaria containing 0.1 ug liter" of the PCB. At a concentra-
tion of 10 yg liter" of the PCB, greater than 75 percent of the animals were
tunicates, indicating dominance changes as PCB concentration increased. Num-
bers of phyla, species, and individuals in the experimental aquaria were re-
duced compared to control aquaria, but there were no apparent effects on
abundance of annelids, brachipods, coelenterates, echinoderms or nemerteans.
The species diversity (H) was not affected by this Aroclor®.
101
-------
Apparently no single statistical index can completely characterize
changes in community structure that occur as a consequence-of pollutant stress
on the marine environment. A combination of dominance measures (McNaughton,
1968), species diversity indices (Pielou, 1966), and evenness measures (Pielou,
1966), together with changes in kinds of species, can provide a basis for
evaluation of effects of pollutants on marine environments. This approach is
proving highly successful for evaluating changes in marine communities after
cessation of pesticide and paper pulp-mill waste input to coastal waters (R.
Livingston, personal communication).
REFERENCES
Bartsch, A. F., and W. M. Ingram. 1966. Biological analysis of water pollu-
tion in North America. Verh. Internat. Verein. Limnol. 16:786-800.
Bechtel, T. J., and B. J. Copeland. 1970. Fish species diversity indices as
indicators of pollution in Galveston Bay, Texas. Contrib. Mar. Sci.
15:103-132.
Boesch, D. F. 1973. Classification and community structure of macrobenthos
in the Hampton Roads area, Virginia. Mar. Biol. 21:226-244.
Booth, B. C. 1975. Growth of oceanic phytoplankton in enrichment cultures.
Limnol. Oceanogr. 20:865-868.
Burns, K. A., and J. M. Teal. 1973. Hydrocarbons in the pelagic Sargassum
community. Deep Sea Res. 20:207-211.
Butler, P. A. 1971. Influence of pesticides on marine ecosystems. Proc.
Royal Soc. London, Ser. B, 177:321-329.
Conover, R. J. 1971. Some relations between zooplankton and Bunker C oil in
Chedabucto Bay, following the wreck of the tanker Arrow. J. Fish. Res.
Bd. Canada 28:1327-1330.
Coppage, D. L. 1972. Organophosphate pesticides: specific level of brain
AChE inhibition related to death in sheepshead minnows. Trans. Amer.
Fish. Soc. 101:534-536.
Costlow, J. 1975. Biological Oceanography: Some Critical Issues, Problems
and Recommendations. National Academy of Sciences, Washington, D.C.
28 pp.
Duke, T. W. 1974. Criteria for determining importance and effects of pesti-
cides on the marine environment; a brief overview. MTS Journal 8:21-22.
Duke, T. W., and D. P. Dumas. 1974. Implications of pesticide residue in the
coastal environment. Pages 137-164 in F. J. Vernberg and W. B. Vernberg,
eds., Pollution and Physiology of Marine Organisms. Academic Press, New
York. 492 pp.
102
-------
Eagle, R. A., and E. I. S. Rees. 1973. Indicator species—a case for caution.
Mar. Poll. Bull. 4:25.
Ekman, S. 1953. Zoogeography of the Sea. Sidgwick and Jackson, Ltd., Lon-
don. 417 pp.
Estes, J. A., and J. F. Palmisano. 1974. Sea otters: their role in struc-
turing nearshore communities. Science 185:1058-1060.
Filice, F. P. 1959. The effect of wastes on the distribution of bottom in-
vertebrates in the San Francisco Bay estuary. Wasmann J. Biol. 17:1-17.
Fisher, N. S. 1975. Chlorinated hydrocarbon pollutants and photosynthesis of
marine phytoplankton; a reassessment. Science 189:463-464.
Fisher, N. S., L. B. Grahm, E. J. Carpenter, and C. F. Wurster. 1973. Geo-
graphical differences in phytoplankton sensitivity to PCB's. Nature
241:548-549.
Goldberg, E. D. 1975. Assessing Potential Ocean Pollutants. National
Academy of Sciences, Washington, D.C. 438 pp.
Golubic, S. 1972. Effect of organic pollution on benthic communities. Mar.
Poll. Bull. 3:56-57.
Grassle, J. P., and J. F. Grassle. 1976. Sibling species in the marine pol-
lution indicator Capitella (polychaeta). Science 192:567-568.
Grassle, J. F., and J. P. Grassle. 1974. Opportunistic life histories and
genetic systems in marine benthic polychaetes. J. Mar. Sci. 32:253-284.
Halcrow, W., D. W. MacKay, and I. Thornton. 1973. The distribution of trace
metals and fauna in the Firth of Clyde in relation to the disposal of
sewage sludge. J. Mar. Biol. Ass. U.K. 53:721-739.
Hansen, D. J. 1972. DDT and malathion; effect on salinity selection by mos-
quito fish. Trans. Amer. Fish. Soc. 101:346-350.
Hansen, D. J. 1974. Aroclor® 1254: effect on composition of developing es-
tuarine animal communities in the laboratory. Contrib. Mar. Sci.
18:19-33.
Hansen, D. J., S. C. Schimmel, and E. Matthews. 1974. Avoidance of Aroclor®
1254 by shrimp and fishes. Bull. Environ. Contain. Toxicol. 12:253-256.
Hansen, D. J., S. C. Schimmel, and J. M. Keltner. 1973. Avoidance of pesti-
cides by the grass shrimp (Palaemonetes pugio) . Bull. Environ. Contain.
Toxicol. 9:129-133.
Hedgpeth, J. W. 1957. Classification of marine environments. In J. W.
Hedgpeth, ed., Treatise on Marine Ecology and Paleoecology 1:17-28.
Mem. 67, Geol. Soc. Amer.
103
-------
Kolkwitz, R. , and M. Marsson. 1908. Okologie der pflanzlichen Saprobien.
Ber. Dt. Botan. Ges. 26 A:505-519.
Lewis, F. G., III. 1974. Avoidance reaction of two species of marine fishes
to Kraft pulp mill effluent. M.S. Thesis, Florida State Univ., Talla-
hassee.
Li, M. F. , and G. S. Traxler. 1972. Tissue culture bioassay method for water
pollution with special reference to mercuric chloride. J. Fish. Res. Bd.
Canada 29:501-505.
Littler, M. M. 1975. Impact of sewage on the distribution, abundance, and
community structure of rocky intertidal macro-organisms. Mar. Biol.
30:277-291.
Livingston, R. J., C. C. Koenig, J. L. Lincer, C. D. McAuliffe, A. Michael,
R. J. Nadeau, R. E. Sparks, and B. E. Vaughn. 1974. Synergism and modi-
fying effects: interacting factors in bioassay and field experiments.
Pages 226-279 in Proceedings, Workshop on Marine Bioassays, Marine Techno-
logical Society, Washington, D.C.
Livingston, R. J. 1975. Impact of Kraft pulp-mill effluents on estuarine and
coastal fishes in Apalachee Bay, Florida, USA. Mar. Biol. 32:19-48.
M-arum, J. P. 1974. Spatial and temporal variations of zooplankton in rela-
tion to offshore oil drilling and estuarine marine faunal exchange.
Ph.D. Dissertation, Florida State Univ., Tallahassee.
McErlean, A. J., S. G. O'Connor, J. A. Mihursky, and C. I. Gibson. 1973.
Abundance, diversity and seasonal patterns of estuarine fish populations.
Estuar. Coast. Mar. Sci. 1:19-36.
McGowan, J. A. 1972. The nature of oceanic ecosystems. In C. B. Miller, ed.,
The Biology of the Oceanic Pacific. Oregon State Univ. Press, Corvallis.
157 pp.
McNaughton, S. J. 1968. Structure and function in California grasslands.
Ecology 49:962-972.
McNulty, J. K. 1961. Ecological effects of sewage pollution in Biscayne Bay,
Florida: sediments and the distribution of benthic and fouling macro-
organisms. Bull. Mar. Sci. 11:394-447.
Menzies, R. J., R. Y. George, and G. T. Rowe. 1973. Abyssal Environment and
Ecology of the World Oceans. John Wiley and Sons, New York. 488 pp.
NAS-NES Committee on Oceanography. 1970. Waste Management Concepts for the
Coastal Zone. Printing and Publishing Office, NAS-NES, Washington, D.C.
126 pp.
Nimmo, D. R., R. R. Blackman, A. J. Wilson Jr., and J. Forester. 1971.
Toxicity and distribution of Aroclor® 1254 in the pink shrimp Penaeus
duorarum. Mar. Biol. 11:191-197.
104
-------
Odum, E. P. 1969. The strategy of ecosystem development. Science
164:262-270.
Paine, R. T. 1966. Food web complexity and species diversity. Amer. Nat.
100:65-75.
Pearson, T. H. 1972. The effect of industrial effluent from pulp and paper
mills on the marine benthic environment. Proc. Royal Soc. London, Ser. B,
180:469-485.
Pielou, E. C. 1966. The measurement of diversity in different types of bio-
logical collections. J. Theoret. Biol. 13:131-144.
Renzoni, A. 1973. Influence of crude oil, derivatives and dispersants on
larvae. Mar. Poll. Bull. 4:9-13.
Rosenberg, R. 1972. Benthic faunal recovery in a Swedish fjord following the
closure of a sulphite pulp mill. Oikos 23:92-108.
Scarratt, D. J. , and V. Zitko. 1972. Bunker C oil in sediments and benthic
animals from shallow depths in Chedabucto Bay, Nova Scotia. J. Fish. Res.
Bd. Canada 29:1347-1350.
Schimmel, S. C., D. J. Hansen, and J. Forester. 1974. Effects of Aroclor®
1254 on laboratory-reared embryos and fry of sheepshead minnows (Cyprini-
don variegatus) . Trans. Amer. Fish. Soc. 103:582-586.
i
Sladecek, V. 1965. The future of the saprobity system. Hydrobiology
25:518-537.
Sparks, A. K. 1972. Invertebrate Pathology. Academic Press, New York.
387 pp.
Sprague, J. B. 1969. Measurement of pollutant toxicity to fish. I. Bioassay
methods for acute toxicity. Water Res. 3:793-821.
Sprague, J. B. 1970. Measurement of pollutant toxicity to fish. II. Utiliz-
ing and applying bioassay results. Water Res. 4:3-32.
Sprague, J. B. 1971. Measurement of pollutant toxicity to fish. III. Sub-
lethal effects and "safe" concentrations. Water Res. 5:245-266.
Thomas, M. L. H. 1973. Effects of Bunker C oil on intertidal and lagoonal
biota in Chedabucto Bay, Nova Scotia. J. Fish. Res. Bd. Canada 30:83-90.
Wade, B., L. Antonio, and R. Mahon. 1972. Increasing organic pollution in
Kingston Harbour, Jamaica. Mar.- Poll. Bull. 3:106-111.
Wass, M. L. 1967. Biological and physiological basis of indicator organisms
and communities. Pages 271-283 in T. A. Olson and F. J. Burgess, eds.,
Pollution and Marine Ecology. Interscience Publishers, New York.
105
-------
Wiebe, P. H., G. D. Grice, and E. Hoagland. 1973. Acid-iron waste as a fac-
tor affecting the distribution and abundance of zooplankton in the New
York Bight. II. Spatial variations in the field and implications for
monitoring studies. Estuar. Coast. Mar. Sci. 1:51-64.
Woodwell, G. M. 1970. Effects of pollution on the structure and physiology
of ecosystems. Science 168:429-433.
Yevich, P. P., and M. M. Berry. 1969. Ovarian tumors in the quahog Mercena-
ria mercenaria. J. Invert. Pathol. 14:266-267.
Zaret, T. M., and R. T. Paine. 1973. Species introduction in a tropical
lake. Science 182:449-455.
106
-------
MICROORGANISMS AS BIOLOGICAL INDICATORS OF COMMERCIAL-
DOMESTIC AND PETROLEUM POLLUTION
by
I. B. Tokin, 0. N. Trunova and T. I. Izgoreva
Institute of Marine Biology of the Academy of Sciences, USSR
Despite a large number of investigations on protecting the marine
aquatoria, the intensity of autopurification, as well as the degree of chemi-
cal and commercial-domestic pollution of the northern seas, have not been
practically studied.
The level of different pollutants in the water can be detected by both
sanitary-chemical and sanitary-bacteriological methods. According to the ob-
servations of A. G. Mironov (1961, 1967, 1970), the sanitary-bacteriological
indicators of sea water pollution are significantly more sensitive than the
sanitary-chemical ones. Thus, proportional to remoteness from the source of
pollution, when the oxidizability and BOD5 of sea water already do not record
pollution of the aquatorium, the latter is detected by microbiological methods
which most reliably show the range of spread of pollution from its source.
The goal of the investigation is to explain the degree of pollution of
two bays of the Barents Sea by commercial-domestic and petroleum pollutants
according to the titer of microbiological indicators—coliform bacteria, en-
terococci, the petroleum oxidizing microorganisms and bacteriophages.
In all, 80 samples of water of varying degrees of pollution were in-
vestigated for the first (most polluted) bay: 40 samples of dirty water (near
the coast) ; 25 samples of water of a moderate degree of pollution (the middle
of the bay); and 15 samples of arbitrarily pure water (at the exit from the
bay) .
Forty-five water samples were investigated for the second bay (less
polluted), 15 samples each of varying degrees of pollution. The investiga-
tions were carried'out with 3 repretitions.
Data on the microbiological indicators of pollution of the bays are
given in Tables 1 and 2.
It is apparent from Tables 1 and 2 that the dirtier the water, the
lower the titer of the sanitary-indicative microorganisms; in this instance,
a change in the titer of indicators of the commercial-domestic pollutants cor-
related with the titer of the petroleum-destroying microorganisms.
107
-------
TABLE 1. TITER OF INDICATOR MICROORGANISMS IN WATER SAMPLES OF THE FIRST BAY
Degree of water pollution
Type of microbe
Coliform bacteria
Enterococci
Petroleum oxidizing
microorganisms
Total microbe number in
thousands of microbial
bodies per ml
TABLE 2. TITER OF INDICATOR
Season
Summer
Winter
Summer
Winter
Summer
Winter
Summer
Winter
. ^ Moderately
Dirty ,J
polluted
10~5-10"6 10"2-10"5
io-3-io-4 lo^-io-2
io-"-io-5 lO-'-io"2
10"2-10"3 10.0-10"1
10~3-10-3 10"1
80.0 4.0
45.0 2.5
MICROORGANISMS IN WATER SAMPLES OF
Arbitrarily
clean
1.0- 10"1
1.0
1.0-lCT1
1.0-10.0
1.0
Over 1.0
0.8
8.2
THE SECOND BAY
Degree of water pollution
Type of microbe
Coliform bacteria
Enterococci
Petroleum oxidizing
mi croorganisms
Total microbial number in
thousands of microbe
bodies per ml
Season
Fall
Winter
Fall
Winter
Fall
Winter
Fall
Winter
^ . . Moderately
Dirty . . *
polluted
lO^-lO"2 1.0-10"1
IO"2 1.0-10"1
1.0-10"1 10.0-1.0
IO"1 10.0
IO"2 10-1
not determined
6.0 0.7
9.0 1.5
Arbitrarily
clean
1.0-10.0
10.0
10.0
10.0-100.0
1.0-10.0
0.1
0.2
In the first bay, investigations were carried out to identify the re-
lationship of the percentage of intestinal phages and the degree of pollution
of the aquatorium with commercial-domestic organic material. A total of 135
water samples was investigated (45 samples each from points with different de-
grees of pollution).
The following cultures were employed for phage nutrients: Escherichia
108
-------
coli, Shigella sonnei, Shigella flexneri, Salmonella new1 and, and Salmonella
breslau. The frequency of isolating phages from the water samples of differ-
ent degrees of pollution is given in Table 3.
TABLE 3. FREQUENCY OF ISOLATING PHAGES FROM SAMPLES OF BAY WATER (in %)
Test microbes
Escherichia coli
Shigella flexneri
Shigella sonnei
Salmonella newland
Salmonella breslau
I
13.
2.
2.
0
0
Without
3
2
2
inoculation
2
0
0
0
0
0
3
0
0
0
0
0
1
55.
31.
22.
11.
4.
With
5
1
2
1
4
inoculation
26
15
6
0
0
2
.6
.5
.6
11
4
0
0
0
3
.1
.4
Note: 1—dirty water; 2—moderately polluted water; 3—arbitrarily pure water
From the analysis of the table, one can see the clear relationship of
the percentage of isolation of phages and the degree of purity of the water.
Without inoculation (which indicates the high concentration of phage-
sensitive bacteria), only phages to E. coli, Shigella flexneri, and Shigella
sonnei were isolated from samples of dirty water with a titer, according to
Appel'man, of 10~2.
With inoculation, phages were isolated from all three categories of
water samples, and the greatest percentage of isolation of phages fell to the
phages to E. coli and Shigella flexneri.
Phages of moderate and strongly lytic activity to coliform bacteria
and the Shigellas (10"^-10~^) and low lytic activity to the Salmonellas
_io~3) were isolated from dirty water samples.
A study of the range of valents and specificity of the isolated phages
showed that the spectrum of action of phages within the confines of its spe-
cies is very high. Thus the bacteriophage of E. coli lysed 100% of the test
cultures (40 strains) .
Most phages were strictly species specific, and only three varieties
of the phage of Shigella sonnei lysed Shigella flexneri. In most of the
phages the colonies were circular. The dimensions of the negative spots
strongly varied from very small (0.1 x 0.1) to large (5x5 mm). In some of
the phages, the formation of sterile spots of various size and morphology was
simultaneously observed (transparent colonies, transparent colonies with a
dark center, dark colonies with a transparent center) , which indicates the
multiplicity of phage types to the sensitive bacteria.
As the result of the investigations, a parallel was established
109
-------
between finding phages in the water, the coli-titer, the titer of enterococci
and the total microbe inoculation.
Hydromechanical (or physico-oceanographic) factors influenced the de-
gree of pollution and the intensity of processes of autopurification of marine
aquatoria to a significant degree. In the second bay the authors followed a
change in the titer of the sanitary-bacteriological and petroleum-oxidizing
microorganisms in connection with the tidal hydrological regime of the aqua-
tor ium.
The titer of the indicator microorganisms was determined during maxi-
mum high and low tides in the southern part of the bay, where waste waters are
continuously discharged, and in the northern part of the bay which washes
through the sound to the open sea.
The relationship of titers of the indicator bacterioflora and the hy-
drological regime of the bay is shown in Table 4 (an average of 6 repetitions
is given).
TABLE 4. THE EFFECT OF HIGH AND LOW TIDES ON THE TITER OF THE SANITARY-
INDICATOR AND PETROLEUM OXIDIZING MICROORGANISMS OF THE MARINE BAY
Type of microbe
At outlet
to the open sea
At discharge point
of waste water
High tide
Low tide High tide
Low tide
Petroleum oxidizing
microorganisms
Coliform bacteria
Enterococci
Total microbe number in
thousands of microbe
bodies per ml
0.1
10.0
1000.0
0.1
0.01
1.0
10.0
0.5
0.01
0.01
1.0
4.0
0.001
0.001
0.1
15.0
It is apparent from Table 4 that the fluctuation in the titer of the
indicator microorganisms during high and low tides indicates the significant
role of the tidal hydrological regime in the autopurification of the studied
aquatorium from the commercial-domestic and petroleum pollutants.
A study was also made of the distribution of petroleum-oxidizing mi-
croorganisms in open regions of the Barents Sea, which is less polluted with
petroleum products than the bays. Sampling was carried out at 141 stations.
In the coastal zones of the sea a number of petroleum oxidizing microorganisms
fluctuated from 1 to 30 specimens per ml, and in the open regions, up to one
specimen per 100 ml.
110
-------
CONCLUSIONS
1. The systematic detection of phages of the coliform group in samples
of polluted water indicates the indicator role of the given microorganisms.
2. The titer of petroleum oxidizing microorganisms in the aqueous en-
vironment can serve as a criterion of pollution of the studied aquatorium from
petroleum pollutants.
REFERENCES
Mironov, O. G. 1961. Materials for achieving a sanitary water mass at the
Theodosian Health Resort. Gigiyena i Sanitarii, No. 4.
Mironov, 0. G. 1967. Convergent winds and several indicators of pollution
in sea water. Naukova Dumka, Kiev.
Mironov, 0. G. 1970. The role of microorganisms growing on oil in the self-
purification process and as an indication of oil pollution in the sea.
Okeanologiya, No. 5.
Ill
-------
PERSISTENCE LIMITS IN ECOLOGICAL SYSTEMS
by
Kenneth T. Perez
Environmental Protection Agency
Environmental Research Laboratory
Narragansett, Rhode Island 02882
ABSTRACT
Establishment of persistence limits based upon previous theoretical,
field, and laboratory studies was found to be either inappropriate or incon-
clusive. Laboratory simulations appear to be the most promising approach to
the approximation of these limits.
THE PROBLEM
Large-scale ecological systems have been and are being subjected to
various disturbances (e.g., hurricanes, forest fires, glaciation). Over geo-
logic time, adaptation to these disturbances has resulted in the present-day
systems. However, within the last two centuries some disturbances, such as
nutrient and metal inputs, have proceeded at rates greater than those observed
in the past (Anonymous, 1975; Goldberg et al., 1976; Elias et al., 1975).
Also, natural systems have never experienced some types of disturbances, such
as radionuclides and polychlorinated biphenyl, at present levels and/or rates
of addition. Much attention has been given to the temporal description of
systems continuously disturbed by anthropogenic inputs (e.g., Menzel, personal
communication; Hall et al., 1970; Odum and Chestnut, 1970; Schindler et al.,
1973; Woodwell, 1962, 1970). While informative, studies of this kind indicate
only what kind of system changes have occurred; they do not, however, address
the fundamental question of the persistence of the system in the face of such
disturbances. The objective of this paper is to define persistence limits and
identify an experimental framework suitable for the measurement of such limits
based on previous studies.
PERSISTENCE LIMITS
As with many ecological terms, persistence has various definitions.
For Holling (1973), a disturbed system persists as long as the interrelation-
ships between the state variables (e.g., species) are maintained: permanent
112
-------
changes in structure but not function are allowed. In the present communique,
a disturbed system persists, provided it will revert both structurally and
functionally to its predisturbed state once the disturbance is relaxed. A
persistence limit of a system is defined as that disturbed state beyond which
the system is irreversible. Each unique disturbance maps a persistence limit;
the family of such limits defines the persistence limits of a system. From a
theoretical and management point of view, the latter definition of persistence
and the identification of persistence limits are considered more important than
Rolling's since the system is maintained in its totality. If the changes in
system-state during anthropogenic disturbances are confined within the persis-
tence limits, then the system or resource will be maintained by definition.
Thus, persistence limits become a logical management tool.
There are a number of prerequisites to and constraints on the above
operational definition of persistence and persistence limits. First, the un-
perturbed system of interest must be adequately defined in time and space and
has holistic properties. Multiple stable points (Sutherland, 1975) are also
inherent in the unperturbed system. Second, the time frame for recovery must
be short relative to possible long-term (1000 years depending upon the genera-
tion time of the slowest biotic element) natural changes in equilibrium states
(see Botkin and Sobel, 1975). Third, the ability to determine a change in the
system during a disturbance will be technology dependent (Woodwell, 1975). A
change in a system exceeding a persistence limit could occur but remain unde-
tected because of the lack of sensitivity in the prevailing technology. How-
ever, it is possible to indirectly detect the presence or absence of thresholds
of systems response by regression techniques applied to disturbance response
data at detectable levels.
PERSISTENCE LIMITS AND PREVIOUS STUDIES
A. THEORETICAL STUDIES
Computer simulation and analytical models of ecosystems abound (May,
1975; Patten, 1971) . Measures of persistence limits for these models might
provide estimates for natural systems. The validity of such estimates assumes
the models are adequate representations of the field or total system. However,
the data base and principles from which most complex models are formulated are
usually derived from isolated components of a system. These components are
experimentally isolated from the total system before experiments are performed.
If systems are holistic (Gallopin, 1971), then even a detailed knowledge of
their parts will not provide the information necessary to describe the total
system. Walters and Efford (1972) showed that a complex model derived from a
ten-year study of isolated components of a lake ecosystem provided limited dy-
namical information. Therefore, the establishment of persistence limits using
previous theoretical models is inappropriate.
113
-------
B. FIELD STUDIES
Experimental studies in the field dealing with the recovery of dis-
turbed, complex systems are few in number and inconclusive. It is difficult
in terms of logistics, cost, and time to disturb and observe the recovery of
large-scale ecosystems. Adequate controls (unperturbed replicates) often do
not exist or are unavailable. Cause-and-effeet relationships in the field
are sometimes difficult to interpret because of unknown and/or uncontrollable
disturbances. These difficulties may explain the paucity of large-scale field
studies. The few studies which deal with recovery usually suffer from con-
straints of time, scale, and control. For example, Bormann et al. (1974) dem-
onstrated the recovery after three years of two functional system properties,
the fluxes of dissolved and particulate material after cutting the large pri-
mary producers and applying chemical defoliants to a deciduous forest. How-
ever, complete structural recovery (if it ever occurs) will take half a cen-
tury or more. Similarly, following fumigation with methyl bromide, Simberloff
and Wilson (1970) showed after two years the incomplete recovery of the ter-
restrial anthropod species found on mangrove islands. After almost thirty
years, sewerage import into Lake Washington ceased. However, four years
later, the relaxation^of the sewerage has not resulted in the complete recov-
ery of the zoo-phytoplankton structure although concomitant watershed changes
may confound this recovery (Edmondson, 1972) .
Through removal of a predatory starfish, Pisaster, from an intertidal
community for five years, a predation disturbance resulted in a shift in the
age structure of its prey, Mytilus (Paine, 1974). Even after re-exposure to
starfish for seven years, the older and larger Mytilus remained (Paine, per-
sonal communication). The starfish, perhaps because of energy constraints,
did not attack these larger Mytilus. Whether this new structure will provide
a refugium for newly settling Mytilus in the future and, thus, propagate this
structure is unknown. This study possibly demonstrates the existence of a
system-state exceeding a persistence limit.
Qualitative and quantitative temporal characteristics of a disturbance
push a system toward its persistence limits. In Paine's (1974) study, the es-
sential characteristic was the duration of the predation disturbance. If the
starfish were excluded for a specific time interval less than five years, then
the resulting smaller increase in the age structure of Mytilus would be re-
versed following starfish re-entry. How critical other temporal properties of
a disturbance, such as intensity, pattern (deterministic-stochastic), speed
and acceleratory rates, are to the displacement of a system beyond its persis-
tence limits has yet to be described.
C. LABORATORY STUDIES
Because of the difficulties inherent in field studies, many ecologists
have attempted to develop laboratory microcosms as analogues of large-scale
systems. However, before attempting to measure persistence limits in micro-
cosms, it is necessary to know how accurately (if at all) the laboratory sys-
tem approximates the natural system. If one could identify properties that
define a system, then the criteria for testing the adequacy of laboratory
simulation would follow. However, because each investigator has differing
114
-------
views concerning those properties, the criteria become dependent upon the in-
vestigator. For example, Paine (personal communication) feels that microcosm
studies excluding large macrofauna are questionable. Alternately, Gordon et
al. (1969) concluded that their microcosms were analogues of a natural system
because some generalized system properties were observed. However, it should
be noted that as the criteria for adequate simulation become more generalized,
a false conclusion that the laboratory system mimics the natural system be-
comes more likely. The importance of macrofauna, turbulence, water turnover
and gas exchange, spatial and temporal heterogeneity, etc., to the simulation
of natural systems in the laboratory remains to be determined. Comparisons,
where possible, of the microcosms and in-situ natural systems being simulated
are first steps to field confirmation of laboratory systems. Perez et al. (in
preparation) have compared various structural (species lists, nutrients, chlo-
rophyll, and ATP) and functional (benthic and pelagic metabolism and nutrient
fluxes) properties of laboratory microcosms to the field and found it possible
to accurately simulate a complex marine system in the laboratory for more than
half a year. The measure of persistence limits on these laboratory systems
will provide estimates of the natural system. No one has attempted to estab-
lish persistence limits for complex microcosms (Levandowsky, in press).
It is my view that the microcosm approach, with adequate experimental
and field control, is the most promising vehicle for providing approximations
of persistence limits of large-scale ecological systems.
LITERATURE CITED
Anonymous. 1975. Assessing Potential Ocean Pollutants. National Academy of
Sciences, Washington, D.C. 438 pp.
Bormann, F. H., G. E. Likens, T. G. Siccama, R. S. Pierce, and J. S. Eaton.
1974. The export of nutrients and recovery of stable conditions following
deforestation at Hubbard Brook. Ecol. Monogr. 44:255-277.
Botkin, D. B., and M. J. Sobel. 1975. The complexity of ecological stability.
Pages 131-140 in S. A. Levin, ed., Ecosystems Analysis and Prediction.
Proc. Slam-Sims Conf.
Edmondson, W. T. 1972. The present condition of Lake Washington. Verh. Int.
Verein. Limnol. 18:284-291.
Elias, R. , Y. Hirao, and C. Patterson. 1975. Impact of present levels of
aerosol Pb concentrations on both natural ecosystems and humans. Int.
Conf. on Heavy Metals in the Environment, October 30, Toronto, Canada.
19 pp.
Gallopin, G. C. 1971. A generalized model of a resource-population system.
I. General properties. Oecologia (Berlin) 7:382-413.
Goldberg, E. D., E. Gamle, J. J. Griffin, and M. Koide. 1976. Strategies
for determining estuarine pollution history. 1st American-Soviet Symp. on
the Biological Effects of Pollution on Marine Organisms, Gulf Breeze, Fla.,
Sept. 20-24.
115
-------
Gordon, R. W., R. J. Beyers, E. P. Odum, and R. G. Eagon. 1969. Studies of a
simple laboratory microecosystem: Bacterial activities in a heterotrophic
succession. Ecology 50:86-100.
Hall, D. J., W. E. Cooper, and E. E. Werner. 1970. An experimental approach
to the production dynamics and structure of freshwater animal communities.
Limnol. Oceanogr. 15:839-928.
Holling, C. S. 1973. Resilience and stability of ecological systems. Ann.
Rev. Ecol. Syst. 4:1-23.
Levandowsky, M. Multispecies cultures and microcosms. In O. Kinne, ed.,
Marine Ecology (in press).
May, R. M. 1973. Stability and Complexity in Model Ecosystems. Princeton
Univ. Press, Princeton, N.J.
Odum, H. T., and A. F. Chestnut. 1970. Studies of Marine Estuarine Ecosys-
tems Developing with Treated Sewerage Wastes. Pages 1-363 in Annual Report
for 1969-1970. Institute of Marine Sciences, Univ. of North Carolina,
Chapel Hill and Moorehead City, N.C.
Paine, R. T. 1974. Intertidal community structure: Experimental studies on
the relationship between a dominant competitor and its principal predator.
Oecologia (Berlin) 15:93-120.
Patten, B. C., ed. 1971. Systems Analysis and Simulation in Ecology, Vol. 1.
Academic Press, New York. 607 pp.
Perez, K. T., C. Oviate, and S. Nixon. 1976. Laboratory simulation of a ma-
rine ecosystem (in preparation).
Schindler, D. W., H. Kling, R. V. Schmidt, J. Prokopowich, V. E. Frost, and
M. Capel. 1973. Eutrophication of Lake 227 by addition of phosphate and
nitrate: The second, third, and fourth years of enrichment, 1970, 1971,
and 1972. J. Fish. Res. Bd. Canada 30:1415-1440.
Simberloff, D. S., and E. O. Wilson. 1970. Experimental zoogeography of is-
lands: A two-year record of colonization. Ecology 51:934-937.
Sutherland, J. P. 1975. Multiple stable points in natural communities.
Amer. Nat. 108:859-873.
Walters, C. J., and I. E. Efford. 1972. Systems analysis in the Marion Lake
IBP Program. Oecologia 11:33-44.
Woodwell, G. M. 1962. The effects of ionizing radiation on terrestrial eco-
systems. Science 138:572-577.
Woodwell, G. M. 1970. The effects of pollution on the structure and physi-
ology of ecosystems. Science 169:429-433.
116
-------
Woodwell, G. M. 1975. The threshold problem in ecosystems. Pages 9-21 in
S. A. Levin, ed., Ecosystems Analysis and Prediction. Proc. Siam-Sims
Conf.
117
-------
THE DEVELOPMENT OF STANDARD LABORATORY AND FIELD
ECOSYSTEM RESEARCH TO DETERMINE THE EFFECTS OF
POLLUTANTS ON THE MARINE ENVIRONMENT
by
K. S. Burdin
Department of Biology, Moscow State University, USSR
For the sake of convenience in examining the problem of the specifics
of research into the effects of pollutants on the marine environment in la-
boratory and field conditions, it is expedient to study the concepts of the
organizational levels of living organisms currently existing in published ma-
terial. Such an approach made it possible for V. D. Fedorov (1974) to develop
fundamental principles of diagnostic and prognostic monitoring. In developing
these principles, we have first undertaken to systematize some of the biologi-
cal parameters (variables) described in the literature (the "biological re-
sponse," in Fedorov1s terminology) which lend themselves to measurement.
Many efforts to classify the structural and functional parameters on
the basis of the organizational levels of living organisms have heretofore
been unsuccessful for a variety of reasons. Primary among them has been a
lack of precision in the concept of correspondence between certain "biological
responses" and the levels of organization. By way of overcoming these obsta-
cles, we have once again undertaken to compare two systems of the levels of
organizing living organisms existing in the literature. According to the Odum
system (1975), the spectrum of organizational levels is characterized as a
horizontal series so that, in his opinion, all levels deserve the attention of
researchers to the same extent. And, with a shift to the right from genetic
systems to ecosystems, some parameters become more important and more change-
able, while the importance of others becomes negligibly small and their
changeability is hardly notable. The picture will appear somewhat different
if the diagram of organizational levels suggested by N. P. Naumov (1972) is
used as the basis for study.
With such an approach, it is possible to isolate the characteristic
level of the structural hierarchy of the organic world in which the integra-
tion of parts reaches such a degree that the system is capable of isolated
existence and independent reproduction of the system, and those relative to
the lower and intermediate levels do not possess such capabilities.
Therefore, it would be justified to examine the structural and func-
tional parameters of the upper but not those of the intermediate and lower
levels. A collection of biological responses for the lower and intermediate
118
-------
levels of the molecular-cellular and organismic stages would depend primarily
on the developmental level of the methodical concepts derived from physics,
chemistry and cybernetics primarily in molecular biology, biochemistry and bio-
physics. An extensive arsenal of methodical concepts of molecular biology and
biophysics enables us to obtain varied information such as the kinetics of in-
tercellular processes and the structure of vitally important macromolecules,
biological membranes or cellular organoids.
TABLE 1. HIERARCHY OF THE STRUCTURE OF THE ORGANIC WORLD
Levels
Molecular-cellular
Stages
Organismic
Superorganismic
Lower
Intermediate
Higher
Molecules of one class
Cellular organoids
Cells
Tissues
Organs, their
systems
Organisms
Populations, species
Biocenotic complexes
Biocenoses, biomes
The theoretical number of morphological and physiological parameters
of an organism is infinite (Serebrovskiy, 1973). In each instance, the quan-
titative measure of the biological response of a morphological and physiologi-
cal parameter will depend on the methodical possibilities. In resolving prac-
tical problems in the fields of medicine and applied biology, those methodical
concepts which can be used without particular complications for the measurement
of biological responses—and having a high correlation with such important
parameters as mortality or viability, reproductivity and the continuity of
life, and so on—will be pressed into service first.
Mechanisms underlying the basis for the dysfunction of various cells,
organisms, whole populations or communities as biological responses to the ef-
fect of one or more harmful factors will be distinguished from one another, if
only because of the various collections of systemic parameters inherent in
each state of a multilevel hierarchy of the structure of living organisms.
It is especially important to bear in mind the considerations outlined
above in setting up standard programs of laboratory and ecosystem field re-
search to determine the effects of pollutants on the marine environment. The
employment of concepts pertaining to the organizational levels of living orga-
nisms allows us to make more precise the sphere of applicability of experimen-
tal results obtained in the laboratory and the potential for using them to ex-
plain the functioning processes of superorganismic systems. The rules which
should be used in structuring programs of laboratory and field ecosystem re-
search are presented below.
Rule I. Laboratory research data on the condition of cells and orga-
nisms from which the entire ecosystem is made up cannot be used unequivocally
to describe the condition of the superorganismic level.
119
-------
Rule II. A parametric description of the status of the superorganis-
mic levels of organization (population, communities, ecosystem) does not neces-
sarily include the parametric description of all its constituent elements.
Rule III. Each level is described with a finite number of systemic
parameters whose measurement is interpreted in accord with a generally ac-
cepted standard method.
Rule IV. Standard programs of laboratory research include a mandatory
collection of test organisms cultured under standard conditions and a detailed
description of standard toxicological methods.
Rule V. At a minimum, standard programs of laboratory research in-
clude experiments on various species of test organisms, on populations and
simulated ecosystems (microcosms).
Rule VI. Standard programs of field research, in addition to measur-
ing systemic parameters for the appropriate level of the superorganismic stage,
should also include the formulation of active experiments based on plans pro-
vided that the gradients of pollution distribution in the aquatic environment
are known.
Rule VII. The conduct of laboratory experiments is warranted only in
cases where the level of aquatic pollution being studied is known and the con-
sequences brought about by its action cannot be established on the basis of
the ecosystem parameters.
THE FEASIBILITY OF DEVELOPING "STANDARD"
LABORATORY STUDIES TO DETERMINE THE EFFECTS
OF POLLUTANTS ON THE MARINE ENVIRONMENT
In order to evaluate the effect of pollutants on various species and
communities inhabiting the open ocean on a quantitative basis, it is necessary
to conduct long-term field observations. However, this is not always possible
because of the problems associated with culture in laboratory conditions. In
this case, the only departure from the established position will be the conduct
of experiments in other sensitive species of marine organisms which can be
easily cultivated in laboratory conditions. Thus, by sensitive organisms we
mean organisms which are quite sensitive to any sort of pollutant; for example,
crustaceans react to parts per million of chlorinated hydrocarbons and certain
metals. They can be considered sensitive "indicator" organisms to these pol-
lutants.
Laboratory research on various species of the organism should be pri-
marily devoted to determining: 1) the rate of accumulation and rate of re-
moval of the pollutant; 2) the relationship between the pollutant concentra-
tion in the organism and its concentration in the water; 3) the permissible
concentration according to generally accepted methods; 4) the remote conse-
quences of small concentrations; 5) the effects of endangering organisms at
the molecular level; and 6) the mechanisms of the toxic effect of pollutants.
120
-------
STUDYING THE EFFECTS OF POLLUTANTS
IN LABORATORY MICROCOSMS
Microcosms are already extensively used to solve the theoretical and
practical problems of ecology. The fact that a large portion of the work is
performed in fresh water microcosms notwithstanding, there are no major ob-
stacles in using laboratory microcosms with salt water and marine organisms.
In view of the low density of populations of pelagic plankton and fish commu-
nities in the open sea, it is difficult to evaluate the effect of pollutants
on their structure and function. A departure from the established position
can be the staging of experiments on microcosms established on small islands,
in inlets (on floating rafts) or on ships. It is possible to use such parame-
ters as growth, reproductivity, physiological process and the variety of spe-
cies as criteria for the biological response to pollutant action in such la-
boratory microcosms. The simplest models of microcosms must meet certain re-
quirements even though it is virtually impossible to suggest a universal micro-
cosm model. In biology and in medicine, experimenters use white strainless
mice or some strains of pure lines as experimental "living test tubes." The
universality of a utilized object certainly serves as a basis for juxtaposi-
tion of the results obtained.
At present it is possible to observe an analogous situation with only
one of the limits of hydrobiology, water toxicology, where monotypical orga-
nisms are rather extensively used in biological experimentation in standard
conditions. In the ecological literature, we did not encounter any works
where an experiment would have created a universal model of a microecosystem
or the figurative "ecological mouse." It is theoretically possible to take
two independent courses in creating an "ecological mouse." One is to "carve
out" a portion of the natural ecosystem and integrate it into the conditions
of laboratory containers, while continuing to imitate natural changes in the
basic physicochemical parameters of the environment while maintaining consis-
tency in others. In this instance, it is possible to transfer the results ob-
tained in a laboratory experiment to the natural ecosystem with some degree of
accuracy as is done in many cases where the results of experiments carried out
on laboratory mice are projected onto higher mammals.
Another method of creating an "ecological mouse" begins with the se-
lection of organisms found in definite feeding relationships and capable of
normal life activity in laboratory conditions. A system consisting of a fish
(Gambusia affinis) , a. crustacean (Daphnia magna) , a snail and algae (Isensee
et al., 1973) can serve as a typical example of a microecosystem. This type
of microcosm can be used to study the movement of pollutants resistant to bio-
logical degrading throughout the entire food chain. The importance of estab-
lishing such experiments lies in the fact that each of the constituent orga-
nisms can serve as a test organism in toxicological experiments, so that data
obtained in various cultures will have an unquestioned value when superimposed
on data obtained in experiments on entire microcosms. Experiments with long-
lived pollutants introduced into the surrounding environment in relatively
small concentrations (mercury, cadmium, and DDT, for example) have a special
importance in this type of experiment.
Establishing such experiments on the basis of standard algorithms al-
lows us to make more precise the sphere of applicability of concepts
121
-------
pertaining to permissible concentrations, with the potential for expansion to
the concept of an ecological norm of pollution.
STANDARDIZATION OF FIELD ECOSYSTEM
RESEARCH TO DETERMINE THE EFFECTS OF
POLLUTANTS ON MARINE ECOSYSTEMS
At the present time, there is an extensive network of field research
programs on the world's oceans at national and international levels. Basic-
ally, specialized scientific-research vessels and, in some cases, ships con-
ducting nonsystematic studies whose primary task is the performance of commer-
cial, transport and other types of work take part in research programs.
Field ecosystem studies to determine the effects of pollutants on mar-
ine ecosystems are, in very rare instances, set up in accord with an extensive
program involving several years' observation of the condition of the biotic
component. In our view, the main reason for this is the absence of precise
concepts relating to systemic indicators characterizing the integral proper-
ties of ecosystems. At best, field studies attempt to include a determination
of the majority of known physicochemical and biological indicators which can
be potentially useful in evaluating the effect of pollutants on marine ecosys-
tems. In terms of form, field studies will be only slightly different from
programs of field research into fresh water ecosystems. And so it goes. The
accumulation of data obtained in field studies does not result in qualitative
progress for evaluating the status of marine ecosystems.
The conduct of field studies should be put into practice by standard
methods proceeding, first of all, from systemic indicators leading to charac-
terization of the integral properties of the superorganismic stage of the
structural hierarchy of the organic world.
Standardization at a given stage should mean the isolation of a frame-
work of systemic indicators which can be used to evaluate the effect of pollu-
tants on the marine ecosystem. Beginning with a minimum framework of systemic
indicators, the unification and intercalibration of determinative methods
should be accomplished during the second stage of field study standardization.
At the third stage, all procedures for the collection of data, its analysis
and storage as well as procedures for retrieval and exploitation should be
unified. At this point, the work should conclude with the development of
standard documentation starting with the collection of field data and ending
with storage.
THE COMBINATION OF LABORATORY AND FIELD RESEARCH
IN ORGANIZING BIOLOGICAL MONITORING OF THE EFFECT
OF POLLUTION ON THE MARINE ENVIRONMENT
The presently existing skepticism regarding the extrapolation of re-
sults obtained in experiments with various test organisms on entire ecosystems
is fully justified for the reasons discussed above. There are some means for
overcoming these obstacles, but all of them are associated with significant
122
-------
material expenditures. It is precisely for this reason that many proposed
projects have not been successfully realized.
While not rejecting the means or methods of combining laboratory and
field research established by tradition in each nation or scientific collec-
tive, it is suggested that some of the obvious approaches be shifted from la-
boratory experiment to field studies. This would enable us to see the degree
of similarity and the possibilities for extrapolating laboratory data onto a
situation in natural conditions.
From the functional point of view, it is possible to study an ecosys-
tem to some degree from various aspects in relation to the stated question and
the types of specialists and equipment. An ecosystem can be characterized by
flows of energy (inward and outward), by the complexity of the food chain, the
diversity of organisms and the cycle of matter. It is obvious that the action
of any disruptive factor, including pollution, can substantially change one of
the functional characteristics of the ecosystem. In natural ecosystems, this
can be established on the basis of lengthy observations over several years
within the conditions of previously planned water areas, if a possible connec-
tion between the level of pollution and the change in the amplitude of one of
the functional characteristics of the ecosystem can be found.
In the overwhelming majority of cases, field ecosystem research is
carried out on a case-to-case basis. In these conditions, the presence of
data on the pollution level in the water areas of interest enables us to pro-
pose a number of standard laboratory ecosystem studies.
MICROCOSM WITH A VOLUME OF 1-5 LITERS
1. A microecosystem consisting of one or several populations and
placed in rigidly controlled conditions, for example, in the conditions of the
conventional chemostat (Carpenter, 1969).
2. Microecosystems consisting of several pure population cultures
with a closed cycle of food substances (Nixon, 1969).
3. Microecosystems consisting of a small portion of a natural ecosys-
tem obtained through multiple reseedings (Beyers, 1963).
4. A microcosm consisting of the minimum collection of food chain or-
ganisms (including fish) assuring a closed cycle of food substances (Isensee
et al., 1973).
5. A microcosm consisting of species or groups of organisms important
from the point of view of the simulated ecosystem living in natural conditions
(Strickland, 1967).
6. A microcosm consisting of abiotic and biotic components combined
in a reservoir or basin populated with basic or ecologically important orga-
nisms or groups of organisms.
7. A microcosm simulating as far as possible the factors of the ex-
ternal environment in laboratory conditions (the combination of species, tem-
perature, illumination, dissolved oxygen, bottom contours), basic biotic
123
-------
components from the point of view of trophies and the rate of introducing pol-
lutants into the microcosm (Falco and Sanders, 1973).
The listed variants of the microcosm can be prepared in laboratory
conditions. In order to carry out studies on these microcosms, the following
is required: a) well-equipped aquatic basins and ponds as experimental bases;
b) experimental basins or continuous-flow canals with an established hydro-
logic regime (simulated water currents, micromodels of reservoirs, and so
forth); c) automated means for observing, collecting and analyzing the infor-
mation obtained from experimental objects; and d) analog machines and third
generation data processing machinery.
The second direction of research is strongly associated with the con-
duct of ecosystem research in field conditions; however, the methods will dif-
fer slightly from those employed in the study of laboratory microcosms. Micro-
cosm size in field conditions will depend primarily on the problem to be solved
and technical and financial capabilities. In all, we propose to distinguish
three types of microcosm variants:
1. The small microcosm with a capacity not exceeding 50 liters enables us
to isolate a small portion of the natural ecosystem with a transparent
membrane (plastic being used most frequently) and to carry out observa-
tions based on a pre-planned program (Fedorov, 1974).
2. The intermediate sized microcosm with a volume of more than 50 liters
isolates a portion of a natural ecosystem into which not all trophies
are introduced (Davis, 1973).
3. The large microcosm enables us to study an isolated ecosystem in all its
complexity (Chem. Eng. News, 1973).
Conducting research in isolated microcosms in field conditions has a
tremer.dous advantage over laboratory studies. The main advantage lies in the
fact that the isolated microecosystem being observed is under the influence of
the same external environmental factors as the entire ecosystem. The intro-
duction of pollutants into an isolated microcosm makes it possible to predict
the fate of the entire ecosystem with the highest degree of accuracy.
In this way, ecosystem research on laboratory microcosms provides a
clarification of the general principles of microcosmic response reactions to
different influences including pollutants of various sorts. Laboratory models
of microcosms, as a rule, remain remote copies of natural ecosystems. There-
fore, the results of experiments performed on them can be extrapolated to
natural ecosystems with such a degree of accuracy that all of their component
parts and the connection between them has been taken into consideration. The
latter requires knowledge of the structure of natural ecosystems and its func-
tional specifics.
Together with the exploitation of the field variants of microcosms,
ecosystem studies enable us to obtain results which can be extrapolated with
complete validity to entire ecosystems, particularly in cases where they are
applied to intermediate and large microcosms.
124
-------
REFERENCES
Beyers, R. J. 1963. The metabolism of twelve aquatic laboratory microecosys-
tems. Ecol. Mongr. 33.
Carpenter, E. J. 1969. A simple, inexpensive algal chemostat. Limnol.
Oceanogr. 14.
Chemical and Engineering News. 1973. Effects of pollutants on marine life
probed. December 17.
Davies, J. M., J. C. Gamble, and J. H. Steel. 1973. Preliminary studies with
a large plastic enclosure. Proc. Conf. Estuarine Ecol., Myrtle Beach,
South Carolina. October.
Falco, J. W., and W. M. Sanders III. 1973. A physical model for simulation
of aquatic ecosystems. In_ Modeling the Eutrophication Process. Proc. of
a Workshop held at Utah State University, Logan, Utah, Sept. 5-6.
Fedorov, V. D. 1974. Towards a strategy for biological monitoring. Scien-
tific report. Biolog. Nauki No. 10.
Isensee, A. R., P. C. Kearney, E. A. Woolson, G. E. Jones, and V. P. Williams.
1973. Distribution of alkyl arsenicals in model ecosystem. Environ. Sci.
Technol. 7:9.
Naumov, N. P. 1972. Ill Biologicheskaya Kibernetika, ed. A. B. Kogan.
Vyshaya Shkola, Moscow.
Nixon, S. W. 1969. A synthetic microcosm. Limnol. Oceanogr. 14.
Odum, G. 1975. The Bases of Ecology. Translated from the 3d English edition.
MIR, Moscow.
Strickland, J. D. H. 1967. Between beakers and bays. New Sci., February 2.
125
-------
IMPACT OF PESTICIDES ON THE MARINE ENVIRONMENT1
by
David J. Hansen
U.S. Environmental Protection Agency
Environmental Research Laboratory
Gulf Breeze, Florida 32561
ABSTRACT
The impact of pesticides on the marine environment can be assessed by
monitoring their occurrence in the marine environment and by evaluating their
toxic effects in laboratory bioassays. Acute static and flow-through bio-
assays generally have been used to set marine water quality criteria, but bio-
assay techniques now can determine effects of long-term exposure to one or
more toxicants on survival, growth, and reproduction of individual species of
mollusks, arthropods and fishes and effects on communities of estuarine orga-
nisms in the laboratory. Bioassays have been lengthened from 96 hours or less
to between one month and two years, and their complexity has also been broad-
ened. Effects of toxicants on the entire life cycle of an oviparous estuarine
fish, Cyprinodon variegatus, can now be studied, and bioassays have been com-
pleted with endrin and heptachlor. Preliminary experiments using this fish
revealed that they typically develop from an embryo to maturity in 10 to 14
weeks, with about 70% survival in the laboratory. Females produce an average
of eight eggs per day and fertilization success exceeds 90%. Effects of a
polychlorinated biphenyl, Aroclor® 1254, and of a pesticide, toxaphene, on de-
veloping communities of estuarine animals have been investigated. These
studies provided data for predicting pollution-induced shifts in composition
of estuarine animal communities.
INTRODUCTION
Pesticides occur in biological and physical components of coastal and
oceanic ecosystems. Some have been implicated in degradation of portions of
the environment because pesticides either adversely affected organisms or were
bioaccumulated in organisms to concentrations deemed excessive for human con-
sumption. However, data on direct toxic effects of pesticides on marine orga-
nisms are limited. Effects induced during chronic pollution are difficult to
Contribution No. 279, Environmental Research Laboratory, Gulf
Breeze.
126
-------
observe and, if observed, only rarely provide data sufficient to implicate a
pesticide directly. Consequently, potential impacts of pesticides and other
pollutants must be assessed through laboratory studies. Laboratory bioassay
data, when properly evaluated, can aid in predicting effects in the field
(Mount, 1974).
The bioassay is perhaps the most useful technique available to the
biologist for predicting the potential hazard of a chemical. It can vary con-
siderably in complexity and utility, and each procedure has advantages and
disadvantages. These tests may be relatively simple acute static and flow-
through bioassays or complex chronic entire life cycle and community bioassays.
Flow-through bioassays usually measure acute stress with more sensitivity than
static bioassays, whereas life cycle and community bioassays provide better
estimates of "safe" concentrations from which water quality criteria can be
derived.
This paper describes bioassays conducted at the Environmental Research
Laboratory at Gulf Breeze, Florida, to test effects of toxicants on estuarine
animals. These bioassays include: (1) acute bioassays conducted at constant
salinity and temperature and include measurements of concentration of the
toxicant in water and test organism and statistical analysis of mortality
data; (2) bioassays on sensitive larval stages of crabs and shrimp; (3) bio-
assays over the reproductive portion of, or the entire life cycle of grass
shrimp (Palaemonetes pugio) and sheepshead minnow (Cyprinodon variegatus); and
(4) bioassays on communities of benthic macroinvertebrates.
BIOASSAY TECHNIQUES
A. ACUTE BIOASSAYS
Acute toxicity experiments are usually conducted to determine the
quantity of chemical that will adversely affect a certain percentage of the
test organisms in a short period of time. The data are used to compare rela-
tive toxicity and relative sensitivity. Comparisons are most reliable when
bioassay methods are uniform and when the tests produce valid statistical data
supported by chemical analyses of test water.
Acute flow-through bioassays have been conducted on some pesticides to
provide 96-hour LCSO's (concentration lethal to 50 percent of the animals)
supported by statistical and chemical analyses. Results of recent experiments
show that penaeid shrimp are usually more sensitive to chemicals than are oys-
ters, grass shrimp or estuarine fishes (Table 1). Acute toxicities of tested
chemicals, except methoxychlor, exceeded those of acute bioassays published in
the Blue Book (NAS-NAE Committee on Water Quality Criteria, 1972).
Acute bioassays have used flowing water of constant temperature and
salinity to improve comparisons of results. Bioassays (Table 1) of DDT, hep-
tachlor (99%), heptachlor epoxide, lindane and methoxychlor were conducted at
25°C and 20 ppt salinity. The salinity was controlled by an inexpensive de-
vice in which appropriate amounts of fresh and saltwater were added through
127
-------
solenoid valves controlled electrically by a photocell that perceives changes
in water density indicated by a floating hydrometer (Bahner and Nimmo, 1975a).
This device has been used successfully for periods of up to 9 months to main-
tain constant (±1 ppt) salinity during bioassays.
TABLE 1. NINETY-SIX HOUR LCSO's AND 95% CONFIDENCE INTERVALS FOR THE SPECIES
OF ESTUARINE ORGANISM MOST SENSITIVE TO SELECTED ORGANIC CHEMICALS
IN FLOW-THROUGH BIOASSAYS. USUALLY, THE AMERICAN OYSTER, TWO FISHES
AND TWO ARTHROPODS WERE TESTED. CONCENTRATIONS IN WATER WERE MEA-
SURED BY ELECTRON-CAPTURE GAS CHROMATOGRAPHY.
Chemical
Chlordane
DDT*
Dieldrin
Endrin
HCB
Heptachlor (74%)
Heptachlor
(99%)*
Heptachlor
Epoxide*
Lindane*
Methoxychlor*
Toxaphene
Sensitive
Species
Pink shrimp
Brown shrimp
Pink shrimp
Pink shrimp
Pink shrimp
Pink shrimp
Pink shrimp
Pink shrimp
Pink shrimp
Pink shrimp
Pinfish
96-hour
LC50 (yg/1)
0.4(0.3-0.6)
0.1(0.1-0.2)
0.7(0.4-1.2)
0.04(0.02-0.05)
>25
0.1(0.07-0.1)
0.03(0.02-0.04)
0.04(0.001-0.1)
0.2(0.1-0.2)
3.5(2.8-4.4)
0.6(0.5-0.7)
Reference
Parrish et al., 1976
Schimmel and Patrick,
unpublished**
Parrish et al. , 1973
Schimmel et al., 1975
Parrish et al. , 1975
Schimmel et al. , 1976
Schimmel et al. , 1976
Schimmel et al. , 1976
Schimmel and Patrick,
unpublished**
Bahner and Nimmo, 1975b
Schimmel et al. , in press
*Less than five species of estuarine animals tested.
**S. C. Schimmel and J. N. Patrick, Jr., Gulf Breeze Environmental Re-
search Laboratory, Gulf Breeze, Florida 32561.
B. SENSITIVE LIFE STAGE AND LIFE-CYCLE BIOASSAYS
Long-term bioassays on sensitive life stages and on life cycles of es-
tuarine organisms are usually conducted to determine the quantity of chemical
that can be tolerated by a species throughout its life cycle or during a
critical portion of its life cycle. Data from this type of bioassay are es-
pecially important in deriving water quality criteria. Water quality criteria
are used to protect that species and, it is hoped, the ecosystem from chronic
effects of a pollutant. Water quality criteria are frequently obtained by
128
-------
multiplying the 96-hour LC50 of the most sensitive species tested by an arbi-
trary application factor. The arbitrary application factor for persistent
pollutants is usually about 0.01 (NAS-NAE Committee on Water Quality Criteria,
1972). Scientifically derived application factors can be obtained by compar-
ing data from acute bioassays with bioassays in which a fish or invertebrate
is exposed to the chemical throughout its life cycle. The factor is obtained
by dividing the concentration of a toxicant that does not affect survival,
growth or reproduction of a species in life-cycle bioassays by the chemical's
96-hour LC50 for that species (Mount and Stephan, 1967; Eaton, 1973).
Sensitive Life Stage Bioassays
Marine toxicologists have not been able to derive experimental appli-
cation factors based on life-cycle exposure because techniques for maintaining
marine cultures were lacking. Therefore, it is necessary to develop and to
use methods that provide toxicity data on sensitive stages of the life cycle
of saltwater species. Our laboratory funded grants and contracts to investi-
gate effects of pesticides on larval development of dungeness crab, Cancer
magister; blue crab, Callinectes sapidus; and the xanthid crab, Rhithropano-
peus harrisii. Curre'nt or past investigations used captan, carbofuran, chlor-
dane, DEF, malathion, methoxychlor, mirex, propanil, trifluralin, 2,4-D and
juvenile hormones. We also supported research on the effects of methoxychlor
and mirex on embryonic, larval, juvenile, and adult striped mullet, Mugil
cephalus (Lee et al., 1975).
Research at Gulf Breeze on sensitive stages of estuarine organisms
concentrates primarily on larval and postlarval grass shrimp (P. pugio), and
embryos and fry of the fishes C. variegatus, Fundulus similis, F. heterocli-
tus, Leiostomus xanthurus, Menidia menidia and Morone saxatilis. Recent
papers on this research include those of Hansen et al. (1975), Middaugh et al.
(1975), Parrish et al. (1976) and Schimmel et al. (1975). This research fo-
cused primarily on effects of toxicants in water on development and survival
of early life-stages of these species. On the other hand, investigations
(Hansen et al., 1973) of effects of a PCB, Aroclor® 1254,* in eggs of the
sheepshead minnow, C. variegatus, indicated that concentrations in excess of
5 yg/g of the PCB in eggs were lethal to embryos and fry (Fig. 1). If this
PCB has similar effects on other fishes residues exceeding 5 yg/g in eggs
would decrease survival of fry.
Life-Cycle Bioassays
Life-cycle bioassays are routinely conducted by freshwater toxicolo-
gists, but saltwater toxicologists have developed similar procedures only re-
cently. Such freshwater bioassays can be conducted with bluegills (Lepomis
macrochirus), fathead minnows (Pimephales promelas), brook trout (Salvelinus
fontinalis), water fleas (Daphnia magna), and other fishes and invertebrates
(Eaton, 1973).
*Registered trademark, Monsanto Company, St. Louis, Missouri. Mention
of trade names does not constitute endorsement by the Environmental Protection
Agency.
129
-------
1OO
50*
HATCH
at
3
V)
AROCLOR 1254 IN
EGGS (jug/g)
EMBRYOS
Figure 1. Effect of Aroclor 1254 in eggs of sheepshead minnows on the sur-
vival of embryos and fry in laboratory bioassays.
Entire life-cycle bioassays are possible with the estuarine fish, C.
variegatus (Schimmel and Hansen, 1975) . In the laboratory, this oviparous
fish develops from an embryo to maturity in about 10 weeks, with about 70%
surviving. The fish spawns readily in an aquarium, producing about 8 eggs per
day (Fig. 2). The size of the fish apparently has no effect on total egg pro-
duction but does influence the frequency of spawning and egg fertility (Schim-
mel and Hansen, 1975). Females begin producing eggs at 27 mm standard length.
In one experiment, 19 fish less than 35 mm long produced an average 8.2 eggs
per day; 15 fish, 35 mm and longer, averaged 7.8 eggs per day. Smaller fish
produced eggs more consistently than larger fish (50% of the days vs 31%) and
with greater fertility (94% fertility vs 79%). These and other observations
lead to a tentative method for entire life-cycle bioassays using this fish
(Hansen and Schimmel, 1975).
130
-------
60
S 50
40
v 30
>•
u
Z
Ul
D 2O
O
10
AVERAGE NUMBER =8.O
1-10 11-20 21-30 31-40 41-50
DAILY EGG PRODUCTION
Figure 2. Frequency distribution of egg production by breeding pairs of
sheepshead minnows (Cyprinodon variegatus).
Recently, sheepshead minnows were exposed to endrin and to heptachlor
to determine the effect of these pesticides on reproduction. 'In one experi-
ment, sheepshead minnows were exposed to 0, 0.025, 0.077, 0.12, 0.31 or 0.77
yg/1 of endrin measured in water during a life-cycle bioassay lasting 25 weeks.
This bioassay consisted of three parts: (1) The exposure began with embryos
and continued through embryonic development, hatching of fry and growth of the
fry to adulthood; (2) exposure continued during monitoring of spawning success,
including egg production and fertility; and (3) the bioassay ended after em-
bryos and fry obtained from spawning fish were exposed for 28 days. The appa-
ratus used in the experiment was developed by Schimmel et al. (1974) and the
methodology was similar to that of Hansen and Schimmel (1975).
Results of this bioassay showed that sheepshead minnows were affected
by endrin (Table 2). Embryos in 0.31 and 0.72 yg/1 hatched in a shorter pe-
riod than embryos in water free of endrin. Fry in 0.72 yg/1 began to die one
day after hatching; more than half were dead by day 12. Survival of juvenile
fish was unaffected. Survival of spawning females in 0.31 yg/1 was reduced
and their eggs were less fertile than those of control females. Survival de-
creased for fry from eggs spawned by fish that were exposed to 0.31 yg/1 of
endrin throughout their lives.
131
-------
TABLE 2. EFFECTS OF ENDRIN ON SHEEPSHEAD MINNOWS (CYPRINODON VARIEGATUS)
EXPOSED THROUGHOUT THE ENTIRE LIFE-CYCLE. CONCENTRATIONS TESTED
WERE: 0, 0.025, 0.077, 0.12, 0.31 AND 0.72 yg/1.
Concentrations
Generation Life Stage Effect , ...
Fl
Embryo
Fry
Early hatching
Death
Decreased growth
0.31, 0.72
0.31, 0.72
0.31
Juveniles No effect
Adults Death of spawning females 0.31
Decreased fertility of eggs 0.31
F Embryos and fry Death 0.31
Effects of technical heptachlor on reproduction and development of C.
variegatus were studied in a similar experiment beginning with juvenile fish,
rather than embryos. Concentrations of technical heptachlor (heptachlor and
trans-chlordane) measured 0.71, 0.97, 1.9, 2.8 and 5.7 yg/1. During the first
four weeks, some juvenile fish died in 2.8 and 5.7 yg/1 of technical hepta-
chlor. Thereafter, few exposed fish died until the reproductive portion of
the experiment in week 8. Heptachlor also reduced number of spawnings, number
of eggs, fertility of the eggs and survival of fry from fertile eggs.
Techniques are being developed for life-cycle bioassays using the
grass shrimp (P. pugio). Studies have been conducted to determine how light
and temperature affect initiation and success of spawning. Larval and post-
larval shrimp were used in bioassays to observe effects of certain PCB's on
larval development and metamorphosis. P. pugio spawns readily, the larvae
develop normally and the species is sensitive to toxic chemicals. Therefore,
the species may be useful for entire life-cycle bioassays.
C. COMMUNITY BIOASSAYS
Bioassays can be used to predict how communities of estuarine orga-
nisms may respond to a toxicant. Bioassays exposing only one species to a
chemical may help predict how a community may respond if a number of species
from various phyla can be tested under similar conditions. However, predic-
tions made from this type of data are questionable, particularly if little is
known about how species interact in the community. Predictions based on data
obtained from field studies also may be questioned because of problems related
to lack of controls and replication. As an alternative, laboratory tests
could study how communities of organisms react when exposed to a chemical.
132
-------
This approach can be valuable if laboratory communities resemble those in the
field and if replicates and concentrations are adequate for statistical analy-
ses.
I have completed two bioassays to determine the effects of Aroclor
1254, and of toxaphene, an organochlorine insecticide, on the development of
estuarine communities. The numbers, species and diversity of animals that
grew from planktonic larvae in contaminated aquaria were compared with those
that grew in identical uncontaminated aquaria. In each bioassay, seawater,
with its natural complement of plankton, flowed into 40 sand-filled aquaria;
10 for each of three toxicant concentrations and a control. Planktonic lar-
vae colonized the sand and walls of each aquarium. At the end of the experi-
ments (4 months for the PCB and 3 months for toxaphene) organisms were col-
lected in a 1 mm-mesh sieve, preserved and identified.
Composition of communities in control aquaria differed from communi-
ties of estuarine animals that developed from planktonic larvae in salt water
that flowed through 10 aquaria contaminated with 1 or 10 yg/1 of Aroclor 1254
(Hansen, 1974). Communities in control aquaria and aquaria that received 0.1
yg/1 of PCB for four months were dominated (>75%) by arthropods, primarily the
amphipod Corophium volutator (Fig. 3). Aquaria that received 1 or 10 yg/1 of
PCB contained fewer arthropods. The dominant species were chordates, pri-
marily the tunicate Molgula manhattensis. Over 75% of the animals in aquaria
that received 10 yg/1 were tunicates. Phyla, species, and individuals (par-
ticularly amphipods, bryozoans, crabs, and mollusks) were present in fewer
numbers in aquaria receiving the PCB. The abundance of annelids, brachiopods,
coelenterates, echinoderms or nemerteans was apparently unaffected (Table 3).
The Shannon-Weaver index of species diversity was not altered by Aroclor 1254
in this experiment. Therefore, its usefulness in assessing PCB-induced
changes in community structure in the environment may be inappropriate.
TABLE 3. EFFECT OF AROCLOR 1254 ON THE NUMBERS OF PHYLA, SPECIES AND INDI-
VIDUALS AND ON THE SHANNON-WEAVER INDEX OF SPECIES DIVERSITY IN
COMMUNITIES OF ESTUARINE ORGANISMS THAT DEVELOPED IN SAND-FILLED
AQUARIA DURING A 4-MONTH BIOASSAY
Control
Aroclor 1254 yg/1
0.1
10
Phyla
Species
Individuals
Species diversity index
9
52
1776
1.82
7
34
2043
1.26
7
43
1421
2.21
5*
25*
657
1.70
*Statistically different from controls, <* = 0.05.
133
-------
tt
111
aa
40
U. 6O
O
80
1OO
ARTHROPODA
CHORDATA
___
ANNELIDA
CONTROL
O.I I.O
AROCLOR 1254 (ng/\)
10.0
Figure 3. Percentage of organisms from various phyla in communities of es-
tuarine organisms that developed in the presence of Aroclor 1254.
TABLE 4. AVERAGE NUMBER OF ANIMALS IN 10 CONTROL AQUARIA AND 10 AQUARIA THAT
RECEIVED 0.1, 1 or 10 jjg/1 OF TOXAPHENE FOR THREE MONTHS. RANGES
IN PARENTHESES.
Phylum
Control
Toxaphene (yg/1)
o.:
10
Mollusca
Annelida
Arthropoda
Coelenterata
Other
124 (65-146)
56 (19-97)
32 (2-257)
3 (0-21)
0.1 (0-1)
170 (98-274)
62 (33-90)
155 (1-523)
3 (0-19)
0.1 (0-1)
142 (65-237)
66 (31-126)
9 (1-63)
10 (0-44)
373 (245-489)
110 (82-182)
0.4 (0-1)
134
-------
In a similar experiment using toxaphene, the structure of communities
that developed in sand-filled aquaria differed from those in control aquaria.
Exposure concentrations were 0, 0.1, 1 and 10 yg/1. The number of mollusks
(primarily gastropods) tripled, annelids (primarily capitellids) doubled, and
arthropods were almost eliminated in aquaria contaminated by 10 yg/1 of toxa-
phene (Table 4). Similar numbers of pelecypods were found in all aquaria;
however, the height (distance from hinge to distal valve edge) of Morton's
cockle (Laevicardium mortoni) was significantly reduced by 10 yg/1 (Fig. 4).
40
20
CONTROL
O.IMS/I
40
20
40
D
o
S 20
N-467
10.0 .ug I
N:431
40
20
0 4 8 12 16 20 0 4 8 12 16 20
HEIGHT (MILLIMETERS)
Figure 4. Height (distance from hinge to distal edge of valve) of Morton's
cockle collected from a community of estuarine organisms that de-
veloped while exposed to toxaphene.
REFERENCES
Banner, L. H., and D. R. Nimmo. 1975a. A salinity controller for flow-
through bioassay. Trans. Amer. Fish. Soc. 104 (2):388-389.
Banner, L. H., and D. R. Nimmo. 1975b. Methods to assess effects of combina-
tions of toxicants, salinity and temperature on estuarine animals. Trace
Substances in Environmental Health—IX, Columbia, Missouri, June 10-12,
1975, pp. 169-177.
135
-------
Eaton, J. G. 1973. Recent development in the use of laboratory bioassays to
determine "safe" levels of toxicants for fish. In G. E. Glass, ed., Bio-
assay Techniques and Environmental Chemistry. Ann Arbor Science Publish-
ers, Inc., Ann Arbor, Mich.
Hansen, D. J. 1974. Aroclor® 1254: Effect on composition of developing es-
tuarine animal communities in the laboratory. Contrib. Mar. Sci. 18:19-33.
Hansen, D. J., and S. C. Schimmel. 1975. Entire life-cycle bioassay using
sheepshead minnows (Cyprinodon variegatus). Fed. Regist. 40 (123) , part II:
26904-26905.
Hansen, D. J., S. C. Schimmel, and J. Forester. 1973. Aroclor® 1254 in eggs
of sheepshead minnows: Effect on fertilization success and survival of
embryos and fry. Pages 420-426 in Proc. 27th Ann. Conf. S.E. Assoc. Game
Fish. Comm. 1973.
Hansen, D. J., S. C. Schimmel, and J. Forester. 1975. Effect of Aroclor®
1016 on embryo, fry, juvenile and adult sheepshead minnow (Cyprinodon
variegatus). Trans. Amer. Fish. Soc. 104(3):584-588.
Lee, J. H., C. E. Nash, and J. R. Sylvester. 1975. Effects of mirex and
methoxychlor on striped mullet, Mugil cephalus L. U.S. Environ. Prot.
Agency, Ecol. Res. Ser. EPA-660/3-75-015. 18 pp.
Middaugh, D. P., W. R. Davis, and R. L. Yoakum. 1975. -The response of larval
fish, Leiostomus xanthurus, to environmental stress following sublethal
cadmium exposure. Contrib. Mar. Sci. 19:13-19.
Mount, D. I. 1974. Testimony in the matter of proposed toxic pollutant ef-
fluent standards for Aldrin-Dieldrin et al. FWPCA (307) Docket #1.
Mount, D. I., and C. E. Stephan. 1967. A method for establishing acceptable
toxicant limits for fish—malathion and the butoxyethanol ester of 2,4-D.
Trans. Amer. Fish. Soc. 96 (2) :185-193.
NAS-NAE Committee on Water Quality Criteria. 1972. Water Quality Criteria,
1972. U.S. Environ. Prot. Agency, Ecol. Res. Ser. EPA-R3-73-033-March
1973. U.S. Govt. Print. Off., Washington, D.C.
Parrish, P. R., G. H. Cook, and J. M. Patrick, Jr. 1975. Hexachlorobenzene:
Effects on several estuarine animals. Pages 179-187 in Proc. 28th Ann.
Conf. S.E. Game Fish. Comm. 1974.
Parrish, P. R., J. A. Couch, J. Forester, J. M. Patrick, Jr., and G. H. Cook.
1973. Dieldrin: Effects on several estuarine organisms. Pages 427-434
in Proc. 27th Ann. Conf. S.E. Assoc. Game Fish. Comm. 1973.
Parrish, P. R., S. C. Schimmel, D. J. Hansen, J. M. Patrick, Jr., and J.
Forester. 1976. Chlordane: Effects on several estuarine organisms. J.
Toxicol. Environ. Health 1:485-494.
136
-------
Schimmel, S. C., and D. J. Hansen. 1975. Sheepshead minnow (Cyprinodon varie-
gatus): An estuarine fish suitable for chronic (entire life-cycle) bio-
assays. Pages 392-398 in Proc. 28th Ann. Conf. S.E. Assoc. Game Fish.
Comm. 1974.
Schimmel, S. C., D. J. Hansen, and J. Forester. 1974. Effects of Aroclor®
1254 on laboratory-reared embryos and fry of sheepshead minnows (Cyprino-
don variegatus). Trans. Amer. Fish. Soc. 103(3):582-586.
Schimmel, S. C., P. R. Parrish, D. J. Hansen, J. M. Patrick, Jr., and J.
Forester. 1975. Endrin: Effects on several estuarine organisms. Pages
187-194 in Proc. 28th Ann. Conf. S.E. Assoc. Game Fish. Comm. 1974.
K
Schimmel, S. C., J. M. Patrick, Jr., and J. Forester. 1976. Heptachlor:
Toxicity to and uptake by several estuarine organisms. J. Toxicol.
Environ. Health 1:955-965.
Schimmel, S. C., J. M. Patrick, Jr., and J. Forester. Uptake and toxicity of
toxaphene in several estuarine organisms. Arch. Environ. Contam. Toxicol.
(in press).
137
-------
OIL AND MARINE ORGANISMS
by
0. G. Mironov
Institute of the Biology of Southern Oceans
Ukrainian SSR Academy of Sciences
Protecting the coean from pollution is now an urgent international
concern. All kinds of wastes originating in man's economic activity sooner or
later end up in seas and oceans. This changes seawater chemistry and upsets
ecological relationships in the ocean; eventually, this can disrupt the
ocean's productivity.
Biologically speaking, the most dangerous and currently the most com-
mon pollutant in seas and oceans is hydrocarbons, primarily petroleum and pe-
troleum products. From tanker cargos alone, each year several million tons of
petroleum and petroleum products are dumped into the oceans; this is compar-
able to the quantities of hydrocarbons that oceans produce in photosynthe-
sis. Petroleum prospecting and recovery in the continental shelves is going
on at a furious pace, leading to more pollution of the ocean. As oil tanker
traffic grows, as bigger tankers are built, the added danger of accidents de-
velops; if accidents happen, tens and hundreds of thousands of tons of petro-
leum products may end up in the ocean at the same time. Besides the direct
damage that these catastrophes inflict, large sums are spent in eliminating
the after-effects of tanker spills. The Torrey Canyon wreck off the British
coast in March 1967 was one such example.
Spilling into the ocean, petroleum can be swept many thousands of
miles from spill sites, gradually penetrate into the ocean depths, accumulate
in the seabed, and then surface again. So we see that petroleum acts on all
groups of marine organisms inhabiting both the surface film and the ocean
depths and the seabed.
Oil threatens primarily the early developmental stages of marine or-
ganisms. Larvae of numerous hydrobionts perish in ocean water containing
petroleum at concentrations of several milligrams per liter. Fish eggs are
particularly susceptible to petroleum. When petroleum is present at a con-
centration of 0.01 mg/liter, the number of nonviable larvae emerging from de-
veloping eggs grows by a factor of severalfold.
Petroleum has a damaging effect on marine organisms in short exposures
(minutes or hours), killing hydrobionts even if they are in clean seawater
after an exposure. Alterations in flora and fauna caused by pollution are
138
-------
well known. These rearrangements in aquatic communities are observed both be-
cause of pollutants continuously spilled into the seawater in small amounts
and in massive spills. In the first case, the changes are gradual and become
noticeable only after many years. This process can be slowed down by fluctua-
tions in the rate of pollution.
Typical of accidental oil spills is the disruption, and even the death,
of entire biocenoses in a short time, for example, in the wreck of tankers,
the rupture of underwater oil pipelines, and so on.
Noteworthy is the fact that petroleum's effects on the rearrangement
of marine communities often stretch beyond the immediate influence of the pol-
lutant; later, alterations in flora and fauna occur even when no hydrocarbons
are directly involved. Diesel fuel spills from the Japanese tanker Timpako
Maru along the California coast significantly cut into the populations of sea
urchins and mollusks feeding on algae. The profuse growth of the giant alga
Macrocystis ensued. When the sea urchin population was restored, the area oc-
cupied by the Macrocystis—the primary food of the sea urchin—was restored to
its original extent.
A system of observation and monitoring of oil contamination that re-
lies on marine organisms, that is, biological monitoring, can be important in
this respect.
By relying on a series of hydrobionts beloning to different systematic
groups, and biocenoses and communities of marine organisms, observations of
oil pollution can be made over practically all the world's oceans. And bio-
logical methods of global observation can and must be combined with several
biological tests available for individual bodies of water; this makes it pos-
sible to track the dynamics of pollution and various transformations of an or-
ganic mixture that is as complex as petroleum, as well as the principal petro-
leum products. Also, this system of biological monitoring permits the evalua-
tion of the role of marine hydrobionts in processes of the natxiral transforma-
tion of hydrocarbons, that is, self-cleaning.
Self-cleaning is a complex process: constituents of pollution break
up and participate in the overall turnover of matter and the transfer of energy
in the ocean. This process, originating in nature, existed long before man
began polluting his environment and even before man appeared on the globe.
However, we must emphasize that the ocean's ability to transform hydrocarbons
and other kinds of pollution is not boundless. Today numerous bodies of water
have already lost their ability to self-clean. In some gulfs and bays, petro-
leum has accumulated in large quantities in the seabed, transforming them into
virtual biological deserts.
Let us look at several examples of biological monitoring of petroleum
pollution. Systematic observations we began (Mironov, 1965, 1970, 1971, 1975)
since 1967 in studying the population, distribution, species composition, and
biochemistry of microorganisms capable of assimilating petroleum hydrocarbons
as the sole source of carbon and energy revealed the following:
1. A direct relationship was noted between the population of
petroleum-oxidizing microorganisms and the rate of petroleum pollution of
139
-------
seawater (Table 1). The largest number of cultures was found in the oil pol-
lution regions; and the population of bacteria feeding on petroleum was 10"-
10^ bacteria per liter of seawater.
2. In addition to the emphasis on population, there is species diver-
sity of microorganisms in the sites of continuous petroleum pollution. The
latter, in all likelihood, can be attributed to the great complexity of the
chemical composition of petroleum; its constituents can be consumed only by
certain microorganism species.
3. The relationship between population level and species diversity of
microorganisms on the one hand, and the rate of oil pollution, on the other,
enables us to examine petroleum-oxidizing microorganisms as indicators of oil
pollution.
The studies, covering all the main regions of the world ocean, allowed
us to place on the second session in the international program of the Joint
Study of the Mediterranean Sea (Dubrovniki, Yugoslavia, 1975), the draft
"Study of the Distribution and Population of Hydrocarbon-Oxidizing Microorga-
nisms in the Aquatic Environment."
TABLE 1. INDICATORS OF THE DISTRIBUTION OF MICROORGANISMS FEEDING ON
PETROLEUM IN THE MEDITERRANEAN SEA
Region
of
investigation
Northern region
Southern region
Total
Number
of
stations
27
27
54
Number of stations at
which growth of micro-
organisms feeding on
petroleum is observed
18
10
28
Number of
cultures
isolated
26
13
39
PROGRAM OF RESEARCH ON "STUDY OF THE DISTRIBUTION
AND POPULATION OF HYDROCARBON-OXIDIZING MICRO-
ORGANISMS IN THE AQUATIC ENVIRONMENT"
Hydrocarbon-oxidizing microorganisms are sensitive indicators of hy-
drocarbon pollution and have the decisive role in the biodegradation of hydro-
carbons in the ocean.
Joint and systematic pursuit of the investigation broadens our knowl-
edge of how far pollution extends in the Mediterranean Sea and allows us to
judge the potentiality of the aquatic environment for self-cleaning.
Experience with similar investigations by the Soviet Union in the
Black Sea and the Mediterranean Sea and some regions of the world ocean shows
that:
140
-------
1. Carrying out these investigations does not require large material
outlays (costs of the laboratory processing of material, including the pur-
chase of equipment, are about 5 million dollars for processing 100 samples and
about 1000 rubles for each subsequent 100 samples); the processing can be car-
ried out by all countries (or by most countries) that are participating in the
Joint Study of the Mediterranean Sea program (JSMS).
2. This program can be carried out both in full volume and by stages,
depending on the equipment available, with unified methods.
3. The present investigation, which should preferably start in 1975,
must later be converted into continual systematic observations, to grow in
volume year by year.
4. Carrying out these observations in one's national coastal waters
can favorably affect the progress of similar studies on an international
scale.
STAGE ONE: MINIMUM PROGRAM
1. Determination of the population (by the method of maximum dilu-
tions) of hydrocarbon-oxidizing microorganisms in mineral media with a single
source of carbon and energy (crude petroleum and simple phenol). Other hydro-
carbons can be used.
a. Only in the surface layer of the sea, 0-1 m
b. Surface layer of the sea plus standard levels
c. Surface layer of the sea plus standard levels plus seabed.
2. The test body of water can include, depending on the opportunities
open to the investigator:
a. Periodic profiles within his country's territorial waters
b. Observations at permanent coastal stations
c. Expeditionary studies in regions of the Mediterranean Sea
(ships in transit can be used)
d. Synchronous studies in bodies of water, and so on.
3. The periodicity of observations can fluctuate from daily to weekly,
monthly, and so on.
Present investigations can pass at once into the second stage of the
program, providing for the isolation from seawater (seabed) of the hydrocarbon-
oxidizing microorganisms and the determination of their hydrocarbon-oxidizing
activity. This makes it possible to conduct approximate calculations of the
self-cleaning activity of the oceans and to predict the extent of pollution of
the aquatic environment.
The program was approved by the countries participating in this con-
ference, and Bulgaria, Greece, Spain, and Romania declared their desire for
141
-------
early affiliation with it. Since this program was executed two months after
it was presented at the JSMS session, an expedition was undertaken to the
Mediterranean Sea aboard the scientific research vessel AKademik Kovalevskiy.
Carrying out these studies at present is of high interest, both because the
Suez Canal was recently opened and because oil shipments across the Mediter-
ranean began.
Besides the microbiological monitoring system in coastal waters, there
is much interest in observing the changes in biocenoses and individual groups
of marine hydrobionts under the effect of pollution. Here we must add the
reservation that in coastal bodies of water changes in the structure of the
marine biota in general are caused not by some single pollutant, for example,
petroleum, but by a combination action of a number of pollutants. In this re-
spect, to carry out monitoring of petroleum pollution, it is necessary to con-
duct observation of the level of accumulation of petroleum hydrocarbons in
aquatic organisms. Naturally, a convenient object here is the filter feeders.
For example, the bivalve mollusk mussel Mytilus can prove promising in this
respect; they are capable of carrying off quite high concentrations of petro-
leum in seawater. However, this system of observation at the present stage
cannot be recommended for broad global use (as microbiological monitoring)
owing to certain technical difficulties. These analyses are within the capa-
bility of well-equipped laboratories that have appropriate specialists on
their staffs.
In discussing observations of a particular pollutant, we must note
that complete information on the pollutant can be obtained from an integrated
study of the environment. As an example, we can point to multi-annual studies
we conducted (Mironov et al., 1975) in the Black Sea. The study of the popu-
lation, species composition, and biochemical features of petroleum-oxidizing
microorganisms in seawater and in the seabed, and physicochemical properties
and processes of the transformation of the organic matter in seabed deposits
helped to produce not only material on the present status of the body of
water, but also helped to evaluate the possibility of self-cleaning in a vast
region of the Black Sea. From our calculations, the self-cleaning potential-
ity of the coastal body of water with respect to petroleum (to depths of
100 m) embracing the region of the Soviet coast of the Black Sea is about 200
tons of petroleum a year.
Thus, in a system of biological monitoring of hydrocarbon pollution of
the ocean environment, one can suggest a method of determining the hydrocarbon-
oxidizing microorganisms that does not require large amounts of money and can
be effected on a global scale and in individual marine bodies of water.
In coastal bodies of water, a method of studying the dynamics of hy-
drocarbon content in marine hydrobionts can prove highly promising. This
method is applicable also in open regions of the sea in observations of acci-
dental oil spills.
142
-------
REFERENCES
Mironov, 0. G. 1969. Microorganisms feeding on hydrocarbons in the Black Sea.
Mikrobiologiya 28(4).
Mironov, 0. G. 1970. Role of petroleum-oxidizing microorganisms in self-
cleaning and indication of petroleum pollution in the sea.
Okeanologiya 5.
Mironov, O. G. 1970. Microorganism growing in oil and oil products in western
and central regions of the Mediterranean Sea. Rev. Int. Ocean. Med. 17.
Mironov, 0. G. 1971. Nefteokislyayushchiye Mikroorganizmy v More [Petroleum-
Oxidizing Microorganisms in the Sea], Naukov Dumka Press, Kiev.
Mironov, 0. G. 1975. Distribution of hydrocarbon-oxidizing microorganisms in
some seas. In Atti del 5° Colloquio Internationale di Oceanografia.
Medica Messina.
Mironov, O. G. , L. N. Kiryukhina, M. I. Kucherenko, and Z. P. Tarkhova. 1975.
Samoochishcheniye v Pribrezhnoy Akvatorii Chernogo Morya [Self-Cleaning in
the Coastal Region of the Black Sea]. Naukova Dumka Press, Kiev.
143
-------
IMPACT OF METALS ON THE MARINE ENVIRONMENT
by
John H. Martin
California State University, San Francisco
Moss Landing Marine Laboratories
Moss Landing, California 95039
ABSTRACT
Scientists have been unable to present evidence that heavy metals are
adversely affecting the inhabitants of the marine environment. Metal-organism
interactions are extremely complex and thus difficult to understand. The re-
sults of small-scale laboratory experiments have not been extrapolated to the
marine environment because the levels used in the laboratory generally have
been of higher magnitude than those occurring in nature. The findings of ele-
vated concentrations of heavy metals in marine organisms has led to the detec-
tion of hot spots. However, it has been impossible to discern whether the or-
ganisms were adversely affected by these metals.
The need for new methodology for evaluating the impact of heavy metals
has led to large-scale ecosystem experiments in which delicate marine orga-
nisms were maintained under near natural conditions. In initial experiments
with plankton, adverse metal effects were observed at concentrations approach-
ing ambient levels—e.g., 5 ppb Cu, 0.25 ppb Hg. These results, in turn, re-
emphasized the need to determine metal concentrations accurately in seawater.
Fortunately, the state-of-the-art for such measurements has constantly im-
proved, primarily because of sensitive new analytical techniques, the use of
clean collection, and analytical procedures. Thus, the continued development
and usage of large-scale ecosystem experiments in conjunction with accurate
measurements of heavy metal levels in seawater should enable scientists to
assess the impact of heavy metals on the world's oceans.
INTRODUCTION
The toxicity of heavy metals has been the subject of a vast amount of
research, primarily because of their threat to human health. For example,
several disastrous cases of mass poisonings by mercury compounds have occurred
in recent years. Between 1953 and 1960, 121 cases of Hg poisoning were re-
corded in Minamata, Japan; 46 people died. In Iraq mass Hg-poisoning epi-
demics occurred in 1956 (100 cases, 14 deaths); in 1960 (1,000 cases), and in
1972 (6,530 cases, 459 deaths) (Eyl, 1971; Bakir et al., 1973).
144
-------
Several other heavy metals are also toxic enough to represent human
health hazards. Cadmium-induced decalcification of the skeleton (Itai Itai
Disease) claimed 56 lives in Japan (Nilsson, 1970). Human lead poisoning has
been known since antiquity (Patterson, 1965).
In addition to the threat to mankind, agricultural scientists have
been concerned as to how the toxicity of certain trace elements affects valu-
able domestic animals. This is best exemplified by selenium, an element whose
toxic effects were apparently first described in 1295 by Marco Polo during his
travels through China (Trelease and Beath, 1949). This element is concentrated
by certain plant species, and herbivores grazing on these plants exhibit symp-
toms commonly called "alkali disease" and "blind staggers." These conditions
can lead to mass mortalities; in 1906 and 1908, 15,000 sheep died of selenium
poisoning in the state of Wyoming (Trelease and Beath, 1949).
Mass mortalities caused by heavy metals have also been observed in the
marine environment. Approximately 2,000 abalone (Haliotis rufescens, H. cra-
cherodii) died from copper poisoning when a power plant cooling system was
tested (Martin et al., 1975b). The copper leached out of copper-nickel tubing
into seawater that had stood in the system for several weeks. When the water
was released, it contained approximately 2,000 ppb Cu, a concentration that
resulted in the deaths of almost all abalone in the immediate discharge area.
However, in comparison to the terrestrial environment, such instances
rarely have been reported in the marine environment. Neverthless, research
efforts on metals in the sea increased in recent years because of a series of
interrelated events. The well-publicized Minamata Bay disaster was caused by
the consumption of marine food items contaminated with large amounts of methyl
mercury. A few years later, analytical capabilities for tracing elements were
advanced by the development of a very sensitive technique for detecting Hg
(Hatch and Ott, 1968). This breakthrough led to the discovery that important
marine food items, such as tuna and swordfish,.had relatively large quantities
of Hg. Consequently, man became increasingly concerned not only with the
quality of food items from the sea, but also with the quality of the marine
ecosystem itself. The combination of these events resulted in increased heavy-
metals research in the marine environment.
PAST RESEARCH EFFORTS
Previous research efforts can be divided into two areas: (1) labora-
tory experiments and (2) measurements of heavy metals in organisms inhabiting
various parts of the marine ecosystem. A comprehensive review of the labora-
tory research on the effects of heavy metals is not feasible in a paper of
this length. Further, it is unwarranted because excellent reviews on this
subject (e.g., Bryan, 1971) and comprehensive bibliographies (i.e., Eisler,
1973, 1975) have been published. However, it is generally apparent that labo-
ratory experiments have been extremely valuable in demonstrating that heavy
metal toxicity in the marine environment is a very complex phenomenon. For
example, Bryan (1971, Table 4) lists as factors influencing the toxicity of
heavy metals: the form of the metal in water (soluble vs. particulate); the
presence of other metals or poisons (antagonistic, additive and synergistic
145
-------
effects); the physiology of the organism (salinity, temperature, dissolved oxy-
gen, pH, etc.); and the condition of the organism (stage in life history, its
size, its activity, acclimatization). It is not surprising, therefore, that
laboratory workers often have been compelled to add relatively large quanti-
ties of toxic elements to observe and determine various effects in the labora-
tory. Nevertheless, this practice drew general criticism that concentrations
necessary for producing measurable effects are of a higher magnitude than
those encountered in all but the most heavily polluted environments. The fail-
ure to detect effects at low levels has prompted some scientists to argue that
metals are not a problem in the marine environment. Others contend that, be-
cause of the complexities, the effects of metals in the marine environment are
not adequately understood. The latter position was considered especially ap-
plicable for delicate organisms that cannot survive under ordinary laboratory
conditions. Thus, it became apparent that new approaches were needed to as-
sess the full impact of metals in the oceans (see below).
In addition to the experimental area, considerable research also has
been conducted to determine the levels of various toxic elements in marine or-
ganisms. These studies led to the detection of hot spots—areas in which en-
vironmental levels are undoubtedly high. They were also useful in determining
the occurrence of food chain amplification processes that might ultimately
threaten man as well as other top carnivores. Although these data reveal noth-
ing about whether the animal assimilating concentrations was adversely affected
by the element, it appears that most scientists erroneously assume that the
higher the levels in an organism, the more likely the chances of adverse ef-
fects. Although this assumption is logical, there is growing evidence that it
is a fallacy. For example, Friberg et al. (1974, p. 107) state: "When a per-
son suffers severe renal damage through the toxic action of cadmium, his kid-
ney concentration of cadmium will decrease considerably. Thus he will have
lower levels in his kidneys than a person with only slight renal disturbances."
In discussing their experiments with a marine gastropod, Betzer and Yevich
(1975, p. 24) note: "In Busycon, experiments using 64-Cu show that under nor-
mal concentrations, copper is taken up and transferred to the internal tissues;
at toxic concentrations, where only the gills (and osphradium) show tissue
damage, the rate of transfer into the internal tissues, particularly the di-
gestive gland, is sharply decreased." Betzer and Yevich also refer to Yager
and Harry (1964), who found the livers of distressed fresh water snails,
Taphius glabratus, had lower copper concentrations than those of normal snails.
These few examples suggest that high concentrations of heavy metals in certain
organs may indicate that the organisms are in good health and that their de-
toxification mechanisms are functioning properly. Obviously, a technique in-
volving more than high concentrations is needed to detect metal damage to the
environment.
Thus, while using both experimental and environmental approaches, sci-
entists have been hard pressed to illustrate that heavy metals are adversely
affecting the marine ecosystem.
146
-------
PRESENT RESEARCH
As mentioned above, small-scale laboratory experiments were valuable
in demonstrating the complex nature of toxic element-organism interrelation-
ships in the marine environment. They also indicated the need for large-scale
total ecosystem experiments, in which delicate organisms could be maintained
and studied when subjected to realistic amounts of various pollutants. Plank-
tonic organisms were likely choices for such experiments since the phytoplank-
ton are the most important organisms in the marine environment. They are ex-
tremely sensitive and thus are likely to be affected by microchanges in their
chemical environment.
For these reasons, the controlled ecosystem pollution experiment
(QEPEX) program was undertaken under the auspices of the National Science
Foundation's International Decade of Ocean Exploration Program (NSF-IDOE).
Initial experiments have been performed in the Saanich Inlet off Vancouver Is-
land, British Columbia (48°39'N, 123°28'W). Approximately 70 tons of natural
water are isolated in control and experimental polyethylene bags (diameter
2.5 m; length 15 m). The bags are cylindrical in shape except for the lower
one-fifth which is conical and tapers to a removable sediment trap at the base.
Initially, bagged plankton populations were compared with those living in the
waters surrounding the bags. The results suggest that the inside and outside
populations and their environments were essentially the same (Takahashi et al.,
1975). Thus, populations under pollutant stress situations can be compared
meaningfully with control populations.
One of the first pollutants to be tested was copper. The choice was
logical since certain aquatic plant species already had responded remarkably
to small amounts of this element. For example, inhibition of Chlorella growth
and photosynthesis was observed at concentrations of one ppb Cu; a reduction
in photosynthesis was also observed when fresh water diatom Nitzschia palea
was subjected to this level. This research was performed by Steemann Nielsen
and his co-workers Kamp-Nielsen and Wium-Andersen. In addition to the find-
ings above, these authors also have discussed the importance of copper in up-
welling systems aud have noted its possible effects on primary productivity
(Steemann Nielsen et al., 1969; Steemann Nielsen and Kamp-Nielsen, 1970, Stee-
mann Nielsen and Wium-Andersen, 1970, 1971, 1972).
The initial CEPEX copper experiemtns demonstrated the value of the
ecosystem approach since several components of the ecosystem could be observed
simultaneously when 10 and 50 ppb Cu were added to experimental bags. The
following events occurred: The number and activity of bacterial heterotrophs
increased markedly, apparently in response to increases in carbon from other
copper-stressed components of the ecosystem (Vaccaro et al., 1975); phyto-
plankton crops and photosynthesis were inhibited by copper; and the excretion
of ^C-labeled organic matter increased (Thomas et al., 1975). The species
composition of the phytoplankton was also affected: Populations of the cen-
trate diatom Chaetoceros sp. declined and were replaced by microflagellates
and Cu-insensitive diatom species such as Nitzschia delicatissima and Navicula
distans (Thomas and Seibert, 1975). Phytoplankton nitrate uptake and the syn-
thesis of nitrate reductase were inhibited. Cell disruption and loss of ac-
cumulated ammonium was observed in Noctiluca sp. (Harrison et al., 1975). The
147
-------
copper caused decreases in the zooplankton standing crop. Larvaceans appeared
to be the most sensitive of the organisms present during the experiments (Gib-
son et al., 1975). Other interesting effects also were noted in the zooplank-
ton population. For example, copepod fecal pellet production dropped markedly
as copper concentrations were increased (Reeve, 1975). An interesting inter-
action between zooplankton predators and their prey also was noted. Cteno-
phores and coelenterates declined in abundance in the experimental bags, while
those in the control bags increased. Consequently the abundance of prey spe-
cies declined markedly. These interactions made it difficult to evaluate the
effect of Cu on the zooplankton (Gibson et al., 1975). Decreases in micro-
plankton were also observed. Copepod nauplii were affected at concentrations
as low as 5 ppb (Beers, 1975).
Although not yet reported, further CEPEX experiments have been con-
ducted with lower concentrations of Cu (5 ppb). Similar results were obtained.
Many identical changes also were observed when concentrations of 0.25 and 1.0
ppb Hg were used (CEPEX Newsletter, 1975) . The CEPEX experiments thus demon-
strate the value of the large ecosystem approach revealing that subtle, often
sublethal effects occur when metals are added to an ecosystem. Clearly more
large-scale ecosystem experiments are needed; and it is encouraging to note
that the construction of another project is now underway at the Environmental
Research Laboratory, Narragansett, Rhode Island, under the auspices of the En-
vironmental Protection Agency.
THE NEED FOR TRACE METAL DATA IN SEAWATER
The effects of heavy metals in the marine environment will never be
adequately understood until accurate measurements of heavy metals in seawater
are obtained: For example, if laboratory experiments indicate an effect at a
concentration of one ppb, it then becomes necessary to define areas where this
concentration is exceeded. These data also are needed to determine transport
pathways, residence times, mixing rates, etc. In recent years, the state-of-
the-art in water chemistry has constantly improved. As a result, metal con-
centrations in seawater are now believed to be much lower than originally es-
timated. For example, copper values for open-ocean seawater formerly were be-
lieved to be approximately 2-3 yg/1. However, in a recent paper, Boyle and
Edmond (1975) observed a range of 0.06-0.21 yg Cu/1 of Antarctic surface water.
Further, they reported a strong correlation between Cu and nitrate. If Cu is
correlated with nitrate throughout the world's oceans, open-ocean surface
waters depleted by nitrate should have less than 10 ng Cu/1. The maximum Cu
concentration in deep ocean water should be 300 ng/1. Martin et al. (1976)
observed a similar correlation between Cd and nitrate and phosphate. Their
Cd value for open-ocean surface water was 0.005 yg/1, a value 20 times lower
than the generally accepted level of 0.1 ppb.
These low values were obtained by using clean sampling and laboratory
techniques. Perhaps sampling is the greatest source of contamination. Large
quantities of trace metals are constantly sloughed off all oceanographic re-
search vessels; the hydrowire attached to samplers is often filthy, and even
the best commercial samplers contain metal parts or are constructed of mate-
rials known for high metal content. These samples are difficult to clean
148
-------
properly at sea, if indeed they are cleaned at all between stations. Because
of contamination problems, scientists have devised new sampling techniques.
Boyle and Edmond collected their samples by lowering a carefully cleaned poly-
ethylene bottle from the bow of a research vessel. The bottle was suspended
on a polypropylene line. Martin et al. (1976) collected samples by rowing a
raft away from the research vessel and then hand-holding a carefully cleaned
polyethylene bottle beneath the sea surface. They also employed a specially
constructed metal-free pump system to collect water samples as deep as 95 m.
Obviously, these methods are not suited for collecting deep-water samples. A
clean reliable deep-water sampler has yet to be built.
Water samples also require special handling in the laboratory. C. C.
Patterson of the California Institute of Technology has led attempts to per-
suade scientists to follow clean collection and laboratory procedures for trace
elements, especially lead. He has long maintained that estimates of Pb in
ocean water were exaggerated by several orders of magnitude because of con-
tamination during the collection, transport, and analyses of these samples.
Patterson convinced many scientists of the need for special procedures during
a lead-in-seawater-intercalibration workshop (see Meeting Report, 1974).
Water collected from-the surface of the heavily polluted Southern California
Bight contained only 0.014 ^g Pb/kg. Other laboratories participating in this
workshop reported concentrations higher than this value. The excess lead was
thought to be introduced during analytical processes in which unpurified re-
agents were used in unclean facilities. Patterson has proven his point. Ma-
rine chemists should adopt his procedures whenever feasible unless they can
demonstrate that such stringent requirements may be safely omitted in certain
situations for certain elements. In any event, scientists involved in trace-
element seawater chemistry are urged to read Patterson and Settle's (1976)
article before beginning or continuing this kind of research.
In conjunction with seawater data, anthropogenic and natural input
rates must be obtained for^heavy metals in the world's oceans. These data are
presently scarce, but steps are being taken in this direction. For example,
Duce and his co-workers in the NSF-IDOE Pollutant Transport program are pro-
viding a great deal of information about the fluxes of metals to the world1s
oceans via atmospheric fallout (e.g., Duce et al., 1975; Hoffman et al., 1974;
Wallace and Duce, 1975). Estimates of total input are available for at least
one geographical arec.. Young and his co-workers provided input rates for sev-
eral heavy metals via wastewater discharge, surface runoff, vessel coating,
ocean dumping, rainfall, and advective transport into the Southern California
Bight (SCCWRP, 1973, 1974, 1975). Such vital studies are needed in other
world population centers that border the marine environment.
WHO KILLED COCK ROBIN?
Scientists involved in environmental pollution research are often re-
quired to perform detective work. For example, mortalities or deleterious
effects are observed for a group of organisms, and the question arises: Was
an environmental pollutant responsible, and if so, which one? Answers to
r;uch questions are very elusive because complexities are almost always en-
countered. E. D. Goldberg of the Scripps Institution of Oceanography has
149
-------
termed this phenomenon the "who-killed-cock-robin syndrome."
This situation is exemplified by studies of premature pupping in the
California sea lions (Zalophus californianus). Large numbers of premature
pups have been counted on the sea lion rookeries since 1968 (see Gilmartin et
al. , 1975; Odell, 1970). Research aimed at determining the causes for this
development have revealed: (1) The mothers of premature pups are usually only
6-8 years old, while the mothers of full-term pups are at least 10 years old
(Gilmartin et al., 1975); (2) many of the abnormal mothers are infected with
Leptospirosis, a virus known to cause abortions (Vedros et al., 1971; Gilmar-
tin et al., 1975); (3) the abnormal mothers have significantly higher amounts
of polychlorinated biphenyls (PCB's) and DDT compounds (DeLong et al., 1973;
Gilmartin et al., 1975); and (4) the normal mothers have significantly higher
amounts of mercury, selenium, and bromine (Martin et al., 1975a). The latter
findings were of interest because each normal mother had equimolar amounts of
Se and Hg in their livers; and, in addition, excess or equimolar amounts of Br
were found in conjunction with these elements. In contrast, the mothers of
premature pups had equimolar amounts of Se and Hg; however, their Br levels
were severely depressed. Perhaps these findings indicate that Br also is in-
volved in the Hg-Se detoxification mechanisms (see Parizek et al., 1971) al-
though for some reason it was not functioning in the abnormal mothers. Whether
it was responsible for the premature pupping is unknown. However, these re-
sults suggest that absolute amounts of elements are not as important as the
relationship of elements.
In addition to demonstrating the complexities involved with environ-
mental detective work, the four factors mentioned above also point to the de-
sirability of simultaneous measurement of different pollutant classes as well
as natural factors within the same samples. Erroneous conclusions can be
reached when only one pollutant is measured. As the Se-Hg interaction indi-
cates, this risk is especially true for one heavy metal.
From the limited overview presented above, we can conclude that the
impact of heavy metals in the marine environment is exceedingly difficult to
understand. Nevertheless, recent developments in large-scale experimental
methods and in analytical capabilities provide us with powerful tools that
will lead to the desired understanding. In the meantime, no one can justifi-
ably state that metals are or are not a problem in the world's oceans.
REFERENCES
Bakir, F., S. F. Damlusi, L. Amin-Zaki, M. Murtadha, A. Khalidi, N. Y. Al-Rawi,
S. Tikriti, and H. E. Dhahir. 1973. Methylmercury poisoning in Iraq.
Science 181:230-241.
Beers, J. R. 1975. Controlled Ecosystem Pollution Experiment: Microzooplank-
ton in the controlled environment pollution experiment. Jn Controlled
Ecosystem Pollution Experiment (CEPEX) Annual Report to the National
Science Foundation's Office of the International Decade of Ocean Explora-
tion (NSF/IDOE). 44 pp.
150
-------
Betzer, S. B., and P. P. Yevich. 1975. Copper toxicity in Busycon canalicu-
latum L. Biol. Bull. 148:16-25.
Boyle, E., and J. M. Edmond. 1975. Copper in surface waters south of New
Zealand. Nature 253:107-109.
Bryan, G. W. 1971. The effects of heavy metals (other than mercury) on ma-
rine and estuarine organisms. Proc. Royal Soc. London, Ser. B.,
177:389-410.
CEPEX Newsletter. 1975. Mercury experiment: Preliminary results from 1/4
scale CEEs, May 28-July 8. Controlled Ecosystem Pollution Experiment,
NSP/IDOE, Newsletter No. 14, July 29, 1975.
DeLong, R. L., W. G. Gilmartin, and J. G. Simpson. 1973. Premature births in
California sea lions: association with high organochlorine pollutant
residue levels. Science 181:1168-1169.
Duce, R. A., G. L. Hoffman, and W. H. Zoller. 1975. Atmospheric trace metals
at remote northern and southern hemisphere sites: pollution or natural?
Science 187:59-61.
Eisler, R. 1973. Annotated Bibliography on Biological Effects of Metals in
Aquatic Environments. U.S. Environmental Protection Agency, Corvallis,
Oregon. 287 pp.
Eisler, R. 1975. Annotated Bibliography on Biological Effects of Metals in
Aquatic Environments, Vol. II. U.S. Environmental Protection Agency, Cor-
vallis, Oregon.
Eyl, T. B. 1971. Organic-mercury food poisoning. New England J. Med.
284:706-709.
Friberg, L., M. Piscator, G. F. Nordberg, and T. Kjellstrom. 1974. Cadmium
in the environment. CRC Press, Cleveland, Ohio. 248 pp.
Gibson, V. R., T. J. Lawson, and G. D. Grice. 1975. Controlled Ecosystem
Experiment: the effect of copper on zooplankton during the June 1974 ex-
periment. In CEPEX Annual Report to NSF/IDOE. 13 pp.
Gilmartin, W. G., R. L. DeLong, A. W. Smith, J. C. Sweeney, B. W. deLappe,
R. W. Risebrough, L. A. Griner, M. D. Dailey, and D. B. Peakall. 1975.
Premature parturition in the California sea lion. J. Wildl. Dis. (in
press).
Harrison, W. H., R. W. Eppley, and E. H. Renger. 1975. Phytoplankton nitro-
gen metabolism, nitrogen budgets and observations on copper toxicity,
CEPEX enclosures. Jn CEPEX Annual Report to NSF/IDOE. 38 pp.
Hatch, W. R., and W. L. Ott. 1968. Determination of sub-microgram quantities
of mercury by atomic absorption spectrophotometry. Anal. Chem 40:2085-
2087.
151
-------
Hoffman, G. L., R. A. Duce, P. R. Walsh, E. J. Hoffman, J. L. Fasching, and
B. J. Ray. 1974. Residence time of some particulate trace metals in the
oceanic surface microlayer: significance of atmospheric deposition. J.
de Recherches Atmospheriques 8:745-759.
Martin, J. H., K. W. Bruland, and W. W. Broenkow. 1976. Cadmium transport in
the California Current. Pages 159-184 in H. L. Windom and R. A. Duce,
eds., Marine Pollutant Transfer. D. C. Heath, Lexington, Mass.
Martin, J. H., P. D. Elliott, V. C. Anderlini, D. Girvin, S. A. Jacobs, R. W.
Risebrough, R. L. DeLong, and W. G. Gilmartin. 1975a. Mercury-selenium-
bromine imbalance in premature parturient California sea lions. (Submit-
ted to Marine Biology.)
Martin, M., M. D. Stephenson, and J. H. Martin. 1975b. Copper toxicity ex-
periments in relation to abalone deaths observed in a power plant's cool-
ing waters. (In preparation.)
Meeting Report. 1974. Interlaboratory lead analyses of standardized samples
of seawater. Mar. Chem. 2:69-84.
Nilsson, R. 1970. Aspects on the toxicity of cadmium and its compounds.
Swedish Natural Science Research Council, Ecological Research Committee
Bull. No. 7.
Odell, D. K. 1970. Premature pupping in the California sea lion. Pages 185-
190 in Proc., 7th Annual Biosonar and Diving Mammal Conf., Stanford Re-
search Institute, Menlo Park, Calif.
Parizek, J., I. Ostadalova, J. Kalouskova, A. Babicky, and J. Benes. 1971.
The detoxifying effects of selenium: interrelations between compounds of
selenium and certain metals. Pages 85-122 in W. Merty and W. E. Cornatzer,
eds., Newer Trace Elements in Nutrition. Marcel Dekker, Inc., New York.
Patterson, C. C. 1965. Contaminated and natural lead environments of man.
Arch. Environ. Health 11:344-360.
Patterson, C. C., and D. M. Settle. 1976. The reduction of orders of magni-
tude errors in lead analyses of biological materials and natural waters by
evaluating and controlling the extent and sources of industrial lead con-
tamination introduced during sample collecting and analysis. To be pub-
lished in P. LeFleur, ed., Accuracy in Trace Analysis. Proc., 7th Mate-
rials Research Symp., National Bureau of Standards Special Publication.
Reeve, M. R. 1975. Controlled Ecosystem Pollution Experiment: the effect of
copper on zooplankton metabolism. In CEPEX Annual Report to NSF/IDOE.
17 pp.
SCCWRP. 1973. The ecology of the Southern California Bight: Implications
for water quality management. Southern California Coastal Water Research
Project, El Segundo, Calif. 531 pp.
152
-------
SCCWRP. 1974. Coastal water research project annual report—1974. Southern
California Coastal Water Research Project, El Segundo, Calif. 197 pp.
SCCWKP. 1975. Coastal water research project annual report—1975. Southern
California Coastal Water Research Project, El Segundo, Calif. 211 pp.
Steemann-Nielsen, E., L. Kamp-Nielsen, and S. Wium-Andersen. 1969. The ef-
fects of deleterious concentrations of copper on the photosynthesis of
Chlorella pyrenoidosa. Physiol. Plant. 22:1121-1133.
Steemann-Nielsen, E., and L. Kamp-Nielsen. 1970. Influence of deleterious
concentrations of copper on the growth of Chlorella pyrenoidosa. Physiol.
Plant. 23:828-840.
Steemann-Nielsen, E., and S. Wium-Andersen. 1970. Copper ions as poison in
the sea and in freshwater. Mar. Biol. 6:93-97.
Steemann-Nielsen, E., and S. Wium-Andersen. 1971. The influence of Cu on
photosynthesis and growth in diatoms. Physiol. Plant. 24:480-484.
Steemann-Nielsen, E., and S. Wium-Andersen. 1972. Influence of copper on
photosynthesis of diatoms, with special reference to an afternoon depres-
sion. Verh. Internat. Verein. Limnol. 18:78-83.
Takahashi, M., W. H. Thomas, D. L. R. Seibert, J. Beers, P. Koeller, and T. R.
Parsons. 1975. The replication of biological events in enclosed water
columns. In CEPEX Annual Report to NSF/IDOE. 29 pp.
Thomas, W. H., 0. Holm-Hansen, D. L. R. Seibert, F. Azam, R. Hodson, and M.
Takahashi. 1975. Effects of copper on enclosed marine plankton communi-
ties: I. Phytoplankton standing crop and productivity. In CEPEX Annual
Report to NSF/IDOE. 21 pp.
Thomas, W. H., and D. L. R. Seibert. 1975. Effects of copper on marine
plankton communities: II. Taxonomic composition of the phytopiankton.
In CEPEX Annual Report to NSF/IDOE. 27 pp.
Trelease, S. F., and O. A. Beath. 1949. Selenium. Trelease and Beath, New
York. 292 pp.
Vaccaro, R. F., F. Azam, and R. E. Hodson. 1975. Response of natural marine
bacterial populations to copper: a controlled ecosystem pollution experi-
ment. In CEPEX Annual Report to NSF/IDOE. 15 pp.
Vedros, N. A., A. W. Smith, J. Schonewald, G. Migaki, and R. C. Hubbard.
1971. Leptospirosis epizootic among California sea lions. Science 172:
1250-1251.
Wallace, G. T., and R. A. Duce. 1975. Concentrations of particulate trace
metals and particulate organic carbon in marine surface waters by a
bubble flotation mechanism. Mar. Chem. 3:157-181.
153
-------
Yager, C. M., and H. W. Harry. 1964. The uptake of radioactive zinc, cad-
mium, and copper by the freshwater snail, Taphius glabratus. Malacologia
1:339-353.
154
-------
APPENDIX
MICROCOSMS AS BIOLOGICAL INDICATORS OF POLLUTION
by
Frank G. Wilkes
U.S. Environmental Protection Agency
Environmental Research Laboratory
Gulf Breeze, Florida 32561
ABSTRACT
•
Microcosms are one method of investigating specific origins, flows,
fates, and/or effects of materials in the environment. The EPA Gulf Breeze
Laboratory conducts and supports research to develop microcosms of many types
and complexities. These microcosms are intended to be simple, easy to apply,
and are designed to investigate specific processes or categories of processes
in the estuarine environment. The objective of this research is to develop a
number of methods to investigate pollutant fate and effects in estuaries. The
results of the individual tests are combined to form a description of the en-
tire system. The ecosystem compartments under investigation include direct
accumulation from water and food by organisms at all trophic levels, bioaccu-
mulation through food chains, direct effects of pollutants on organisms, i.e.,
mortality, reproduction and behavior and indirect effects of sublethal levels
of pollutants such as changes in predator-prey relationships. Microbial pro-
cesses at both air-water and sediment-water interfaces are investigated as
well as physical and chemical transformations.
Specific tests under development include:
1. Predator/Prey Effects Test in which prey selection and the ability
of the prey to avoid predation is affected by the pollutant.
2. Lugworm Benthic System in which pollutants are accumulated by these
important benthic organisms and mobilized into the soil through their activity.
3. Model Salt Marsh Ecosystems in which pollutant effects are deter-
mined on microcosms which simulate seasonal and tidal conditions.
4. Microbial microcosms in which the degradation of pollutants by
naturally occurring microbial populations is investigated as well as the ef-
fect of the pollutant on population diversity and composition.
5. A "Slow-Flow" System in which the fate of pollutants is determined
in a sealed, continuous air and water flow, aquarium containing a representa-
tive estuarine community.
155
-------
6. A Gradient Controlled System in which the avoidance and/or prefer-
ence of estuarine organisms to pollutant gradients is determined.
7. A Food Chain System in which the bioaccumulation of pollutants
through typical estuarine food chains is determined.
REPORT OF THE JOINT AMERICAN-SOVIET EXPEDITION
ON THE SCIENTIFIC RESEARCH VESSEL "MOSCOW UNIVERSITY"
FOR THE FIRST STAGE OF INTERCALIBRATION OF METHODS
FOR THE HYDROBIOLOGICAL ANALYSIS OF AQUATIC ECOSYSTEMS
by
Richard L. Iverson
Department of Oceanography
Florida State University
Konstantin C. Burdin
Victor Maximov
Michal Lymin
Biological Faculty
Hydrobiological Department
M. V. Lomonosov Moscow State University
INTRODUCTION
In accordance with the American-Soviet Environmental Agreement within
the framework of Project VI-2.1, "Effect of Pollutants on Marine Organisms,"
a protocol was signed expressing the desire to conduct an intercalibration of
hydrobiological methods for the analysis of aquatic ecosystems. The first
stage of the intercalibration was a comparison of American and Soviet methods
for measuring phytoplankton productivity which was conducted aboard the Re-
search Vessel Moscow University in the western Atlantic Ocean and eastern Gulf
of Mexico during summer, 1975.
Previous international intercalibration of the carbon-14 phytoplankton
primary productivity method devised by Steemann-Nielsen (1952) had revealed
differences in productivity data obtained in the Indian Ocean by American,
Soviet, Japanese, and Australian investigators (Doty et al., 1965). The dif-
ferences were related to variations in choice of sampling depth, in-situ incu-
bation apparatus, standardization of carbon-14 stock solutions, and in the de-
termination of Geiger counter efficiencies. Since we were aware of the re-
sults of the Indian Ocean phytoplankton productivity intercalibration, and
since new methods have been developed for processing carbon-14 labeled phyto-
plankton material since 1965, attention was focused in the intercalibration
reported here on effects of differences in in-situ incubation gear, filters
and filtration equipment, and liquid scintillation processing of samples.
Liquid scintillation methods have been developed to the point where they are
156
-------
in wide use for estimation of phytoplankton productivity (Wolfe and Schleske,
1967). Liquid scintillation counting of carbon-14 labeled phytoplankton can
be accurately accomplished if absorption by phytoplankton is assessed and cor-
rected for (Pugh, 1970).
SAMPLING, SAMPLE PREPARATION,
AND INCUBATION OF SAMPLES
Water samples were collected with well-aged 5-liter Niskin water bot-
tles from 5 arbitrarily chosen depths at each of 10 stations in the western
Atlantic Ocean, Florida Straits, and eastern Gulf of Mexico. Only light bot-
tles were used for this intercalibration since Morris et al. (1971) suggested
there is little justification for dark bottle corrections to light bottle
carbon-14 fixation in the open ocean. The American method used three 180 ml
glass bottles which were filled with water from a Niskin bottle for each sam-
ple depth. Two milliliters of solution containing 1 x 107 disintegrations per
minute (DPM) of carbon-14 as NaHCOa were then added to each bottle with a Corn-
wall automatic syringe. The bottles were capped, shaken, and placed within a
clear plastic tube attached vertically to the hydrographic line with a clamp.
The samples were lowered into the sea and incubated for periods of about 3
hours at the depths where they were collected. After the incubation period,
the samples were returned to the deck of the ship where one ml of a 2% solu-
tion of HgCla was added to each bottle to kill the phytoplankton. Care was
taken to avoid exposure of water samples to direct sunlight while the bottles
were filled and before phytoplankton were killed.
Soviet productivity bottles were about 125 ml in volume and were filled
alternately and in triplicate from the same Niskin bottle from which corre-
sponding American samples were obtained. A carbon-14 stock solution supplied
by the Americans was used by both the Soviets and Americans in order to remove
carbon-14 standardization problems from consideration in the intercalibration.
Soviet bottles were suspended on a horizontal platform made of Plexiglas which
was attached to the hydrographic wire at each sample depth. Soviet and Ameri-
can suspension devices were placed in the top position on the hydrographic
line in alternate experimental incubations. Soviet samples were kept in the
shade after incubation until formalin was added to kill the phytoplankton.
The titration alkalinity method given in Strickland and Parsons (1972)
was performed for samples obtained from each Niskin bottle by American and
Soviet scientists using American equipment.
SAMPLE PROCESSING
Poisoned American productivity samples were filtered through 24 mm
diameter Whatman GF/C glass fiber scintillation grade filters under low vacuum.
The filters were exposed to HCl fumes for 10 seconds before being frozen for
transport to shore. On shore, samples were thawed and placed in liquid scin-
tillation vials to which 10 ml of Aquasol® (New England Nuclear) were added.
Radiocarbon activity was measured with a Picker liquid scintillation spectrome-
ter using the channels ratio method of standardization. A channels ratio
curve was prepared using commercial quenched toluene carbon-14 standards and
157
-------
also using the phytoplankton biomass method of Pugh (1970). The Pugh curve
was prepared using phytoplankton from a series of different volumes of water
collected in the Atlantic Ocean. Quantities of phytoplankton were so low that
little quenching was observed. The carbon-14 stock solution was standardized
by adding aliquots of a 1:1000 dilution of the 2-ml radiocarbon "spike" to
Aquasol® containing one ml of phenethylamine as given by Iverson et al. (1976).
The Soviets used "Synpor" (Czechoslovakia) membrane filters with an
effective pore size of 0.3 microns. After filtration, samples were held in
HC1 fumes for several minutes before freezing. Filters were kept frozen before
analysis in Moscow where filters were thawed, desiccated, and placed in liquid
scintillation vials together with 4 ml of scintillation solution which con-
sisted of 4 g PPO and 0.1 g POPOP in 1.0 liter of toluene. Radioactivity of
the samples was measured with a Nuclear Chicago Mark II liquid scintillation
spectrometer (USA) equipped with an Algotronic computer (Diehl, Germany). The
method of Pugh (1970) was used to prepare a standard channels ratio curve.
Since phytoplankton on filters constitutes a heterogeneous system, the
filters were removed from the scintillation solution and dried before solubi-
lization to prepare a homogeneous system. The scintillation solution was re-
counted to measure radioactivity which had eluted from each filter. The dried
filters were placed in clean scintillation vials with 0.15 ml of distilled
water. After allowing the filters to soak for a short time period, 1.0 ml of
NCS solubilizer (Amersham, Searle, USA) was added to each vial. Filters were
allowed to remain in the NCS solution for 30 minutes after which 0.03 ml of
glacial acetic acid was added to each sample to lower chemiluminescence. The
samples were cooled, 15 ml of scintillation solution was added to each vial,
and the vial contents were carefully mixed. The samples were held in the dark
at 4°C for 12 hours before measurement of radioactivity of the homogeneous so-
lution. For standardization, a quench curve for a solution with a known spe-
cific activity was prepared by adding carbon-14 ethanol to filters which con-
tained labeled phytoplankton. The filters were solubilized by the method
given above. The activity of the carbon-14 ethanol solution was determined
with a standard quench curve obtained with carbon-14 quenched standards (Amer-
sham, England).
RESULTS AND DISCUSSION
There did not appear to be consistent differences in carbon-14 ac-
tivity for samples processed with American and Soviet methods except for Sta-
tion 9 where Synpor filters were used by both sides (Table 1). American ac-
tivity was consistently lower than Soviet activity at all depths for Station 9
suggesting that it is necessary to solubilize the Synpor membrane filters in
order to minimize radioisotope absorption problems. Anomalously high activity
values (for example, Station 3, Soviet, and Station 7, American) were usually
characterized by high standard deviations which reflected the effects of high
activity values for one or two of the three replicate samples. These high
activity values are unexplained but may be the result of uneven distribution
of phytoplankton between replicates or the result of problems in correcting
for chemiluminescence. The mean coefficient of variation for American carbon-
14 activity values was 21 (range 2 to 123) while the mean coefficient of varia-
tion for Soviet carbon-14 activity values was 29 (range 0 to 145).
158
-------
TABLE 1. CARBON-14 ACTIVITY AND PHYTOPLANKTON PRODUCTIVITY VALUES
Station Information
Incubation
Period (hr)
#1
07/28/75
40 53N
71 27W
2.0 hr
#2
07/28/75
40 32N
71 27W
3.0 hr
#3
07/29/75
40 04N
71 34W
3.0 hr
#4
07/29/75
34 42N
71 54W
3.0 hr
#7
08/05/75
24 09N
81 49W
3.0 hr
#8
08/05/75
24 ION
81 SOW
3.0 hr
#9*
08/06/75
24 04N
81 57W
3.1 hr
#19**
08/07/75
24 33N
83 45W
3.0 hr
Depth
(m)
0.5
1.3
9.5
25.0
35.0
0.5
1.3
3.5
9.5
25.0
0.5
1.5
3.5,
9.5
25.0
0.5
1.3
3.5
9.5
25.0
1
8
15
25
50
1
8
15
25
50
1
8
15
25
50
1
10
25
50
75
C-14
mean
467
515
382
501
559
430
425
313
229
208
233
313
345
210
1096
315
411
335
206
343
130
99
1026
44
151
189
138
129
113
84
-
107
26
30
103
175
39
49
39
43
American
DPM+ Productivity
ls++
53
16
20
94
11
97
10
52
149
58
12
27
20
10
28
14
8
23
11
49
5
21
126
5
7
51
42
10
11
16
-
69
32
14
84
43
3
12
3
14
mg C m~ 3 hr l
0.60
0.66
0.49
0.64
0.72
0.35
0.36
0.27
0.20
0.18
0.20
0.26
0.30
0.18
0.95
0.26
0.34
0.27
0.17
0.28
0.11
0.09
0.89
0.04
0.13
0.17
0.12
0.11
0.10
0.08
-
0.09
0.02
0.03
0.09
0.15
0.03
0.04
0.03
0.06
C-14
mean
357
-
444
461
630
135
258
171
90
210
218
809
332
1214
3239
181
298
233
161
357
147
113
115
203
166
121
113
636
102
102
2281
202
142
111
136
74
57
222
65
76
Soviet Union
DPM+ Productivity
ls++
33
-
51
120
750
100
25
102
30
13
24
498
58
1763
364
21
20
11
7
69
90
14
2
197
1
4
11
875
8
6
1851
58
25
23
9
9
8
4
11
7
mg C m~ 3 hr~ l
0.46
-
0.57
0.59
0.81
0.11
0.22
0.15
0.08
0.18
0.19
0.68
0.28
1.04
2.81
0.15
0.24
0.18
0.13
0.29
0.13
0.10
0.10
0.18
0.14
0.11
0.10
0.56
0.09
0.10
1.94
0.17
0.12
0.09
0.12
0.06
0.05
0.19
0.06
0.10
(continued)
159
-------
TABLE 1 (continued)
Station Information
Incubation
Period (hr)
#20**
08/08/75
24 31N
83 40W
3.0 hr
#21
08/09/75
24 35N
83 52W
3.2 hr
Depth
(m)
1
8
15
25
50
1
8
15
25
50
C-14
mean
51
35
33
54
46
88
55
86
79
99
American
DPM+ Productivity
1S++
2
12
5
18
6
41
17
9
9
55
mg C m 3 hr l
0.05
0.03
0.03
0.05
0.04
0.08
0.05
0.05
0.07
0.09
C-14
mean
162
140
90
71
65
88
49
44
44
52
Soviet Union
DPM+ Productivity
ls++
34
100
5
9
2
30
23
8
3
0
mg C m 3 hr~ 1
0.46
0.08
0.08
0.06
0.06
0.08
0.04
0.04
0.04
0.05
+Carbon-14 activity on filters in disintegrations per minute.
++0ne standard deviation from the sample mean.
*Synpor filters used by both sides.
**Whatman GF/C filters used by both sides.
Primary, productivity values obtained in continental shelf waters off
Rhode Island were of the same order as values reported by Ryther and Yentsch
(1958). Kabanova and Baluja (1977) obtained primary productivity values in
the Straits of Florida similar to those reported here.
Primary productivity values were integrated for each station using
linear interpolation between productivity values at discrete sample depths.
Integrated productivity values were treated by linear regression (Fig. 1).
Station 3 was deleted from the regression due to high variance in Soviet sam-
ples. Station 9 was deleted from the regression due to consistent underesti-
mation of productivity by the American method using Synpor filters. Since the
large activity value at 15 m, Station 8, obtained by the Soviets was due to a
high activity for one sample, that sample was deleted from the treatment. The
slope of the regression of American and Soviet productivity values was not
significantly different from 1.0 (p=0.05) while the intercept was not sig-
nificantly different from 0.0 (p = 0.05). This suggests that the American and
Soviet methods for measuring phytoplankton primary productivity used in this
intercalibration gave similar values. Differences in the design of in-situ
incubation apparatus and in position of American and Soviet incubation appa-
ratus on the hydrographic wire did not affect the productivity estimates.
160
-------
_
mg Cm
Figure 1. Linear regression of Soviet and American integrated productivity
values. The regression equation is Y=0.70 + 0.92 X with r2 = 0.94.
Numbers indicate stations, 95 percent confidence limits are drawn.
ACKNOWLEDGMENTS
Financial support for this work was provided in the United States by
the Environmental Protection Agency and in the Soviet Union by Moscow State
University.
Soviet liquid scintillation counting was performed by Drs. Ozrina and
Pastuchenko from the Isotope Analysis Laboratory of the Moscow State Univer-
sity Biological Faculty.
REFERENCES
Doty, M. D., H. R. Jitts, O. J. Koblentz-Miske, and Y. Saito. 1965. Inter-
calibration of marine plankton primary productivity techniques. Limnol.
Oceanogr. 10:282-286.
Iverson, R. L., H. F. Bittaker, and V. B. Myers. 1976. Loss of radiocarbon
in direct use of Aquasol for liquid scintillation counting of solutions
containing ^C-NaHCOa- Limnol. Oceanogr. 27:756-758.
161
-------
Kabanova, V., and L. L. Baluja. 1973. Produccion primaria en la region meri-
dional del Golfo de Mexico y cerca de la costa noroccidental de Cuba.
Serie Oceanologia No. 16, Institute de Oceanologia, Academia de Ciencias
de Cuba.
Morris, I., C. M. Yentsch, and C. S. Yentsch. 1971. Relationship between
light carbon dioxide fixation and dark carbon dioxide fixation by marine
algae. Limnol. Oceanogr. 16:854-858.
Pugh, P. R. 1970. Liquid scintillation counting of 1'*C-diatom material on
filter papers for use in productivity studies. Limnol. Oceanogr.
15:652-655.
»Ryther, J. H., and C. S. Yentsch. 1958. Primary production of continental
shelf waters off New York. Limnol. Oceangr. 3:327-335.
Steemann-Nielsen, E. 1952. The use of radioactive carbon (1'*C) for measuring
organic production in the sea. J. Conseil, Conseil Perm. Intern. Explora-
tion Mer. 18:117-140.
Strickland, J. D. H., and T. R. Parsons. 1972. A Practical Handbook of Sea-
water Analysis. Bull. No. 167 (2d ed.), Fisheries Research Board of
Canada.
Wolfe, D. A., and C. L. Schelske. 1967. Liquid scintillation and geiger
counting efficiencies for carbon-14 incorporated by marine phytoplankton
in productivity measurements. J. Conseil, Conseil Perm. Intern. Explora-
tion Mer. 31:31-37.
MICROBIOLOGY
Soviet and American investigators have developed "standard" procedures
for estimating bacterioplankton uptake. Although a valuable insight into phy-
toplankton-bacterioplankton relationships, bottle assays are subject to ques-
tion because they prevent natural diffusion of nutrients and autoinhibitory
substances and fail to exclude the effect of predation. A developing method-
ology based on diffusion cultures using selectively filtered populations as an
inoculum and as an estimate of biomass appears worthy of further development.
Both methods, together with the analysis of dissolved organic matter during
the diurnal cycle to determine gross release and uptake rates, should be use-
ful in investigating the gradation of stress to a specific pollutant in sev-
eral environments.
It is proposed that an American ship and a Soviet ship visit selected
sites to compare the standard C bottle assays with the diffusion culture and
dissolved organic matter release and uptake for primary and heterotrophic pro-
ductivity.
Dr. Sieburth expressed the desire that the R/V Endeavor be used to
study Soviet waters noted for hydrocarbon pollution, while a Soviet vessel
conducts similar studies in a comparable U.S. area.
162
-------
Laboratory studies should be conducted as a supplement and perhaps as
a forerunner to the cruises. Personnel from the laboratory of Professor Sie-
burth, as guests of Professor M. Gusev of the Biological Science Department of
MGU in Moscow, would learn his biochemical indicators, which could be used in
later research. This group could perhaps visit Yuri I. Sorokin at Galenzhik
in the Crimea, at some time in the future. Conversely, Professor Gusev and
colleagues would be invited by Professor Sieburth and the Marine Ecosystem
Research Laboratory to study energy-related pollution at Narragansett, Rhode
Island, in the future.
Arrangements for the laboratory visits should start immediately, with
plans for cruises to be formulated in the future.
RADIOACTIVITY
The present levels of radioactive contamination of surface waters of
world oceans is about 1000 times lower than the levels of natural radionuclides.
In coastal waters and enclosed seas, the concentration of artificial radio-
nuclides is 5-10 times higher than in oceanic water. In fresh water (rivers,
lakes, and reservoirs) the concentration of artificial radionuclides is com-
parable with the concentration of natural radioactive substances. At the same
time, it was noted that concentrations of artificial radionuclides have reached
significant levels in several marine environments where the discharge of radio-
active wastes takes place.
The calculated values of absorbed doses of ionizing radiation which may
affect hydrobiota in water bodies of various types presently are lower than
background dose rates in the open ocean and approximately equal for freshwater
organisms. In some areas where direct discharge of radioactive wastes is tak-
ing place, the dose to fish may reach 0.1 rad/day.
Data collected thus far indicate that doses of ionizing radiation in
the marine environment have not adversely affected systemic groups, including
fish. However, long residence times of fish in a contaminated environment
where the dose may approach 0.1 rad/day, may be followed by the lowering of
resistance to harmful factors, the destruction of reproductive functions and
perhaps a decrease in population size.
As the use of atomic energy increases and as there is increased dis-
charge of radioactive wastes, the area of contamination in the ocean also will
grow proportionally. This provides a need to initiate a program of forecast
monitoring in regions where radioactive wastes are being discharged. Special
attention should be paid to the more potentially dangerous of the long-lived
radioisotopes such as plutonium-238, plutonium-239, strontium-90, cesium-137
and several others.
Due to the anticipated growth of nuclear industry, the U.S. National
Academy of Sciences has estimated that the world-wide inventory of plutonium
will approach 94 x 106 Ci by the year 2000. Thus it is imperative that re-
163
-------
search on migration, accumulation and potential biological effects of pluto-
nium as well as other transuranics and fission products in the marine environ-
ment be expanded. In order that this research be conducted most efficiently,
we recommend that a joint U.S.-U.S.S.R. workshop on the cycling and biological
effects of transuranics and other radionuclides in the marine environment be
held. Such a workshop would allow an exchange of views and research informa-
tion among scientists from both countries.
In addition, perhaps an intercalibration study or cooperative research
project could be accomplished prior to the workshop and the data exchanged at
that time. If such a recommendation receives favorable consideration, we be-
lieve that the workshop should take place in the near future.
ECOSYSTEMS
There is a need to develop procedures for the systematic study of pol-
lutant effects at the ecosystem level. Such studies should include investiga-
tions of both unstressed and stressed systems, and must range in objectivity
and complexity from simple laboratory microcosms to field ecosystem analysis.
At this time, no generally applicable concepts exist for developing such tests
under both laboratory and field conditions. Therefore, investigators have no
standards for designing such tests or interpreting data derived from them. A
standardized methodology is required to provide the ability to compare results
from similar systems in different areas and with different pollutants. Micro-
cosms are becoming increasingly recognized as a necessary tool for understand-
ing and finding solutions to both theoretical and practical problems. There-
fore, we recommend that the US and USSR continue to emphasize the development
of marine and estuarine microcosms that can provide information for water
quality management decisions. We further recommend that the US and USSR co-
ordinate their investigations to avoid duplication of effort and enhance the
exchange of experience and information. To fulfill this mutual research ob-
jective, we propose that:
a. Over the next two years, the US and USSR agree upon and develop common
laboratory microcosms (i.e., microbiological; predator-prey, etc.).
b. For the development of these microcosms, provisions should be made for
consultation (exchange of information and site visits).
c. To provide a common basis for system design and data interpretation and
to develop the common ability to predict environmental impacts, the US
and USSR should exchange pollutants (i.e., persistent pesticides), which
have global usage. Such pollutants would include those manufactured in
each country for distribution throughout the biotic component of the
biosphere.
164
-------
MODELS
Models, viewed as simplified representations of a real-world system,
subsystem or function, may be constructed as biological models of the "micro-
cosm-type," words, diagrams, or mathematical equations. As analytical tools,
they may be used qualitatively for conceptual and planning purposes or quanti-
tatively for simulation purposes to understand the behavior of the system of
interest or for both purposes. Although models merely reflect certain aspects
of the system studied and may represent theories or hypotheses, they also have
some intrinsic predictive qualities. Formal models, either qualitative or quan-
titative, become increasingly important scientific tools in posing tractable
questions about complex systems, lending emphasis to key factors and eliminat-
ing non-essential detail.
A wide variation of opinion seems to exist concerning the number of
factors or systems essential to a simulation model (as a system of equations
and coefficients) and the degree of investigator control over the operational
simulation model (i.e., pre-determined boundary limits vs intrinsic limits
established by the model). In contrast, the mathematical techniques are not
questioned, although the mathematical approach to modeling may vary (deter-
mined models, stochastic models). These opinions range from the inclusion and
evaluation of all known components and interactions, to the inclusion and
evaluation of only components and interrelationships known or likely to be im-
portant in resolving the question(s). Both approaches are considered useful
in addressing pollution problems. A distinction was also made between models
simulated to solve a problem (an infinite number of constructions) versus
those designed to understand the causes of the dynamic behavior of systems
(only one best approximation model construction). Regardless, models have a
distinct purpose in formulating research goals. The true value of a modeling
effort is measured by subsequent verification.
It is proposed that participants in a future US-USSR joint research
program use conceptual models and follow through with their parameterization,
subsequent simulation, and verification.
American and Soviet specialists should attempt to develop a mathemati-
cal model and to provide a quantitative expression of a simple ecosystem. For
this purpose, experimental microcosm systems may be used (these could be a pe-
lagic microcosm version or microcosm that includes the substrate and overlying
water). Future tasks might include comparison of such models as well as their
development and improvement.
165
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO
EPA-600/9-78-007
2.
4. TITLE ANDSUBTITLE
FIRST AMERICAN-SOVIET SYMPOSIUM ON THE BIOLOGICAL
EFFECTS OF POLLUTION ON MARINE ORGANISMS
7. AUTHOR(S)
Thomas W. Duke and Anatoliy I. Simonov
9. PERFORMING ORGANIZATION NAME AND ADDRESS
U.S. Environmental Protection Agency
Environmental Research Laboratory
Gulf Breeze, Florida
12. SPONSORING AGENCY NAME AND ADDRESS
U.S. Environmental Protection Agency
Office of Research and Development
Washington, DC
3. RECIPIENT'S ACCESSION NO.
5. REPORT DATE
6. PERFORMING ORGANIZATION CODE
8. PERFORMING ORGANIZATION REPORT NO.
10. PROGRAM ELEMENT NO.
1BA608
11. CONTRACT/GRANT NO.
Joint Project VI-2.1
13. TYPE OF REPORT AND PERIOD COVERED
Final. Sept.pmhpr ?n_?d, 1977
14. SPONSORING'AGENCY CODE
EPA/600/4
15. SUPPLEMENTARY NOTES
16. ABSTRACT
This symposium was conducted under a US-USSR Environmental Agreement,
Project 02.06-21 titled "Influence of Pollutants on Marine Organisms." American
and Soviet specialists discuss state-of-the-art for hydrobiological analysis of
basic structural components of marine ecosystems and the influence of various
pollutants on these components. Participants define problems related to methods
for modeling the influence of pollutants on the marine environment, long-term
forecasting and determination of permissible loads of pollutants, and the
unification and intercal ibration of methods for determining production of micro-
organisms of ocean bacterioplankton and phytoplankton. Results or laboratory
research on the influence of pollution on the marine environment are presented.
Proceedings were published in English and Russian in compliance with the Memorandum
from the 4th Session of the Joint US-USSR Committee on Cooperation in the Field
of Environmental Research.
17.
a. DESCRIPTORS
Water pollution
Marine organisms
Modeling
Oceanic ecosystems
Photoplankton
Oil
Microcosms
Padi^a^ti vi f y
18. DIStRIBUtlON STATEMENT
Unlimited
KEY WORDS AND DOCUMENT ANALYSIS
b. IDENTIFIERS/OPEN ENDED TERMS C. COSATI Field/Group
Expotentially broken rod
US-USSR Agreement on Coop-
eration in the Field of
Environmental Protection
Project 02.06-21
19 SECURITY CLASS (This Report) 21. NO. OF PAGES
Unclassified 155
2O SECURITY CLASS (This page) 22. PRICE
Unclassified
EPA Form 2220-1 (Rev. 4-77) PREVIOUS EDI TION is OBSOLETE
------- |