United States
Environmental Protection
Agency
Environmental Research
Laboratory
Athens GA 30613
-A'
Research and Development
EPA-600/S3-83-084 Nov. 1983
ŁEPA Project Summary
Prediction of Ecological Effects
of Toxic Chemicals: Overall
Strategy and Theoretical Basis
for the Ecosystem Model
Ray R. Lassiter
A strategy was developed for model-
ing ecosystems to permit the assess-
ment of toxic chemical effects on
element cycling and other ecosystem
processes. The strategy includes the
use of multi-species representations of
biotic communities and mathematical
descriptions of the processes that are
important in aquatic ecosystems.
Direct effects of toxicants are assigned
to the species comprising the biotic
community, in a manner suggested by
available toxicological information.
Effects are calculated as the difference
between selected measures of
processes from unaffected systems and
systems affected by the presence of a
toxic chemical. Ecological effects cal-
culated in this manner are considered to
be heuristically useful.
This Project Summary was developed
by EPA's Environmental Research
Laboratory, Athens. GA, to announce
key findings of the research project that
is fully documented in a separate report
of the same title (see Project Report
ordering information at back).
Introduction
Analysis of environmental risks posed
by the release of toxic chemicals neces-
sarily would require predictions of the
fate of a chemical and its effects. Pre-
diction of a chemical's fate presupposes
an understanding of transport and
transformation processes, or "what the
environment does to the chemical."
Prediction of a chemical's effects presup-
poses a wider knowledge of the
chemical's fate, and by contrast, "what
the chemical does to the environment." It
is also necessary to know which effects
are of concern, because very many poten-
tial effects exist. Ecological effects are the
concern of this research, and effects that
occur in ecosystems, and whose charac-
teristics are influenced by ecological
interactions, are considered to be ecolog-
ical effects. A particular class of potential
ecological effects that arguably could be
of major significance are effects on cycles
of elements. Although results of this
research will not be limited to considera-
tion of the effects of toxicants on element
cycles, they are the class of effects that
have motivated most of the work. A major
reason for focusing on element cycle
effects is that there is apparently no on-
going work in this area despite the appar-
ent potential for such effects.
The Nature of Element Cycles
and Importance of Considering
Their Response to Toxicants
Biotic processes in the proportions and
magnitudes that we know them depend
upon concentrations of chemicals
remaining approximately at their current
levels. The energy for biotic processes is
derived from catalysis of redox reactions
between reduced and oxidized forms of
elements. The primary source of the dis-
equalibria that drive these reactions is
photosynthesis in which both highly
reduced organic compounds and
molecular oxygen are produced. The
remaining biotic processes feed off this
disequilibrium, and in doing so contribute
to maintenance of the pool sizes of geo-
-------
chemicals. These include the atmospher-
ic gases, the composition of the ocean,
dissolved constituents of fresh waters,
and the earth's soils.
These geochemical pools have been
remarkably stable over geological time
scales. This fact attests to the historical
stability of the contributory processes. No
comparative analysis of the relative roles
of abiotic and biotic processes in
maintaining the pools of geochemicals
has been made. The number and
importance of the processes known to be
carried out by organisms, however, leaves
little doubt that the characteristics of the
earth depend on the continuation of biotic
processes at about their current levels.
Whether these processes will remain
stable in the presence of continued
introduction of xenobiotic chemicals is
unknown. It is prudent, therefore, to
consider potential conditions that could
lead to serious disturbances of element
cycle transformations. Conditions that
would lead to alterations in ecosystem
structure and function are expected to
lead to corresponding alterations in
element cycling.
Strategy of Ecosystem Modeling
for Analyzing Alterations in
Element Cycling
Element cycling is a system phenom-
enon. Organisms use pools of chemicals
that are products of other organisms'
metabolism. These and other phenomena
that occur as fluxes that change chemical
forms and redox states are processes. For
a given set of elements one can describe
the major processes that occur to cycle
the elements through their various forms.
They can be grouped into those pertain-
ing to microrganisms or macroorganisms
and to autotrophs or heterotrophs. The
level of resolution at which these
processes are described is a critical
choice for accomplishing the desired
goals. These goals are to represent the
aquatic ecosystem such that models of
the processes respond to external
influences, including the introduction of
toxicants approximately the same as do
real systems. If this goal is reached, it will
be possible to analyze for ecological
effects of toxic chemicals, including their
effects on element cycling processes.
Carrying out analyses of this sort is, in a
special sense, a predictive (or prognostic)
exercise that is fully comprehensible only
if the usage of the term "effect" is under-
stood, and if the special sense of
prediction is made clear.
An effect implies a deviation from some
nominal state or rate. In this model, there-
fore.the effect is referred to as a deviation
from a nominal condition. This deviation
can be obtained by first using the model to
obtain the nominal condition, then using
it again in the same manner, but with the
presence of a toxicant represented
additionally. To normalize effects so that
measurement units do not influence the
magnitude of the values, the difference
should be divided by the nominal value.
Any quantity of interest can be used to
calculate an effect, from the amount of a
material present to the rate at which a
process is occurring, regardless of
whether there is any direct influence of
the toxicant on the quantity of interest.
Because the effect is obtained by model
calculation in which comparable
conditions can be strictly maintained, and
it is known that the only difference
between the nominal and affected condi-
tions is the presence of the chemical, any
differences can be attributed to the
presence of the chemical and referred to,
therefore, as an effect. These calculations
are referred to as predictions of ecological
effect of the toxic chemical.
The special sense in which "predic-
tions" is used is that the predicted values
have heuristic value but do not refer to an
expected effect in a particular real
ecosystem at some time. The predicted
value is intended to be heuristically useful
by permitting assessment of the probable
effects in aquatic ecosystems of selected
characteristics without necessarily refer-
encing specific real aquatic systems. The
strategy of modeling is designed to make
this use possible.
The strategy of modeling makes use of
representation of multiple species, each
carrying out a particular process. It is
often observed that concentrations of
environmental chemicals that are known
to turn over rapidly are fairly stable,
whereas any particular species that uses
or produces the chemical may fluctuate
widely. This suggests that compensation
occurs among species carrying out the
same function. That is, reduction in levels
of one species may be compensated by
corresponding increases in the levels of
one or more others, so that the processes
responsible for the level of the chemical
are relatively stable. The possibility of
stability resulting from species
redundancy and compensation is
important when considering the effects
of chemicals on the cycling of major
elements. If this phenomenon is
important, it is probable that representing
a group of organisms by a single variable
(as opposed to a multi-species represen-
tation) would lead to conclusions that
element cycling processes are more
fragile than they, in fact, are.
Another important component in this
modeling strategy is the way that element
cycling processes are represented by
species populations. In element cycling, it
is the processes that are important. They
are carried out, however, by particular
populations. Each individual species pop-
ulation is characterized by size, energy
source, source of materials for
biosynthesis, motility, and several other
descriptors. The descriptors will define a
particular population as carrying out a
specific process. Representations of
multi-species populations will differ
among themselves in the particular
values of the descriptors. For example,
one population may consist of large
individuals that move slowly and use
biomass of microorganisms for both
energy and biosynthesis, while another
may differ in being smaller and moving
more rapidly. These organisms are
represented differently from carrying out
different processes by differences in the
energy source and source of materials for
biosynthesis. For example, another group
of organisms might use NH4+and 02 for
energy and inorganic chemicals for
biosynthesis.
Representing the organisms that exist
in systems to carry out all the important
functions using this approach requires
that a large amount of descriptive data be
available. That amount of data will not be
available for all the organisms necessary
to represent the processes of any given
ecosystem. Therefore, the use of this
approach requires an additional strategic
component to make it possible to parame-
terize the model for the large number of
species representations necessary to the
approach. Essentially this will be
accomplished by first describing a large
number of species for the initial set of
organisms that occupy the environment.
Selection of the descriptors will be gu ided
by what is known about the kinds of
organisms carrying out the various
ecosystem processes, /. e., they will be
assigned to the model representations
over a range that includes what is known.
It is unlikely that an assignment can be
made such that the entire initial set will
persist for very long (not approach zero).
It is anticipated, however, that some of
the variables will be persistent. These will
represent a set of organisms that are
inter-compatible and that are "adapted"
to the environment. It is extremely
unlikely that this set will well represent
real populations (in the sense of close
correspondence of their descriptors to
-------
those of real populations). Yet, their
descriptors necessarily will be in the
neighborhood of those of real populations,
because only such descriptors are to be
used. Repeated selection of compatible
sets of organism representations for a
given environment beginning with differ-
ent initial sets is expected to build a
distribution of descriptors that would
include the characteristics of real organ-
isms that would live in the environment
as described.
This synthetic approach appears to be
the best way to obtain descriptions of
biotic communities that are appropriate
for an environment that is selected for
assessment in anticipation of a toxic
chemical's possible introduction. With
this approach a nominal behavior of
element cycling processes can be
obtained. The same type of uncertainty
exists for the direct effects of a toxic
chemical on the organisms of a
community, however, as exists for the
characteristics of types of organisms that
inhabit the community. The final element
of the strategy defines the process by
which toxicological information is used to
obtain affected behavior.
For each set of compatible species
representations, a set of direct effects will
be assigned from the range of direct
effects suggested by the available
toxicological information. Recalculation
of the behavior in the presence of the
toxicant, then, will build another
distribution of behaviors. Normalized
differences between the two behaviors,
as already discussed, are the ecological
effects of the chemical.
It is readily apparent that ecological
effects calculated in this manner are not
expected to be observed exactly in any
particular system. They are expected,
however, to include values for real
systems similar to the one analyzed.
Furthermore, if the nature of the distribu-
tion of descriptors of the compatible sets
of species and of the direct effects of toxi-
cants on them is realistic, a measure of
the uncertainty associated with the
predicted effects can be obtained. The
value of the calculated effects will not be
primarily quantitative, however, but
rather, will more appropriately be used as
a learning exercise for identification of
possible ecological effects that are too far
removed from the direct effects to be
apparent without an ecosystem model.
Descriptions of the Processes
Much of the research effort was
devoted to assimilation of existing and
development of new mathematical de-
scriptions of the processes. Uptake of
materials by microorganisms is described
by a rectangle hyperbola that is derived in
a way that rationalizes its use for all
uptake processes. Depletion of materials
being taken up from a zone in the vicinity
of cells is considered. Photoautotrophy is
described in terms of both bioenergetics
and uptake of materials, so that limitation
by light and materials is possible. Chemo-
autotrophs and photoautotrophs share
the same biosynthetic pathways. Chemo-
autotrophs, however, use a variety of
terminal electron acceptors. The bioener-
getics of their energy reactions is used to
calculate the rate of energy production,
and as with photoautotrophs and for that
matter all organisms, limitation of growth
is considered to be the minimum of the
potential growth rates for each of the
requirements. Photoautotrophs produce
organic matter using solar energy,
whereas, chemoautotrophs produce
organic matter using energy of redox
reactions between reduced forms of
inorganic compounds and oxygen. In the
process, oxidized forms of elements are
formed. The reduced forms of the
inorganic compounds are produced by
heterotrophs using oxidized species of
elements in anaerobic zones.
Heterotrophic processes are repre-
sented for both microorganisms and
macroorganisms. All use organic com-
pounds for both energy and for synthesis.
Microorganisms are found in both
aerobic and anaerobic zones of aquatic
systems. In aerobic zones they use
molecular oxygen as their terminal
electron acceptor (oxidizing agent) for
oxidizing organic compounds. In anaero-
bic zones oxidized forms of the major
elements are used as terminal electron
acceptors. These oxidized forms are those
produced by chemoautotrophy. In this
manner, the element cycles are
completed. Growth of heterotrophs is
described as the difference between the
available energy and materials and their
need for maintenance and activity. Macro-
heterotrophs differ from microhetero-
trophs in being limited to aerobic envi-
ronments and m their use of specific
structures and behaviors to obtain food.
Two major kinds of feeding are described
for the model: filter feeding and pursuit
feeding. General model descriptions for
both types are developed for the general
case of any number of consumers feeding
on any number of prey. Rates of con-
sumption are calculated directly from
characteristics of predator and prey,
rather than assigned as coupling constants.
Outlook
The strategy of modeling and the
requisite theory to carry it out exist.
Additional work is needed to bring the
expression of toxicological theory to the
state of development (in the context of
this model) as the theory for ecosystem
modeling. Further work is needed to
represent fermentation as an ecological
process. Coding is underway for the
ecosystem model. Whether the sets of
compatible representations of biotic
communities and the effects as computed
will be usefully close to real systems
needs to be tested. The fact that no speci-
fic real system is represented presents a
difficulty in that regard. Laboratory eco-
systems, however, can be run repeatedly
under the same conditions and with or
without the presence of a toxicant. One
potential testing scheme would make use
of this capability and compare distribu-
tions and effects from laboratory ecosys-
tems with distributions and effects from
the model as described.
-------
The Project Summary was authored by Ray R. Lassiter who is also the EPA
Project Officer (see below).
The complete report, entitled "Prediction of Ecological Effects of Toxic Chemicals:
Overall Strategy and Theoretical Basis for the Ecosystem Model," (Order No. PB
83-261 685; Cost: $10.00, subject to change) will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Environmental Research Laboratory
U.S. Environmental Protection Agency
Athens, GA 30613
4U.S. GOVERNMENT PRINTING OFFICE: 1M3-459-017/7227
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
BULK RATE
U.S. POSTAG
PAID
Cincinnati, Oh
Permit No. G3
Official Business
Penalty for Private Use $300
------- |