Trends Expected in Stressed Ecosystems

Author(s): Eugene P. Odum

Source: BioScience, Vol. 35, No. 7, Managing Stressed Ecosystems (Jul. - Aug., 1985), pp.
419-422
Published by: Oxford University Press on behalf of the American Institute of Biological
Sciences
Stable URL: https://www.jstor.org/stable/1310021
Accessed: 22-11-2019 00:50 UTC

REFERENCES
Linked references are available on JSTOR for this article:
https://www.jstor.org/stable/1310021?seq=1&cid=pdf-reference#references_tab_contents
You may need to log in to JSTOR to access the linked references.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms

American Institute of Biological Sciences, Oxford University Press are collaborating

with JSTOR to digitize, preserve and extend access to BioScience

This content downloaded from 24.11.88.25 on Fri, 22 Nov 2019 00:50:10 UTC
All use subject to https://about.jstor.org/terms
 Trends Expected in Stressed Ecosystems

Eugene P. Odum

When ecosystems are not suffering from unusual external perturbations, we observe In 1969, I published a table, "Trends to
certain well-defined developmental trends. Since disturbance tends to arrest, or even be Expected in Ecosystem Develop-
reverse, these autogenic developments, we can anticipate some ecosystem responses to ment," which contrasted early and late
stress. Trends expected in stressed ecosystems include changes in energetics, nutrient stages of succession in terms of 24 eco-
cycling, and community structure and function.
system properties. This tabular model
was subsequently revised (Odum 1971,
1983). My hypothesis is that a disorder-
The term stress has been widely used that may be injured or killed, but it is not ing disturbance to which a community is
for both cause and effect, stimulus and a stress at the ecosystem level; the ab- not adapted arrests and, in many cases,
response, or input and output. Although sence of fire would be a stress at this reverses these autogenic developments
some authors worry about this dual use level. Furthermore, disturbance at the (Odum 1981). Accordingly, I have pre-
of the word, I view stress as a syndrome same level may sometimes have a posi- pared a table, "Trends Expected in
comprising both input and output. One tive effect or produce both positive andStressed Ecosystems" (see box, page
cannot have a response without a stimu- 420), whose format is similar to my mod-
lus. One can easily identify which part of I
el of autogenic succession.
the syndrome is being singled out by In general, this model covers negative
using modifying adjectives, prefixes, or responses in the longer term at the eco-
clauses; thus, stress as an input can be
Stress is a syndrome system level. Many of the trends listed
designated the stressor as contrasted to comprising both are well supported by observation and
stress, the response, or output. (In refer- experiments as useful indices of stress,
input and output, but others are hypothetical and require
ring to stress as a syndrome, I do not
mean to imply a strict analogy with Se- stimulus and response. further study and testing. As already
lye's [19731 "general stress syndrome," indicated, responses at the ecosystem
although as recently noted by Rapport et level can be expected to be not only
al. 1985, there are interesting parallels different but also more diffuse and longer
between the physiological and communi- negative responses, hence the "subsidy-termed than responses at the population
ty levels.) stress" syndrome (see Odum et al. level. Moreover, chronic stress that con-
I believe that common words in every- 1979). tinues for a long time may have a differ-
day language, such as stress, competi- In summary, I am using the term stress ent effect than an acute stress that is
tion, or community should not be given the way the dictionary defines it and the quickly followed by recovery and return
highly technical meanings, since this way the public understands it-to mean to an unstressed state. Keeping these
only serves to confuse the general read- a detrimental or disorganizing influence. qualifications and boundary constraints
er; as already indicated, modifiers can I prefer to use the term subsidy for an in mind, let us now examine the 18
restrict meanings. But we must recog- input that produces a positive response, components in the model.
nize that these common terms have dif- even though it may be accompanied or
ferent meanings, and often different followed by negative responses. In this ENERGETIC RESPONSES
manifestations at different levels of orga- paper I focus on negative responses to
nization. Allen (1984), in discussing hier- unusual external disturbances, or stress- Theoretically, an increase in commu-
archical organization, pointed out that a ors of low probability to which a commu- nity respiration (item 1) should be the
disturbance detrimental at one level may nity of organisms is not preadapted. first early-warning sign of stress since
be beneficial at a higher level. For exam- When ecosystems are not affected by repairing damage caused by the distur-
ple, periodic fire in the fire-adapted strong external perturbations, such as bance requires diverting energy from
chaparral is a stress to many organisms storms or human disturbances, we ob- growth and production to maintenance.
serve certain well-defined developmen- Hence, the R/B ratio (the maintenance to
tal trends. For example, the biomass and biomass ratio, item 3) increases. Odum
Odum is with the Institute of Ecology, University of
Georgia, Athens, GA 30602. ? 1985 American Insti- sizes of organisms tend to increase and (1967) speaks of this response as an
tute of Biological Sciences. All rights reserved. net community production to decrease. "energy drain," or the process of

419
July/August 1985

This content downloaded from 24.11.88.25 on Fri, 22 Nov 2019 00:50:10 UTC
All use subject to https://about.jstor.org/terms
 I

greater than one. Any input of toxic

substances or silt reduces the ratio to
Trends Expected in Stressed Ecosystems less than one. The PIR ratio is sometimes
a good indicator of stress in laboratory
Energetics microcosms set up to monitor the effects
of a potential pollutant. In one of our
1. Community respiration increases (H. T. Odum's "pumping out" of
experiments with cadmium (Hendrix et
disorder [Odum 1967] or Prigogine's increase in the "dissipative al. 1982), the stressor decreased the ratio
structure" [Prigogine et al. 1972])
in static microcosms with no input or
2. P/R (production/respiration) becomes unbalanced (< or > 1)
output of culture media but not in flow-
3. P/B and R/B (maintenance:biomass structure) ratios increase
through microcosms, where an increase
4. Importance of auxiliary energy increases (Margalef's [1975] exoso-
in photosynthesis caused by poisoning of
matic metabolism)
grazers was balanced by an increase in
5. Exported or unused primary production increases
microbial respiration. More often than
not, however, disturbance produces an
Nutrient cycling increase in respiration greater than the
change in the rate of production, thus
6. Nutrient turnover increases
altering the P/R ratio.
7. Horizontal transport increases and vertical cycling of nutrients de-
Margalef (1975) pointed out that the
creases (cycling index decreases) drain of productive energy in dissipating
8. Nutrient loss increases (system becomes more "leaky")
entropy "opens up" the ecosystem so
auxiliary energy from the outside be-
Community structure comes more important in the continued
survival of the system. The altered and
9. Proportion of r-strategists increases
unbalanced metabolism may result in an
10. Size of organisms decreases increase in unused resources that may
11. Lifespans of organisms or parts (leaves, for example) decrease
then be stored within the system or ex-
12. Food chains shorten because of reduced energy flow at higher
ported (items 4 and 5).
trophic levels and/or greater sensitivity of predators to stress
To summarize, community respiration
13. Species diversity decreases and dominance increases; if original per unit of biomass tends to increase and
diversity is low, the reverse may occur; at the ecosystem level, biomass accumulation to decrease as or-
redundancy of parallel processes theoretically declines
ganisms cope with the disorder created
by unusual exogenous disturbance. Ac-
General system-level trends cordingly, stressed ecosystems will tend
to exhibit unbalanced P/R ratios and
14. Ecosystem becomes more open (i.e., input and output environments increased P/B and R/B ratios-in other
become more important as internal cycling is reduced) words, a decreased ratio of biomass to
15. Autogenic successional trends reverse (succession reverts to earlier
energy flow, or a low efficiency of con-
stages) verting energy to organic structure.
16. Efficiency of resource use decreases
17. Parasitism and other negative interactions increase, and mutualism
and other positive interactions decrease
NUTRIENT CYCLING
18. Functional properties (such as community metabolism) are more
robust (homeostatic-resistant to stressors) than are species compo- The three trends listed in the box are,
sition and other structural properties of course, interdependent. Increased
turnover and decreased cycling frequent-
ly appear in stressed ecosystems. To-
gether they result in accumulation of
"pumping out the disorder." In the con- within hours after introducing a pollutant nutrients which, like unused production,
text of Prigogine's theories of open, far- into floating containers. Since microor- may be lost from the system. O'Neill et
from-equilibrium systems (see Prigogine ganisms respond quickly to any change, al. (1977) and Van Voris et al. (1980)
et al. 1972), an increase in respiration they may be the level at which we should have found that leaching of calcium is a
represents an increase in activity of the search for early-warning signs (see good index of stress severity in terrestri-
"dissipative structure." In practice, res- Ivanovici and Wiebe 1981). al microcosms and also in the field; the
piration increase does not provide a very The P/R, or production/respiration, ra-more intense the stressor, the more cal-
good eary-warning distress signal be- tio (item 2) is, of course, affected by any cium lost.
cause it is difficult to detect small in- change in the partitioning of energy be- In a recent symposium edited by
creases in large open systems. Sometween production and maintenance. Mooney and Godron (1983), several au-
indirect measure, however, might be fea- Since the P/R ratio tends toward balance thors, including M. Rapp and P. M.
sible. Menzel (1980), in reviewing the in undisturbed ecosystems, we can ex- Vitousek, agree that leaching losses of
results of the CEPEX (Controlled Eco- pect the ratio to become unbalanced withmajor nutrients are normally much less
system Pollution Experiment) mesocosm stress. Experience with coral reefs is a than the amounts recycled within forests
experiments, states that changes in indi- good example. The P/R ratio of an undis-but that disturbance may result in in-
ces of bacterial activity were detected turbed coral reef is generally slightly creased losses.

420 BioScience Vol. 35 No. 7

This content downloaded from 24.11.88.25 on Fri, 22 Nov 2019 00:50:10 UTC
All use subject to https://about.jstor.org/terms
 Margalef (1975) reminds us that in- predatory birds resulting from DDT pol- large biomass forests, and also in ecosys-
creasing horizontal transport (i.e., one- lution is an example. tems adapted to nutrient-poor condi-
way flow) at the expense of internal tions. My hypothesis is that, in either
The effect of stress on diversity is also
cycling is one of the principal ways hu- complex and not well understood, partly case, a disordering input should disrupt
mans disturb natural ecosystems. Min- because responses may be different these at intricate homeostatic positive in-
ing, soil erosion, stream pollution, and different levels. Species diversity, forteractions and increase the likelihood
fertilizer runoff from croplands are famil- example, may be affected differentlythat parasitism, overgrazing, and other
iar examples. from food chain diversity or nutrientnegative interactions will develop, as
diversity. Species diversity is almost al-when forests are subjected to air pollu-
ways reduced (item 13) by toxic inputs tion or ionizing radiation. I realize one
COMMUNITY STRUCTURE (insecticides, for example) unless the di-can argue instead that stressful environ-
versity is very low to begin with (Barrettments tend to promote mutualism as a
As long as conditions remain favorable 1969). But a decrease in species diversitystrategy for survival, as for example,
for a variety of organisms, competition is not a reliable index of stress in general
lichens on rock outcrops or mycorrhizal
for resources will tend to favor K-select- because a disturbance affecting the pine trees on poor soils. My view is that
ed species as nutrients and energy be- such mutualistic systems are the adapt-
come fully used, especially in the mature ed, or normal, state, which is especially
system. Accordingly, with stress we vulnerable to additional or unusual exog-
would expect a decrease in such species When stress is detectable enous stress. Indeed, lichens are sensi-
and an increase in opportunistic, or r- tive to air pollution. One can cite numer-
selected, species (item 9); this frequently at the ecosystem level, ous examples of increased parasitism
happens when forests are clear-cut or
there is real cause for and predation resulting from stress. For
subjected to ionizing radiation, or in just example, game managers find that an
about any case of severe disruption. alarm, for it may signalincrease
a in stomach worms in deer is a
Woodwell (1983) points out that general- breakdown in good indicator of an overbrowsed range
ly, "disturbance favors communities (Eve and Kellogg 1977).
dominated by small-bodied, rapidly re-
homeostasis.
That functional properties may be
producing, hardy species." The "weedy more robust than structural properties
type plants" and "pest-type animals," (item 18) is an intriguing and important
as it were, take over. In the CEPEX possible trend. Schindler's whole-lake
experiments, organism size decreased structure of the system (e.g., patch experiments
cut- certainly support this hy-
(item 10) no matter what kind of pollut- ting in a forest) often increases the diver-
pothesis (Schindler 1974, Schindler and
ant was added to the mesocosms. In the sity of species of both plants and ani- Turner 1982). When large amounts of
plankton, for example, large diatoms mals. In theory, redundancy of parallel acid or nutrients were introduced into
were replaced by small green algae,processes (or what Hill and Wiegert lakes, 1980 primary productivity and other
which, in turn, could reduce the number call "congeneric homeotaxis") should aspects of community metabolism were
and size of organisms in higher trophic be reduced by stress at the ecosystem remarkably homeostatic, but species
levels. A good example of the reduction level (item 13), but I know of nothing composition of the plankton and fish
in lifespan (item 11) is the response of documenting this possibility. were greatly altered. We are reminded
pine forests to air pollution. We often seeInterestingly, human settlement at here of Margalefs statement (1981) that:
the first sign of stress when needles,first tends to make the landscape more "Stress is something that puts into ac-
which normally remain on the branches diverse (patchy), but with increasing ur- tion the mechanism of homeostasis." In
banization and monoculture in agricul- this case, species replacement and other
for two years, fall off after the first year
(Williams 1980). If pollution stress con-ture, the landscape becomes less di- adjustments keep the overall function of
tinues, the lifespan of the trees them- verse, unless parks, open space, and the system steady. Accordingly, early
selves is reduced. nature reserves are planned during warning of stress will be more easily
The impact of stress on the food web development.
is seen at the species level, although shifts
not clear-cut. A shortening of food here should be accompanied by changes
chains with a reduction in top predatorsGENERAL SYSTEM-LEVEL TRENDS in the rate of respiration and/or decom-
(item 12) frequently follows eutrophica- position, which are more difficult to de-
tion and toxic waste pollution in lakes. The last five items (14 through 18) tect in large systems. When stress is
The reasons for this are not understood,summarize the overall trends that might detectable at the ecosystem level, there
although two different explanations havebe expected in stressed ecosystems. is real cause for alarm, for it may signal a
been suggested: (1) Small organisms out- Items 14-16 follow logically from the breakdown in homeostasis.
compete large organisms, both under discussion of trends 1-13, but the last
conditions of enrichment and toxic two statements are hypothetical and re-
stress. (2) Although large organisms are
quire discussion.
often more efficient feeders when re- Cooperation and mutualism seem to REFERENCES CITED
sources are scarce (the size-efficiency
develop when resources become scarce,
Allen, T. F. H. 1984. The abuse of the con-
hypothesis; see Dodson 1974), they areand working together for mutual benefit
cept "disturbance": a scaling problem.
subject to bioaccumulation of toxins, has a high selective value (see Axelrod Bull. Ecol. Soc. Am. 65: 54.
have vulnerable life history stages, or are 1984, Axelrod and Hamilton 1981). Ac- Axelrod, R. 1984. The Evolution of Cooper-
otherwise more sensitive to disturbance cordingly, we find that mutualism is im- ation. Basic Books, New York.
than are small organisms. The demise of
portant in mature ecosystems, such as Axelrod, R., and W. D. Hamilton. 1981. The

July/August 1985 421

This content downloaded from 24.11.88.25 on Fri, 22 Nov 2019 00:50:10 UTC
All use subject to https://about.jstor.org/terms
 evolution of cooperation. Science 211:
Publishing and Documentation, Wagenin- tion and Marine Ecology. Wiley-Intersci-
1390-1396. gen, Netherlands. ence, New York.
Barrett, G. W. 1969. The effects of an acute . 1981. Stress in ecosystems: a future O'Neill, R. V., B. S. Ausmus, D. R. Jackson,
insecticide stress on a semi-enclosed grass- approach. Pages 281-289 in G. W. Barrett R. I. Van Hook, P. Van Voris, C. Washburn,
land ecosystem. Ecology 49: 1019-1035. and R. Rosenberg, eds. Stress Effects on and A. P. Watson. 1977. Monitoring terres-
Dodson, S. I. 1974. Zooplankton competition Natural Ecosystems. John Wiley & Sons, trial ecosystems by analysis of nutrient ex-
and predation: an experimental test of the New York. port. Water Air Soil Pollut. 8: 274-277.
size-efficiency hypothesis. Ecology 55: Menzel, D. W. 1980. Applying results derivedPrigogine, I., G. Nicolis, and A. Babloyantz.
605-613. from experimental microcosms to the study 1972. Thermodynamics and evolution.
Eve, J. H., and F. E. Kellogg. 1977. Manage- of natural pelagic marine ecosystems. Physics Today 25: 22-38; 138-144.
ment implications of abomasal parasites inPages 742-751 in J. B. Giesy, ed. Micro- Rapport, D. L., H. A. Regier, and T. C.
southeastern white-tailed deer. J. Wildl. cosms in Ecological Research. DOE Sym- Hutchinson. 1985. Ecosystem behavior un-
Manage. 41: 169-177. posium Series 52 CONF-781101. National der stress. Am. Nat., in press.
Hendrix, P. F., E. P. Odum, and C. L. Technical Information Service, Springfield,Schindler, D. W. 1974. Experimental studies
Langner. 1982. Cadmium in aquatic micro- VA. of eutrophication and lake recovery: some
cosms: implications for screening the eco- Mooney, H. A., and M. Godron, eds. 1983. implications for lake management. Science
logical effects of toxic substances. Environ.Disturbance and Ecosystems. Springer- 184: 897-899.

Manage. 6: 543-553. Verlag, Berlin. Schindler, D. W., and M. A. Turner. 1982.

Hill, J., and R. G. Wiegert. 1980. Microcosms Odum, E. P. 1969. The strategy of ecosystem Biological, chemical and physical respons-
in ecological modelling. Pages 138-163 in development. Science 164: 262-270. es of lakes to experimental acidification.
J. P. Giesy, ed. Microcosms in Ecological 1971. Fundamentals of Ecology, 3rd Water Air Soil Pollut. 18: 259-271.
Research. DOE Symposium Series 52, ed. W. B. Saunders Co., Philadelphia, PA. Selye, H. 1973. The evolution of the stress
CONF-781101. National Technical Infor- . 1981. The effects of stress on the concept. Am. Sci. 61: 692-699.
mation Service, Springfield, VA. trajectory of ecological succession. Pages Van Voris, P., R. V. O'Neill, W. R. Eman-
Ivanovici, A. M., and W. J. Wiebe. 1981. 43-47 in G. W. Barrett and R. Rosenberg, uel, and H. H. Shugart. 1980. Functional
Towards a working definition of "stress": a eds. Stress Effects on Natural Ecosystems.complexity and ecosystem stability. Ecolo-
review and critique. Pages 13-27 in G. W. John Wiley & Sons, New York. gy 61: 1352-1360.
Barrett and R. Rosenberg, eds. Stress Ef- 1983. Basic Ecology. W. B. SaundersWilliams, W. T. 1980. Air pollution disease in
fects on Natural Ecosystems. John Wiley & Co., Philadelphia, PA. the California forests. Environ. Sci. &
Sons, New York. Odum, E. P., J. T. Finn, and E. H. Franz. Technol. 14: 179-182.

Margalef, R. 1975. Human impact on trans- 1979. Perturbation theory and the subsidy- Woodwell, G. M. 1983. The blue planet: of
portation and diversity in ecosystems. How stress gradient. BioScience 29: 349-352. wholes and parts and man. Pages 2-10 in
far is extrapolation valid? Pages 237-241 in Odum, H. T. 1967. Biological circuits and the H. A. Mooney and M. Godron, eds. Distur-
Proceedings of the First International Con- marine systems of Texas. Pages 99-157 in bance and Ecosystems. Springer-Verlag,
gress of Ecology. Centre for Agricultural T. A. Olson and F. J. Burgess, eds. Pollu- Berlin.

1
AIBS, the one organization representing biologists with
f * BioScience * Forum * Societal and Industrial Affiliates
* Public Responsibilities Program * Scientific meetings
* Special Science Programs
Your membership brings you BioScience each month, offering
* Articles on current research * News and feature stories
THE AMERICAN * Book reviews * People and places
INSTITUTE * Issues in science policy * Hardware and software evaluations
* Professional opportunities * AIBS news
OF BIOLOGICAL
I SCIENCES
Just complete this form and send to us with your payment. We will take care
of the rest.

YES! I would like an AIBS Membership. MEMBERSHIP DUES

Name D Individual (? $32.50 per year

I
[L Sustaining , $47.00 per year
Address _
(Includes $20.00 per year for BioScience)

[- Student Ca $17.00 per year

I Please charge my VISA

(Zip Code)
MasterCard
L[ Emeritus ( $17.00 per year
(Includes $10.50 per year for BioScience)

Card number Expiration date

Signature

Make checks payable to AIBS. Please remit in US dollars

only. Mail to: AIBS, 730 1 lth St., NW Washington, DC
20001-4584

L
U

422 BioScience Vol. 35 No. 7

This content downloaded from 24.11.88.25 on Fri, 22 Nov 2019 00:50:10 UTC
All use subject to https://about.jstor.org/terms