20080868 Ver 2_Supporting Documentation for WQ Conclusion_20180830

Scarbraugh, Anthony
From: Julia Berger <[e-mail]>
Sent: Thursday, August 30, 2018 9:30 AM
To: 'Steffens, Thomas A CIV USARMY CESAW (US)'; 'Lekson, David M CIV USARMY CESAS
(US)'; Scarbraugh, Anthony; Staples, Shane
Cc: 'Jeff Furness'; 'Sam Cooper'; 'Tyler Peacock'
Subject: [External] Grimm et al Babal and ecological stability and Gideon 1983
Attachments: Grimm-Wissell997_Article_BabelOrTheEcologicalStabilityD.pdf, Gideon.pdf
Good morning:
as requested yesterday at the creeks meeting, please see attached article Grimm et al 1997 which Dr. Reyes supplied
to elucidate his perspective and conclusions about water quality changes.
Although not requested yesterday, attached also is the other article he added in support (Gideon 1983).
Thank you.
Julia Kirkland Berger
Technical Director I
CZR Incorporated
4709 College Acres Drive, Suite 2
Wilmington NC 28403
910.392.9253 (o)
910.392.9139 (fl
www.czr-inc.com

Oecologia (1997) 109:323-334
Volker Grimm • Christian Wissel
© Springer -Verlag 1997
Babel, or the ecological stability discussions: an inventory and analysis
of terminology and a guide for avoiding confusion
Received: 4 June 1996 / Accepted: 5 November 1996
Abstract We present an inventory and analysis of dis-
cussions of ecological stability, considering 163 defini-
tions of 70 different stability concepts. Our aim is to de-
rive a strategy that can help to dispel the existing "confu-
sion of tongues" on the subject of "stability" and prevent
its future recurrence. The strategy consists of three ques-
tions that should be kept in mind when communicating
about stability properties. These three questions should
overcome the three main sources of confusion in termi-
nology. Firstly, which stability properties are being ad-
dressed in the stability statement? Our analysis shows
that the general term "stability" is so ambiguous as to be
useless. It can be replaced by the stability properties
"staying essentially unchanged" (constancy), "returning
to the reference state (or dynamic) after a temporary dis-
turbance" (resilience), and "persistence through time of
an ecological system" (persistence). Second, to what
ecological situation does the statement refer? An ecolog-
ical situation is defined by a set of features that, taken as
a whole, determine the domain of validity of a stability
statement. The six most important features form the
"ecological checklist", which serves to classify ecologi-
cal situations and thereby provides a system of coordi-
nates for communication. The six points are: variable of
interest, level of description, reference state, disturbance,
spatial scale and temporal scale. Thirdly, is the statement
anchored in the situation in question, or is there unac-
ceptable generalisation by inferring "stability" of the
whole system from a certain stability property in a cer-
tain ecological ecological situation? This question sepa-
rates the scientifically valuable content of a statement
from the desire for general statements which is often
projected through stability statements.
Key words Terminology • (Ecological) stability
Resilience • Persistence • Generality
V. Grimm (®) • C. Wissel
UFZ Centre for Environmental Research Leipzig -Halle,
Department of Ecological Modelling, P.O. Box 2,
D-04301 Leipzig, Germany
Introduction
Human concepts are signposts through the confusing
complexity of nature. We need them to narrow down the
never-ending tally of possible questions that we empiri-
cally or theoretically ask of nature. Without concepts it is
impossible to work scientifically. The price for this,
however, is that the concepts determine the ways and
methods in which we perceive nature. Critical examina-
tion of the concepts of their field is therefore part and
parcel of every scientist's obligations. This is particularly
the case in ecology: nearly all ecological concepts are
hotly contested (e.g. the concepts "competition" and
"density dependence"), but the contested point is the role
that these concepts actually play in nature — the concepts
themselves are relatively clear and simple to define.
How, though, should one critically evaluate a concept
if there are many overlapping or even contradictory defi-
nitions of it? This is the situation that confronts us when
dealing with the concept of "stability". "Stability" is, as
we will see, one of the most nebulous terms in the whole
of ecology. The aim of this paper is to enable the proper
critical evaluation of the term "stability". We do not in-
tend to introduce "our" version of the stability definition,
and thereby only add to the problem, but rather we will
proceed from the most quoted works on the subject of
stability (Lewontin 1969; Holling 1973; Botkin and So-
bel 1975; Orians 1975; Westman 1978; Connell and Sou-
sa 1983; Pimm 1984), which are all incisive reflections
and illustrate the diversity of stability concepts. The
point of departure of this paper is therefore an inventory
of the stability discussion in the ecological literature.
The central thesis of this paper is that the fundamental
cause of the terminological confusion stems from the fol-
lowing conflict. Stability concepts derived from mathe-
matics and physics are only suited to characterising the
dynamic behaviour of simple dynamic systems, but eco-
logical systems are not simple dynamic systems (Grimm
et al. 1992). By "simple dynamic system", we mean a
system in which, for instance, state variable, reference
states, and possible disturbances are, because of the sim-

324 0 E C 0 L 0 G I A 109 (1997) © Springer -Verlag
plicity of the system, unambiguously defined. There are
few degrees of freedom in the description of simple dy-
namic systems. Classical examples are the pendulum and
the rolling ball. Ecological systems are not simple dy-
namic systems because many degrees of freedom are al-
ways available for their description. A forest, for exam-
ple, may be described using many different variables
such as biomass, diversity, density of certain dominant
species, nutrient pools, nutrient cycling rates, buffer ca-
pacity of soil, spatial patterns, and many more. Distur-
bances may affect any of these variables in a different
way. Reactions to disturbances may depend on strength,
frequency, spatial and temporal pattern, regularity and
many other aspects of the disturbances.
The use of stability concepts makes the projection of
ecological systems on to a simple dynamic system inevi-
table, in that one is forced to decide on a particular state
variable, reference state, a particular type of disturbance
and so on. But because even the simplest ecological sys-
tems have at least a few different possible descriptions,
the use of stability concepts, without additional instru-
ments, is bound to lead to the current state of confusion.
We introduce an "ecological checklist" as an additional
instrument, which serves to structure and classify state-
ments about stability properties of ecological systems. In
the light of the ecological checklist it becomes clear that
stability concepts are no miracle cure for all the real and
imaginary "diseases" of ecology, but intellectual and
practical tools which help to improve the description, un-
derstanding and protection of ecological systems
(Grimm 1996).
Table 1 A list of stability
terms to be found in the litera-
Stability (25)
ture. Adjectives (e.g. stable,
Persistence (15)
persistent) are changed into
Constancy (5)
substantives. The numbers in
Domain of attraction (2)
parentheses denote the number
Ecological stability (6)
of definitions to be found for
Elasticity (8)
each expression. Terms marked
Resilience (17)
with an asterisk (*) are defined
Resistance (9)
in the original German. The
Functional stability
terms are classified as: (1) Con-
Global stability
ventional terms (first column);
k -Stability*
(2) newly invented terms (sec-
Lagrange stability
ond column); (3) "Stability",
Local stability
plus an adjective (third column)
Mathematical stability
Terms and definitions: an inventory
How many definitions of stability concepts and measures
are there in the literature? The answer to this question
depends on what we understand by the term "stability
concepts". Initially therefore, we want to make do with
the usual blurred definitions; that stability concepts are
concepts that have something to do with stability — what-
ever we mean by stability. A more exact definition will
be given in the next paragraph. Our collection of stability
concepts has, in the meantime, grown to 163 definitions
from 70 different stability concepts and more than 40
measures (c£ Grimm et al. 1992). Appendix 1 contains
the literature references for all the stability definitions
evaluated in this article.
Table 1 shows that ecologists usually have three ways
of defining a stability concept: Either they use a known
term and redefine it (e.g. resilience — left column in Ta-
ble 1), or they invent a new term (e.g. attractor block —
middle column in Table 1), or they extend the term "sta-
bility" by one adjective (e.g. biomass stability — right
column in Table 1). As there is no unified terminology in
ecology the authors do the only thing possible to the
terms from Table 1: they explicitly name and define
those properties of ecological systems they want to ex-
amine. But do we need so many terms and definitions?
How can one maintain a perspective with this kind of
conceptual diversity? The key question that, alone, can
establish clarity is: how many different properties of eco-
logical systems occur in all of the many definitions? If
there really were more than 50 different "stability prop-
Attractor block
Adjustment [stability]
Amplitude (4)
Anthropogenic stability
Cyclicity
Biomass stability
Damping
c -Stability*
Dynamic boundedness
Connective stability
Dynamic fragility (2)
Cyclical stability
Dynamic robustness (3)
D -stability
Ecological lability
Essential stability
Ecosystem health
Functional stability
Existence
Global stability
Hysteresis (2)
k -Stability*
Inertia (4)
Lagrange stability
Malleability (2)
Local stability
Maturity
Mathematical stability
Mutual invasibility
Multi -stability*
Permanence
Natural stability
Persistence at fixed densities
Neutral stability
Persistence in the wide sense
o -Stability*
Recurrence
Perceived stability
Regulation
Practical stability
Repellor
Qualitative stability
Resiliency (2)
Relative stability
Responsiveness
r -Stability*
Semi -stable attractor
Resistance stability (2)
Sensitivity (2)
Species deletion stability
Stable attractor
Structural stability (2)
Strictly persistent
t -Stability*
Strongly persistent
Temporal stability
Vulnerability (2)
Terminal stability
Weakly persistent
Total stability
Trajectory stability
Ultra -stability*

OECOLOGIA 109 (1997) © Springer -Verlag 325
Table 2 The basic vocabulary of the ecological stability discus- for example, means that the corresponding definition is to be
sion and classification of the definitions from the literature. First found in Connell and Sousa (1983) on page 97. Third column: lists
column: the six stability concepts in ecology which remain after of the definitions (together with references) with essentially the
"distillation" of the definitions from Table 1. Second column: liter- same properties as those in the first column, but which use a dif-
ature references for the definitions which agree in the main with ferent tern for the property
the definitions from the first column. "Connell and Sousa 83:97".
Stability term and
definition
Authors who use the term in the
first column in more or less the
same way
Terms with definitions mainly the same as in the first
column
(1) Constancy:
Connell and Sousa 83:97
Biomass stability — King and Pimm 1983:329
Returning to the
Gigon 83:97
Ecological stability* — Zw6lfer 78:15
Staying essentially
Harrison 79:661
Functional stability — Rejmanek 92:455
unchanged
Lewontin 69:21
Perceived stability — Begon et al. 90:802
temporary
Orians 75:141
Persistence — Rahel 90:328
disturbance
Remmert 89:286
Stability* — Haber 79:24
Stability — Murdoch 70:497
Stability — Putman and Wratten 85:338
Temporal stability — Preston 69:9
(2) Resilience:
Harrison 79:660
Stability — Hallet 91:383
Leps et al. 82:54
Stability — Holling 73:17
Returning to the
Putman and Wratten 85:339
Stability — Pimm 84:322
reference state (or
Ulrich 92:181
Stability — Steele 74:180
dynamic) after a
Westman 78:705
Adjustment — Connell and Sousa 83:790
temporary
Connective stability — Siljak 74:280
disturbance
Elasticity — Gigon 83:98
Elasticity* — Remmert 84:286
[Global, local] stability — Begon et al. 90:792
Mathematical stability — Danielson and Stenseth 92:83
Regulation — Murdoch 70:497
Resiliency — Kuss and Hall 91:715
Species deletion stability — Pimm 80:142
(3) Persistence:
Allen 83:4
Stability — Begon et al. 90:792
Armstrong and McGhee 76:320
Stability — Chesson and Huntly 89:293
Persistence through
Botkin and Sobel 75:629
Stability — Connell and Slatyer 77:1129
time of an ecological
Connell and Sousa 83:791
Stability — Crowley 92:246
system
DeAngelis and Waterhouse 87:7
Stability — Preston 69:7
Estberg and Patten 76:151
Stability — Roff 74:246
Harrison 79:660
Stability — Wu 76:156
Hastings 88:1666
Ecological stability — Nisbet and Gurney 82:10
Strong 90:421
Ecological stability — Wu 77:347
Warner and Chesson 85:772
Essential stability — Wu 77:352
Yodzis 89:128
Existence — Bossel 92:267
Lagrange stability — Thornton and Mulholland 74:479
Mutual invasibility — Yodzis 89:128
Persistence at fixed densities — Armstrong and McGhee 76:319
Persistence in the wide sense — Royama 77:3
Permanence — Law and Blackford 92:568
Practical stability — Thornton and Mulholland 74:483
Strictly persistent — Royama 77:2
Strongly persistent — Li 88:353
Terminal stability — Wu 76:159
Total stability — Wu 76:159
Weakly persistent — Li 88:353
(4) Resistance:
Begon et al. 90:792
Stability — Hurd and Wolf 74:465
Boesch 74:109
Stability — MacArthur 55:534
Staying essentially
Connell and Sousa 83:790
Stability — Margalef 68:12
unchanged despite the
Gigon 83:98
Stability* — Remmert 89:286
presence of
Harrison 79:660
Ecological stability — Mulholland 76:167
disturbances
Harwell et al. 81:108
Ecological stability — Rutledge et al. 76:356
Kuss and Hall 91:715
Inertia — Murdoch 70:500
Leps et al. 82:54
Inertia — Orians 74:64
Steinman et al. 90:80
Inertia — Orians 75:141
Inertia — Westman 78:705
Malleability — Westman 91:213
Resilience — Holling 73:17
Resistance stability — Sutherland 90
Responsivness — Roughgarden 75:6
Sensitivity — Estberg and Patten 76:152
Sensitivity* — Remmert 84:286
Vulnerability — Vincent and Anderson 79:218

326 0 E C 0 L 0 G I A 109 (1997) © Springer -Verlag
Table 2 (continued)
Stability term and Authors who use the term in the Terms with definitions mainly the same as in the first
definition fust column in more or less the column
same way
(5) Elasticity:
Connell and Sousa 83:790
Ecological stability — Danielson and Stenseth 92:38
Amplitude — Orians 75:141
Orians 74:64
Resilience
— Begon et al. 90:792
Speed of return to the
Orians 75:141
Resilience —
Carpenter et al. 92:784
reference state (or
Westman 78:706
Resilience —
Crowley 92:247
dynamic) after a
Westman 91:213
Resilience
— DeAngelis 80:764
temporary disturbance
Resilience
— Hallet 91:384
Elasticity — Ulrich 92:181
Resilience
— Harwell et al. 81:108
Semi -stable attractor — Byers et al. 92:25
Resilience
— Nakajima and DeAngelis 89:502
Resilience
— Pimm 84:322
Resilience
— Steinman et al. 90:80
Resilience
— Steinman et al. 91:1299
Resiliency
— Boesch 74:109
(6) Domain of Holling 73:3
Amplitude — Connell and Sousa 83:790
attraction: Pimm 84:322
Amplitude — Orians 75:141
Amplitude — Westman 78:706
The whole of states
Amplitude — Westman 91:213
from which the
Attractor block — Armstrong and McGhee 76:320
reference state (or
Dynamic fragility — Begon et al. 90:792
dynamic) can be
Dynamic fragility — May 75:163
reached again after a
Dynamic robustness — Begon et al. 90:792
temporary disturbance
Dynamic robustness — Danielson and Stenseth 92:38
Dynamically bounded — Lewontin 69:18
Dynamical robustness — May 75:163
Elasticity — Ulrich 92:181
Repellor — Byers et al. 92:26
Semi -stable attractor — Byers et al. 92:25
Stable attractor — Byers et al. 92:10
erties" then our attempt to clarify the ecological stability
discussion would be superfluous. The wealth of terms
would only reflect the multitude of properties that exist
and would therefore be irreducible. If however, there are
really only a few different properties at the root of all
these terms, then there is still hope. The answer to the
key question is surprising and encouraging: out of all of
the 163 definitions of the 70 terms there are only three
fundamentally different properties: (1) "staying essen-
tially unchanged", (2) "returning to the reference state
(or dynamic) after a temporary disturbance" and (3)
"persistence through time of an ecological system". Note
the difference between properties 1 and 3. "Staying es-
sentially unchanged" refers to a certain reference state or
dynamics (see next section), which may be an equilibri-
um, or oscillations, or irregular but limited fluctuations.
"Persistence through time", on the other hand, does not
refer to any particular dynamic but only to the question
whether a system persists as an identifiable entity (Shra-
der-Frechette and McCoy 1993; Grimm 1996). Almost
all of the 163 definitions can be categorised under one of
these three fundamental properties (see categories in Ap-
pendix 1). Only 22 definitions cannot be categorised
(Appendix 2), either because they are built on their own
terminology which is difficult to transpose into everyday
terminology (e.g. c-, t- and o- stability, multi- and ultra -
stability), or because they describe a particular ecologi-
cal situation (e.g. cyclical stability, anthropogenetic sta-
bility), or perhaps because they describe properties other
than dynamic ones (e.g. maturity, ecosystem health).
According to our assessment it should be sufficient to
work from the three fundamental properties in ecological
stability discussions. What we want, however, is not to
impose "our" assessment on the existing selection of
terms and definitions, but to inventory it. The fact is that
a few aspects of the three fundamental properties are
named so frequently in the literature that they must be
accepted as individual concepts (Table 2). The evaluation
of the collection of definitions shows that this is the case
for the following three aspects: (4) "staying essentially
unchanged despite the presence of disturbances" (an in-
terpretation of property 1), (5) "speed of return to the
reference state (or dynamic) after a temporary distur-
bance" (an aspect of property 2) and (6) "the whole of
states from which the reference state (or dynamic) can be
reached again after a temporary disturbance" (a further
aspect of property 2).
Table 2 summarises the results of our inventory and
analysis. The six definitions in the first column are for-
mulated as the commonest denominators, i.e. they
match up, either partially or wholly, with as many of the
definitions that appear in the literature as possible. This
allows us to include the majority of the 163 definitions
that we evaluated in the scheme. The second column
lists the authors whose terms and definitions agree, for
the most part, with the definitions in the first column.
The third column shows the definitions which, in the
main, describe the properties listed in the first column
but use a different term (i.e. synonyms). Table 2 can be
regarded as a kind of "interpreter" between the termi-

nologies of various authors. Apart from the 22 defini-
tions mentioned above that did not fit into the scheme,
any definitions in which two or even three of the funda-
mental stability properties were summarised together
(see Appendix 3 and the next section), were also exclud-
ed from Table 2.
For the six properties in Table 2 we chose the follow-
ing names: (1) constancy, (2) resilience, (3) persistence,
(4) resistance, (5) elasticity and (6) domain of attraction.
Table 2 shows that the terms constancy, persistence and
resistance agree with the dominant terminological usage.
With constancy and resistance however, it must be noted
that often the inverse property is considered as well, i.e.
variability instead of constancy and sensitivity instead of
resistance, but that variability and sensitivity are also
used as descriptions for the dimensions of constancy and
resistance (Grimm et al. 1992).
The choice of expression for properties 2, 5 and 6 is
more difficult. As we wish to refrain from using the term
"stability" (see next section) we are left with no alterna-
tive but to describe property 2 ["returning to the refer-
ence state (or dynamic) after a temporary disturbance"]
as resilience. Unfortunately, we must then select the term
elasticity for property 5 ["speed of return to the reference
state (or dynamic) after a temporary disturbance"] al-
though property 5 is named resilience in most defini-
tions. This confusion is the price for dispensing with
"stability" as an individual expression; a price which is,
however, worth paying (see next section). Apart from
this, properties 2 and 5 (in our terminology resilience
and elasticity) are so closely related that our choice of
terms should not cause too much confusion. No descrip-
tion of property 6 ["The whole of states from which the
reference state (or dynamic) can be reached again after a
temporary disturbance"] has generally dominated the lit-
erature. As a result we selected the expression which we
found the most descriptive: domain of attraction.
The confusion of terminology shown in Tables 1 and
2 is impressive. How could this confusion arise even
though there are only three fundamentally different sta-
bility properties? Only if we really understand the sourc-
es of the confusion will we be able to overcome it.
Causes of terminological confusion
The first source of confusion: the term "stability"
This paper deals with stability concepts. But what, in
fact, is "stability"? The literature contains more defini-
tions for this term than for any other (see Table 1) and it
is impossible to assign a universal name to these defini-
tions (see Table 2 and Appendix). It is for this reason
that "stability" does not appear in the basic vocabulary of
Table 2. The definitions of "stability" or "ecological sta-
bility" can be divided into two groups. The first group
includes the definitions in which "stability" is identified
by one of the properties constancy, resilience, persis-
tence, resistance or elasticity. None of these definitions
has a chance of succeeding because if "stability" can be
OECOLOGIA 109 (1997) © Springer -Verlag 327
identified with different properties then the expression
must have more than one meaning. The second group of
definitions takes account of ambiguity, insofar that "sta-
bility" is simultaneously placed on an equal footing with
two or even three properties. Appendix 2 shows that
here, too, different authors attribute different properties
to "stability" according to their own particular point of
view. In this way, the meaning of "stability" becomes
even more ambiguous. A further hindrance to salvaging
the term "stability" which comes to light is the view that
"stability" refers to a property. But how can one property
be simultaneously two (or even three) properties? If a
system in a certain situation displays resilience but no re-
sistance, whilst the relationship in a second system is ex-
actly the opposite, which of the two systems is then "sta-
ble" or "unstable"?
We therefore go one vital step further than the authors
of the expressions in Appendix 2 and use "stability" only
as a common qualifier for the six properties listed in Ta-
ble 2. We call a property a "stability property" if it can
be filed under any of the three fundamental properties
constancy, resilience or persistence. The vital difference
from previous definitions of "stability" is that we no lon-
ger define "stability" directly (because it is impossible)
but via the three fundamental properties. "Stability" it-
self is not a property but a terminological link that em-
phasises the close connection between the three funda-
mental properties (what this connection looks like de-
pends on the individual case). "Stability" makes sense as
a collective term because it prevents us from regarding
properties like, for example, resilience and persistence,
in isolation. "Stability" is thus only a generic term which
cannot be used as a qualifier in particular statements
about ecological systems. The most important conse-
quence of our central definition of stability properties is
the simple but important conclusion: "stability" is not a
stability property!
Many authors try to overcome the uncertainty of the
expression "stability" by adding an adjective to it (Ta-
ble 1). Unfortunately the adjective is not enough to over-
come the ambiguity of the term "stability". In this way
"stability" in "biomass stability" means constancy (King
and Pimm 1983), in "essential stability" persistence (Wu
1977), in "elastic stability" resilience (Gigon 1983), and
so on (see Table 2). These expressions may suffice as lo-
cal terminology, i.e. within one article, but for a proper
understanding of the ecological stability discussion they
are an unnecessary obstacle in that they cannot remove
the ambiguity from "stability".
It might be better to delete the descriptions "stable",
"stabilising" and "stability" as individual expressions
completely from the ecological vocabulary. It would be
naive, however, to believe that the word "stability"
could, or should be deleted from ecology. Therefore we
propose to use "stability" only as a short form or substi-
tute for "stability properties". This makes it possible to
continue working with expressions that contain the word
"stability", without having to identify "stability" with
one single property: A "stability definition" is the defini-
tion of a stability property and a "stability term" is the

328 0 E C 0 L 0 G I A 109 (1997) © Springer -Verlag
description of this property. Stability terms and defini-
tions together form the "stability concept". A "stability
measure" is a measure for the quantitative assessment of
a stability property. There are generally several measures
to a stability property. The results of the use of stability
concepts in ecology are "stability statements", i.e. as-
sessments of stability properties with the aid of stability
measures. Finally, by "stability mechanisms" we mean
the mechanisms that are responsible for certain stability
properties.
Let us summarise the first results of our analysis of
the causes of confusion. "Stability" cannot be unambigu-
ously defined and is therefore unusable as a concept. The
ambiguity of the term "stability" is one of the main
sources of confusion in the ecological stability discus-
sion. To express oneself clearly one must abandon "sta-
bility" as an individual expression. The central role in
discussions of ecological stability must now be adopted
by the individual stability properties instead by "stabili-
ty" itself.
The second source of confusion: the diversity
of ecological situations
If "stability", as an individual expression, were to disap-
pear from ecological literature tomorrow, a great deal
would already be gained. Instead of "stability" everyone
would, at last, say what they actually mean: constancy,
resilience or persistence. But the ambiguity of "stability"
only partly explains the diversity of definitions. Even if
we were to delete "stability" from Tables 1 and 2, there
would still be enough confusion to require an explana-
tion. Where does this confusion come from if stability
concepts are so simple? The source must lie in ecology
itself. The critical point with the use of stability concepts
in ecology is the fact that simple definitions of stability
concepts provide no information as to how they should
be applied, in the anything -but -simple world of ecology
(Botkin 1981). The problem is that, in ecological sys-
tems there is never just one but always a multitude of po-
tential ways to apply stability concepts!
To clarify this point let us consider a simple dynamic
system which is often used for observing stability con-
cepts: a ball rolling over valleys and hills (Botkin 1981
uses a pendulum as an example). The whole system is so
simple that the state variable (position), reference state
(equilibrium) and disturbance (displacement) are all un-
ambiguously determined. In ecology the opposite always
applies: none of these features is automatically deter-
mined. Even the simplest ecological systems are so com-
plex that there are always many different possible vari-
ables, reference states and disturbances. Consider, for
example, the following three systems:
1. A metapopulation inhabiting a landscape consisting
of many small and few large habitats. An assessment of
the persistence of the population strongly depends on the
choice of the state variable: overall population size, oc-
cupation of patches, or spatially explicit distribution of
occupied and empty patches.
2. A forest where outbreaks of herbivorous insects occur
(for example, larch bud moth in alpine forests: Bal-
tensweiler and Fischlin 1988). An assessment of con-
stancy and resilience requires the definition of a refer-
ence state, but what is considered as a reference state (or
dynamic) depends on the spatial and temporal scale, on
the species considered, whether soil properties are part
of the system description, and so on (Gigon and Grimm
1997).
3. A benthic community in a tidal flat. It may be charac-
terized by, for instance, species diversity, by dominance
relationships, interactions with predators, biodeposition,
hydrodynamic processes, regional larvae pools, storms,
or ice drifts. These examples illustrate what has long
been known:
"Ashby (1952) pointed out that every situation in nature
can be described in an infinite number of ways. An ob-
server can choose any arbitrary set of variables and pa-
rameters to define an abstract system, which then can be
the object of study by empirical and theoretical methods.
Of course, some systems are more advantageous than
others for making further progress."
(Hall and DeAngelis 1985, p.340; see also Schaffer
1981), which however has largely been ignored when ap-
plying stability concepts in ecology.
We call a certain combination of features which char-
acterize a "situation in nature" an "ecological situation".
With the introduction of ecological situations we can
now formulate another important sentence. The domain
of validity of a stability statement is delimited by the eco-
logical situation under observation. If, for example, the
resilience of the numbers of individuals in a population
is assessed during a small short-term disturbance in the
numbers of individuals (e.g. through hunting), then the
assessment is only valid for that particular situation. It
gives no indications about how, for example, the age
structure of the population would respond to large, peri-
odic disturbances in the environment. Rahel (1990) dem-
onstrates how the assessment of constancy in a commu-
nity depends on the variables chosen. Communities qual-
ified as highly variable when using the variables "abso-
lute abundance" or "abundance ranking" may still be
constant when using the variable "presence or absence of
species over time".
Stability statements can only relate to particular eco-
logical situations. As soon as the situation changes (e.g.
consideration of another state variable, different distur-
bances, different reference states), the stability statement
is, ipso facto, no longer valid. The widespread practice in
ecology of transferring stability statements from a spe-
cial ecological situation to the whole system is psycho-
logically understandable, but scientifically wrong.
The assessment of a stability property is inseparably
linked to the ecological situation in question. This fact
has two consequences. Firstly, a lot of authors integrate
"their" ecological situation into the definition of "their"

Table 3 An ecological checklist. The list consists of the six char-
acteristics of an ecological situation which delimit the domain of
validity of a stability statement. The checklist points are listed as
OECOLOGIA 109 (1997) © Springer -Verlag 329
questions in the second column. A few typical answers to these
questions are listed in the third column
Features of the ecological Checklist question for this feature Example answers
situation
(1) Level of description
On what level of description is the
Individual, population, community,
stability property examined?
ecosystem, landscape, ...
(2) Variable of interest
Which ecological variable of interest
Biomass, population size, age
is being considered?
structure, nutrient cycling rate, spatial
patterns, ...
(3) Reference state or reference
What is the reference state or dynamic
Equilibrium, trend, cycles, high or low
dynamic, respectively
of the variable of interest without
spatial or temporal variability, ...
external influences?
(4) Disturbance What does the disturbance look like?
What is being disturbed?
(5) Spatial scale To which spatial scale does the
stability statement refer?
(6) Temporal scale To which temporal scale does the
stability statement refer?
stability concepts (e.g. biomass as variable, equilibrium
as reference state). This is the second main reason for the
flood of definitions referred to above. Secondly, the com-
plete characterisation of the ecological situation in ques-
tion is an essential part of every stability statement.
Without this the statement cannot be evaluated and real
communication is rendered impossible.
How can one bring these two problems under control?
We have already introduced the solution to consequence
1 (first column in Table 2). Stability concepts are con-
cepts for characterising certain dynamic properties. Their
definitions should not refer to ecological systems (except
the concept of persistence: cf. Grimm 1996). For conse-
quence 2 we fall back on the aviator's tried and tested
solution — a checklist. Pilots cannot afford to forget vital
points before take -off because they may crash. In the
same way, ecologists cannot afford to forget essential
points for characterising the ecological situation, other-
wise their ability to communicate about stability proper-
ties may "crash".
The "ecological checklist" consists of the features
that are most important for characterising an ecological
situation (Table 3). We compiled the list by asking our-
selves which features of an investigation in which stabili-
ty properties are to be evaluated determine the domain of
validity of the resulting stability statements. Or, to put it
another way, what features of the ecological situation un-
der examination must be communicated in order that the
stability statements can be fully interpreted and com-
pared to others? The six points on the checklist are brief-
ly discussed below (c£ Orians 1975; Connell and Slatyer
Disturbance of the state variable or of
a system parameter, lasting
disturbance or short term effect,
intensity of the disturbance, frequency
of the disturbance, ...
Size of the researched area, ability of
the researched species to spread,
typical lengths in the spatial
heterogeneity of the research area, ...
Time horizon of the statement,
longevity of the examined organisms,
temporal structure in the
environmental heterogeneity,...
1977; Connell and Sousa 1983; Gigon 1983, 1984;
Pimm 1984; Schwegler 1985; Grimm et al. 1992; Grimm
1996; Gigon and Grimm 1997):
1. Level of description. Evaluations of stability proper-
ties are bound to the observed level of description (e.g.
individual, population, ecosystem).
2. Variable of interest. Regardless of the ecological
system under observation there are always various de-
scriptive variables which can show different dynamic be-
haviour. The variable of interest is thus an integral part
of every stability statement.
3. Reference state or reference dynamic. In order to de-
cide what is to be regarded as a disturbance and whether
a system returns to its "normal" state after a disturbance,
one needs an idea of what the system's "normal" state or,
"reference dynamic" actually is. Often there are different
definitions of a reference dynamic, and the evaluation of
the stability properties resilience and resistance vary cor-
respondingly.
4. Disturbance. "Disturbances" are factors which do not
belong to the system (in contrast to "natural distur-
bances", which are an integral part of the system: Pickett
and White 1985). It is not however sufficient simply to
talk about "disturbances". Statements must be made
about what is being disturbed and what are the spatial
and temporal characteristics of the disturbance. In preda-
tor -prey systems, for example, the reaction to distur-
bances strongly depends on the frequency of the distur-
bance and the parameter of the model being disturbed
(Schmidt and Wissel 1991; Grimm et al. 1992).

330 0 E C 0 L 0 G I A 109 (1997) © Springer -Verlag
5. Spatial scale. Stability statements are initially only
valid within the observed spatial framework. The stabili-
ty concepts constancy and resilience have, as a rule, no
single "correct" framework or standard for a particular
investigation (cf. Levin 1992). For the evaluation of per-
sistence, though, the "correct" scale is determined by the
system definition which is being employed. The research
area must be big enough for the processes which com-
prise the system to occur within it (Connell and Sousa
1983).
6. Temporal scale. The same applies here (mutatis mu-
tandis) as for the spatial framework.
The checklist points are also formulated as questions
in Table 3. Answering all six questions defines an eco-
logical situation to which a stability statement can relate.
If only one of the features is changed (e.g. a different
disturbance or time scale) then we have a new ecological
situation and the old stability statements will, in general,
no longer be valid.
The checklist fulfils three main functions. Firstly, it
acts as a safety gauge which constantly monitors the do-
main of validity of statements during their construction
and interpretation. It particularly helps us to test whether
the conclusions from a stability investigation are an-
chored in the ecological situation in question or whether
there has been unacceptable generalisation, i.e. a transfer
to other situations or even the whole system. Secondly, it
forms a sort of system of coordinates for communication
into which stability statements of similar, but also of
quite different, investigations can be organised. In this
way we achieve comparable statements and thereby ob-
tain a chance of leaving the domain of validity of indi-
vidual statements behind in favour of a holistic view of
ecological systems.Thirdly, our checklist helps with the
planning of stability investigations, whether empirical or
theoretical, in that it exposes the gaps in previous re-
search programmes (Grimm et al. 1992). The checklist
enables us to tackle the problem of evaluating stability
properties systematically.
Complete stability statements are also called for in
other papers (Orians 1975; Connell and Slatyer 1977;
Connell and Sousa 1983; Gigon 1983, 1984; Pimm
1984; Schwegler 1985). An explanation for the lack of
effect of these papers on the woolly state of the ecologi-
cal stability discussion may be that none of them explic-
itly focused on an terminological and conceptual analy-
sis of the stability discussion. Apart from this, the most
important reason, in our opinion, that necessitates a
checklist was not explicitly addressed, i.e. that stability
concepts were not invented for the investigation of eco-
logical systems but for the characterisation of simple dy-
namic systems. Without an additional instrument their
application in ecology causes the kind of chaos that we
have documented in Table 2. The ecological checklist is
the additional instrument that should prevent this chaos.
It is, of course, quite possible to have a completely
different opinion from us on the matter of the relation-
ship between ecological and dynamic systems. Berryman
(1991), for example, reaches exactly the opposite conclu-
sion: "The continuing confusion over the concepts of
regulation, stability and density -dependence results from
our refusal to recognize that ecosystems obey the same
rules as all other dynamic systems, both natural and en-
gineered" (Berryman 1991, p.142). Berryman would be
right, if there were such a thing as an unambiguous con-
vertible representation of ecological systems to simple
dynamic systems. However, there is no such thing, — nei-
ther for populations, nor for communities, nor for eco-
systems. Our point of view and Berryman's are not how-
ever mutually exclusive. They can, to some extent, be
reconciled with the aid of "dynamic constraints" (R.
Law, personal communication). Dynamic constraints on-
ly come into play in extreme situations, e.g. with very
high densities in a population, or with extreme condi-
tions in the abiotic environment. The extreme conditions
in such situations limit the relational possibilities of the
ecological system in such a way as to allow an unambig-
uous projection into a simple dynamic system. The con-
cept of density -vagueness suggested by Strong (1986a,
b) is an example of an attempt to couple the dynamic
concept, "density dependence", with the idea of dynamic
constraints. Density dependence only comes into play
when the density of a population exceeds certain ceilings
or falls below certain floors. There is no unambiguous
relationship between density and growth rate of the
population for densities between these limits, because
one and the same density can lead to quite different lev-
els of reproductive success depending on supply of re-
sources and variability within a population (Grimm and
Uchmanski 1994).
The third source of confusion: "stability"
and the hope for a more mature ecology
If the widespread misconception that scientists are only
guided in their work by their rationality were true, then
there would probably be no need to write this paper. The
results of our analysis to date, which are no more than
simple guidelines, show that: (1) there are few genuinely
different stability concepts; (2) "stability" alone is so
ambiguous as to be unusable as a concept; and (3) stabil-
ity statements can only relate to clearly defined ecologi-
cal situations. One really would not need to work
through mountains of ecological literature to arrive at
these conclusions; a little healthy human intuition and,
when in doubt, a flick through the dictionary, would suf-
fice. How is it that ecologists pay so little heed to these
guidelines? The answer is that scientists are no less sub-
ject to their desires, hopes and dreams in their work than
anyone else (Grimm 1996). In this section we want to
show how the desires, hopes and dreams which are at-
tached to the term "stability", form a significant barrier
to the fruitful communication of stability properties.
The fact is that the term "stability" has an enormous
attraction for ecologists. One cannot help but notice that
"ecologists are happy to talk about `stability' (Pimm

1991, p. xii), and politicians, managers and naturalists
really love it. Ecological research is no less reputable or
intensive than research in other disciplines, but it is frus-
trating, in the context of the successes in other disci-
plines, to see so many ecological studies standing more
or less in isolation whilst a general ecological theory lies
far off on the horizon (McIntosh 1985). Most ecological
studies only allow statements about a particular system
and even these statements are, as a rule, of extremely
limited validity. The spatial and temporal scales of the
experiments are almost always small; only a tiny section
of the species spectrum can be encompassed, the influ-
ence of abiotic factors cannot be systematically analysed
and so on. The list of limitations can quite easily be ex-
tended. The results of ecological studies fit together
more like a kaleidoscope than jigsaw pieces, in that it is
often a distorted picture of nature that we see and not a
meaningful, consistent picture which would allow us to
reach predictively into reality.
The preoccupation with "stability" seems to offer a
way out of this. "Stability" stands for the hope of over-
coming the current state in ecology. Schwegler (1985)
expresses this hope clearly: "Stability belongs to the ex-
pressions (like information and energy) of which, some-
times, a global explanatory power is expected and which
is supposed to make tedious attention to detail more or
less superfluous." (p. 263; translated from German).
Working on "stability", or at least speculating about
it, gives one the feeling of assisting in a "master plan"
(Whittaker and Levin 1977): "Ecologists have sought a
theory or master plan of evolution permitting interpreta-
tion of communities through a limited number of strong-
ly linked and widely significant relationships. Such a
theory is naturally desired by ecologists as scientists"
(Whittaker and Levin 1977, p.136).
We are convinced that there is an important nucleus
buried amongst the mixture of woolliness, unrealised
dreams and ambition that characterises the current state
of communication of stability properties (Grimm 1996).
As Pimm (1991) rightly indicates,"just because the terms
are fuzzy, this does not mean that the underlying ideas
are unimportant."(p. 4). But in order to solidify the "un-
derlying ideas", we must first discover why the expres-
sions are unclear and second, how clear expressions are
to be achieved. Exactly this is the aim of this paper.
A strategy for avoiding confusion
The results of our inventory and analysis of the ecologi-
cal stability discussion can be summarised as a simple
strategy which can help avoid confusion in the communi-
cation of stability properties. The strategy consists of
questions that help to analyse statements one may en-
counter in publications or in discussions and to classify
them so as to enable comparison with other stability
statements. The confusion -avoidance strategy is as fol-
lows. Every time we encounter either the word "stabili-
ty", a stability concept or a statement about a stability
OECOLOGIA 109 (1997) (D Springer -Verlag 331
property in the literature or in a discussion, we try to find
answers to the following three questions:
1. What stability property is being addressed?
2. To which ecological situation does the stability state-
ment refer?
3. Is the statement anchored in the situation under con-
sideration or is there unacceptable generalisation?
The answer to question 1 sweeps away all of the ter-
minological uncertainty by concentrating on the central
elements in the ecological stability discussion — the sta-
bility properties. The basic vocabulary "distilled" by us
helps to answer this question (first column in Table 2).
The ecological checklist is an instrument for answering
question 2. The uncertainty about the stability state-
ments' validity limits can be clarified as much as possi-
ble by answering the individual points on the checklist.
If no answer is possible for some of the points on the
checklist (for example, the questions about spatial and
temporal scales in many mathematical models), then the
statement is encumbered with uncertainty that cannot be
further diminished. Question 3 is for the most part cov-
ered by question 2 but should, once again, encourage the
explicit evaluation of requirement and reality in a publi-
cation or a discussion. It is designed as a safety mecha-
nism against the above-named irrational sources of con-
fusion. We believe that this confusion -avoidance strategy
can, despite or even because of its simplicity, be an ef-
fective instrument in clearing up the current confusion in
the ecological stability discussion as well as preventing
its future recurrence.
Acknowledgements We thank R. Brandl, A. Gigon, A. Huth, K.
Jax, F. Jeltsch, R. Law, R.P. McIntosh, C. Neuert, A. Ratz, E.
Schmidt, D.R. Strong, and M. Williamson for valuable comments
on earlier drafts of this paper.
Appendix 1
Literature references for all stability definitions which have been
evaluated in this paper
For each reference, the terms defined in that paper are listed (in-
cluding the page number where the definition can be found). The
abbreviations in parentheses denote which stability property (ac-
cording to the first column in Table 2) is described by that particu-
lar definition. Abbreviations: ili resilience, ist resistance, con con-
stancy, per persistence, dom domain of attraction, ela elasticity, ?
cannot be classified, + two or more properties are grouped togeth-
er in one definition, mat the language of mathematics is used for
the definition, mats a pure mathematical definition.
This list of references does not claim to be complete, nor is in-
clusion in or exclusion from the list linked to any evaluation of
single stability definitions. The only aim of this list is to document
the extreme diversity of terms and definitions (references to fur-
ther terms and definitions are given in Williamson 1972, 1984,
1987; Connell and Sousa 1983; Pimm 1991; Hutson and Schmitt
1992; Shrader-Frechette and McCoy 1993). The full definitions of
the references listed below are available via WWW server
http://www.oesa.ufz.de.
Allen (1983) persistenceA (per, mat)
Armstrong and McGhee (1976) persistence at fixed densities:319
(per, mat), attractor block:320 (dom, mat), persistence:320
(per, mat)

332 0 E C 0 L 0 G I A 109 (1997) © Springer -Verlag
Begon et at. (1990) dynamic fragility:792 (dom), dynamic robust-
ness:792 (dom), global stability:792 (ili), local stability:792
(ili), resilience:792 (ela), resistance:792 (ist), stability:792
(per), perceived stability:802 (con)
Boesch (1974) persistence:109 (per+con), resiliency:109 (ela), re-
sistance:109 (ist), stability: 109 (ili+ist),
Bossel (1992) existence:267 (per, mat)
Botkin and Sobel (1975) persistence:629 (per, mat), persis-
tence:630 (per, mat), recurrence:631 (per, mat)
Byers et al. (1992) stable attractor:10 (dom, mat), semi -stable at-
tractor:25 (dom, mat), repellor:26 (dom, mat)
Carpenter et al. (1992) resilience:784 (ela)
Chesson and Huntly (1989) stability:293 (per)
Connell and Slayer (1977) stability: 1129 (per)
Connell and Sousa (1983) adjustment [stability]:790 (ili), ampli-
tude:790 (dom), constancy:790 (con), elasticity:790 (ela), re-
sistance:790 (ist), stability:790 (ili+ist), persistence:791 (per)
Crowley (1992) resilience:247 (ela), stability:246 (per),
Danielson and Stenseth (1992) dynamic robustness:38 (dom), eco-
logical stability:38 (ela), mathematically stable:38 (ili)
DeAngelis (1980) relative stability:764 (ist+ela), resilience:764
(ela)
DeAngelis and Waterhouse (1987) persistence:7 (per)
Diamond (1976) stability:287 (?)
Estberg and Patten (1976) persistence:151 (per, mat), sensitivi-
ty:152 (ist)
Gigon (1983) ecological lability:96 (?), ecological stability:96
(per+ili), anthropogenetic stability:97 (?), constancy:97 (con),
cyclicity:97 (?), elasticity:98 (ili), natural stability:97 (?), re-
sistance:98 (ist)
Gigon (1984) 6kologische Stabilitat:14 (per+ili), 6kologische
Labilitat:15 (?), Konstanz:17 (con), Elastizitat:20 (ili), Resi-
stenz:20 (ist), Zyklizitat:20 (?)
Gob (1980) vulnerability:3 (?, mat)
Haber (1979) Persistenz (=k-Stabilitat):22 (per con), Resilienz
(=r-Stabilitat):22 (ili+con), Stabilitat:24 (con)
Hallet (1991) stability:383 (ili), resilience:384 (ela), D-stabili-
ty:386 (matt)
Harrison (1979) stability:659 (ili+ist), persistence:660 (per), resil-
ience:660 (ili), resistance:660 (ist), constancy:661 (con)
Harwell et al. (198 1) resilience: 108 (ela), resistance: 108 (ist)
Hastings (1988) persistence: 1666 (per)
Holling (1973) domain of attraction:3 (dom), neutral stability:3
(?), resilience: 17 (ist), stability: 17 (ili)
Hurd and Wolf (1974) stability:465 (ist)
King and Pimm (1983) biomass stability:229 (con)
Kuss and Hall (1991) resiliency:715 (ili), resistance:715 (ist)
Law and Blackford (1992) permanence:568 (per, mat)
Leps et al. (1982) resilience:54 (ili), resistance:54 (ist)
Lewontin (1969) dynamically bounded:18 (dom, mat), constan-
cy:21 (con, mat), structural stability:22 (mat')
Li (1988) strongly persistent:353 (per, mat), weakly persistent:353
(per, mat)
MacArthur (1955) stability:534 (ist)
Margalef (1968) stability: 12 (ist)
Margalef (1974) maturity: 66 (?)
May (1973) qualitative stability:639 (mat')
May (1975) dynamic fragility:163 (dom), dynamical robust-
ness: 163 (dom)
Minns (1992) ecosystem health:110 (?)
Mulholland (1976) ecological stability: 167 (ist)
Murdoch WW (1970) regulation:497 (ili), stability:497 (con), in-
ertia:500 (ist)
Nakajima and DeAngelis (1989) resilience:502 (ela)
Nisbet and Gurney (1982) ecological stability: 10 (per)
Orians (1974) elasticity:64 (ela), inertia:64 (ist)
Orians (1975) amplitude: 141 (dom), constancy: 141 (con), elastici-
ty:141 (ela), inertia: 141 (ist), stability: 141 (ili+con), cyclical
stability: 143 (?), trajectory stability: 143 (?)
Pimm (1980) species deletion stability: 142 (ili)
Pimm (1984) domain of attraction:322 (dom), resilience:322 (ela),
stability:322 (ili)
Preston (1969) stability:7 (per), temporally stable:9 (con)
Putman and Wratten (1985) stability:338 (con), resilience:339 (ili)
Rahel (1990) persistence:328 (con)
Rejm'anek (1992) functional stability:455 (con)
Remmert (1989) Elastizitat:286 (ili), Empfindlichkeit:286 (ist),
Konstanz:286 (con), Stabilitat:286 (ist)
Roff(1974) stability:246 (per)
Roughgarden (1975) responsivness:6 (ist)
Royama (1977) strictly persistent:2 (per), persistence in the wide
sense:3 (per)
Rutledge et al. (1976) ecological stability:356 (ist)
Schwegler (1981) c-Stabilitat:131 (?), t-Stabilitat:134 (?), o-Stabi-
litat:136 (?)
Siljak (1974) connective stability:280 (ili)
Smedes and Hurd (198 1) resistance stability: 1561 (ist+ili)
Steele (1974) stability:180 (ili)
Steinman et al. (1990) resistance:80 (ist), resilience:80 (ela)
Steinman et al. (199 1) resilience: 1299 (ela)
St6cker (1974) Stabilitat:241 (ili+ist), Ultrastabilitat:245 (ist+?),
Multistabilitat:245 (?)
Strong (1990) persistence:421 (per)
Sutherland (1990) resistance stability: 1990 (ist)
Suttman and Barrett (1979) stability:637 (ili+ist+ela)
Thornton and Mulholland (1974) Lagrange stability:479 (per,
mat), practical stability:483 (per)
Ulrich (1992) elasticity: 181 (dom), resilience: 181 (ili)
Vincent and Anderson (1979) vulnerability: 218 (ist)
Warner and Chesson (1985) persistence:772 (per, mat)
Westman (1978) resilience:705 (ili), inertia:705 (ist), ampli-
tude:706 (dom), elasticity:706 (ela), hysteresis:706 (?), mal-
leability:706 (?)
Westman (1991) amplitude:213 (dom), damping:213 (?), elastici-
ty:213 (ela), hysteresis:213 (?), malleability: 213 (ist)
Wu (1976) stability:156 (per), terminal stability:159 (per), total
stability: 159 (per)
Wu (1977) ecological stability:347 (per), essential stability:352
(per, mat)
Yodzis (1989) structural stability:32 (matt), mutual invasibi-
lity:128 (per) persistence: 128 (per)
Zw6lfer (1978) 6kologische Stabilitat:15 (con)
Appendix 2
Definitions which do not fit into our classification scheme
in Table 2
Asterisk * denotes definitions which are in original German, mat
denotes definitions that were not included in Table 2 because they
are purely mathematical
Anthropogenetic stability — Gigon (1983):97
Cyclical stability — Orians (1975):143
Cyclicity — Gigon (1983):97
D -stability— Hallet (1991):386 (mat)
Ecological lability — Gigon (1983):96
Ecosystem health — Minns (1992):110
Hysteresis — Westman (1978):706
Hysteresis — Westman (1991):213
Malleability — Westman (1978):706
Maturity — Margalef (1974):66
Natural stability — Gigon (1983):97
Neutral stability — Holling (1973):3
c -stability* — Schwegler (1981):131
Trajectory stability — Orians (1975):143
t -stability* — Schwegler (1981):134
o -stability* — Schwegler (1981):136
Qualitative stability — May (1973):639 (mat)
Recurrence — Botkin and Sobel (1975):631
Structural stability — Lewontin (1969):22, Yodzis (1989):32 (mat)
Ultra -stability* — Stocker (1974):245
Multi -stability* — Stocker (1974):245
Vulnerability — Goh (1980):3 (mat)

Appendix 3
Definitions that combine two or three different stability properties
For abbreviations see Appendix 1.
Resilience* (=r -stability*) — Haber (1979):22 (ili+con)
Stability — Orians (1975):141 (ili+con)
Stability — Boesch (1974):109 (ili+ist)
Stability — Diamond (1976):287 (ili+per)
Stability — Connell and Sousa (1983):790 (ili+ist)
Stability — Harrison (1979):659 (ili+ist)
Resistance stability — Smedes and Hurd (1981):1561 (ili+ist)
Stability* — St6cker (1974):241 (ili+ist)
Stability — Suttman and Barrett (1979):637 (ili+ist+ela)
Relative stability — DeAngelis (1980):764 (ist+ela)
Persistence* (=k -stability*) — Haber (1979):22 (per+con)
Persistence — Boesch (1974):109 (per+con)
Ecological stability* — Gigon (1983):96 (per+ili)
References
Allen LJS (1983) Persistence and extinction in Lotka-Volterra re-
action -diffusion equations. Math Biosci 65:1-12
Armstrong RA, McGehee R (1976) Coexistence of species com-
peting for shared resources. Theor Popul Biot 9:317-328
Baltensweiler W, Fischlin A (1988) The larch bud moth in the
Alps. In: Berryman AA (ed) Dynamics of forest insect popula-
tions. Plenum Press New York, pp 332-351
Begon M, Harper JL, Townsend CR (1990) Ecology, 2nd edn.
Blackwell, Oxford
Berryman AA (1991) Stabilization or regulation: what it all
means. Oecologia 86:140-143
Boesch DF (1974) Diversity, stability and response to human dis-
turbance in estuarine ecosystems. In: Proceedings of the first
international congress of ecology. Pudoc, Wageningen, pp
109-114
Bossel H (1992) Real -structure process description as the basis of
understanding ecosystems and their development. Ecol Model
63:261-276
Botkin DB (198 1) A grandfather clock down the staircase: stabili-
ty and disturbance in natural ecosystems. In: Waring RH (ed)
Forests: fresh perspectives from ecosystem analysis. Oregon
State University Press, Corvallis, pp 1-10
Botkin DB, Sobel MJ (1975) Stability in time -varying ecosystems.
Am Nat 109:625-646
Byers RE, Hanssell RIC, Madras N (1992) Stability -like proper-
ties of population models. Theor Popul Biol 42:10-34
Carpenter SR, Kraft CE, Wright R, He X, Soranno RA, Hodgson
JR (1992) Resilience and resistance of a lake phosphorus cycle
before and after food web manipulation. Am Nat 140:781-798
Chesson PL, Huntly N (1989) Short-term instabilities and long-
term community dynamics. Trends Ecol Evo14:293-298
Connell JH, Slatyer RO (1977) Mechanisms of succession in natu-
ral communities and their role in community stability and or-
ganiziation. Am Nat 111:1119-1144
Connell JH, Sousa WP (1983) On the evidence needed to judge
ecological stability or persistence. Am Nat 121:789-824
Crowley PH (1992) Densitity dependence, boundedness, and at-
traction: detecting stability in stochastic systems. Oecologia
90:246-252
Danielson BJ, Stenseth NC (1992) The ecological and evolution-
ary implications of recruitment for competitively structured
communities. Oikos 65:34-44
DeAngelis DL (1980) Energy flow, nutrient cycling, and ecosys-
tem resilience. Ecology 61:764-771
DeAngelis DL, Waterhouse JC (1987) Equilibrium and nonequi-
librium concepts in ecological models. Ecol Monogr 57:
1-21
Diamond P (1976) Domains of stability and resilience for biologi-
cal populations obeying difference equations. J Theor Biol
61:287-306
OECOLOGIA 109 (1997) © Springer -Verlag 333
Estberg GN, Patten BC (1976) The relation between sensitivity
and persistence under small perturbations. In: Levin SA (ed)
Ecosystem analysis and prediction. Society for Industrial and
Applied Mathematics, Philadelphia, pp 151-154
Gigon A (1983) Typology and principles of ecological stability
and instability. Mountain Res Development 3:95-102
Gigon A (1984) Typologie and Erfassung der 6kologischen Stabi-
litat and Instabilitat mit Beispielen aus Gebirgs6kosystemen.
Verh Ges Okol 12:13-29
Gigon A, Grimm V (1997) Stabilitatskonzepte in der Okologie:
Typologie and Checkliste fdr die Anwendung. In: Franzle O,
Muller F, Schr6der W (eds) Handbuch der Okosystemfor-
schung. Ecomed, Landsberg (in press)
Gob BS (1980) Management and analysis of biological popula-
tions. Elsevier, Amsterdam
Grimm V (1996) A down-to-earth assessment of stability concepts
in ecology: dreams, demands, and the real problems. Sencken-
bergiana Mar 27:215-226
Grimm V, Uchmanski J (1994) Dichteabhangigkeit and asymme-
trische Konkurrenz zwischen Individuen: ein individuenbasier-
tes Modell. Verh Ges Okol 23:311-319
Grimm V, Schmidt E, Wissel C (1992) On the application of sta-
bility concepts in ecology. Ecol Model 63:143-161
Haber W (1979) Theoretische Anmerkungen zur "6kologischen
Planung". Verh Ges Okol 7:19-30
Hall CAS, DeAngelis DL (1985) Models in ecology: paradigms
found or paradigms lost? Bull Ecol Soc Am 66:339-346
Hallet JG (1991) The structure and stability of small mammal fau-
nas. Oecologia 88:383-393
Harrison GW (1979) Stability under environmental stress: resis-
tance, resilience, persistence, and variability. Am Nat 113:
659-669
Harwell MA, Cropper WP, Ragsdale HL (198 1) Analyses of tran-
sient characteristics of a nutrient cycling model. Ecol Model
12:105-131
Hastings A (1988) Food web theory and stability. Ecology 69:
1665-1668
Holling CS (1973) Resilience and stability of ecological systems.
Annu Rev Ecol Syst 4:1-23
Hurd LE, Wolf LL (1974) Stability in relation to nutrient enrich-
ment in arthropod consumers of old -field successional ecosys-
tems. Ecol Monogr 44:465-482
Hutson V, Schmitt K (1992) Permanence and the dynamics of bio-
logical systems. Math Biosci 111:1-72
King AW, Pimm SL (1983) Complexity, diversity, and stability: a
reconciliation of theoretical and empirical studies. Am Nat
122:145-149
Kuss FR, Hall CN (1991) Ground flora trampling studies five
years after closure. Environ Manage 15:715-728
Law R, Blackford JC (1992) Self -assembling food webs: a global
viewpoint of coexistence of species in Lotka-Volterra commu-
nities. Ecology 73:567-578
Leps J, Osbornova-Kosinova J, Rejmanek M (1982) Community
stability, complexity and species life history strategies. Ve-
getatio 50:53-63
Levin SA (1992) The problem of pattern and scale in ecology.
Ecology 73:1943-1967
Lewontin RC (1969) The meaning of stability. Brookhaven Symp
Biol 22:13-24
Li J (1988) Persistance in discrete age -structured population mod-
els. Bull Math Biol 50:351-366
MacArthur RH (1955) Fluctuations of animal populations and a
measure of community stability. Ecology 36:533-536
Margalef R (1968) Perspectives in ecological theory. University of
Chicago Press, Chicago
Margalef R (1974) Diversity, stability and maturity in natural eco-
systems. In: Proceedings of the first international congress of
ecology. Pudoc, Wageningen, pp 66
May RM (1973) Qualitative stability in model ecosystems. Ecolo-
gy 54:638-641
May RM (1975) Stability in ecosystems: some comments. In:
Dobben WH van, Lowe -McDonnell RH (eds) Unifying con-
cepts in ecology. Junk, The Hague, pp 161-168

334 0 E C 0 L 0 G I A 109 (1997) © Springer -Verlag
McIntosh RP (1985) The background of ecology — concept and
theory. Cambridge University Press, Cambridge
Minns CK (1992) Use of models for integrated assessment of eco-
system health. J Aqua Ecosyst Health 1:109-118
Mulholland RJ (1976) Stability analysis of the response of ecosys-
tems to perturbations. In: Levin SA (ed) Ecosystem analysis
and prediction. Society for Industrial and Applied Mathe-
matics, Philadelphia, pp 166-181
Murdoch WW (1970) Population regulation and population iner-
tia. Ecology 51:497-502
Nakajima H, DeAngelis DL (1989) Resilience and local stability
in a nutrient -limited resource -consumer model. Bull Math Biol
51:501-510
Nisbet RM, Gurney WSC (1982) Modelling fluctuating popula-
tions. Wiley, Chichester
Orians GH (1974) Diversity, stability and maturity in natural eco-
systems. In: Proceedings of the first international congress of
ecology. Pudoc, Wageningen, pp 64-65
Orians GH (1975) Diversity, stability and maturity in natural eco-
systems. In: Dobben WH van, Lowe -McDonnell RH (eds)
Unifying concepts in ecology. Junk, The Hague, pp 139-149
Pickett STA, White PS (eds) (1985) The ecology of natural distur-
bance and patch dynamics. Academic Press, New York
Pimm SL (1980) Food web design and the effect of species dele-
tion. Oikos 35:139-49
Pimm SL (1984) The complexity and stability of ecosystems. Na-
ture 307:321-326
Pimm SL (1991) The balance of nature? University of Chicago
Press, Chicago
Preston FW (1969) Diversity and stability in the biological world.
Brookhaven Symp Biol 22:1
Putman RJ, Wratten SD (1985) Principles of ecology. Croom
Helm, London
Rahel FJ (1990) The hierarchical nature of community persis-
tence: a problem of scale. Am Nat 136:328-344
Rejmanek M (1992) Stability in a multi -species assemblage of
large herbivores in East Africa: an alternative interpretation.
Oecologia 89:454-456
Remmert H (1989) Okologie. Eine Einfiihrung, 4th edn. Springer,
Berlin Heidelberg New York
Roff DA (1974) Spatial heterogeneity and the persistence of popu-
lations. Oecologia 15:245-258
Roughgarden J (1975) Population dynamics in a stochastic envi-
ronment: spectral theory for the linearized N -species Lotka-
Volterra competition equations. Theor Popul Biol 7:1-12
Royama T (1977) Population persistence and density -dependence.
Ecol Monogr 47:1-35
Rutledge RW, Basore BL, Mulholland RJ (1976) Ecological sta-
bility: an information theory viewpoint. J Theor Biol 57:
335-371
Schaffer W (1981) Ecological abstraction: the consequences of re-
duced dimensionality in ecological models. Ecol Monogr
51:383-401
Schmidt E, Wissel C (199 1) Modelle zur Klassifizierung and Qua-
ntifizierung 6kologischer Stabilitat. Verb Ges Okol 19:709-718
Schwegler H (1981) Stabilitatsbegriffe fair biologische Systeme.
Angew Bot 55:192-137
Schwegler H (1985) Okologische Stabilitat. Verb Ges Okol 13:
263-270
Shrader-Frechette KS, McCoy ED (1993) Method in ecology —
strategies for conservation. Cambridge University Press, Cam-
bridge
Siljak DD (1974) Connective stability of complex systems. Nature
249:280
Smedes GW, Hurd LE (198 1) An empirical test of community sta-
bility: resistance of a fouling community to a biological patch
forming disturbance. Ecology 62:1561-1572
Steele J (1974) Stability in plankton ecosystems. In: Usher MG,
Williamson MH (eds) Ecological stability. Chapman and Hall,
London, pp 179-191
Steinman AD, Mulholland PJ, Palumbo AV, Flum TF, Elwood JW,
DeAngelis DL (1990) Resistance of lotic ecosystems to a light
elimination disturbance: a laboratory stream study. Oikos
58:80-90
Steinman AD, Mulholland PJ, Palumbo AV, Flum TF, DeAngelis
DL (1991) Resilience of lotic ecosystems to a light -elimina-
tion disturbance. Ecology 72:1299-1313
St6cker G (1974) Zur Stabilitat and Belastbarkeit von Okosyste-
men. Arch Natursch Landschaftsforsch 14:237-261
Strong DR (1986a) Density -vague population change. Trends Ecol
Evol 1:39-42
Strong DR (1986b) Densitiy vagueness: abiding the variance in
the demography of real populations. In: Diamond JM, Case TJ
(eds) Community ecology. Harper and Row, New York, pp
257-268
Strong DR (1990) Realistic models of persistence. Ecology 71:421
Sutherland JP (1990) Perturbations, resistance, and alternative
views of the existence of multiple stable points in nature. Am
Nat 136:270-275
Suttman CE, Barrett GW (1979) Effects of serin on arthropods in
an agricultural and an old field plant community. Ecology
60:628-641
Thornton KW, Mulholland RJ (1974) Lagrange stability and eco-
logical systems. J Theor Biol 45:473-485
Ulrich B (1992) Forest ecosystem theory based on material bal-
ance. Ecol Model 63:163-183
Vincent TL, Anderson LR (1979) Return time and vulnerability
for a food chain model. Theor Popul Biol 15:217-231
Warner RR, Chesson PL (1985) Coexistence mediated by recruit-
ment fluctuations: a field guide to the storage effect. Am Nat
125:769-787
Westman WE (1978) Measuring the inertia and resilience of eco-
systems. BioScience 28:705-710
Westman WE (1991) Ecological restoration projects measuring
their performance. Environ Professional 13:207-215
Whittaker RH, Levin SA (1977) The role of mosaic phenomena in
natural communities. Theor Popul Biol 12:117-139
Williamson M (1972) The analysis of biological populations. Ar-
nold, London
Williamson M (1984) The measurement of population variability.
Ecol Entomol 9:239-241
Williamson M (1987) Are communities ever stable ? Br Ecol Soc
Symp 26:353-371
Wu LS (1976) On the stability of ecosystems. In: Levin SA (ed)
Ecosystem analysis and prediction. Society for Industrial and
Applied Mathematics, Philadelphia, pp 155-165
Wu LS (1977) The stability of ecosystems — a finite -time ap-
proach. J Theor Biol 66:345-359
Yodzis P (1989) Theoretical ecology. Harper and Row, New York
Zw6lfer H (1978) Was bedeutet "6kologische Stabilitat'? In: Oko-
logie and Zukunftssicherung. Bayreuther Hefte 3:13-33

Mountain Research
and Development
Typology and Principles of Ecological Stability and Instability
Author(s): Andreas Gigon
Source: Mountain Research and Development, Vol. 3, No. 2, Workshop on the Stability and
Instability of Mountain Ecosystems. Berne-Riederalp, 14-19 September 1981 [Part One]
(May, 1983), pp. 95-102
Published by: International Mountain Society
Stable URL: http://www.jstor.org/stable/3672989
Accessed: 15-06-2018 16:02 UTC
REFERENCES
Linked references are available on ,TSTOR for this article:
http://www. jstor.org/stable/3672989?seq=1$cid=pdf-ref erence#ref erences_tab_contents
You may need to log in to JSTOR to access the linked references.
.TSTOR 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 TSTOR, please contact [e-mail].
Your use of the .TSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
http://about.jstor.org/terms
JSTOR
International Mountain Society is collaborating with TSTOR to digitize, preserve and extend
access to Mountain Research and Development
This content downloaded from 150.216.21.176 on Fri, 15 Jun 2018 16:02:22 UTC
All use subject to http://about.jstor.org/terms

Mountain Research and Development, Vol. 3, No. 2, 1983, pp. 95-102
TYPOLOGY AND PRINCIPLES OF ECOLOGICAL STABILITY AND INSTABILITY
ANDREAS GIGON
Geobotanical Institute, The Rubel Foundation
Swiss Federal Institute of Technology (ETH)
8044 Zurich, Switzerland
ABSTRACT The concepts of ecological stability, instability, and lability are discussed and defined; they should be used only as general
concepts. Ecological stability is the continued existence of an ecological system or its capability to restore the original state after a
change. It is important to distinguish between natural and anthropogenic stability and instability. According to the presence or absence
of disturbance factors (extraordinary factors) and the amplitude of the oscillations of the parameters under consideration the following
types of ecological stability can be distinguished: constancy, cyclicity, resistance, and elasticity. For every investigation of stability
special criteria and scales of observation must be elaborated.
Based on the example of plant communities of the Alps general conclusions are drawn: (1) no ecological system is stable in respect
to all possible disturbance factors; (2) different types of stability can occur within the same ecological system; (3) the same disturbance
factor can have completely different effects in very similar ecological systems; (4) depending on the criteria and scales of observation
the same ecological phenomenon can be considered stable or unstable and evaluated negatively or positively. Thus statements about
stability are only of limited value unless it is clearly specified which criteria and scales of observation are used, which disturbance
factor is being considered, and which type of stability is occurring.
A list of principles contributing to ecological stability is presented; it can be used as a checklist for assessing the risk of instability.
RtSUME Typologie et principes de la stabiliti et 11instabiliti icologiques. Les concepts de stabilite, d'instabilite et de labilits ecologiques sont
discutss et definis; ils devraient etre employes seulement comme concepts generaux. La stabilite ecologique est 1'existence continue
d'un systeme ecologique ou sa capacite de revenir a 1'etat original apres un changement. Il est important de faire une distinction
entre stabilite (instabilite) naturelle et anthropogene. Les types de stabilite ecologique suivants sont distingues en raison de la presence
ou absence de facteurs de perturbation (facteurs extraordinaires) et de ]'amplitude des oscillations du parametre considers: Constance,
cyclicite, resistance et elasticite. Il est discute que chaque analyse d'un cas de stabilite exige ]'elaboration de criteres appropries. Basses
sur 1'example de 1'6cologie de prairies des Alpes les conclusions generales suivantes sont tirees: (1) Il n'existe aucun systeme ecologique
qui Boit stable envers tous les facteurs de perturbation possibles. (2) Dans le meme systeme ecologique des types de stabilite differents
peuvent etre realises. (3) Des 6cosystemes semblables peuvent reagir de maniere tres differente a 1'6gard du meme facteur de perturbation.
(4) En fonction des criteres choisis le meme phenomene ecologique peut etre considers comme stable ou instable et jugs de maniere
positive ou negative. Cela prouve que les affirmations gensrales concernant la stabilite ne peuvent pas etre employees sans specifier
clairement les criteres employes ainsi que le facteur de perturbation et le type de stabilite.
Une liste de principes contribuants a la stabilite ecologique est presentee; elle peut etre employee comme "checklist" pour juger
le risque d'instabilite.
zuSAMMENFASSUNG Typologie and Prinzipien der okologischen Stabilitat and Instabilitat. Die Begriffe 6kologische Stabilitat, Instabilitat and
Labilitat werden diskutiert and definiert; sie soldten nur als allgemeine Oberbegriffe verwendet werden. Oekologische Stabilitat ist
das Bestehenbleiben eines 6kologischen Systems and die Fahigkeit nach Veranderungen in die Ausgangslage zurfickzukehren. Es
ist wichtig zwischen natiirlicher and anthropogener Stabilitat (Instabilitat) zu unterscheiden. Aufgrund des Vorhandenseins oder Fehlens
von Fremdfaktoren and aufgrund der Schwankungen des betrachteten Merkmals werden folgende Typen bkologischer Stabilitat
unterschieden: Konstanz, Zyklizitat, Resistenz, and Elastizitat. Fur jede Beurteilung der Stabilitat sind spezielle Kriterien and MaQstibe
zu erarbeiten. Am Beispiel von Pflanzengesellschaften der Alpen wird dargelegt: (1) Kein 6kologisches System ist beziiglich allen
m6glichen Fremdfaktoren stabil. (2) Verschiedene Typen von Stabilitat k5nnen im selben okologischen System verwirklicht sein.
(3) Derselbe Fremdfaktor kann sick in ahnlichen okologischen Systemen ganz verschieden auswirken. (4) Je nach den gewahlten
Kriterien and Magstaben kann dasselbe 6kologische Phanomen ads stabil oder als instabil angesehen and positiv oder negativ bewertet
werden. Somit rind allgemeine Aussagen fiber Stabilitat wissenschaftlich and praktisch nur beschrankt brauchbar, sofern nicht dargelegt
ist, welche Beurteilungskriterien und-mal1stabe gewahlt wurden, sowie welcher Fremdfaktor and welcher Stabilitatstyp betrachtet
wurden.
Es wird eine Liste von Prinzipien, die zu 6kologischer Stabilitat beitragen, dargelegt; sie kann u. a. als Checkliste fiir die Beurteilung
des Risikos von Instabilitat dienen.
This content downloaded from 150.216.21.176 on Fri, 15 Jun 2018 16:02:22 UTC
All use subject to http://about.jstor.org/terms

96 / MOUNTAIN RESEARCH AND DEVELOPMENT
INTRODUCTION
The aims of this paper are: to develop a comprehensive
and practical typology for ecological stability and insta-
bility; to discuss some aspects that become important when
the concept of stability is applied to practical problems; and
to list some ecological principles relevant to stability.
The words stability and instability are so frequently used
in current language, and in the different branches of
science, that they must have different meanings for each
person. In ecology, and particularly in planning, they are
becoming more and more a kind of magical concept. Stable
ecosystems and landscapes are what should be planned for,
and unstable ones should be stabilized. Is it really as sim-
ple as that? What, after all, is stability?
This paper shows, as have many before, that it is neces-
sary to distinguish different types of stability and instability
(Holling, 1973; Orians, 1975; Remmert, 1980). The spe-
cial feature of the typology presented here is that it is more
comprehensive than those found in the earlier literature.
The approach is phenomenological and not theoretical; it
is described in more detail by Gigon (1981, in prep.). Using
as examples the stability and instability of plant commu-
nities in the Alps, practical conclusions are drawn which
can be applied to many other ecological systems.
TYPOLOGY OF ECOLOGICAL STABILITY AND INSTABILITY
STABILITY, INSTABILITY, AND LABILITY
AS OVERALL CONCEPTS
A comparison of various textbooks as well as conference
and symposium reports shows that the term stability is used
in many different ways even within the single field of ecol-
ogy (Woodwell and Smith, 1969; Van Dobben and Lowe -
McConnell, 1975). The diversity of meanings is so large
that stability can no longer be used for describing a very
specific state of an ecological system, as proposed, for in-
stance, by Rolling (1973) and Remmert (1980), but must
be considered as a broad overall concept, as proposed by
Orians (1975) and Gigon (1981). In the present paper the
following definition is used:
ecological stability: the continued existence of an ecological
system and its capability to restore its original state after
a change.
Examples of stable ecosystems are the redwood forests
in California and Alpine grasslands. Stability is also im-
plied in the recovery of the population of deer (Capreolus
capreolus) after decimation through hunting, as observed
in Central Switzerland.
The definition of stability proposed here comprises two
quite different aspects: the static aspect of the continued
existence of the system; and the dynamic aspect of the capa-
bility for restoration of the original state after certain
changes. It is not necessary for both the static and the
dynamic aspects to be realized in an ecosystem to warrant
it being designated as stable; the presence of the static
aspect alone is sufficient. For instance, certain types of
tropical rain forest are described as stable because they do
not change much in species composition over decades or
centuries. But it is well known that after a large-scale felling
of the forest the species composition changes irreversibly;
that is, the original state is not restored (G6mez-Pompa
et al., 1972).
The opposite of stability is instability and is defined as fol-
lows:
ecological instability: the process of irreversible change of an
ecological system and the absence of a capability for restoring
its original state after a change.
Examples include the ecological succession on glacial
moraines and in old fields; the increasing erosion of over-
grazed pastures on steep slopes, which leads to bare rock,
as seen in many mountains of the Mediterranean region;
the inability for restoration of the original species composi-
tion in dry limestone grasslands in Switzerland after two
years of heavy fertilizer application such that many plant
and insect species disappear and do not "return" for at least
several decades; and, of course, the mostly irreversible
effects of large-scale felling of the "stable" tropical rain for-
ests can be interpreted also as instability.
It is important to realize that after shorter or longer
phases of instability most ecological systems develop into
a new stable phase: instability does not last for ever.
Discussion of the concepts of stability and instability be-
comes complicated because of the existence of many eco-
systems, mostly anthropogenic, which have characteristics
falling between the stated definitions of stability and in-
stability. For example, planted monocultures of spruce
(Picea abies (L.) H. Karsten) in the Swiss midlands are not
really unstable, as is often said by ecologists; these forests
are not necessarily in "the process of irreversible change."
On the other hand, it is also misleading to call them stable
because they show a strong inclination to irreversible change
through storm damage or pests or disease infestation. This
pattern can be described as:
ecological lability: the pronounced inclination of an ecologi-
cal system to change irreversibly.
Ecological lability is the phenomenon of general potential
instability of a stable ecological system.
The examples briefly discussed in this paper are rele-
vant to the extensive use of the concepts of stability and
instability. It is evident that for practical purposes, such
as ecological planning and environmental impact assess-
ment, more precise concepts must be elaborated; the ques-
tions of criteria and scales of observation and evaluation
must be discussed in detail.
TYPOLOGY OF ECOLOGICAL STABILITY
Natural and Anthropogenic Stability
For applied purposes it is particularly important to know
This content downloaded from 150.216.21.176 on Fri, 15 Jun 2018 16:02:22 UTC
All use subject to http://about.jstor.org/terms

whether the stability of an ecological system is realized
when man does not interfere, and whether anthropogenic
inputs are necessary to ensure its continued existence. This
leads to the distinction between: natural stability: stability
realized without the interference of man; and anthropogenic
stability: stability which is realized only with anthropogenic
inputs and which will change into instability when these
inputs cease.
Examples of natural stability include Alpine grasslands,
peat bogs, and certain tropical rain forests. Anthropogenic
stability is exemplified by fertilized hay -meadows, cotton
plantations, rice paddies, and many other agro-ecosystems.
The distinction between natural and anthropogenic stabil-
ity is not always easy; and man-made ecosystems do not
always indicate anthropogenic stability, or vice versa.
Examples include man-made pastures through the cutting
and burning of forests at timberline in the Alps which show
a stability that can be called natural. Even if man does not
interfere, certain types of these pastures would persist; they
would simply be grazed more intensively by deer, chamois,
and other animals of the natural fauna.
Constancy, Cyclicity, Resistance, and Elasticity
Many different typologies of ecological stability have
been elaborated already (Woodwell and Smith, 1969;
Holling, 1973; Orians, 1975; Haber, 1979; Remmert,
1980). Based on these investigations a phenomenological
and simple typology has been proposed and described in
detail by Gigon (1981). The following criteria were used:
1. Presence or absence of a disturbance factor (perturba-
tion or impact factor) that is, a factor not being a nor-
mal part of the ecological system under consideration
(Ellenberg, 1972).
2. Small or large amplitude of the changes (oscillations)
of the ecological factor under consideration.
For understanding ecological stability it is important to
know whether there is (or has been) a disturbance factor or
not. Sulphur dioxide emissions, fertilizer application to an
oligotrophic ecosystem, introduction of foreign plants or
animals, and a natural landslide in an ecosystem where
it had never occurred before are examples of disturbance
factors. There are many cases where the distinction be-
tween normal and disturbance factors is unclear.
The amplitudes of the changes (oscillations) in the ecological
system are another parameter important for distinguish-
ing different types of stability. These changes, or oscilla-
tions, can be small, such as those occurring in the plant
community of alpine meadows or mixed forests in Central
Europe, or they can be large. Large oscillations, for in-
stance, are occurring in the plant species composition in
certain boreal forests of North America (Remmert, 1980)
and in many mediterranean-type ecosystems (Walter,
1973). For detailed study, of course, the terms small and
large oscillations must be defined. The causes of the oscil-
lations can be intrinsic in the ecological system or they can
be due to disturbance factors.
Using these criteria four types of ecological stability can
be distinguished; they are schematically represented in Fig-
ure 1. It is also stressed that all four types of stability can be
present simultaneously in the same ecological system.
A. GIGON / 97
Types of natural ecological stability:
constancy resistance
d
v
E,
N 11
NU S
o`
Tt t2 time 0 12 time
cyclicity elasticity
11 t2 time 0 12 time
Natural ecological instability:
endogenous exogenous
v
o�
e-�?72
OCt2 time ttime
0 = ecological system V" =changed ecol. system
FIGURE 1. Schematic representation of some types of ecological
stability. For this typology the criteria used were the presence
or absence of disturbance factors and the amplitude and reversi-
bility of changes (oscillations) of the ecological parameter under
consideration.
Constancy
Constant stability may be defined as a type of stability char-
acterized by an absence of disturbance factors and by no,
or only small, changes (oscillations) in the ecological
parameter under consideration (Figure 1). This does not
mean, however, that the ecological system is completely
static. A Central European mixed forest shows constant
stability even if old trees die here and there and small oscil-
lations in the population densities of certain insects or birds
occur. Evergreen tropical rain forests and Alpine grasslands
are other examples of ecosystems with this type of stability.
cyclicity
Cyclic stability may be defined as a type of stability char-
acterized by the absence of disturbance factors and by the
presence of large cyclic oscillations of the ecological param-
eter(s) under consideration (Figure 1). These oscillations
are due to normal factors belonging within the ecological
system. A classical example of an ecosystem with cyclic sta-
bility is the Californian chaparral. In the chaparral, fire,
This content downloaded from 150.216.21.176 on Fri, 15 Jun 2018 16:02:22 UTC
All use subject to http://about.jstor.org/terms

98 / MOUNTAIN RESEARCH AND DEVELOPMENT
usually due to lightning, is a natural ecological factor
(Walter, 1973). It causes large cyclic oscillations in biomass
and floristic composition. Fire also enhances germination
of many species, destroys certain phytotoxic substances
accumulated in the soil, and mobilizes nutrients fixed in
the dead organic material. Cyclic burning is essential for
the "continued existence" of the chaparral, that is, for its
stability.
Other ecosystems with cyclic stability are coniferous for-
ests, such as the central Alpine larch forests (Larix decidua
Miller), which show large oscillations in the populations
of the larch budmoth (Zeiraphera diniana Guenee) (Baltens-
weiler, 1964). Remmert (1980) mentions the phenomenon
of cyclic stability in certain tundra ecosystems of Alaska
and in East African savannas.
Resistance
Resistant stability may be defined as the type of stability
characterized by the presence of a disturbance factor, which
has no, or only a small, effect upon the ecological parameter
under consideration (see Figure 1). The parameter being
considered withstands the disturbance.
Examples include a forest ecosystem showing no change
in its floristic composition after a single application of
sewage sludge, and a limestone grassland in the Jura
Mountains, where the turf is not altered by trampling by
tourists in summer or autumn.
Elasticity
Elastic stability may be defined as a type of stability
characterized by large changes (oscillations) due to a dis-
turbance factor. When the disturbance factor is no longer
present the original state is restored (Figure 1).
For example, decimation (disturbance) of the deer popu-
lation will have effects on the composition of the forb layer
in a forest, but after a certain time the deer population re-
covers and the original state is restored. Similarly, two
years of irrigation of a fertilized, species -poor meadow in
Central Europe will change its floristic composition; some
years after irrigation has been terminated, however, the
original composition is restored.
Resilience is a concept introduced into ecology by Holling
(1973). He defines it as follows: "Resilience determines the
persistence of relationships within a system and is a
measure of the ability of these systems to absorb changes
of state variables, driving variables and parameters, and
still persist. In this definition resilience is the property of
the system and persistence or probability of extinction is
the result.... Stability on the other hand, is the ability
of a system to return to an equilibrium state after a tem-
porary disturbance." In the terms of the present paper,
resilience corresponds more or less to constancy and re-
sistance, whereas stability, as defined by Holling, cor-
responds to elasticity as it is defined in this paper. Resil-
ience and stability in the sense of Holling are not used
further in the present paper because these terms have been
used in many different ways by different authors and be-
cause resistance is a better understood term than resilience.
TYPOLOGY OF ECOLOGICAL INSTABILITY
As with ecological stability, ecological instability also can
be subdivided into different types. A distinction must be
TYPOLOGY OF ECOLOGICAL
STABILITY and 11NSTABILITY
potential actual
LABILITY
natural anthropogenic natural anthropogenic
1:7� (stabilizing input)
no with nat, with anthr, no with nat, with anthr,
disturbance disturbance disturbance disturbance disturbance disturbance
factor factor factor factor factor factor
I
no with cyclic no with irreversible irreversible irreversible
oscillation oscillation oscillation oscillation change change change
l
nat,or anthr. nat,or anthr. nat,or anthr. nat,or anthr. nat,endogen. [nat.exI ogen. anthropogen.
CONSTANCY CYCLICITY RESISTANCE ELASTICITY INSTABILITY INSTABILITY INSTABILITY
FIGURE 2. Hierarchic synopsis of the types of stability and instability described in the text.
This content downloaded from 150.216.21.176 on Fri, 15 Jun 2018 16:02:22 UTC
All use subject to http://about.jstor.org/terms

made between natural instability—instability realized without
the interference of man—and anthropogenic instability—in-
stability due to the influence of man.
Examples of anthropogenic instability include the irre-
versible floristic changes in lake or bog ecosystems due to
man-made eutrophication, and the increasing disruption
of mountain ecosystems due to over -exploitation (lumber
cutting and overgrazing, for example, leading to erosion).
The distinction between anthropogenic and natural insta-
bility is not always easy. For instance, is the succession
occurring on abandoned arable fields to be considered
anthropogenic or natural?
Natural ecological instability can be subdivided into
endogenous and exogenous, the latter being caused by a natural
disturbance factor (see Figure 1). Examples of natural
A. GIGON / 99
endogenous instability are the irreversible changes occur-
ring during the natural succession on a glacial moraine.
These changes are "organized" mainly by the ecosystem
itself. Examples of exogenous instability are the irreversible
changes occurring due to a natural landslide or a volcanic
eruption in ecosystems which are not usually subject to
these disturbance factors.
.SYNOPSIS OF THE DIFFERENT TYPES OF ECOLOGICAL
STABILITY AND INSTABILITY
The different types of stability and instability proposed
in this paper can be combined into the hierarchic synopsis
shown in Figure 2. For each logical step leading to the dif-
ferent types, special criteria and both time -scale and area
must be defined, as will be shown below.
CRITERIA OF OBSERVATION AND EVALUATION OF ECOLOGICAL STABILITY
Different types of stability and instability have been
described so far in a broad and qualitative manner. It is
now necessary to determine the choice of criteria and scale,
in terms of both time and space, for more precise and quan-
titative investigation. Because each ecological system and
disturbance factor differs from every other, special criteria
and scales must be elaborated for each case to be studied. For exam-
ple, in an investigation of the stability of a peat bog, the
criteria and scales could be the annual growth rate of the
Sphagnum species. In a sylvicultural investigation, however,
the biomass changes in tons per tens of hectares, and in
terms of decades, or even centuries, must be considered.
In an ornithological study the absence of a single critical,
so-called indicator species for two or three years can already
be used as a criterion for instability. In most investigations
the criteria and scales for assessing stability or instability
comprise the following aspects:
—temporal aspect: centuries, years, months, vegetation
period
— spatial aspect: square kilometre or metre; size of the field,
stand, lake, or landscape -ecosystem
—other quantifiable aspects: number of species, biomass,
amount of nitrogen fixed per hectare and per year
—non -quantifiable (qualitative) aspects: which species dis-
appear? how is the behaviour of the animals changed?
As indicated above, the details of the criteria and the time-
scale and size of area are different for every case studied.
They must be defined by the investigator. Thus, the same
ecological phenomenon can be classified as stable or un-
stable depending on the criteria and scales chosen. How-
ever, the investigator is not completely free to define his
criteria and scales of observation. (Here the problem of sub-
jectivity or objectivity of the concept of ecological stability
is introduced—a problem which is discussed in more de-
tail by Gigon (1981).) In order to be useful, the criteria
and scales of observation must fulfil the following scien-
tific requirements:
— they must be adequate for the phenomenon investigated
—they must be practical (operational) and, if possible,
replicable
—they must be described in such a way that other scientists
can understand and use them.
As an example, Table 1 lists some criteria and scales use-
ful for assessing stability or instability of grassland com-
munities in the Alps.
In general, stability is evaluated positively by man,
whereas instability has a negative connotation. The rea-
sons for this are not only practical and scientific but quite
often have psychological components. Moreover, these rea-
sons are seldom clearly specified. From a scientific stand-
point, of course, this is not acceptable. As with the other
aspects of stability the criteria and scales, or at least the
point of view for the evaluation, must be described. To do
so will show that stability sometimes has a positive value
for man and sometimes a negative value. Moreover, the
same stable (or unstable) ecological phenomenon can be
evaluated positively or negatively, depending on the view-
point of the observer.
As an example, the absence of significant dry matter in-
crease in certain Nardus grasslands in the central Alps, even
if a total of 150 kg P205 per hectare during 7 years is
applied (Geering, 1968), can be evaluated very differently.
This resistance -stability in respect to fertilizer effects is eval-
uated negatively by the farmers and the fertilizer firms,
but positively by the nature conservationists and tourists
because the beautiful and species -rich grasslands are not
changed.
TABLE 1
Examples of criteria and scales of observation for assessing
stability of grassland communities in the central Alps
Temporal aspects:
5-20 years
Spatial aspects:
0.1-0.5 ha of homogenous
stand
Other quantifiable aspects:
Plant community defined
stable if:
—change of number of species
< 20%
—total biomass change < 30%
—changes of biomass of grass,
legume, forb and moss com-
partments each < 207o
Non -quantifiable (qualitative) Which species disappear or
aspects:
immigrate?
This content downloaded from 150.216.21.176 on Fri, 15 Jun 2018 16:02:22 UTC
All use subject to http://about.jstor.org/terms

100 / MOUNTAIN RESEARCH AND DEVELOPMENT
APPLICATION OF THE TYPOLOGY TO ANALYSIS OF STABILITY
OF SOME PLANT COMMUNITIES OF THE ALPS
As an example of the application of the typology pre-
sented here, Table 2 shows the effects of different dis-
turbance factors on three plant communities of the Swiss
Alps. The ecology of these communities has been described
by Ellenberg (1978). From Table 2 some general conclu-
sions can be drawn which are relevant also for many other
ecological systems:
1. No ecological system is stable in respect to all possible
disturbance factors.
2. Different types of stability can occur within the same
ecological system (e.g., constancy, resistance, and elas-
ticity occur simultaneously in the Nardetum pasture).
3. The same disturbance factor can have completely dif-
ferent effects, even in very similar ecological systems
(e.g., the effect of trampling on the Trisetetum and on
the Caricetum ferrugineae meadows).
4. The same stability (or instability) behaviour can be eval-
uated positively or negatively depending on the point
of view of the observer (e.g., the effect of ski -run con-
struction on the Nardetum pasture).
It has already been argued that the same ecological phe-
nomenon may be classified as stable or unstable depend-
ing on the criteria and scale of observation chosen by an
investigator. This can be summarized as follows: statements
on stability or instability are only of limited scientific and practical
value unless it is clearly specified which criteria and scale of ob-
servation are used, which disturbance factor is considered, and which
type of stability or instability is realized.
PRINCIPLES OF ECOLOGICAL STABILITY
Many different ecological structures and functions which
contribute to ecological stability can be distinguished.
According to the level of organization it is possible to group
them into autecological, population -ecological (demecologi-
cal), and synecological stability principles (Ligon, 1981).
Only the latter will be treated in here. Stability on a
particular level of organization, of course, very often has
effects also on other levels. Furthermore, the same stability
principle can occur on more than one level of organiza-
tion such as the negative feedback principle.
SYNECOLOGICAL STABILITY PRINCIPLES
Synecological stability principles can be defined as eco-
logical structures and functions immediately (directly) con-
tributing to ecological stability and in which more than one
species participates. Table 3 gives a list of some of these
TABLE 2
Stability and instability (as defined in Table I and in the text) of 3 grassland communities in the central Swiss Alps;
evaluations are positive (+) or negative (-)
Plant communities,
instability ( t
resistant
anthropogenic
elevation, slope
elastic stability
(Trisetetum) at 1,700 in and 200 S
Anthropogenic
and exposure Disturbance
(perturbation, impact) factors
disturbance factors
(regeneration)
reforestation)
Ski -run con -
(constancy)
3 seasons of struction by re -
No natural
Large ava-
Normal
intensive moving 20 cm
ecological dis-
lanche followed
management
trampling by of top soil,
turbance and
by normal
(grazing, hay-
cattle, then no reseeding and
no management
management
making, etc.)
more skiing
Subalpine pasture (Nardetum) constant resistant constant t elastic instability
at 2,300 m and 300 S stability stability stability stability (new ecosystem
regeneration or erosion)
Evaluation. of the stability or in general: + in general: + in general: + in general: + skiers: + sum -
instability: mer tourists
and farmers: —
Subalpine fertilized meadow
instability ( t
resistant
anthropogenic
elastic stability
elastic stability
(Trisetetum) at 1,700 in and 200 S
slow natural
stability
stability
(regeneration)
(regeneration)
reforestation)
(constancy)
Evaluation of the stability or
in general: -
in general: +
in general: +
in general +
in general: +
instability
foresters: ( + )
Subalpine meadow (Caricetum fer-
instability
a) resistant
anthropogenic
instability
instability
rugineae) at 1,600 in and 450 N
(erosion)
stability
stability
(erosion)
(erosion)
b) instability
(constancy)
(erosion)
Evaluation of the stability or
in general: —
a) in general: +
in general: +
in general: —
in general: —
instability
b) in general: —
This content downloaded from 150.216.21.176 on Fri, 15 Jun 2018 16:02:22 UTC
All use subject to http://about.jstor.org/terms

TABLE 3
Synecological stability principles and risks of instability resulting
from the introduction of a new seed mixture
into subalpine meadows
Examples of structures and Possible risks of instability
functions contributing to eco- resulting from the introduc-
logical stability (synecological tion of a new seed mixture
stability principles). for subalpine meadows.
1. Steady state (dynamic
—Higher yield (nutrient out -
equilibrium) of energy,
put) needs higher nutrient
chemical elements and pop-
input: How? Fertilizers?
ulations
Leaching?
2. Inertia and reservoir func-
— Nutrient pools slowly used
tion of large pools of energy,
up? Depletion?
chemical elements and popu-
lations
3. Buffering capacity of de-
veloped soil, microclimate,
biocenosis
4. Biogeochemical cycle
making possible life without
growing or shrinking: no
accumulation, no depletion,
no dependence from outside
5. Negative feedback
6. Spreading of risks (if
—Species compensating each
parts are interchangeable)
other so that sward is stable
richness of species and other
(constant) also in particularly
ecosystem parts
dry or wet summer or long
winter?
7. Numerical and functional
response in predator -prey
relationships
8. Co -adaptations of species
—Mycorrhiza? Pollination?
(other than in 1-7) symbio-
Diseases? Digestibility for
sis, commensalism, para-
local cattle race?
sitism ...
A. GIGON / 101
principles. From this it can be seen that many well-known
ecological structures and functions can be interpreted as
stability principles. Examples include, buffering capacity;
biogeochemical cycle; and negative feedback. Since these
principles are described in most ecology books (Odum,
1971; Krebs, 1978; Remmert, 1980) it is not necessary to
discuss them further. They are described in detail in terms
of their relevance to the concept of stability by Gigon (1981,
in prep.).
A practical application of Table 3 is that it can be used
as a checklist for assessing the risk of instability. For in-
stance, possible risks of instability due to the introduction
of a new seed mixture for subalpine meadows is listed.
OVERALL STABILITY OF AN ECOLOGICAL SYSTEM
The co-operation of the different stability principles in
an ecological system gives rise to what can be called over-
all stability. However, despite the many different stability
principles, no ecological system is stable in respect to all
possible disturbance factors (see also Table 2). How can
the different stability behaviours of an ecological system
be understood and predicted in respect to the different dis-
turbance factors? As a hypothesis it can be stated that this
differential behaviour is related to the evolutionary and
co -evolutionary past of the ecological system (Orians, 1975;
Gigon, 1981). This hypothesis is not discussed further here,
but it indicates that for understanding stability it is neces-
sary to study not only the present structures and functions
but also the evolutionary history of the ecological system.
SUMMARY
A detailed typology of ecological stability and instabil-
ity, some applications of this typology to Alpine grasslands,
and, briefly, some principles contributing to ecological sta-
bility have been presented. Results are summarized:
1. The terms stability, instability, and lability are discussed
and defined; they should be used only as overall con-
cepts.
For practical application it is important to distinguish
between natural and anthropogenic stability and insta-
bility.
Four main types of stability are distinguished: constant,
cyclic, resistant, and elastic. Disturbance factors are de-
fined as factors that are not normal components of the
ecological system under consideration. A synopsis of the
different types of stability and instability is shown in
Figure 2.
For every investigation of stability, special criteria and
scales must be elaborated. These must comprise tem-
poral, spacial, other quantifiable, and often also non -
quantifiable (qualitative) aspects. The criteria and scales
must be adequate for the phenomenon investigated,
practical, and described in such a way that they can be
understood and used by other scientists.
Examples of stability/instability in some plant commu-
nities of the Alps lead to the following general conclu-
sions:
— No ecological system is stable in respect to all possible
disturbance factors
— Different types of stability can occur within the same
ecological system
— The same disturbance factor can have completely dif-
ferent effects in very similar ecological systems
—Depending on the criteria and scales of observation
the same ecological phenomenon can be considered
stable or unstable and evaluated positively or nega-
tively in respect to man.
Thus it follows that statements on stability are only of
limited scientific and practical use unless it is clearly
This content downloaded from 150.216.21.176 on Fri, 15 Jun 2018 16:02:22 UTC
All use subject to http://about.jstor.org/terms

102 / MOUNTAIN RESEARCH AND DEVELOPMENT
specified which criteria and scales of observation are
used, which disturbance factor is being considered, and
which type of stability is occurring.
6. A list of structures and functions contributing to eco -
REFERENCES
Baltensweiler, W., 1964: Zeiraphera griseana Hubner (Lepidoptera:
Tortricidae) in the European Alps. A contribution to the problem
of cycles. Can. Ent., 96: 792-800.
Ellenberg, H., 1972: Belastung and Belastbarkeit von Oeko-
systemen. 19-26. In Steubing, L., Kunze, C., and Jaeger,
J. (eds.), Belastung and Belastbarkeit von Oekosystemen. W.
Blasaditsch, Augsburg. 246 pp. (Verh. Ges. f. Okologie I).
, 1978: Vegetation Mitteleuropas mit den Alpen. 2nd ed. Ulmer,
Stuttgart. 982 pp.
Geering, J., 1968: Ueber die Ausniitzung and Wirtschaftlichkeit
von Handelsdiingern im Naturfutterbau. Schweiz. landw.
Forsch., 7: 266-293.
Gigon, A., 1981: Oekologische Stabilitat; Typologie and
Realisierung. Fachbeitr. Schweiz. MAB -Information 7. 42 pp.
(Bundesamt. f. Umweltschutz, Bern.)
, in prep.: Oekologische Stabilitat and biologisches
Gleichgewicht. UTB, Ulmer, Stuttgart.
G6mez-Pompa, A., Vasquez-Yanes, C., and Guevara, S., 1972:
The tropical rain forest: a non-renewable resource. Science, 177:
765-769.
logical stability is presented. It can be used as a check-
list for assessing the risk of instability.
7. An analysis of the (co)evolutionary history of the eco-
logical system is essential for understanding its stability
and instability.
Haber, W., 1979: Theoretische Anmerkungen zur "6kologischen
Planung." Verh. Ges. f. Oekologie, VII: 19-30.
Holling, C.S., 1973: Resilience and stability of ecological sys-
tems. Ann. Rev. Ecology and Systematics, 4: 1-23.
Krebs, Ch.J., 1978: Ecology: the Experimental Analysis of Distribu-
tion and Abundance. 2nd ed. Harper & Row, New York. 678 pp.
Odum, E. P., 1971: Fundamentals of Ecology. 3rd ed. Saunders,
Philadelphia. 574 pp.
Orians, G.H., 1975: Diversity, stability and maturity in natural
ecosystems. In Van Dobben, W.H. and Lowe -McConnell,
R.H. (eds.), Unifying Concepts in Ecology, pp. 139-150.
Remmert, H., 1980: Ecology' Springer, Berlin. 289 pp.
Van Dobben, W.H. and Lowe -McConnell, R.H. (eds.), 1975:
Unifying Concepts in Ecology. Junk, The Hague. 302 pp.
Walter, H., 1973: Vegetation of the Earth. The English Universities
Press, London, and Springer, New York. 237 pp.
Woodwell, G.M. and Smith, H.H. (eds.), 1969: Diversity and
stability in ecological systems. Brookhaven Symposia in
Ecology, 22. 264 pp.
This content downloaded from 150.216.21.176 on Fri, 15 Jun 2018 16:02:22 UTC
All use subject to http://about.jstor.org/terms