Concepts in restoration ecology - Restoration Ecology - page 20

system is not defined, and in the second definition
the same is true for the minimal structure of an
ecological unit. Yet, disturbance has no real meaning
without some notion of the normal functioning or
minimal structure of a reference system. Both these
definitions link causes and effects in one way or
another, and indeed they are both relative definitions.
White and Jentsch (2001) suggested measuring dis-
turbance in absolute terms, for instance by the reduc-
tion of biomass (see also Grime 1979). In that way
problems related to the relative definitions of disturb-
ance could be avoided. For the time being we prefer
a relative definition of disturbance, thus challenging
the need to relate it to a well-defined reference system.
Indeed, the terms resilience and resistance, applied to
indicate the return to a steady state (stability, see below)
are even defined in a relative sense.
For purposes of ecological restoration, reference
systems have to be defined to determine whether a
disturbed ecosystem is being restored or remains in a
state of degradation. If a reference system is being
defined in terms of a 'minimum structure' (Pickett
et al . 1989), it may mean either that the system can be
confined to a few keystone species, the other species
being more or less redundant, or that all the species
in the community are to be included in the minimum
structure in order to restore ecosystem functioning, with
no species being redundant. These two contrasting
points of view are still being discussed. Some authors
favour the redundant-species hypothesis (only a few
keystone species contribute to the functioning of the
ecosystem), while others advocate the rivet hypo-
thesis (all or almost all species essentially contributing
to some ecosystem function); see for example Ehrlich
and Ehrlich (1981, 1992), Lawton (1997), Prins and
Olff (1998), van Andel (1998a), Waide et al . (1999)
and Chapter 3. It has long been assumed that an
increase in the number of species in a community or
ecosystem would imply an increase in connectance and
interaction strength; such an increase in complexity
would increase stability. But this seems to be the excep-
tion rather than the rule. It is still far too early to draw
any general conclusions.
To avoid confusion, it is useful to distinguish
between kind, frequency, intensity and scale (or
extent) of disturbance (e.g. Connell & Slatyer 1977,
Grubb 1985). The kind of disturbance depends on
the environmental factor concerned, whether biotic or
abiotic. The degree of disturbance is determined by

DISTURBING
EVENT

steady state (stable)

resistance

resilience

unstable

new steady state

Time

Fig. 2.3 Ecosystem responses to a disturbing event:
resistance and resilience are modes of recovery, the
unstable response implies a long-term disturbance
which may or may not be reversible through
restoration practices. Ecosystem state may be
represented by productivity, species richness or other
characteristics. Modified after Aber and Melillo (1991).

2.3.1 Disturbance

An external change of environmental conditions may
or may not result in a disturbance of the system under
consideration. In the case of disturbance, there is no
way of returning to the initial state of the system, at
least not in the short run, but the system may arrive
at an alternative state, which also may appear to be
a steady state (see Fig. 2.3). It is a task for restora-
tion ecologists to judge the differences in quality
between the two or more steady states, for example
between a forest and a clear-cut forest (which then
may be transformed to a semi-natural meadow).
In addition to the aforementioned general descrip-
tion of disturbance, various definitions of disturbance
exist in the literature. Two definitions of disturbance
are worth mentioning here.

1 Disturbance is a change in conditions, which inter-
feres with the normal functioning of a biological
system (van Andel et al . 1987).
2 Disturbance is a change in the minimal structure
of an ecological unit caused by a factor external
to the level of interest (Pickett et al . 1989).

Both definitions have their weaknesses. In the first
definition the normal functioning of a biological