the role of humans, largely reflecting the approach
of the research literature in the field. Yet, as clear
from particular parts of the discussion, human soci-
eties have had profound impacts on island ecolo-
gies, nowhere more so than on some of the more
remote oceanic islands. We are only beginning to
understand how this may have impacted on island
diversities. This caution serves to reinforce the take
home message of this section that it is not enough
to assume that a system is in equilibrium, any more
than to assume that it is not. This should be a
matter of empirical investigation, particularly if
other theories are to be constructed on the basis of
equilibrial assumptions (cf. Weins 1984). We need
to remember this when we turn to the application
of island theories to conservation problems in the
final section of the book.
6.6 Summary
Although the equilibrium model of island biogeog-
raphy (EMIB) has served to advance understanding
of island ecology, the work reviewed in the previ-
ous two chapters demonstrates that the dynamic
equilibrium framework is no longer sufficient to
encompass the field. We argue in the present chap-
ter that the EMIB has a narrower frame of reference
than original envisioned, working for only a subset
of island species data sets. This subset is largely
determined by the scale parameters of the target
system relative to the space usage and dispersal
powers of the target taxon.
Haila distinguishes four coupled space-time
scales, which he terms: (1) the individual scale, (2)
the population dynamics scale, (3) the population
differentiation scale, and (4) the evolutionary scale.
Systems that correspond to the population dynam-
ics scale are most likely to show dynamics in tune
with the EMIB. We illustrate how even within a
single taxonomic group, such as birds or mammals,
different species can in practice have radically
different spatial scales of interaction with an island
system. Temporally, the dynamics of different com-
ponents of an ecosystem also respond on varying
wavelengths. For instance, whereas bird communi-
ties may respond fairly rapidly to changing island
carrying capacity, or disturbance events, the succes-
sional dynamics inherent in forested ecosystems
imply that vegetation systems may take decades
to hundreds of years to establish a meaningful
equilibrium condition, always assuming that no
further disturbance occurs. The importance of tak-
ing account of hierarchical relationships across
trophic levels is also stressed, such that for instance,
the equilibrium point for birds might change over
time in response to longer term dynamics in the
vegetation.
To accommodate these complexities, we present
a framework in which the dynamic equilibrium
(EMIB) model is placed alongside dynamic non-
equilibrium, static-equilibrium, and static non-
equilibrium concepts, and we discuss the
applicability of these ideas to case study systems.
The lack of long-term high resolution data series
hampers our ability to distinguish between these
competing hypotheses, but we go on to evaluate
evidence for how and why longer term variation in
island carrying capacities may occur. Natural dis-
turbance phenomena, such as tsunami, hurricanes,
volcanic eruptions, and even marked fluctuations
in weather conditions linked to ENSO events, may
all impact on island ecologies, with varying poten-
tial significance to island species numbers and com-
position.
Increasing the complexity of island ecological
models to incorporate such factors will improve
the realism of the models but at the cost of gener-
ality. An alternative approach is to develop
species-based models, which may allow us to
avoid some of the limitations of the EMIB, but it
has yet to be established that species-based models
have predictive power across the range of a
species, or that they will lead to improved models
for whole assemblages. Given the importance of
understanding the emergent ecological behaviour
of assemblages of species, we suggest that further
efforts be devoted to exploring the scale sensitivity
of island ecological models, the degree to which
systems demonstrate equilibrium dynamics, and
how hierarchical links across trophic levels influ-
ence emergent properties, such as species richness
and turnover.164 SCALE AND ISLAND ECOLOGICAL THEORY: TOWARDS A NEW SYNTHESIS