Geoscience Reference
In-Depth Information
ponents of this idea were Nicholson (1933, 1957) and Lack (1954). In con-
trast, Andrewartha and Birch (1954) argued that most populations are not
held at equilibrium density. Rather, densities merely fluctuate. In their view
most species avoid extinction because they comprise what we now call
metapopulations (Levins 1969). These consist of a series of subpopulations
whose densities fluctuate independently of one another, but are linked by dis-
persal. Extinction of subpopulations occurs quite often, but these are recolo-
nized by individuals dispersing from other subpopulations, allowing the
species to persist indefinitely over the entire region. This process has been
called spreading of risk (den Boer 1968; Reddingius 1971). In recent years
metapopulation dynamics have been explored by way of simulations (Hanski
1989) that have revealed that such systems eventually go extinct in the absence
of density dependence.
The debate about the ubiquity of density-dependent processes has per-
sisted to the present day despite the efforts of various ecologists to terminate
the discussion either because it was bankrupt (Krebs 1991; Wolda 1995) or
because they deemed that prevalence of density dependence was too obvious
to deny (Royama 1977, 1992). Turchin (1995) provides a comprehensive
review of the current status of the debate. Nevertheless, many ecologists have
insisted that no conclusion regarding the existence of density dependence in a
population system can be made unless their action can be demonstrated in
data collected from the populations. This has proved difficult to achieve. Until
recently, adequate methods for detecting density dependence in population
systems have been lacking and earlier methods have been shown to be statisti-
cally invalid. Several new methods have been proposed over the last decade,
most of which involve a variety of computer-based resampling procedures. I
review the most promising or widely used of these tests here and discuss their
limitations.
Ecologists used to assume that populations governed by density-dependent
processes had simple dynamics. They supposed that densities either remained
close to equilibrium or exhibited regular oscillations about the equilibrium
value. The pioneering work on deterministic chaos by Robert May (1974,
1976) taught us otherwise. May studied the behavior of the discrete logistic,
arguably the simplest possible density-dependent population model, and
showed that densities would fluctuate erratically and unpredictably if the
reproductive rates were sufficiently high. Before this ecologists had assumed
that the erratic fluctuations characteristic of most natural populations were
caused by random influences such as weather conditions that disturbed the
system away from equilibrium. There ensued an effort to determine whether
