Concepts and problems - Nonequilibrium Ecology - page 20

Figure 1.1. Metapopulation dynamics of the Glanville fritillary butterfly. Size of the
metapopulation measured as a fraction of the occupied patches P. The dotted line in the
upper panels gives the number of distinct patches; the continuous line gives the pooled
area of these patches. Panel (a) (lower) gives the predicted equilibrium metapopulation
size (thick line) with ten replicate predicted trajectories before, during, and following
an observed reduction in habitat area over a 20-year period (shown in upper plot). Panel
(b) shows similar results, but there was a further 50% reduction in area of each of the
remaining patches (upper plot); in this case, the equilibrium moves to metapopulation
extinction although a substantial amount of habitat remains. The simulated trajectories
show a slow (over many years) decline with much variation. Note the delay in
metapopulation dynamics, as indicated by the difference between P (the fraction of
occupied patches), and the calculated equilibrium values given in the thick line. From
Hanski ( 1999 ). Reprinted by permission of Oxford University Press and the author.
stochasticity is spatially often correlated). Results show that there are
alternative (more than two) equilibria. Over longer time scales, often
due to changes in landscape structure which may occur on or around the
same time scale as extinctions and recolonizations, metapopulations may
not be at stochastic equilibrium with landscape conditions but are tracking
the changes with some delay. Such nonequilibrium conditions are likely
for the fritillary populations in a 25 km 2 area in northern ˚ land. Projections
demonstrated that, after habitat reduction, a new equilibrium would
be reached after many (tens or even hundreds of ) years (Figure 1.1 ) .
Generally, ''transient time,'' i.e., the time between disturbance and
re-establishment of equilibrium, is longest when the disturbance is large,
when a species is close to the extinction threshold and has a slow turnover,
and when there are only few dynamically important habitat patches
(Ovaskainen and Hanski 2002 ).
As shown above, the effects of environmental disruptions on population
dynamics are clear, but it should be emphasized that nonequilibrial