Environmental Engineering Reference
In-Depth Information
populations in a balance between local extinctions and
recolonizations, and thus necessary for long-term
persistence (Hanski
et al
. 1996). Apart from minimum
viable metapopulation size, one has also to consider
the minimum amount of suitable habitat necessary for
metapopulation persistence (Andrén 1994); not all
suitable habitat may be occupied simultaneously.
As there is not so much useful information about
minimum viable metapopulation sizes, we focus on
the concept of the minimum viable population size,
which is defined as the minimum number of indi-
viduals in a local population that has a good chance
of surviving for some relatively long period of time;
for instance a 95% chance of surviving for at least
100 years (Soulé 1980, 1987, Lande 1988).
While asking the question, how small is too small?,
we should not only take into account the minimum
number of individuals of a species under concern, but
also the minimum size of the area, biotope or wildlife
reserve. Gurd
et al.
(2001) derived estimates of the
minimum area requirement from species-area curves;
for mammals in parks and nature reserves in the
Alleghenian-Illinoian mammal province (Canada) the
minimum area would be 5037 km
2
(ranging from
2700 to 13,296 km
2
), whereas the majority of reserves
are smaller. Similarly, Fritts and Carbyn (1995) estim-
ated the area required for a minimum viable popula-
tion of 100 grey wolves in North America as 3000 km
2
;
under otherwise favourable circumstances 500 -1000 km
2
could be adequate.
From a genetic point of view, the minimally
required number of genotypes, which is in general the
minimum amount of genetic variation, can be more
important than the number of individuals. However,
Lande (1988) argued that demography may usually
be of more immediate importance than population
genetics in determining the minimum viable sizes
of wild populations. In many species, individuals in
populations declining in low numbers experience
diminished viability and reproduction by non-genetic
causes, and there may be a threshold density or num-
ber of individuals from below which the population
cannot recover. Examples are known of viable popu-
lations, despite a very low level of genetic variation.
The northern elephant seal (
Mirounga angustirostris
),
in a region off the coast of California and Mexico,
experienced an extreme population bottleneck of less
than 20 -30 seals during a period of 20 years due to
exploitation in the 19th century. Legislative pro-
tection in the United States and Mexico resulted in a
dramatic recovery in number, although their genomic
diversity was greatly reduced (Bonnell & Selander 1974,
Hoelzel
et al.
1993). Having mentioned this, it would
be wrong to conclude that genetic variability is of
minor importance in general. A large amount of
literature is available on the effects of genetic erosion
(see Vergeer
et al
. 2003), but these results have not
been considered in view of a minimum viable
(meta)population. A few examples are available of the
rescue of depauperate populations by the (re)intro-
duction of new genes, which may point at genetic
bottlenecks. Madsen
et al.
(1999) published on the
successful restoration of an inbred adder population.
Their population data from 1983 to 1995 indicate that
the Smygehuk adders (
Vipera berus
) were on the
brink of extinction, with falling numbers and negli-
gible recruitment. Introducing new genes from a dif-
ferent population enabled the adders to make a
dramatic recovery. Bijlsma
et al.
(2000) have shown
experimentally in
Drosophila melanogaster
that popu-
lations with a low rate of genetic variation may remain
stable as long as they are cultivated under optimal
conditions. Imposed stress, however, has a much
stronger impact; high temperature stress and ethanol
stress upon small vital populations differing in the level
of prior inbreeding under optimal conditions revealed
increasing extinction probabilities with increasing
inbreeding levels, even for low levels of inbreeding.
The authors emphasized the need for further research
on the interaction between genetic and non-genetic
processes. It is recommended that breeding pro-
grammes not only avoid inbreeding, but also are done
under conditions that mimic future natural situations
(Bijlsma
et al
. 1999).
Results of recent empirical studies suggest that
whereas genetic variation may decrease with reduced
remnant population size, not all fragmentation events
lead to genetic losses and different types of genetic
variation, as for example allozyme and quantitative
variation, may respond differently. In some circum-
stances, fragmentation actually appears to increase gene
flow among remnant populations, breaking down
local genetic structure (Young
et al
. 1996). There are
many causes of extinction, the fate of a specific popu-
lation cannot generally be predicted and there is no
single answer to the problem of what the minimum
