Hostname: page-component-7c8c6479df-hgkh8 Total loading time: 0 Render date: 2024-03-27T14:05:09.919Z Has data issue: false hasContentIssue false

Extinction risk in fragmented habitats

Published online by Cambridge University Press:  11 May 2004

David H. Reed
Affiliation:
The University of Mississippi, Department of Biology, University, MS 38677-1848, USA
Get access

Abstract

Population models incorporating demographic, environmental and genetic stochasticity were created from long-term data on natural populations of 30 species of vertebrates. The probability of extinction for a single population in a continuous habitat was compared to that of multiple isolated, or semi-isolated, populations occupying a fragmented landscape with an equivalent total carrying capacity. Populations occupying a fragmented landscape were modelled for a range of dispersal rates and levels of asynchrony in the effects of environmental disturbances. Dispersal among subpopulations in the fragmented landscape partially alleviates the effect of fragmentation on extinction rates, despite the models explicitly incorporating disease epidemics which spread between subpopulations through dispersal. Even moderate environmental correlations among subpopulations greatly reduces the viability of the metapopulation relative to the case where the populations are totally independent. Whether a population performed better as a single population or as a metapopulation was strongly affected by the carrying capacity assumed, the time frame examined and the initial fitness of the population. A single population always fared better when the total habitat available was capable of supporting ≤1000 adults. Thus, continued habitat fragmentation can be expected to fuel the ongoing global extinction crisis and conservation efforts should be aimed at interconnecting isolated habitat patches.

Type
Research Article
Copyright
2004 The Zoological Society of London

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)