148
S. Harrison
1 Introduction
Natural habitats range from relatively continuous to extremely patchy, and from
relatively permanent to highly ephemeral. For the past several decades, ecologists
have struggled to incorporate this natural spatial and temporal variation into our
thinking about species interactions, population persistence and biological diversity. Meanwhile, human impacts continue to push natural systems in the direction
of increasing patchiness and variability, and ecologists are increasingly called upon
to identify strategies for mitigating these impacts. While we are blessed with a
growing abundance of theory about how patchy habitats might affect ecological
patterns and processes, we still lack a solid empirical foundation from which to
evaluate these theories. This chapter describes an attempt to add one small piece to
such a foundation, through a study of an ecological community that is naturally
patchy and contains a number of rare taxa.
An impressive variety of theoretical perspectives may be brought to bear on how
the patchiness of a habitat might affect ecological patterns and processes. MacArthur
and Wilson's (1967) theory of island biogeography predicts that the smaller and
more isolated a habitat, the lower an equilibrium level of diversity it will achieve.
Metapopulation theory (Hanski 1997; Hanski and Simberloff 1997) predicts that
species may become extinct regionally if patches of their habitat become fewer or
more isolated. Models of the coexistence of predators and prey, or strong competitors, identify a more positive side to patchiness; a discontinuous environment can
promote coexistence because temporarily vacant patches provide refuges for the
victim species (reviewed in Harrison and Taylor 1997; Nee et al. 1997). Extending
the latter idea to the "metacommunity" level, models by Caswell and Cohen (1991,
1993) show that local extinction, colonization and competition in a patchy environment can lead to high total species richness by promoting variation in species composition among patches (beta diversity, Whittaker 1960). However, Tilman et al.
(1994) demonstrated that fragmentation can lead to the loss of the dominant species in a metacommunity, because these species are assumed to be the worst dispersers. A metacommunity model by Holt (1997) considers species that can either
be patch specialists or patch-and-matrix generalists, and predicts that rare patch
types will support a lower ratio of specialist to generalist species than will commoner types. This is closely related to the idea, basic to the field of landscape
ecology, that diversity is shaped by the flow of organisms through mosaic habitats
(Wiens 1997).
This brief and far from exhaustive review illustrates that ecological theory does
not offer conservationists a simple, single message about patchiness and fragmentation. Moreover, in considering how such theories might apply to real species and
communities, there are additional dimensions of biological realism that must be
kept in mind. In plants, for example, long-distance movements are difficult to observe, and local extinction and recolonization are hard to determine because so
many species have persistent seed banks. However, pollination represents a second
avenue by which spatial isolation might affect the survival of plant popUlations
(Rathcke and Jules 1993; Aizen and Feinsinger 1994a, 1994b; Groom 1998).
S. Harrison
1 Introduction
Natural habitats range from relatively continuous to extremely patchy, and from
relatively permanent to highly ephemeral. For the past several decades, ecologists
have struggled to incorporate this natural spatial and temporal variation into our
thinking about species interactions, population persistence and biological diversity. Meanwhile, human impacts continue to push natural systems in the direction
of increasing patchiness and variability, and ecologists are increasingly called upon
to identify strategies for mitigating these impacts. While we are blessed with a
growing abundance of theory about how patchy habitats might affect ecological
patterns and processes, we still lack a solid empirical foundation from which to
evaluate these theories. This chapter describes an attempt to add one small piece to
such a foundation, through a study of an ecological community that is naturally
patchy and contains a number of rare taxa.
An impressive variety of theoretical perspectives may be brought to bear on how
the patchiness of a habitat might affect ecological patterns and processes. MacArthur
and Wilson's (1967) theory of island biogeography predicts that the smaller and
more isolated a habitat, the lower an equilibrium level of diversity it will achieve.
Metapopulation theory (Hanski 1997; Hanski and Simberloff 1997) predicts that
species may become extinct regionally if patches of their habitat become fewer or
more isolated. Models of the coexistence of predators and prey, or strong competitors, identify a more positive side to patchiness; a discontinuous environment can
promote coexistence because temporarily vacant patches provide refuges for the
victim species (reviewed in Harrison and Taylor 1997; Nee et al. 1997). Extending
the latter idea to the "metacommunity" level, models by Caswell and Cohen (1991,
1993) show that local extinction, colonization and competition in a patchy environment can lead to high total species richness by promoting variation in species composition among patches (beta diversity, Whittaker 1960). However, Tilman et al.
(1994) demonstrated that fragmentation can lead to the loss of the dominant species in a metacommunity, because these species are assumed to be the worst dispersers. A metacommunity model by Holt (1997) considers species that can either
be patch specialists or patch-and-matrix generalists, and predicts that rare patch
types will support a lower ratio of specialist to generalist species than will commoner types. This is closely related to the idea, basic to the field of landscape
ecology, that diversity is shaped by the flow of organisms through mosaic habitats
(Wiens 1997).
This brief and far from exhaustive review illustrates that ecological theory does
not offer conservationists a simple, single message about patchiness and fragmentation. Moreover, in considering how such theories might apply to real species and
communities, there are additional dimensions of biological realism that must be
kept in mind. In plants, for example, long-distance movements are difficult to observe, and local extinction and recolonization are hard to determine because so
many species have persistent seed banks. However, pollination represents a second
avenue by which spatial isolation might affect the survival of plant popUlations
(Rathcke and Jules 1993; Aizen and Feinsinger 1994a, 1994b; Groom 1998).