Chapter 26 A Landscape Perspective
635
abundance of fish and epibenthic crustaceans. Instead they suggested that processes operating beyond
the landscape (e.g. proximity to inlets, larval delivery) might need to be considered (Hovel et al., 2002).
Similar ideas were proposed by Bell et al. (1988)
based upon studies of seagrass, decapods, and juvenile fish. Bell and Hall (1997) also reported that
spatial location within an estuary best explained patterns of macroalgal accumulation in seagrass beds in
Florida, USA. These studies suggest that some ecological attributes may reflect processes that operate
over spatial scales that exceed the dimensions of a
local landscape.
IV. Terrestrial vs. Marine Landscapes
Researchers have come to recognize that the landscape approach commonly applied to terrestrial studies may be useful in marine studies. However, a number of findings from the seagrass landscape studies
discussed above reveal marked differences in factors
controlling seagrass vs. terrestrial landscapes. Notably, seagrass landscapes, although also formed by
angiosperms, appear to be dynamic on a far shorter
temporal scale than the forests that are often a focus
of terrestrial landscape studies, making it critical that
a sampling regimen be adequately replicated in the
short-term. Robbins and Bell (2000) also suggested
that many of the responses to disturbance typically
recorded in terrestrial systems (e.g. succession) may
not be common in at least some seagrass ecosystems, highlighting further the interesting divergence
in some aspects of offshore vs. onshore landscapes.
Dissimilarities between terrestrial and subtidal
landscapes may be a result of limited seed production by seagrasses and/or the highly monospecific
character of the seagrass beds in the some of the areas studied. In contrast to terrestrial systems, where
gaps serve as sites for seed recruitment and offer
areas for invasion by additional plant species (e.g.
Grubb, 1977), Bell et al. (1999) found that gaps
within monospecific seagrass beds in Tampa Bay,
Florida were not invaded by other species (see also
Robbins and Bell, 2000; but see Williams, 1990).
Additionally, the limited studies available on disturbance gaps and boat propeller scarring within seagrass beds show little evidence that gaps expand
and propagate in these systems (Dawes et al., 1997;
Kirkman and Kirkman, 2000; Bell et al., 2002; but
see Whitfield et al., 2002), resulting in the familiar
wave regeneration patterns often seen in forests. In
contrast, subtidal marine landscapes are more likely
to be regularly disturbed than many terrestrial settings due to the increased kinetic energy of the fluid
medium in which they grow, resulting in short return
intervals of disturbances. A final difference is that
much of the change in cover in seagrass landscapes
(e.g. gap closure) in many areas purportedly occurs
via growth of rhizomes belowground (Robbins and
Bell, 2000; Jensen and Bell, 2001), rather than by the
closing of the canopy or by germinating seedlings.
Thus, some of the basic ideas pertaining to disturbance and habitat heterogeneity (particularly with
respect to recruitment and spread of disturbance—
e.g. percolation), which appear to work adequately
for terrestrial systems like forests and arid grasslands, do not appear to be as straightforward in subtidal landscapes.
High levels of disturbance lead to fragmentation, and impacts of seagrass habitat fragmentation on fauna, resulting in a decrease in connectivity and an increase in patch number, appear also
to differ from those that have been reported for
fauna from terrestrial systems. A major distinction is that spacing between underwater seagrass
patches, in contrast to patch spacing in terrestrial
landscapes, may often be insufficient to hinder dispersal of organisms between patches given the important role of passive transport with water currents
in the dispersal of marine organisms (Sogard and
Able, 1994; Bell et al., 2001). Conceivably, some
of the impacts of fragmentation on fauna may only
be observed at a greater degree of patch isolation
than that typically examined in past studies or in
terrestrial landscapes, or when patches are sufficiently large that edge and core areas are functionally established. However, little information exists
presently on resource availability or predator distribution for fauna found at edges of seagrass beds (e.g.
Bologna and Heck, 2002; also see below). Findings
on seagrass bed fragmentation by propeller scarring,
combined with other studies on faunal response to
seagrass plantings (e.g. Fonseca et al., 1996) or faunal abundance in seagrass patches of different sizes
(McNeill and Fairweather, 1993) seemingly agree
that many of the dominant epibenthic fauna in seagrass beds do not appear to be limited by the amount
of seagrass cover/density, at least over the scale at
which many studies have been conducted and at the
degree of fragmentation examined (see also Attrill
et al., 2000; Hovel et al., 2001; Uhrin and Holmquist,
635
abundance of fish and epibenthic crustaceans. Instead they suggested that processes operating beyond
the landscape (e.g. proximity to inlets, larval delivery) might need to be considered (Hovel et al., 2002).
Similar ideas were proposed by Bell et al. (1988)
based upon studies of seagrass, decapods, and juvenile fish. Bell and Hall (1997) also reported that
spatial location within an estuary best explained patterns of macroalgal accumulation in seagrass beds in
Florida, USA. These studies suggest that some ecological attributes may reflect processes that operate
over spatial scales that exceed the dimensions of a
local landscape.
IV. Terrestrial vs. Marine Landscapes
Researchers have come to recognize that the landscape approach commonly applied to terrestrial studies may be useful in marine studies. However, a number of findings from the seagrass landscape studies
discussed above reveal marked differences in factors
controlling seagrass vs. terrestrial landscapes. Notably, seagrass landscapes, although also formed by
angiosperms, appear to be dynamic on a far shorter
temporal scale than the forests that are often a focus
of terrestrial landscape studies, making it critical that
a sampling regimen be adequately replicated in the
short-term. Robbins and Bell (2000) also suggested
that many of the responses to disturbance typically
recorded in terrestrial systems (e.g. succession) may
not be common in at least some seagrass ecosystems, highlighting further the interesting divergence
in some aspects of offshore vs. onshore landscapes.
Dissimilarities between terrestrial and subtidal
landscapes may be a result of limited seed production by seagrasses and/or the highly monospecific
character of the seagrass beds in the some of the areas studied. In contrast to terrestrial systems, where
gaps serve as sites for seed recruitment and offer
areas for invasion by additional plant species (e.g.
Grubb, 1977), Bell et al. (1999) found that gaps
within monospecific seagrass beds in Tampa Bay,
Florida were not invaded by other species (see also
Robbins and Bell, 2000; but see Williams, 1990).
Additionally, the limited studies available on disturbance gaps and boat propeller scarring within seagrass beds show little evidence that gaps expand
and propagate in these systems (Dawes et al., 1997;
Kirkman and Kirkman, 2000; Bell et al., 2002; but
see Whitfield et al., 2002), resulting in the familiar
wave regeneration patterns often seen in forests. In
contrast, subtidal marine landscapes are more likely
to be regularly disturbed than many terrestrial settings due to the increased kinetic energy of the fluid
medium in which they grow, resulting in short return
intervals of disturbances. A final difference is that
much of the change in cover in seagrass landscapes
(e.g. gap closure) in many areas purportedly occurs
via growth of rhizomes belowground (Robbins and
Bell, 2000; Jensen and Bell, 2001), rather than by the
closing of the canopy or by germinating seedlings.
Thus, some of the basic ideas pertaining to disturbance and habitat heterogeneity (particularly with
respect to recruitment and spread of disturbance—
e.g. percolation), which appear to work adequately
for terrestrial systems like forests and arid grasslands, do not appear to be as straightforward in subtidal landscapes.
High levels of disturbance lead to fragmentation, and impacts of seagrass habitat fragmentation on fauna, resulting in a decrease in connectivity and an increase in patch number, appear also
to differ from those that have been reported for
fauna from terrestrial systems. A major distinction is that spacing between underwater seagrass
patches, in contrast to patch spacing in terrestrial
landscapes, may often be insufficient to hinder dispersal of organisms between patches given the important role of passive transport with water currents
in the dispersal of marine organisms (Sogard and
Able, 1994; Bell et al., 2001). Conceivably, some
of the impacts of fragmentation on fauna may only
be observed at a greater degree of patch isolation
than that typically examined in past studies or in
terrestrial landscapes, or when patches are sufficiently large that edge and core areas are functionally established. However, little information exists
presently on resource availability or predator distribution for fauna found at edges of seagrass beds (e.g.
Bologna and Heck, 2002; also see below). Findings
on seagrass bed fragmentation by propeller scarring,
combined with other studies on faunal response to
seagrass plantings (e.g. Fonseca et al., 1996) or faunal abundance in seagrass patches of different sizes
(McNeill and Fairweather, 1993) seemingly agree
that many of the dominant epibenthic fauna in seagrass beds do not appear to be limited by the amount
of seagrass cover/density, at least over the scale at
which many studies have been conducted and at the
degree of fragmentation examined (see also Attrill
et al., 2000; Hovel et al., 2001; Uhrin and Holmquist,