Chapter 26 A Landscape Perspective
631
Some recent examples of extensive destruction of
wide swaths of seagrass beds by large aggregations
of sea urchins have been reported. For example, Rose
et al. (1999) documented the removal of 82% of seagrass biomass and >95% of apical meristems along
the path of a relatively uncommon urchin aggregation in Florida, USA during a 1-year period. In a
Kenyan lagoon, Alcoverro and Mariani (2002) projected the recovery time of seagrasses from urchin
disturbance at approximately 99 months, while they
estimated the return interval of the disturbance itself at only 44 months, thereby identifying urchin
activity as an important structuring agent in selected
locations. Dugongs, which feed on both above- and
below-ground components of plants, are reported
to leave feeding trails and thus enhance edge formation within some seagrass beds, especially those
of Halophila ovalis (Nakaoka and Aioi, 1999; see
also Masini et al., 2001) Lefebrve et al. (1999) also
reported that the West Indian manatee, Trichechus
manatus, fed extensively on patches of H. wrightii,
T. testudinum, and Syringodium filiforme in Florida
and Puerto Rico but that even heavily grazed patches
recovered from seasonal grazing impacts. Therefore, in some locations and at selected times, herbivore grazing events may remove significant amounts
of seagrass and alter the size, shape, and isolation/connectivity of patches.
3. Role of Abiotic Factors
In addition to biotic factors that may contribute to
the spatial arrangement of the seagrass landscape,
nutrient availability or physical factors such as sediment movement may also play an organizing role
(see also Marb` a et al., Chapter 6). The patchiness
of C. nodosa off of northeastern Spain was found
to be strongly correlated to the inshore migration
of sand waves (Marb` a and Duarte, 1995). Robbins
and Bell (2000) examined the effect of water depth
over a small range (30–100 cm) in mixed patches of
H. wrightii and T. testudinum on a shallow shoal in
Florida, USA. Seasonal species-specific depth preferences were noted, but hydrodynamic factors were
suggested as potentially more important for seagrass
distribution than depth preferences alone (Robbins
and Bell, 2000). The effects of nutrient availability at
the landscape scale are less straightforward. Nutrient
availability was found to be unrelated to the patch
dynamics of Z. marina in a fjord in the Baltic Sea
(Worm and Reusch, 2000). While both slow-release
NPK treatments and treatments with live mussels
added to plots increased porewater nutrient levels,
neither treatment showed consistent results in terms
of seagrass shoot growth or shoot addition (Worm
and Reusch, 2000).
The idea that physical energy may mold the largescale patterns of vegetation structure within seagrass
landscapes has been suggested with regard to both
gap and patch formation. For example, Patriquin
(1975) and Wanless (1981) discussed the appearance
of “blowouts” in tropical seagrass systems occurring
with intense wave and current energy during periods of elevated onshore winds associated with cold
fronts in the winter and tropical disturbances in the
summer. Kirkman and Kirkman (2000) recently reported blowouts in Australian seagrass beds. Strong
physical forces on seagrasses may also affect longterm patterns of seagrass patch dynamics by influencing sexual reproduction. Inglis (2000) noted that
the formation of persistent seed banks was reduced in
seagrass habitats in Australia that experience strong
water flow and suggested that large scale patterns of
seed distributions may reflect historical patterns of
seed production and disturbance.
The influence of the general hydrodynamic regime
on the seagrass landscape has been described in detail in a number of recent papers and is further discussed in Section VI, Chapter 8 (Koch et al.). Current
speeds, depth, and a relative exposure index (REI)
that incorporates wind speed, direction, and fetch
were used to describe areas of seagrass coverage off
North Carolina, and landscape pattern was found
to be strongly correlated with these physical factors (Fonseca and Bell, 1998). This relationship with
physical factors led to the development of a predictive model that can be used to simulate the effect of
development and restoration on seagrass landscapes
(Kelly et al., 2001; Fonseca et al., 2002). Although
predictive modeling of seagrass patch dynamics is
only in its earliest stages of development, a few other
attempts have also been made. Kendrick et al. (1999)
used aerial photographs to show that coalescence of
seagrass patches had occurred over a 20-year period
(1972–1993) on Success Bank in Australia. They
then developed a model for radial increase of patches
based on published rhizome elongation rates and
aerial photographs from 1972 that was moderately
successful in predicting the changes that were seen in
successive aerial photographs (1982 and 1993). Fonseca et al. (2004) also used spatially explicit models
to predict recovery (recolonization) of aboveground
631
Some recent examples of extensive destruction of
wide swaths of seagrass beds by large aggregations
of sea urchins have been reported. For example, Rose
et al. (1999) documented the removal of 82% of seagrass biomass and >95% of apical meristems along
the path of a relatively uncommon urchin aggregation in Florida, USA during a 1-year period. In a
Kenyan lagoon, Alcoverro and Mariani (2002) projected the recovery time of seagrasses from urchin
disturbance at approximately 99 months, while they
estimated the return interval of the disturbance itself at only 44 months, thereby identifying urchin
activity as an important structuring agent in selected
locations. Dugongs, which feed on both above- and
below-ground components of plants, are reported
to leave feeding trails and thus enhance edge formation within some seagrass beds, especially those
of Halophila ovalis (Nakaoka and Aioi, 1999; see
also Masini et al., 2001) Lefebrve et al. (1999) also
reported that the West Indian manatee, Trichechus
manatus, fed extensively on patches of H. wrightii,
T. testudinum, and Syringodium filiforme in Florida
and Puerto Rico but that even heavily grazed patches
recovered from seasonal grazing impacts. Therefore, in some locations and at selected times, herbivore grazing events may remove significant amounts
of seagrass and alter the size, shape, and isolation/connectivity of patches.
3. Role of Abiotic Factors
In addition to biotic factors that may contribute to
the spatial arrangement of the seagrass landscape,
nutrient availability or physical factors such as sediment movement may also play an organizing role
(see also Marb` a et al., Chapter 6). The patchiness
of C. nodosa off of northeastern Spain was found
to be strongly correlated to the inshore migration
of sand waves (Marb` a and Duarte, 1995). Robbins
and Bell (2000) examined the effect of water depth
over a small range (30–100 cm) in mixed patches of
H. wrightii and T. testudinum on a shallow shoal in
Florida, USA. Seasonal species-specific depth preferences were noted, but hydrodynamic factors were
suggested as potentially more important for seagrass
distribution than depth preferences alone (Robbins
and Bell, 2000). The effects of nutrient availability at
the landscape scale are less straightforward. Nutrient
availability was found to be unrelated to the patch
dynamics of Z. marina in a fjord in the Baltic Sea
(Worm and Reusch, 2000). While both slow-release
NPK treatments and treatments with live mussels
added to plots increased porewater nutrient levels,
neither treatment showed consistent results in terms
of seagrass shoot growth or shoot addition (Worm
and Reusch, 2000).
The idea that physical energy may mold the largescale patterns of vegetation structure within seagrass
landscapes has been suggested with regard to both
gap and patch formation. For example, Patriquin
(1975) and Wanless (1981) discussed the appearance
of “blowouts” in tropical seagrass systems occurring
with intense wave and current energy during periods of elevated onshore winds associated with cold
fronts in the winter and tropical disturbances in the
summer. Kirkman and Kirkman (2000) recently reported blowouts in Australian seagrass beds. Strong
physical forces on seagrasses may also affect longterm patterns of seagrass patch dynamics by influencing sexual reproduction. Inglis (2000) noted that
the formation of persistent seed banks was reduced in
seagrass habitats in Australia that experience strong
water flow and suggested that large scale patterns of
seed distributions may reflect historical patterns of
seed production and disturbance.
The influence of the general hydrodynamic regime
on the seagrass landscape has been described in detail in a number of recent papers and is further discussed in Section VI, Chapter 8 (Koch et al.). Current
speeds, depth, and a relative exposure index (REI)
that incorporates wind speed, direction, and fetch
were used to describe areas of seagrass coverage off
North Carolina, and landscape pattern was found
to be strongly correlated with these physical factors (Fonseca and Bell, 1998). This relationship with
physical factors led to the development of a predictive model that can be used to simulate the effect of
development and restoration on seagrass landscapes
(Kelly et al., 2001; Fonseca et al., 2002). Although
predictive modeling of seagrass patch dynamics is
only in its earliest stages of development, a few other
attempts have also been made. Kendrick et al. (1999)
used aerial photographs to show that coalescence of
seagrass patches had occurred over a 20-year period
(1972–1993) on Success Bank in Australia. They
then developed a model for radial increase of patches
based on published rhizome elongation rates and
aerial photographs from 1972 that was moderately
successful in predicting the changes that were seen in
successive aerial photographs (1982 and 1993). Fonseca et al. (2004) also used spatially explicit models
to predict recovery (recolonization) of aboveground