Chapter 21 Seagrasses, Fish, and Fisheries
521
Manipulative experiments are difficult to perform at
meaningful scales and often are regarded as socially
unacceptable (Nagelkerken et al., 2001; Vanderklift and Jacoby, 2003). When events result in loss of
seagrass, these perturbations should be used as an
opportunity for experiments (see for example, Underwood, 1996).
B. Correlations Between Seagrass Cover
and Fisheries Catch
Several studies have reported a significant, positive linear relationship between commercial landings of penaeid shrimp and the area of vegetated
wetlands, but similar correlations generally have not
been shown for other species (see references within
Orth and van Montfrans, 1990). Orth and van Montfrans (1990) found no significant relationship between landings of blue crab (20 year averages) and
habitat quantity (total seagrass, total marsh, or total vegetated area) across broad latitudinal distances
(four major regions of the USA), but similar comparisons within the Gulf of Mexico showed a significant positive relationship of blue crab landings
with total vegetated area. When a log-linear regression model was used to estimate the relationship between catch per unit effort of blue crab and abundance of submerged aquatic vegetation (SAV) in the
previous year the relation between catch per unit
effort (CPUE), SAV lagged by 1 year, effort and
time was highly significant suggesting a relationship between seagrass cover and subsequent blue
crab catches (Anderson, 1989). Although there are
problems with using fishery dependent landings, few
fisheries-independent data are available to use for
making comparisons.
The relationship between seagrass habitat and
fisheries production has also been investigated in
terms of the possible use of seagrass as a nursery,
feeding, or spawning area by fish. A seagrass residency index (SRI) was constructed, which modeled
residence time of particular commercially and recreationally important fish in seagrass habitats in South
Australia (Scott et al., 2000). The SRI indicated that
garfish (Hyporhamphus melanochir, Hemiramphidae) and King George whiting (Sillaginodes punctata, Sillaginidae) were more likely to be affected
by changes in abundance of seagrass beds than
Australian salmon (Arripis truttacea, Arripididae)
or snapper (Pagrus auratus, Sparidae). A SRI was
the first step toward developing a spatial model of
seagrass–fish habitat that would allow quantitative
assessment of the economic value of seagrass habitats, particularly for fisheries production (Scott et al.,
2000).
C. Trophic Linkages
Many of the small fish which are found in seagrass beds may be important prey for species utilizing other habitats. For example, more than 30%
of the 93 fish species sampled from a variety of
habitats in Groote Eylandt (Gulf of Carpentaria,
Australia) were primarily piscivorous with the main
fish species consumed being Mugilidae, Clupeidae,
and Monacanthidae (Brewer et al., 1995). However,
few studies have sampled both predatory fishes and
their prey over the same spatial and temporal scales.
Abundances of small fishes were negatively correlated with the abundances of a transient piscivorous
fish Arripis truttacea (Arripididae) over the same
temporal scales (Hindell et al., 2000b). Based on
the diet of A. truttacea, Hindell et al. (2000b) suggested that piscivory may influence the assemblage
structure of small fishes in seagrass beds. They also
investigated the influence of predatory fish on the
abundance of small fish in seagrass (structured) and
non-vegetated (unstructured) habitats by manipulating the presence of piscivorous fishes using exclusion and enclosure cages (Hindell et al., 2000a).
When predators were excluded, the relative increase
of small fish was greater over non-vegetated habitats
than seagrass, suggesting that the structural complexity of the seagrass may somehow mediate predation (Hindell et al., 2000a, 2002). At the level of an
individual species, predation strongly influenced the
distribution of King George whiting (Sillaginodes
punctata, Sillaginidae) within a location, but this was
likely a behaviorally mediated response rather than
direct predation (Hindell et al., 2002). Experiments
were conducted at multiple (n = 3) locations within
an estuary (Port Phillip Bay), but the results differed
among locations. For example, at one location predation was not important, but at another there was a
strong predation effect (Hindell et al., 2001). These
results were thought to reflect differences in abundances of predatory fish between the two locations.
In another experimental study, Eggleston et al.
(1998b) investigated trophic links between seagrass
and patch reefs within a large seagrass bed in
the tropics. Prior to deployment of patch reefs,
densities of crabs were similar at experimental
521
Manipulative experiments are difficult to perform at
meaningful scales and often are regarded as socially
unacceptable (Nagelkerken et al., 2001; Vanderklift and Jacoby, 2003). When events result in loss of
seagrass, these perturbations should be used as an
opportunity for experiments (see for example, Underwood, 1996).
B. Correlations Between Seagrass Cover
and Fisheries Catch
Several studies have reported a significant, positive linear relationship between commercial landings of penaeid shrimp and the area of vegetated
wetlands, but similar correlations generally have not
been shown for other species (see references within
Orth and van Montfrans, 1990). Orth and van Montfrans (1990) found no significant relationship between landings of blue crab (20 year averages) and
habitat quantity (total seagrass, total marsh, or total vegetated area) across broad latitudinal distances
(four major regions of the USA), but similar comparisons within the Gulf of Mexico showed a significant positive relationship of blue crab landings
with total vegetated area. When a log-linear regression model was used to estimate the relationship between catch per unit effort of blue crab and abundance of submerged aquatic vegetation (SAV) in the
previous year the relation between catch per unit
effort (CPUE), SAV lagged by 1 year, effort and
time was highly significant suggesting a relationship between seagrass cover and subsequent blue
crab catches (Anderson, 1989). Although there are
problems with using fishery dependent landings, few
fisheries-independent data are available to use for
making comparisons.
The relationship between seagrass habitat and
fisheries production has also been investigated in
terms of the possible use of seagrass as a nursery,
feeding, or spawning area by fish. A seagrass residency index (SRI) was constructed, which modeled
residence time of particular commercially and recreationally important fish in seagrass habitats in South
Australia (Scott et al., 2000). The SRI indicated that
garfish (Hyporhamphus melanochir, Hemiramphidae) and King George whiting (Sillaginodes punctata, Sillaginidae) were more likely to be affected
by changes in abundance of seagrass beds than
Australian salmon (Arripis truttacea, Arripididae)
or snapper (Pagrus auratus, Sparidae). A SRI was
the first step toward developing a spatial model of
seagrass–fish habitat that would allow quantitative
assessment of the economic value of seagrass habitats, particularly for fisheries production (Scott et al.,
2000).
C. Trophic Linkages
Many of the small fish which are found in seagrass beds may be important prey for species utilizing other habitats. For example, more than 30%
of the 93 fish species sampled from a variety of
habitats in Groote Eylandt (Gulf of Carpentaria,
Australia) were primarily piscivorous with the main
fish species consumed being Mugilidae, Clupeidae,
and Monacanthidae (Brewer et al., 1995). However,
few studies have sampled both predatory fishes and
their prey over the same spatial and temporal scales.
Abundances of small fishes were negatively correlated with the abundances of a transient piscivorous
fish Arripis truttacea (Arripididae) over the same
temporal scales (Hindell et al., 2000b). Based on
the diet of A. truttacea, Hindell et al. (2000b) suggested that piscivory may influence the assemblage
structure of small fishes in seagrass beds. They also
investigated the influence of predatory fish on the
abundance of small fish in seagrass (structured) and
non-vegetated (unstructured) habitats by manipulating the presence of piscivorous fishes using exclusion and enclosure cages (Hindell et al., 2000a).
When predators were excluded, the relative increase
of small fish was greater over non-vegetated habitats
than seagrass, suggesting that the structural complexity of the seagrass may somehow mediate predation (Hindell et al., 2000a, 2002). At the level of an
individual species, predation strongly influenced the
distribution of King George whiting (Sillaginodes
punctata, Sillaginidae) within a location, but this was
likely a behaviorally mediated response rather than
direct predation (Hindell et al., 2002). Experiments
were conducted at multiple (n = 3) locations within
an estuary (Port Phillip Bay), but the results differed
among locations. For example, at one location predation was not important, but at another there was a
strong predation effect (Hindell et al., 2001). These
results were thought to reflect differences in abundances of predatory fish between the two locations.
In another experimental study, Eggleston et al.
(1998b) investigated trophic links between seagrass
and patch reefs within a large seagrass bed in
the tropics. Prior to deployment of patch reefs,
densities of crabs were similar at experimental