Once excavating species have bitten off fragments of
substrate, the fragments pass into the muscular pharyngeal
mill at the beginning of the gut. The mill consists of grinding surfaces covered in teeth which are continuously
replaced, and as these surfaces are moved back and forwards the coral fragments are broken down into a fine
powder releasing the algae and breaking the plant cells
releasing the nutrients which are then absorbed by the fish
and the fine powder is released into the water column
(Bellwood and Choat, 1990) (Figure 2f). It has been
shown that the gut contains the necessary enzymes to
breakdown the cellulose plant walls (Choat et al., 2002).
Rates of grazing
Rates of grazing by parrotfishes (Figure 2e) have been estimated from 0.61 to 1.68 kg CaCO 3 m
À2 year
À1
in Barbados
(Frydyl and Stearn, 1978), 0.05 to 0.9 kg CaCO 3 m
À2
year
À1
at La Réunion, Indian Ocean to 0.7–3.30 CaCO 3 m
À2
year
À1
at Moorea, French Polynesia (Peyrot-Clausade et al.,
2000), 0.9–3.89 kg CaCO 3 m
À2 year
À1
– on inner reefs,
increasing to 5.2–8.4 kg CaCO 3 m
À2 year
À1
on mid shelf
reefs and 32.3 kg CaCO 3 m
À2 year
À1
on the outer shelf reef
crest and 23.1 kg CaCO 3 m
À2 year
À1 on the reef flat with
very little erosion occurring on the outer shelf reef slope
or back reef habitats (08.–1.8 kg CaCO 3 m
À2 year
À1
) on
the Great Barrier Reef (Hoey and Bellwood, 2008). In part
this is a reflection of the distribution of grazers across the
reef (Russ, 1984) which has been well documented for the
Great Barrier Reef. A variety of methods have been used
to estimate these rates so that some caution needs to be taken
in interpreting these results and obviously rates depend on
species, size of individuals, location of study both within
and between reefs and methods used. Some workers have
measured the amounts of calcium carbonate in the gut
(Peyrot-Clausade et al., 2000) whereas others (Bellwood,
1986; Bruggemann et al., 1996; Frydl and Stearn, 1978;
Rotjan and Lewis, 2005, 2006) have measured the size of
the bite marks, depth of excavation and observed the frequency of feeding and calculated the amount of substrate
removed and factored in size and densities of the fish
populations to obtain rates of loss over the reef.
On healthy reefs, bioerosion by parrotfishes is the dominant agent of grazing, and typically loss of substrate by
these fishes across the reef tends to be balanced by net calcification (Hoey and Bellwood, 2008).
Evidence of grazing from fossil reefs
Parrotfishes are well preserved on fossil reefs and the
oldest ones including a species of Bolbometopon, an
eroder are all of Miocene age (Bellwood and Schultz,
1991). Molecular data suggest that the basal parrotfish
division into seagrass and reef clades occurred approximately 42 million years ago (Streelman et al., 2002).
Although the feeding mode of the reef clade is equivocal,
the origin of this lineage at 42 Ma provides an independent
estimate of the maximum age of parrotfish bioerosion. The
impact of herbivores, therefore, may have had two phases,
with a rise in nonexcavating grazing prior to the early
Eocene and the advent of deep excavating fish herbivory
sometime later, between 42 and 5 Ma (Bellwood, 2003).
Obviously interactions between herbivory and coral reefs
have been occurring for a very long time with changes in
the composition of the coralline algal crusts of substrates
associated with an increase in the density of fish grazers
(Bellwood, 2003, Figure 7). Much of this evidence is from
analyzing the mouth parts of the fish present in various fossil reefs, as actual bite marks or evidence of grazing are
unlikely to have been preserved. More recent molecular
studies are confirming these timelines (Read et al., 2006).
Predation on live coral
A great variety of organisms prey on live coral predators
and Rotjan and Lewis (2008) provides a detail list by species (includes both vertebrates and invertebrates) and geographical regions. They distinguish between species
which remove only mucus, coral tissue, or skeleton, and
they provide rates of consumption reported in the literature, which were based either on the number of bites per
minute or the % of live coral in the gut.
All this grazing activity which removes the surface
layer of the coral substrate exposes a new surface which
is then rapidly recolonized again. Grazers while targeting
the endolithic algae also collect any other type of boring
organisms living in the surface layers of the coral skeleton
(Rotjan and Lewis, 2005).
Determining rates of bioerosion
Experimental studies have measured rates of bioerosion
using blocks of recently killed colonies of Porites which
have been attached to the substrate for varying lengths of
time (Figure 3a). Blocks are then sliced and measured and
changes in dimensions determined, to determine loss of
substrate by external grazing, increases in dimensions by
accretion of coralline algae and losses by internal erosion
by macro and microborers (Figure 3b). Rates of loss by
borers involve calculating the volume of the burrows and
because of their characteristic shapes and sizes they can be
apportioned to each of the major groups of borers. By
knowing the density of the coral substrate the amount of calcium carbonate lost can be calculated and then figures
extrapolated to amounts per m
À2 year
À1 although obviously
the distribution of substrate available for colonization is not
uniform across a reef. These rates also include loss of substrate by physical and chemical erosion but separating them
from losses caused by bioerosion is almost imposible
(Peyrot-Clausade et al., 1995). All these processes also act
synergistically, for example as the substrate becomes
honeycombed by borers this facilitates physical erosion as
water is flushed through the substrate, or when pieces of
substrate are removed by grazers, new surfaces are exposed
facilitating further losses by chemical and physical erosion.
Rates of loss will vary across the reef depending on the
amount of live coral cover present as well as the morphological types present (massive, plate, branching, encrusting)
and the amount of dead coral substrate of varying ages
BIOEROSION
149
substrate, the fragments pass into the muscular pharyngeal
mill at the beginning of the gut. The mill consists of grinding surfaces covered in teeth which are continuously
replaced, and as these surfaces are moved back and forwards the coral fragments are broken down into a fine
powder releasing the algae and breaking the plant cells
releasing the nutrients which are then absorbed by the fish
and the fine powder is released into the water column
(Bellwood and Choat, 1990) (Figure 2f). It has been
shown that the gut contains the necessary enzymes to
breakdown the cellulose plant walls (Choat et al., 2002).
Rates of grazing
Rates of grazing by parrotfishes (Figure 2e) have been estimated from 0.61 to 1.68 kg CaCO 3 m
À2 year
À1
in Barbados
(Frydyl and Stearn, 1978), 0.05 to 0.9 kg CaCO 3 m
À2
year
À1
at La Réunion, Indian Ocean to 0.7–3.30 CaCO 3 m
À2
year
À1
at Moorea, French Polynesia (Peyrot-Clausade et al.,
2000), 0.9–3.89 kg CaCO 3 m
À2 year
À1
– on inner reefs,
increasing to 5.2–8.4 kg CaCO 3 m
À2 year
À1
on mid shelf
reefs and 32.3 kg CaCO 3 m
À2 year
À1
on the outer shelf reef
crest and 23.1 kg CaCO 3 m
À2 year
À1 on the reef flat with
very little erosion occurring on the outer shelf reef slope
or back reef habitats (08.–1.8 kg CaCO 3 m
À2 year
À1
) on
the Great Barrier Reef (Hoey and Bellwood, 2008). In part
this is a reflection of the distribution of grazers across the
reef (Russ, 1984) which has been well documented for the
Great Barrier Reef. A variety of methods have been used
to estimate these rates so that some caution needs to be taken
in interpreting these results and obviously rates depend on
species, size of individuals, location of study both within
and between reefs and methods used. Some workers have
measured the amounts of calcium carbonate in the gut
(Peyrot-Clausade et al., 2000) whereas others (Bellwood,
1986; Bruggemann et al., 1996; Frydl and Stearn, 1978;
Rotjan and Lewis, 2005, 2006) have measured the size of
the bite marks, depth of excavation and observed the frequency of feeding and calculated the amount of substrate
removed and factored in size and densities of the fish
populations to obtain rates of loss over the reef.
On healthy reefs, bioerosion by parrotfishes is the dominant agent of grazing, and typically loss of substrate by
these fishes across the reef tends to be balanced by net calcification (Hoey and Bellwood, 2008).
Evidence of grazing from fossil reefs
Parrotfishes are well preserved on fossil reefs and the
oldest ones including a species of Bolbometopon, an
eroder are all of Miocene age (Bellwood and Schultz,
1991). Molecular data suggest that the basal parrotfish
division into seagrass and reef clades occurred approximately 42 million years ago (Streelman et al., 2002).
Although the feeding mode of the reef clade is equivocal,
the origin of this lineage at 42 Ma provides an independent
estimate of the maximum age of parrotfish bioerosion. The
impact of herbivores, therefore, may have had two phases,
with a rise in nonexcavating grazing prior to the early
Eocene and the advent of deep excavating fish herbivory
sometime later, between 42 and 5 Ma (Bellwood, 2003).
Obviously interactions between herbivory and coral reefs
have been occurring for a very long time with changes in
the composition of the coralline algal crusts of substrates
associated with an increase in the density of fish grazers
(Bellwood, 2003, Figure 7). Much of this evidence is from
analyzing the mouth parts of the fish present in various fossil reefs, as actual bite marks or evidence of grazing are
unlikely to have been preserved. More recent molecular
studies are confirming these timelines (Read et al., 2006).
Predation on live coral
A great variety of organisms prey on live coral predators
and Rotjan and Lewis (2008) provides a detail list by species (includes both vertebrates and invertebrates) and geographical regions. They distinguish between species
which remove only mucus, coral tissue, or skeleton, and
they provide rates of consumption reported in the literature, which were based either on the number of bites per
minute or the % of live coral in the gut.
All this grazing activity which removes the surface
layer of the coral substrate exposes a new surface which
is then rapidly recolonized again. Grazers while targeting
the endolithic algae also collect any other type of boring
organisms living in the surface layers of the coral skeleton
(Rotjan and Lewis, 2005).
Determining rates of bioerosion
Experimental studies have measured rates of bioerosion
using blocks of recently killed colonies of Porites which
have been attached to the substrate for varying lengths of
time (Figure 3a). Blocks are then sliced and measured and
changes in dimensions determined, to determine loss of
substrate by external grazing, increases in dimensions by
accretion of coralline algae and losses by internal erosion
by macro and microborers (Figure 3b). Rates of loss by
borers involve calculating the volume of the burrows and
because of their characteristic shapes and sizes they can be
apportioned to each of the major groups of borers. By
knowing the density of the coral substrate the amount of calcium carbonate lost can be calculated and then figures
extrapolated to amounts per m
À2 year
À1 although obviously
the distribution of substrate available for colonization is not
uniform across a reef. These rates also include loss of substrate by physical and chemical erosion but separating them
from losses caused by bioerosion is almost imposible
(Peyrot-Clausade et al., 1995). All these processes also act
synergistically, for example as the substrate becomes
honeycombed by borers this facilitates physical erosion as
water is flushed through the substrate, or when pieces of
substrate are removed by grazers, new surfaces are exposed
facilitating further losses by chemical and physical erosion.
Rates of loss will vary across the reef depending on the
amount of live coral cover present as well as the morphological types present (massive, plate, branching, encrusting)
and the amount of dead coral substrate of varying ages
BIOEROSION
149