trapped in embayments (e.g., Pioneer Bay, Orpheus Island
[Hopley et al., 1983]) leeside island spits (e.g., Rattlesnake
Island, GBR [Hopley, 1982]), and deltaic gravels (e.g.,
Myall Reef, Cape Tribulation, GBR [Partain and Hopley,
1989]) (Figure 5c). Typically (but not exclusively) these
reefs initiate near the shoreline in relatively shallow water
and prograde seaward as vertical accommodation space is
limited, often by late Holocene sea-level fall. In such cases,
as the reef progrades into deeper water only the upper few
meters of its internal structure may be distinctly reefal, with
a shallow reef framework capping a detrital and often
muddy deposit below. This is the case of many fringing
reefs surrounding the high islands of the GBR and in Torres
Strait (Woodroffe et al., 2000; Hopley et al., 2007), and this
has also been documented for fringing reefs off Mozambique (Perry, 2005) and in South East Asia. Tudhope and
Scoffin (1994) demonstrated that the fringing reef at
Phuket, which is dominated by a Porites framework (see
Porites), has prograded over mud by a process of block toppling at the reef front as sea level has fallen since the
mid-Holocene.
Pulsed progradation
This model includes fringing reefs developed by episodically building a new reef structure parallel to the existing
reef front, and then backfilling the intervening space with
unconsolidated reef sediments (Figure 5d). Antecedent
topographic controls are inferred as a probable control of
this accretion pattern, but other factors may also be
involved. The fringing reef at Yam Island, Torres Strait
has developed in this way, indicated by dated subsurface
corals from cores and fossil microatolls across the reef flat
(Woodroffe et al., 2000). A similar pattern of reef development is described at Mangaia in the Cook Islands where
two separate reef crests occur over the reef flat – an older
reef crest over the emergent inner reef flat and the younger
lower reef crest to seaward; behind each lies a backreef
filled with detrital sediment (Yonekura et al., 1988). Sealevel fall is proposed as the principal driving factor for this
episodic accretion, with the first phase concluding around
4,000 years ago, and the younger reef crest forming
1.7 m below the first around 2,000 years later (Yonekura
et al., 1988). At the resort reef on Dunk Island, GBR,
two separate phases of reef development can also be identified, one between 7,000 and 4,500 years BP and another
after 1,600 years. In this lower energy setting, reef growth
in both phases occurred as massive colonies grew to sea
level and inter-colony spaces were infilled with reef rubble
and terrigenous sediment (Perry and Smithers, 2010).
Reef crest and backfill
In this growth model the reef crest reaches sea level more
rapidly than the backreef to form a shallow lagoon that is
later infilled (Figure 5e). This mode of growth is inferred
for fringing reefs in the West Indies where sea level has
only recently been achieved (Adey, 1978; Kennedy and
Woodroffe, 2002), but also appears relevant to fringing
reefs in areas where there is no evidence of higher Holocene sea levels or where reef growth rates have been slow
so that fringing reefs have only recently been sea-level
constrained. Coring from fringing reefs at Mauritius and
the Seychelles in the Western Indian Ocean, for example,
indicate growth histories of infill behind a reef crest that
reached sea level first, forming a shallow backreef lagoon
(Montaggioni and Faure, 1997; Braithwaite et al., 2000).
Similarly, at Lord Howe Island, a high-latitude reef, the
broad reef crest around 200 m wide reached sea level
around 5,000 years ago forming backreef lagoon (around
2 km wide and averaging 1.5 m deep) that has been slowly
infilling since (Kennedy and Woodroffe, 2000), and
Kan et al. (1995) suggest the fringing reef at Minna Island,
Japan has a comparable history, complying with this
growth model.
Storm-reworked framework and infill
This growth model occurs in storm affected settings, and
involves the destruction and reworking of reef framework
by episodic storms, between which detrital material may
be stabilized and form the substrate for the next phase of
reef construction by calcifying reef organisms
(Figure 5f ). This model has been argued to be applicable
to the fringing reefs surrounding Grand Cayman
(Blanchon and Jones, 1995; Blanchon et al., 1997); storms
have been linked to fringing reef growth and morphology
in the Seychelles (Braithwaite et al., 2000). An internal
structure dominated by detrital carbonate clasts, often
deposited in distinct units was interpreted as signifying
the importance of episodic accretion events such as
cyclones on normally low-energy inshore reefs on the
GBR (Smithers and Larcombe, 2003; Perry and Smithers,
2006), and given the importance of detrital material in
most fringing reef structures it is probable that this model
will also fit other areas affected by episodic high-energy
storms.
Other models
The six models above are applicable to relatively stable
and “typical” shorelines. Fringing reefs occur in other tectonic and physical settings where additional but less common modes of reef growth can occur. On tectonically
rising or subsiding coastlines the effect is to reduce or
increase the amount of accommodation space respectively. As discussed earlier, where accommodation space
is limited – as may occur on uplifted coasts – progradation
dominates, whereas subsidence at rates higher than vertical reef accretion provides increased accommodation
space – and vertical accretion tends to dominate reef
growth (see Figure 4). Cabioch et al.’s study (1999) of
fringing reefs around Grand Terre, New Caledonia elegantly demonstrates these effects, with different parts of
the coastline subsiding and uplifting at rapid and slow
rates. Where subsidence is rapid, relatively thick Holocene reefs have developed, where subsidence is relatively
slow, only thin Holocene reefs have developed, and where
uplift has occurred Holocene reefs comprise only a thin
FRINGING REEFS
439
[Hopley et al., 1983]) leeside island spits (e.g., Rattlesnake
Island, GBR [Hopley, 1982]), and deltaic gravels (e.g.,
Myall Reef, Cape Tribulation, GBR [Partain and Hopley,
1989]) (Figure 5c). Typically (but not exclusively) these
reefs initiate near the shoreline in relatively shallow water
and prograde seaward as vertical accommodation space is
limited, often by late Holocene sea-level fall. In such cases,
as the reef progrades into deeper water only the upper few
meters of its internal structure may be distinctly reefal, with
a shallow reef framework capping a detrital and often
muddy deposit below. This is the case of many fringing
reefs surrounding the high islands of the GBR and in Torres
Strait (Woodroffe et al., 2000; Hopley et al., 2007), and this
has also been documented for fringing reefs off Mozambique (Perry, 2005) and in South East Asia. Tudhope and
Scoffin (1994) demonstrated that the fringing reef at
Phuket, which is dominated by a Porites framework (see
Porites), has prograded over mud by a process of block toppling at the reef front as sea level has fallen since the
mid-Holocene.
Pulsed progradation
This model includes fringing reefs developed by episodically building a new reef structure parallel to the existing
reef front, and then backfilling the intervening space with
unconsolidated reef sediments (Figure 5d). Antecedent
topographic controls are inferred as a probable control of
this accretion pattern, but other factors may also be
involved. The fringing reef at Yam Island, Torres Strait
has developed in this way, indicated by dated subsurface
corals from cores and fossil microatolls across the reef flat
(Woodroffe et al., 2000). A similar pattern of reef development is described at Mangaia in the Cook Islands where
two separate reef crests occur over the reef flat – an older
reef crest over the emergent inner reef flat and the younger
lower reef crest to seaward; behind each lies a backreef
filled with detrital sediment (Yonekura et al., 1988). Sealevel fall is proposed as the principal driving factor for this
episodic accretion, with the first phase concluding around
4,000 years ago, and the younger reef crest forming
1.7 m below the first around 2,000 years later (Yonekura
et al., 1988). At the resort reef on Dunk Island, GBR,
two separate phases of reef development can also be identified, one between 7,000 and 4,500 years BP and another
after 1,600 years. In this lower energy setting, reef growth
in both phases occurred as massive colonies grew to sea
level and inter-colony spaces were infilled with reef rubble
and terrigenous sediment (Perry and Smithers, 2010).
Reef crest and backfill
In this growth model the reef crest reaches sea level more
rapidly than the backreef to form a shallow lagoon that is
later infilled (Figure 5e). This mode of growth is inferred
for fringing reefs in the West Indies where sea level has
only recently been achieved (Adey, 1978; Kennedy and
Woodroffe, 2002), but also appears relevant to fringing
reefs in areas where there is no evidence of higher Holocene sea levels or where reef growth rates have been slow
so that fringing reefs have only recently been sea-level
constrained. Coring from fringing reefs at Mauritius and
the Seychelles in the Western Indian Ocean, for example,
indicate growth histories of infill behind a reef crest that
reached sea level first, forming a shallow backreef lagoon
(Montaggioni and Faure, 1997; Braithwaite et al., 2000).
Similarly, at Lord Howe Island, a high-latitude reef, the
broad reef crest around 200 m wide reached sea level
around 5,000 years ago forming backreef lagoon (around
2 km wide and averaging 1.5 m deep) that has been slowly
infilling since (Kennedy and Woodroffe, 2000), and
Kan et al. (1995) suggest the fringing reef at Minna Island,
Japan has a comparable history, complying with this
growth model.
Storm-reworked framework and infill
This growth model occurs in storm affected settings, and
involves the destruction and reworking of reef framework
by episodic storms, between which detrital material may
be stabilized and form the substrate for the next phase of
reef construction by calcifying reef organisms
(Figure 5f ). This model has been argued to be applicable
to the fringing reefs surrounding Grand Cayman
(Blanchon and Jones, 1995; Blanchon et al., 1997); storms
have been linked to fringing reef growth and morphology
in the Seychelles (Braithwaite et al., 2000). An internal
structure dominated by detrital carbonate clasts, often
deposited in distinct units was interpreted as signifying
the importance of episodic accretion events such as
cyclones on normally low-energy inshore reefs on the
GBR (Smithers and Larcombe, 2003; Perry and Smithers,
2006), and given the importance of detrital material in
most fringing reef structures it is probable that this model
will also fit other areas affected by episodic high-energy
storms.
Other models
The six models above are applicable to relatively stable
and “typical” shorelines. Fringing reefs occur in other tectonic and physical settings where additional but less common modes of reef growth can occur. On tectonically
rising or subsiding coastlines the effect is to reduce or
increase the amount of accommodation space respectively. As discussed earlier, where accommodation space
is limited – as may occur on uplifted coasts – progradation
dominates, whereas subsidence at rates higher than vertical reef accretion provides increased accommodation
space – and vertical accretion tends to dominate reef
growth (see Figure 4). Cabioch et al.’s study (1999) of
fringing reefs around Grand Terre, New Caledonia elegantly demonstrates these effects, with different parts of
the coastline subsiding and uplifting at rapid and slow
rates. Where subsidence is rapid, relatively thick Holocene reefs have developed, where subsidence is relatively
slow, only thin Holocene reefs have developed, and where
uplift has occurred Holocene reefs comprise only a thin
FRINGING REEFS
439