Geology Reference
In-Depth Information
effect was not suffi cient to return shallow-water
accretion to a position of primacy.
It is interesting to note that a weak but positive
correlation was found between the rates of sea-
level rise and accretion. Again this was probably
infl uenced by a lack of accommodation space for
shallow-water reefs during intervals of slowed
sea-level rise. However, the higher rates of accre-
tion during periods of rapidly rising sea level were
not found at mid-depths but rather in some of
the deepest calculated palaeowater depths.
Core PAR-19 (Fig. 10b) built at an average rate
exceeding 9 m kyr
1
. Despite starting off in 15 m
of water, a depth generally considered too deep
to allow signifi cant accretion by branching corals,
this
A. palmata
reef was actually catching up with
rapidly rising sea level when accretion suddenly
stopped
c
. 9000 years ago; it was in water only 5 m
deep by that time. Similarly, the submerged reefs
off Barbados cored by Fairbanks (1989) accreted
at rates higher than 6 m kyr
1
during perhaps the
most rapid sea-level rise in the Holocene. All of
this suggests reconsideration of existing Holocene
reef models, which are based largely on strong
depth- and species-dependant accretion.
2
The decrease in average accretion rate from
3.58 m kyr
1
in shallow water to 2.76 m kyr
1
between depths of 20 and 25 m was less than
the standard deviations measured at any depth
(2.20-3.13 m kyr
1
), and was not statistically
signifi cant.
Average accretion rates in reefs dominated by
3
branching
A. palmata
(3.83 m kyr
1
) were not
statistically different than those from intervals
of massive species (3.07 m kyr
1
).
Bioerosion and export of detritus into deeper
4
reef environs undoubtedly exert some control
on this pattern. Whether they can make up the
difference between previously assumed trends
and those shown here awaits further study.
Nevertheless, the role of bioerosion emerges as
a factor that is quantitatively important in not
only infl uencing the character of internal reef
fabrics, but also affecting the rate at which reefs
build as well.
Few reefs have built faster than the rate of
5
sea-level rise in the early and mid-Holocene.
Therefore, drowned reefs should be more
expected than has been suggested by Schlager's
(1981) 'drowning paradox'. This has an impor-
tant bearing on understanding the relationship
between reef accretion and rising sea level, both
past and future.
Coral reefs are structures that are built as much
CONCLUSIONS
6
The fi ndings of this study appear to confl ict with
two of the most basic rules of Holocene coral-reef
accretion. There are undoubted inaccuracies in
the methods used to log the cores and the assump-
tions used in analysing the data from the literature.
However, it is diffi cult to imagine any combina-
tion of factors that, if changed, could make the
pattern seen in Figs 7-10 look like those in Figs 2
and 3. The next step is to revisit the studies from
which the existing concepts were developed, and
determine whether the reef-wide patterns that
they examined were actually different than those
shown here or whether the models derived from
them were simply a very good explanation of what
was known at the time, but now need to recon-
sider. Similarly, the results reported above should
be used to resolve how such different conclusions
can be reached and whether these should be used
to formulate new models.
Based on the fi ndings discussed above, the
following conclusions are offered.
by physical sedimentation as by calcifi cation by
corals, coralline algae and other large calcifi ers.
In short, reef corals grow; coral reefs accrete.
ACKNOWLEDGEMENTS
This paper is indirectly the result of nearly
30 years of core investigations that were supported
by the National Science Foundation, the National
Oceanic and Atmospheric Administration
(National Undersea Research Program and Sea
Grant), and the National Park Service. More
recently, our understanding of Holocene and
older fossil reefs in outcrop have been made pos-
sible through funding by the Petroleum Research
Fund. I am indebted to an army of colleagues
and students who have toiled with me to collect
many of the samples upon which these conclu-
sions are based, in particular, Ivan Gill, Randy
Burke, Heinrich Zankl, Hank Tonnemacher and
Wilson Ramirez. Their names are absent from this
paper only to protect the innocent. I also want to
acknowledge the support of colleagues at West
Indies Laboratory who served as my mentors or
1
No strong depth-related decrease in reef-accre-
tion rate was found in cores from reefs that built
in water depths between 0 and 25 m.
