OCEANOGRAPHIC PROCESSES OF CORAL REEFS: Physical and Biological Links in the Great Barrier Reef - Chapter 9 potx

Biodiversity on the Great
Barrier Reef: Large-Scale
Patterns and Turbidity-
Related Local Loss of
Soft Coral Taxa
Katharina Fabricius and Glenn De’ath
CONTENTS
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
Methods. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130
Field Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130
Analytical Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131
Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132
Spatial Patterns in Soft Coral Richness, and the Influence
of Turbidity and Sedimentation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132
Spatial Distribution of Turbidity and Sedimentation . . . . . . . . . . . . . . . . . . . 133
Patterns in Soft and Hard Coral Cover. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133
Depth-Related Patterns. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134
Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
Acknowledgments. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138
INTRODUCTION
Indo-Pacific coral reefs contain globally the highest level of biodiversity of any
marine ecosystem, with the centre of this biodiversity located around the archipelago
of Malaysia, Indonesia, and the Philippines. The Great Barrier Reef (GBR) is part of
the Indo-Pacific biogeographic region, and contains a subset of the Indo-Pacific taxa
found in the most species-rich areas farther north, as well as species that are not found
anywhere else but on the GBR (Veron, 1995). Around 2800 coral reefs, extensive sea-
grass areas, species-rich soft- and hard-bottom inter-reefal and lagoonal ecosystems,
9
127
© 2001 by CRC Press LLC

continental slopes, and pelagic ecosystems are all represented within the Great
Barrier Reef Marine Park, which is the world’s largest World Heritage Area
(Wachenfeld et al., 1998).
Because of its vast size (348,000 km
2
area, stretching over 2000 km or 14° of
latitude) and its high biodiversity, surveys and species inventories have been carried
out only on a few taxonomic groups in small proportions of the marine park. Some
areas are still uncharted even for shipping purposes. Large-scale systematic mapping
of the major biotic groups such as scleractinian corals and fishes only began on a
large scale in the 1990s. Other groups which are extremely species-rich, such as
sponges, crustaceans, echinoderms, or molluscs, remain largely unmapped, although
some of these taxa are likely to hold key positions in the ecosystem.
In this chapter we summarise the patterns in biodiversity for an abundant and
species-rich group of organisms, commonly known as soft corals and sea fans, or
octocorals (class: Octocorallia, Order Alcyonacea). Soft corals are sessile, perennial,
and often long-lived corals. In contrast to the hard corals, they do not possess a mas-
sive external skeleton made of calcium carbonate; instead their colonies are sup-
ported by small calcareous needles or a hydroskeleton. Most “true” soft corals are
phototrophic, i.e., they contain symbiotic algae (zooxanthellae) in their tissue which,
depending on light, convert carbon dioxide into sugars, and thus supply the soft corals
with energy. Most “sea fans” do not host zooxanthellae, thus their food depends
entirely on material suspended in the water, a strategy called heterotrophy. Soft corals
occur in high abundances on many types of coral reefs. They may numerically dom-
inate reefs in turbid in-shore regions, as well as clear water reefs away from coastal
influences (Benayahu & Loya, 1981; Tursch & Tursch, 1982; Dinesen, 1983; Dai,
1990; Fabricius, 1997).
Soft coral abundances and the number of soft coral taxa found at any location (rich-
ness) are subject to relatively strong physical control (Fabricius & De’ath, 1997). Like
plants, they are inescapably subject to the light, wave, water quality, and sedimentary

environment where they settled as larvae. Biotic controls, such as predation, or over-
growth by neighbours appear to be relatively ineffective for soft coral abundances. In
contrast to the mass predation of hard corals by Acanthaster planci (De’ath & Moran,
1998), or mass “predation” of bêche-de-mer, trochus, giant clams, lobster, mud crabs,
sharks, predatory fishes, turtles, and dugong (to name just some) by Homo sapiens, no
large-scale mass mortalities are know for soft corals. The reasons for low biotic control
are their high concentrations of toxic or feeding-deterrent metabolites (e.g., Coll et al.,
1983; Sammarco et al., 1985; Maida et al., 1995) and low commercial value.
On the GBR, several hundred soft coral species coexist with around 350 species
of hard corals (Cnidaria: Scleractinia; Veron, 1995). Space competition between the
two groups may be important in areas of high densities but appears inconsequential
in regulating abundances before crowding sets in (Bak et al., 1982; Fabricius, 1997).
Competition is reduced because both groups occupy different trophic and physical
niches. Differences between the trophic niches of hard and soft corals are related to
two important morphological characteristics: First, efficient stinging cells allow hard
corals to actively capture zooplankton as food. In contrast, the stinging cells of soft
corals are poorly developed, hence their diet consists of predominantly small
128 Oceanographic Processes of Coral Reefs
© 2001 by CRC Press LLC
suspended particulate matter and picoplankton (Fabricius et al., 1995a and b; Ribes
et al., 1998; Fabricius & Dommisse, 2000). Second, the light-reflecting massive
skeleton in hard corals is covered only by a thin layer of zooxanthellae-loaded tissue,
providing for a high surface-area/volume ratio and hence very efficient photosynthe-
sis in hard corals. In contrast, the photosynthetic efficiency of the phototrophic soft
corals is low, due to the lack of a light-reflecting massive internal skeleton, and an
unfavourably low surface-area volume ratio (Fabricius & Klumpp, 1995).
This chapter presents the large-scale patterns of biodiversity in soft corals (here
used synonymously with taxonomic richness), and total hard and soft coral cover.
Both abundances (cover) and biodiversity are being used to assess the state of ecosys-
tems: low biodiversity and cover are both direct results of severe environmental con-

ditions, and low cover also indicates a recent disturbance (Done et al., 1996). Low
biodiversity can be the result of a high, or very low, frequency of episodic distur-
bance. In a frequently disturbed environment, speed of recolonisation determines
whether a taxon survives or not, as slow-colonising or slow-maturing taxa will be
unable to persist (Done, 1997). Under such circumstance, communities are charac-
terised by low biodiversity and low cover, with an overrepresentation of young, fast
colonising but competitively weak taxa. Occasionally, extended periods without
disturbance allow competitively strong taxa to monopolise areas by slowly outcom-
peting and replacing the less defensive neighbouring taxa. Under such rare circum-
stance, the communities are characterised by low biodiversity but a high level of
space occupancy, generally by large, old, and competitively strong individuals. The
maintenance of a high level of biodiversity of tropical coral reefs is often attributed
to an “intermediate” exposure to natural disturbances such as cyclones, floods, preda-
tors, or extreme temperatures, which relieve competition for space and facilitate the
coexistence of a high number of species (Connell, 1976).
Water pollution and overfishing are the two major types of chronic man-made
disturbance in coral reefs. Chronically, increased levels of runoff of sediments, nutri-
ents, and pesticides impinge on coastal reefs, with wide-ranging effects on corals and
other reef organisms (reviews in Pastorok & Bilyard, 1985; Rogers, 1990; Gabic
& Bell, 1993; Wilkinson, 1999). Sometimes responses to these chronic disturbances
are not obvious for several decades; however, a single severe disturbance event in a
chronically disturbed area can trigger a phase shift from reef-building hard corals to
non-reef-building taxa such as macro algae (Hughes, 1994; Done, 1992). Soft corals
also established and monopolised space on some reefs after disturbance of hard
corals, but such space monopolisation is restricted to a few taxa and a distinct type of
reef habitat (shallow in-shore fringing reefs in moderately clear water: reviewed in
Fabricius, 1998; Fabricius & Dommisse, 2000). It appears intuitive that chronic dis-
turbance reduces diversity, because only few taxa will be robust enough to persist.
The present study demonstrates that indeed the generic richness both of zooxanthel-
late and azooxanthellate soft corals is depressed in areas of reduced water clarity, one

of the consequences of terrestrial runoff of nutrients and soils (Rogers 1990;
Wolanski & Spagnol, in press). Such reduction in biodiversity will have to be con-
sidered in the debate of effects of chronic nutrient enrichment of in-shore reefs in
regions of intense land use.
Biodiversity on the Great Barrier Reef 129
© 2001 by CRC Press LLC
METHODS
F
IELD METHODS
A large-scale biodiversity survey and species inventory program were carried out on
the GBR between latitude 10 and 25°S. The surveys were designed to characterise
patterns of biodiversity and physical conditions within the GBR, as a baseline for
determining future trends and as a basis for identification of areas of highest protec-
tion value. The soft coral surveys were conducted on 161 reefs (~6% of the 2800
GBR reefs; Figure 1). On each reef, generally one to three sites (each in a different
location, depending on time and accessibility) were inspected. Up to five transects
were surveyed per site, each at a pre-defined depth-range (18 to 13 m, 13 to 8 m, 8 to
3 m, 3 to 1 m, and reef flat). All surveys were conducted by the first author, by scuba
diving over a transect typically 200 to 300 m long and 1 to 3 m wide, for 10 to 15 min,
or until no new taxa were encountered for several minutes. Longer transects were sur-
veyed in areas of low visibility to compensate for a narrower field of view. A total of
1346 transects at 361 sites were investigated.
The surveys were carried out using a rapid ecological assessment technique
(REA), based on abundance ratings of estimates of substratum cover in six ranked cat-
egories (initially developed for vegetation analyses by Braun-Blanquet (1964). REA
was chosen rather than the more conventional belt and line transects because of its
advantages in terms of area surveyed, time requirements, and the superior representa-
tion of rare and heterogeneously distributed taxa (the majority of taxa are rare in
highly diverse communities). A wide variety of REA methods have been developed,
assessed, and successfully applied to coral reef benthos surveys since the 1970s (e.g.,

Kenchington, 1978; Done, 1982; Dinesen, 1983; Miller & De’ath, 1995; Devantier
et al., 1998); we followed a protocol similar to that of Devantier et al. (1998).
During the survey and after completion of each transect, the following data were
recorded:
1. Relative abundances of taxa: 0 ϭ absent; 1 ϭ one or few colonies;
2 ϭ uncommon; 3 ϭ common; 4 ϭ abundant; and 5 ϭ dominant. Soft
corals were surveyed mostly at generic rather than species level because a
substantial proportion of species are still undescribed, and species identi-
fication requires a microscopic examination, which is unsuitable for large-
scale field surveys. Samples of unknown or uncertain colonies were
collected and later identified. Of the 61 genera recorded on the GBR,
only the 40 most common taxa were recorded in the early phase of the sur-
veys, and for consistency only these 40 taxa were included in the present
analyses.
2. Visual estimates of overall abundance (percent total cover) of soft corals
and hard corals. Cover was estimated in 2.5% increments from 1 to 10%,
in 5% increments from 10 to 30%, and in 10% increments for Ͼ30%
cover. An assessment of the precision of visual estimates of life coral cover
indicated that differences between experienced observers were not signif-
icant (Miller & De’ath, 1996).
130 Oceanographic Processes of Coral Reefs
© 2001 by CRC Press LLC
3. The following abiotic variables were estimated at all sites:
a. Sediment deposit on the reef substratum (particle sizes ranging from very
fine to moderately coarse), rated on a 4-point scale: 0 ϭ none, 1 ϭ thin
layer, 2 ϭ considerable amount of sediment which could be completely
resuspended by fanning, and 3 ϭ thick, deep layer of sediment.
b. Turbidity (measured as visibility, in meters). The method was a modi-
fied Secchi disc technique, in that the maximum visible distance of a
bright object was estimated horizontally at each survey site. A hori-

zontal distance was preferred over the traditional vertical Secchi dis-
tance, as the former is not affected by shallow depths (on outer-shelf
reefs, the bottom is often visible from the surface), and by surface
refraction (thus estimates are less affected by the azimuth of the sun,
cloud cover, and wave height).
ANALYTICAL METHODS
The first set of analyses was carried out on reef-averaged data, which is the relevant
scale for management and conservation of biodiversity. We modelled spatial variation
in richness, soft and hard coral cover, and physical variables using generalised addi-
tive models (Hastie & Tibshirani, 1990). Loess smoothers (Hastie & Tibshirani,
1990) were used to fit smoothed effects of both spatial and physical variables. The
degree of smoothness was minimised but sufficient to account for both spatial effects
and spatial correlation. The statistical software S-PLUS was used for all data analy-
ses (Statistical Sciences, 1995).
Latitude and longitude would normally be used for the spatial component of such
models. However, the GBR runs from ~SE to NW, and physical and ecological gra-
dients, which run typically across and (to a lesser degree) along the shelf, are there-
fore tilted 45° to the geodesic system. To improve the analysis and graphical
representation of the spatial patterns, the latitude/longitude data were converted into
relative distance across and along the GBR (Figure 2). Relative distance across the
GBR (henceforth: “across”) is defined as the distance of a site to the coast, divided
by the sum of distances to the coast and to the outer edge of the GBR. Relative dis-
tance along the GBR (henceforth: “along”) is similarly defined as the distance to the
northern end of the GBR divided by the sum of distances to the northern and south-
ern ends of the GBR. This has the effect of mapping the GBRMP to a rectangle, or
unit square if we assume that units across equate to units along (Figure 2). The coor-
dinates of the across–along system are locally orthogonal and run at right angles and
parallel to the coast, taking advantage of the fact that many processes are affected by
the natural geometry of the GBR. Such presentation gives better resolution particu-
larly of the steep gradients across the narrow shelf of the northern GBR.

Depth-related patterns were investigated at transect level, after dividing the data
into groups representing six GBR regions (Figure 1): the northern and southern reefs,
and three cross-shelf categories. The along-shore split was set at 19.5° latitude a zone
of transition for soft coral communities (unpublished data). The northern 55% along
included 901 transects, and the more homogenous southern 45% contained 445
Biodiversity on the Great Barrier Reef 131
© 2001 by CRC Press LLC
transects. Splits across the shelf were set at 38 and 85% of across, with those
reefs Ͻ38% representing the in-shore reefs, 38 to 85% representing the mid-shelf,
and those Ͼ0.85% across classified as outer-shelf reefs.
RESULTS
SPATIAL PATTERNS IN SOFT CORAL RICHNESS, AND THE INFLUENCE
OF
TURBIDITY AND SEDIMENTATION
Total soft coral richness was consistently highest on mid-shelf reefs, and declined
steeply toward the in-shore, and to a lesser extent toward the outer-shelf reefs
(Figure 3). Along the shelf, richness was highest between 14 and 18°S, and declined
slightly toward the north, and more strongly toward the south. The north-to-south
decline was most pronounced on mid- and outer-shelf reefs (reduction from 25 to 15,
and 22 to 13, respectively), and less so on near-shore reefs (reduction from ~12 gen-
era in the north to ~8 in the south). The “hotspot” of soft coral biodiversity on the
GBR was north of Cairns on the mid-shelf reefs, where up to 27 of the genera were
recorded on individual reefs. Richness was lowest (~8 genera per reef) on the south-
ern in-shore sites. Richness was well explained spatially, with 59.6% of variation
explained by the smooth surface (df ϭ 14.4).
Both water turbidity (visibility) and sediment deposits showed significant rela-
tionships with total richness when added to the spatial model. The required degree of
spatial smoothing decreased substantially (from 14.4 to 8.4 df), suggesting the phys-
ical variables accounted for local variation previously unexplained by the more flex-
ible spatial smoother. Visibility affected soft coral richness particularly strongly.

Richness was highest in areas where visibility was 10 m or greater, and declined
sharply in areas of Ͻ10 m visibility (Figure 3). The relationship of total richness to
visibility was non-linear and negligible in areas of Ͼ10 m visibility. In all other areas,
richness declined by ~1 taxon for each meter reduction in visibility. For example, two
reefs in the same region (i.e., similar across and along location), with visibilities of
10 and 5 m, would be likely to differ in richness by five genera. Visibility on its own
explained ~22% of the total variation in richness (Figure 4). Sedimentation also
affected generic richness, although to a weaker extent than visibility (Figure 3).
Richness increased linearly with increasing sedimentation, with an increase in sedi-
ment deposits of 1 rating increasing richness by 1.5 genera (4.3% of variation was
explained by sedimentation alone; Figure 4).
The pattern in richness of zooxanthellate taxa largely matched that of the rich-
ness of all taxa (Figure 5), as was expected since 28 of 40 taxa were zooxanthellate.
Richness of zooxanthellate taxa was greatest on the mid-shelf north of Cairns, and
depressed on in-shore reefs north of Townsville (relative distance along ϭ 0.45 to
0.8). Richness of the zooxanthellate taxa was affected by turbidity, similar to the
effect on the total richness: richness declined by ~0.7 taxa m
Ϫ1
at a visibility Ͻ10 m
(Figure 5). Again, effects were negligible at levels of Ͼ10 m visibility. However, sed-
iment had no effect on the richness of zooxanthellate taxa.
There were pronounced spatial differences in richness of the light-dependent
zooxanthellate taxa and the light-independent zooxanthellae-free taxa. Richness of
132 Oceanographic Processes of Coral Reefs
© 2001 by CRC Press LLC
zooxanthellae-free taxa almost continuously declined from north to south (Figure 6).
Cross-shelf patterns were weak, but richness was slightly higher on the mid- and inner-
shelf than on the outer-shelf. The in-shore area north of Townsville (poor in zooxan-
thellate taxa) was particularly rich in zooxanthellae-free taxa. Very few
zooxanthellae-free taxa were recorded in the southern mid- and outer-shelf reefs.

Effects of visibility on richness of the zooxanthellae-free taxa were again similar to
those on the total richness, being negligible at levels of Ͼ10 m. Richness of zooxan-
thellae-free taxa declined by ~0.2 taxa m
Ϫ1
decrease in visibility (Figure 6). Sediment
again had no effect on richness. Of a total of 54% variation of richness of the zooxan-
thellae-free taxa explained by a combined spatial and physical model, 9% were
explained by visibility on its own (Figure 7). In contrast, visibility explained 33%
of variation in richness of the zooxanthellate taxa (total variation explained by the
combined model: 63%).
SPATIAL DISTRIBUTION OF TURBIDITY AND SEDIMENTATION
Visibility was strongly related to relative distance across the shelf (Figure 8). This
pattern was stable and emerged despite the noise of natural seasonal and wind-related
variability. Lowest visibility (~4 m) was recorded on the innermost part of the shelf
between Townsville and Cape Flattery (0.45 to 0.75 along, 0 to 0.2 across). Visibility
was also low across the inner 40% of the wide shelf of the Broad Sound/Keppel
Island region (~0.1 to 0.25 along, 0 to 0.4 across), where tidal ranges exceed 5 m.
Both on mid- and outer-shelf reefs, the southern reefs tended to be more turbid than
those farther north. Along-shore differences were most pronounced on the outer-shelf
reefs: in the north, outer-shelf reefs had visibility of ~30 m, compared with only
~10 m in the south. Visibility was also related to the amount of sediment deposited
on a reef. While the combined spatial and sediment model explained 85% of the vari-
ation in visibility, the spatial model on its own accounted for 83%, and sediment on
its own for 57% of the variation (Figure 9).
Sediment deposits showed somewhat complementary cross-shelf patterns to that
of visibility, with highest values along the coast between Bowen and the Daintree
River, and lowest values on the outer-shelf reefs of the northern half of the GBR
(Figure 8). The spatial model explained 66% of variation in the data (Figure 9).
PATTERNS IN SOFT AND HARD CORAL COVER
Soft coral cover was greatest (~30%) in the area covering the Whitsunday to Palm

Islands groups (Figure 10). It was low in-shore in the area extending from south of
Townsville to the Daintree River in the north. It was also low on the outermost reefs
in the southern corner (Pompeys and Swains), and on the northern mid- and outer-
shelf reefs. The spatial model explained 46% of variation in soft coral cover. Soft
coral cover was unrelated to sedimentation and visibility.
Hard coral cover tended to increase with increasing distance from the coast
(Figure 10). It was highest (mean: 35%) on outer-shelfs between Townsville and
Cape Tribulation (0.5 to 0.7 along). The area of lowest cover was between the
Whitsundays and the Broad Sound (0.3 to 0.4 along). Hard coral cover aver-
aged Ͻ5% on 11 of the 161 surveyed reefs. The contribution of soft corals to the total
Biodiversity on the Great Barrier Reef 133
© 2001 by CRC Press LLC
coral cover was high in the central GBR on reefs located 30 to 40% across the shelf.
The spatial model explained only 21% of the variation in hard coral cover, and sedi-
mentation and visibility did not explain additional variation (Figure 11). The propor-
tion of soft coral cover to total cover followed from the patterns of soft and hard coral
cover; it was again unrelated to sedimentation and visibility (Figures 10 and 11).
The relationship between site-specific soft coral richness and soft and hard coral
cover within individual transects is shown in Figure 12. This analysis was done on tran-
sect level, since biotic variables are expected to interact locally. Soft coral richness
increased with soft coral cover in areas Ͻ8% soft coral cover, but remained at ~10 gen-
era per transect, independently of soft coral cover, where soft coral cover exceeded 8%.
Soft coral richness was highest in areas of ~10 to 20% hard coral cover, and gradually
declined toward 50% lower values in areas with very low or very high hard coral cover.
DEPTH-RELATED PATTERNS
Soft coral richness, defined here as the number of soft coral genera found at a site
(depth zone), varied to a greater extent within the upper 18 m of depth, than within
the 40- to 200-km distance across, and 2500 km along the shelf (Figure 13). Variation
(22%) was explained by depth on its own, whereas across and along together
explained 15%. Site-specific soft coral richness increased with depth at all shelf posi-

tions both in the north and the south (Figure 14).
Visibility was purely a function of shelf position and was independent of depth
(Figure 14). On in-shore reefs both in the north and the south, visibility averaged 7 m
(range: 0.5 to 18 m). Visibility on the northern mid-shelf and the southern mid- and
outer-shelf was about twice those of the in-shore, and on northern outer-shelf reefs,
visibility averaged 26 m.
The amount of sediment deposited on a reef site increased with depth and with
decreasing distance to the shore (Figure 14). The thickest sediment deposits were
found at or below 10 m depth on inner-shelf reefs. Sediment on deeper (10 to 15 m)
mid- and outer-shelf sites was similar (mean of 0.8 to 1.2) to those of very shallow
in-shore sites (0.8 to 1.2). The outer-shelf sites in the north had very low sediment
levels at any depth, and in the southern half there was no difference in sediment
deposits between mid- and outer-shelf reefs.
Mean total cover of soft corals and hard corals was strongly related to depth
(Figure 15). On the northern GBR, the depth of greatest mean soft coral cover moved
down-slope with increasing distance from the coast: on inner-shelf reefs, cover was
greatest at 2 to 5 m depth (20% cover), on mid-shelf reef at 10 m (12%), and on outer-
shelf reefs at ~15 m depth (17%). On the southern GBR, this pattern was maintained,
except for inshore-sites at 15 m with high cover, attributable to dense soft coral stands
at many sites on the Whitsunday Islands group.
Hard coral cover was a function of depth at any location across and along the
shelf (Figure 15). Differences between the northern and southern sector were small.
At all locations, hard coral cover was highest on the reef crests, dropping to lower val-
ues on the reef flat, and decreasing continuously with increasing depth on the outer
reef slopes. Highest values were 43 and 38% on crests of outer-shelf of the northern
and the southern reefs, respectively.
134 Oceanographic Processes of Coral Reefs
© 2001 by CRC Press LLC
The contribution of soft coral cover to total coral cover (soft corals plus hard
corals) increased with depth (Figure 15). It varied from around 40% to over 60% on

the inner-shelf, between 15 and 43% on mid-shelfs, and 10 to 40% on the outer-shelf
reefs. On mid- and outer-shelf reefs, it tended to be higher in the northern than in the
southern part of the GBR.
DISCUSSION
The generic richness of soft corals changed over three spatial dimensions: down the
reef slopes, across the continental shelf, and along the GBR. Additional differences
in richness between neighbouring reefs were explained by water turbidity, and to a
weaker extent by sediment deposition on the reef. Within any given area, more turbid
reefs had lower soft coral richness than reefs closeby in clearer water. Similarly, reefs
with higher sediment deposits were associated with slightly higher soft coral rich-
ness. The richness of both zooxanthellate and zooxanthellae-free taxa was affected
by turbidity, indicating that effects were not only related to reduced light exposure
(the larvae of zooxanthellae-free taxa often settle in low-light environments).
On coral reefs of the GBR, two well-described patterns in the distribution and
abundance of biota are (1) depth zonations and (2) zonations across the shelf accord-
ing to distance to the land and the edge of the continental shelf (Done, 1982;
Williams, 1982; Dinesen, 1983). Depth and distance to the land are in themselves not
causal but act as proxies for a range of co-varying abiotic and biotic environmental
variables such as turbidity, sedimentation, light, and wave exposure. The complex
interactions between such variables are not easily separated.
Patterns parallel to the coast and along the GBR are less understood than depth
zonations and cross-shelf differences, as few consistent and spatially comprehensive
data sets exist to date. A general decline in biodiversity on coral reefs away from the
equator has long been recognised. Gradients in sea surface temperature and associ-
ated seawater chemistry, restricted larval transport through ocean currents after the
ice age, and variations in rates of recruitment have been discussed as underlying
mechanisms (Veron, 1995). In the soft corals, three taxa (Clavularia,
Pachyclavularia, and Heliopora) were not recorded south of ~20° latitude, however
another three taxa (Pinnigorgia flava, Plumigorgia, and Isis hippuris) were common
in the south but rarely encountered north of 16° latitude (Fabricius & De’ath, 2000).

Therefore, the decline in soft coral richness toward the south was not generally due
to the complete absence of particular genera, but due to less frequent encounters of a
wide range of taxa. The centre of richness of zooxanthellate soft corals coincided
with the area of the GBR where a branch of the South Equatorial Current brings trop-
ical water from the Solomon Islands and Vanuatu across the eastern Coral Sea. The
current splits and bifurcates between 14 and 18° latitude (depending on season), and
water (and larval) movement is unidirectional toward both the north and the south
from the point of bifurcation (reviewed in Wolanski, 1994). It is unknown how these
currents affect dispersal and richness, but interestingly, species richness in hard
corals on the GBR is also highest in the same region, declining slightly toward the
north and steeply toward the south (Veron, 1995). The issue is complicated by the fact
Biodiversity on the Great Barrier Reef 135
© 2001 by CRC Press LLC
that the richness of zooxanthellae-free taxa showed a clear north–south gradient,
unrelated to that of the zooxanthellate hard and soft corals.
The total cover of soft and hard corals on the GBR remained uninfluenced by tur-
bidity and sedimentation. This finding indicates the potential for species replace-
ments: in certain circumstances, turbidity-tolerant taxa fill in the space for less
tolerant taxa, so cover remains the same but diversity declines. It also highlights the
need for detailed taxonomic inventories when conditions of coral reefs are to be
assessed. Total cover, which is the only parameter assessed in some environmental
studies, appears unsuitable for indicating changes such as increasing turbidity in the
reef environment until high very levels are reached (e.g., Devantier et al., 1998;
Morton, 1994). This is an important finding to consider when environmental impact
studies or reef monitoring data are interpreted.
Water quality is a key parameter in the ecology of reef benthos and may account
for differences in distribution and abundance of filter feeders such as soft corals.
Annual mean concentrations of particulate nutrients and chlorophyll increase toward
the shore (Furnas & Mitchell, 1986; Liston et al., 1992; Revelante & Gilmartin, 1982)
and toward the more temperate southern parts of the GBR (Furnas, in preparation).

Many octocorals are relatively inefficient in photosynthesis and depend on high lev-
els of irradiance and additional food intake to cover their carbon demand (Fabricius
& Klumpp, 1995). Turbidity negatively affects light availability but may represent a
gain of suspended particulate food for organisms which are able to use it (Anthony
& Fabricius, in press). The relationship with sedimentation is more complex: reefs
completely free of sediment are generally also particularly wave-exposed or have
steep slopes so sediment accumulation is reduced (Fabricius & De’ath, in press),
which could contribute to the lower richness found on low-sediment reefs than neigh-
bouring reefs with the same visibility but more sediment.
The question whether increased runoff affects turbidity on the GBR is still con-
troversial. Larcombe and Woolfe (1999) suggest that turbidity and rates of sedimen-
tation do not increase with runoff, because rates are driven by the physical
environment (wave-related resuspension) and are limited by the surface area of depo-
sition. On the other hand, water clarity in a flood plume is severely reduced, although
the suspended material adds relatively little to the overall sediment weight (a “visu-
ally spectacular” plume often contains only a few mg l
Ϫ1
suspended solids at greater
distance from the river mouth; discussed in Larcombe and Woolfe, 1999, based on
data from Taylor, 1996). While the coarse fraction settles out close to the river mouth,
the muddy, light, and nutrient-enriched sediment fraction may remain in the system
for months after discharge, where it will go through many cycles of deposition and
resuspension before being metabolised or trapped in a north-facing embayment.
Enhanced phytoplankton production due to the release of nutrients contributes fur-
ther to increase turbidity.
Wolanski and Spagnol (in press) reported of the declining visibility on Low Isles,
a coastal reef off Cairns (~16° 23Ј S, 145° 34Ј E). This island was investigated in
detail in 1927/1928, and a mean visibility of ~11 m was recorded over a 6-month
observation time. Today, maximum visibility rarely exceeds 8 m, and the mean is esti-
mated to be around 6 m (Wolanski & Spagnol, in press; Bell & Elmetri, 1995; and

136 Oceanographic Processes of Coral Reefs
© 2001 by CRC Press LLC
our own observation over 10 days in three visits). Such change in visibility equals a
loss of ~5 soft coral genera (Figure 4). We produced a simple and tentatively soft
coral biodiversity response model to visualise the long-term effects of this change in
water clarity on the generic diversity of reefs around Low Isles (Animation 1). The
model was based on the following assumptions: pollution originated at the wet trop-
ics coast at 0.6 along-shore distance, events were discrete pulse discharges of sus-
pended particles, which were diluted with distance from the source while spreading
radially. The response was modelled based on the non-linear relationship depicted in
Figure 4. Wave- and depth-dependent settlement/resuspension cycles were ignored
for simplicity. We started at the status of present-day visibility using our recorded vis-
ibility and richness data, and created a scenario in which coastal visibility dropped
progressively to Ͻ3 m. Reduction in richness was noticeable well into the mid-shelf
region. The present-day centre of soft coral diversity, located on the mid-shelf north
off Cairns, diminished progressively, and disappeared except on the far northern edge
of the GBR at increasing levels of turbidity. Although such decrease in visibility is
hypothetical, the model nevertheless points at the importance of protecting the water
quality in the wet tropics for a long-term preservation of biodiversity on the GBR.
The world presently faces a global biodiversity crisis, with highest levels of
species extinctions recorded at least since the Cretaceous period. An estimated 100
species of animals and plants are being eradicated every day in terrestrial systems.
Next to nothing is known about species extinctions in marine realms, and the under-
standing of patterns in biodiversity of coral reefs is rudimentary at best. Coral reefs
are under increasing pressure worldwide, with a large proportion of coral reefs being
already severely degraded, or at risk of degradation (Wilkinson, 1999). Three types
of human activities are principal causes for reef degradation: Firstly, extensive land
clearing, sewage discharge, and agricultural runoff affect coastal reefs by means of
increased sediment and nutrient loads. Secondly, fishing is so intense and destructive
in more densely populated regions that recruitment overfishing and downstream

effects on abundances of macroalgae and corals have been recorded (Hughes, 1994;
McClanahan et al., 1996). Thirdly, the frequency of bleaching and often death in all
zooxanthellate organisms, including hard and soft corals, is currently increasing due
to increasing maximum summer sea surface temperatures as a result of greenhouse
gas emissions (Hoegh-Guldberg, 1999). Many taxa have pelagic larvae, thus reefs of
the GBR which are numerous and connected by ocean currents may be replenished
by larvae from undisturbed areas farther upstream. More isolated reefs are not as
likely to experience recolonisation by pelagic larvae, and local extinctions in such
oceanic atolls are likely (Wilkinson, 1999). The establishment of protected areas,
which act as sources of larvae for exploited or disturbed areas, is the most promising
approach for the local protection of coral reef biodiversity. At the same time, the
health of coastal reefs is intricately linked with land management, and protected areas
can only fulfil their role if deterioration of water quality is avoided by appropriate
coastal zone and catchment management.
We do not know whether any keystone taxa are represented among the soft
corals which are missing in areas of high turbidity (these are, in particular, members
of the family Xeniidae). We also do not know how key functional processes (e.g., the
Biodiversity on the Great Barrier Reef 137
© 2001 by CRC Press LLC
chemical micro-environment on the reef, as soft corals constantly release anti-
fouling substances [Maida et al., 1995], or competition with other benthos groups)
are affected by the presence or absence of certain soft coral taxa. The study may serve
as an example of the complexity of responses and relationships in coral reefs. In the
presence of such sparse knowledge the precautionary principle in managing the adja-
cent land and preventing influx of nutrients and soils should prevail.
ACKNOWLEDGMENTS
The study was funded by the Commonwealth of Australia Cooperative Research
Centres Program through the Cooperative Research Centre for the Great Barrier Reef
World Heritage Area, and supported by the Australian Institute of Marine Science. We
greatly appreciate helpful comments and suggestions on the manuscript by Jon Brodie.

REFERENCES
Anthony, K. & Fabricius, K.E. Shifting roles of heterotrophy and autotrophy in coral energy
budgets at varying turbidity. Journal of Experimental Marine Biology and Ecology, in
press.
Bak, R.P.M., Termaat, R.M., & Dekkar, R. 1982 Complexity of coral interactions: influence of
time, location of interactions and epifauna. Marine Biology 69, 215–222.
Bell, P.R.F. & Elmetri, I. 1995 Ecological indicators of large-scale eutrophication in the Great
Barrier Reef Lagoon. Ambio 24, 208–215.
Benayahu, Y. & Loya, Y. 1981 Competition for space among coral-reef sessile organisms at
Eilat, Red Sea. Bulletin of Marine Science 31, 514–522.
Braun-Blanquet, J.J. 1964 Pflanzensoziologie, Grundzüge der Vegetationskunde, 3rd ed.
Springer Press, New York.
Coll, J.C., Tapiolas, L.M., Bowden, B.F., Webb, L., & Marsh, H. 1983 Transformation of soft
coral (Coelenterata, Octocorallia) terpenes by Ovula ovum (Mollusca, Prosobranchia).
Marine Biology 74, 35–40.
Connell, J.H. 1976 Competitive interactions and the species diversity of corals. In Mackie,
G.O. (ed) Coelenterate Ecology and Behavior. Plenum Press, New York.
Dai, C.F. 1990 Interspecific competition between Taiwanese corals with special reference to
interactions between alcyonaceans and scleractinians. Marine Ecology Progress Series
60: 291–297.
De’ath, G. & Moran, P. 1998 Factors affecting behaviours of crown-of-thorns starfish
(Acanthaster planci). II. Feeding preferences. Journal of Experimental Marine Biology
and Ecology 220, 107–126.
Devantier, L., Suharsono, M., Budiyanto, A., Tuti, J., Imanto, P., & Ledesma, R. 1998 Status
of coral communities of Pulau Seribu, 1985–1995. Proceedings: Coral Reef Evaluation
Workshop Pulau Seribu, Jakarta, Indonesia. UNESCO, Jakarta Office, 1–24.
Devantier, L.M., De’ath, G., Done, T.J., & Turak, E. 1998 Ecological assessment of a complex nat-
ural system: a case study from the Great Barrier Reef. Ecological Applications 8, 480–496.
Dinesen, Z.D. 1983 Patterns in the distribution of soft corals across the central Great Barrier
Reef. Coral Reefs 1, 229–236.

Done, T.J. 1982 Patterns in the distribution of coral communities across the central Great
Barrier Reef. Coral Reefs 1, 95–107.
138 Oceanographic Processes of Coral Reefs
© 2001 by CRC Press LLC
Done, T.J. 1992 Phase shifts in coral communities and their ecological significance.
Hydrobiologia 247, 121–132.
Done, T.J. 1997 Decadal changes in reef-building communities: implications for reef growth
and monitoring programs. pp. 411–416 in Lessios, H.A. (ed) Proceedings of the 8th
International Coral Reef Symposium 1. Smithsonian Tropical Research Institute, Balboa,
Panama.
Done, T.J. 1999 Coral community adaptability to environmental change at the scales of
regions, reefs and reef zones. American Zoologist 39, 66–79.
Done, T.J., Odgen, J.C., Wiebe, W.J., & Rosen, R.B. 1996 Diversity and ecosystem function of
coral reefs. pp. 393–423 in Mooney, H.A., Cushman, J.H., Medina, E., Sala, O.E.,
& Schulze, E.D., (eds) Functional Roles of Biodiversity: Global Perspectives. John Wiley
& Sons, London.
Fabricius, K.E. 1997 Soft coral abundance in the central Great Barrier Reef: effects of
Acanthaster planci and the physical environment. Coral Reefs 16, 159–167.
Fabricius, K.E. 1998 Reef invasion by soft corals: Which taxa and which habitats? pp. 77–90
in Greenwood, J.G. & Hall, N.J. (eds) Proceedings of the Australian Coral Reef Society
75th Anniversary Conference, Heron Island, October 1997. School of Marine Science,
University of Queensland, Brisbane.
Fabricius, K.E., Benayahu, Y., & Genin, A. 1995a Herbivory in asymbiotic soft corals. Science
268, 90–92.
Fabricius, K.E. & De’ath, G. 1997 The effects of flow, depth and slope on cover of soft coral
taxa and growth forms on Davies Reef, Great Barrier Reef. pp. 1071–1076 in Lessios,
H.A. (ed) Proceedings of the 8th International Coral Reef Symposium 2. Smithsonian
Tropical Research Institute, Balboa, Republic of Panama.
Fabricius, K.E. & De’ath, G. 2000 Soft Coral Atlas of the Great Barrier Reef. Australian
Institute of Marine Science, Townsville, 57 pp.

Fabricius, K.E. & De’ath, G. Environmental factors associated with the spatial distribution of
crustose coralline algae on the Great Barrier Reef. Coral Reefs, in press.
Fabricius, K.E. & Dommisse, M. 2000 Depletion of suspended particulate matter over coastal
reef communities dominated by zooxanthellate soft corals. Marine Ecology Progress
Series 196, 157–167.
Fabricius, K.E., Genin, A., & Benayahu, Y. 1995b Flow-dependent herbivory and growth in
asymbiotic soft corals. Limnology and Oceanography 40, 1290–1301.
Fabricius, K.E. & Klumpp, D.W. 1995 Wide-spread mixotrophy in reef-inhabiting soft corals:
the influence of depth, and colony expansion and contraction on photosynthesis. Marine
Ecology Progress Series 125, 195–204.
Furnas, M. & Mitchell, A.W. 1986 Phytoplankton dynamics in the central Great Barrier Reef.
I. Seasonal changes in biomass and community structure and their relation to intrusive
activity. Continental Shelf Research 6, 363–384.
Gabic, A.J. & Bell, P.F. 1993 Review of the effects of non-point nutrient loading on coastal
ecosystems. Australian Journal of Marine and Freshwater Research 44, 261–283.
Ginsburg, R.N. 1994 Global Aspects of Coral Reefs: Health, Hazards and History.
Proceedings of a Colloquium. Rosenstiel School of Marine and Atmospheric Science,
University of Miami, June 10–11th, 1993, 420 pp.
Hastie, T.J. & Tibshirani, R.J. 1990 Generalized Additive Models. Chapman & Hall, London.
Hoegh-Guldberg, O. 1999 Climate change, coral bleaching, and the future of the world’s coral
reefs. Marine and Freshwater Research 50, 839–866.
Hughes, T.P. 1994 Catastrophes, phase shifts, and large-scale degradation of a Caribbean coral
reef. Science 265, 1547–1551.
Biodiversity on the Great Barrier Reef 139
© 2001 by CRC Press LLC
Kenchington, R.A. 1978 Visual surveys of large areas of coral reefs. pp. 149–161 in Stoddart,
D.R. & Johannes, R.F. (eds) Coral Reefs: Research Methods. UNESCO, Paris.
Larcombe, P. & Woolfe, K.J. 1999 Increased sediment supply to the Great Barrier Reef will
not increase sediment accumulation at most coral reefs. Coral Reefs 18: 163–169.
Liston, P., Furnas, M.J., Mitchell, A.W., & Drew, E.A. 1992 Local and mesoscale variability of

surface water temperature and chlorophyll in the northern Great Barrier Reef, Australia.
Continental Shelf Research 12, 907–921.
Maida, M., Sammarco, P.W., & Coll, J.C. 1995 Effects of soft corals on scleractinian coral
recruitment. I. Directional allelopathy and inhibition of settlement. Marine Ecology
Progress Series 121, 191–202.
McClanahan, T.R., Kamukuru, A.T., Muthiga, N.A., Yebio, M.J., & Obura, D. 1996 Effects of sea
urchins reduction on algae, coral and fish populations. Conservation Biology 10, 136–154.
Miller, I.R. & De’ath, G. 1996 Effects of training on observer performance in assessing ben-
thic cover by means of the manta tow technique. Marine Freshwater Research 47, 19–26.
Morton, B. 1994 Hong Kong’s coral communities: status, threats and management plans.
Marine Pollution Bulletin 29, 74–83.
Pastorok, R.A. and Bilyard, G.R. 1985 Effects of sewage pollution in coral-reef communities.
Marine Ecology Progress Series 21, 175–189.
Platnick, N.I. 1992 Patterns of biodiversity. pp. 15–24 in Elredge, N. (ed), Systematics,
Ecology, and the Biodiversity Crisis. Columbia University Press, New York.
Ray, G.C. & Grassle, J.F. 1991 Marine biological diversity. Bioscience 41, 453–469.
Revelante, N. & Gilmartin, M. 1982 Dynamics of phytoplankton in the Great Barrier Reef
Lagoon. Journal of Plankton Research 4, 47–76.
Ribes, M., Coma, R., & Gili, J.M. 1998 Heterotrophic feeding by gorgonian corals with sym-
biotic Zooxanthela. Limnology and Oceanography 43, 1170–1179.
Rogers, C.S. 1990. Responses of coral reefs and reef organisms to sedimentation. Marine
Ecology Progress Series 62, 185–202.
Sammarco, P.W., Coll, J.C., & LaBarre, S. 1985 Competitive strategies of soft corals
(Coelenterata, Octocorallia). II. Variable defensive responses and susceptibility to scler-
actinian corals. Journal of Experimental Marine Biology and Ecology 91, 199–215.
Statistical Sciences 1995 S-PLUS, Version 3.3 for Windows. A division of Mathsoft Inc.,
Seattle.
Taylor, J. 1996 Sediment input to the Great Barrier Reef lagoon via river discharge: the Barron
River. pp. 152–154 in Larcombe, P., Woolfe, K., & Purdon, R.G. (eds) Great Barrier
Reef: Terrigenous Sediment Flux and Human Impacts. Proceedings of a Research

Symposium, CRC Reef Research Centre, Townsville,
Tursch, B. & Tursch, A. 1982 The soft coral community on a sheltered reef quadrat at Laing
Island (Papua New Guinea). Marine Biology 68, 321–332.
Veron, J.E.N. 1995 Corals in Space and Time. University of New South Wales Press, Sydney,
321 pp.
Wachenfeld, D.R., Oliver, J.K., & Morrissey, J.I. 1998 State of the Great Barrier Reef World
Heritage Area. Great Barrier Reef Marine Park Authority, Townsville, 140 pp.
Wilkinson, C.R. 1999 Global and local threats to coral reef functioning and existence: review
and predictions. Marine and Freshwater Research 50, 867–878.
Williams, D. McB. 1982 Patterns in the distribution of fish communities across the central
Great Barrier Reef. Coral Reefs 1, 35–43.
Wolanski, E. 1994 Physical Oceanographic Processes of the Great Barrier Reef. CRC Press,
Boca Raton, FL, 194 pp.
Wolanski, E. & Spagnol, S. Pollution by mud of Great Barrier Reef coastal waters. Journal of
Coastal Research, in press.
140 Oceanographic Processes of Coral Reefs
© 2001 by CRC Press LLC
Biodiversity on the Great Barrier Reef 141
FIGURE 2 A spatial plot of soft coral richness, using
the traditional geodesic coordinate system
(latitude–longitude), and for easier viewing, in the
coordinate system based on relative distance of a reef
across and along the GBR shelf (right). A local
regression spatial smoother was used to model
richness, and the fitted surface was then mapped back
to latitude–longitude coordinates.
FIGURE 3 Left: Spatial plot of soft coral richness
(number of genera encountered per reef). Local
regression spatial smoothers were used for the spatial
plots. Middle and right: Partial effects of visibility and

sedimentation on soft coral richness. The red line is
the partial effect (i.e., the effect of the explanatory
variable holding all other explanatory variables
constant), estimated by a local regression smoother
(loess, span of 0.5) (left panel), or by a linear model
(right panel). The blue dashed lines represent 95%
confidence intervals, and the orange dashed line
indicates the no-effects level. The points represent the
residuals.
FIGURE 4 Proportion of variation in total soft coral
richness explained by spatial (left arrows), physical
(right arrows), and a combination of spatial and
physical variables (central arrow).
FIGURE 1 Map of the GBR indicating the locations of
the sampled reefs. Colour codes define the position of
the sampling points on the continental shelf: inner-shelf
reefs are located on the innermost 38% of the shelf
width, mid-shelf reefs are at 38 to 85%, and outer-shelf
reefs are Ͼ85% across the shelf. Southern reefs are all
reefs Ͻ45% along the shelf, with the northern reefs
representing the remaining 55%.
© 2001 by CRC Press LLC
142 Oceanographic Processes of Coral Reefs
FIGURE 5 Left: Spatial plot of richness of
zooxanthellate soft coral taxa. Local regression spatial
smoothers were used for the spatial plots. Right:
Partial effects of visibility on soft coral richness. For
detailed legend see Figure 3.
FIGURE 6 Left: Spatial plot of richness of
zooxanthellae-free soft coral taxa. Local regression

spatial smoothers were used for the spatial plots.
Right: Partial effects of visibility on soft coral
richness. For detailed legend see Figure 3.
FIGURE 7 Proportion of variation in generic richness
of zooxanthellate (left) and zooxanthellate-free (right)
soft corals explained by the spatial variables and
visibility.
FIGURE 8 Spatial plot of turbidity (measured as
Secchi visibility) and of sediment deposits on the
reefs.
FIGURE 9 Proportion of variation in visibility and
sedimentation explained. Variation in visibility was
related to spatial variables (left arrows) and
sedimentation (right arrows). Variation in sediment
was explained by only spatial variables.
© 2001 by CRC Press LLC
Biodiversity on the Great Barrier Reef 143
FIGURE 10 Spatial plot of soft coral cover, hard coral
cover, and the proportion of soft corals of the total
coral cover (soft coral plus hard coral cover).
FIGURE 11 Soft coral cover, hard coral cover, and
the proportion of soft corals to total cover explained
by spatial variables. Physical variables had no effect
on cover.
FIGURE 12 Relationship between site-specific soft
coral richness, and soft coral cover (right) or hard
coral cover (left). The solid line represents smooth fit
(df ϭ 4, R
2
ϭ 28.9). Dashed lines are 95% confidence

intervals.
FIGURE 13 Proportion of variation in site-specific
soft coral richness explained by spatial (left arrows),
depth (right arrows), and a combination of spatial and
physical variables (central arrow).
FIGURE 14 Mean levels of site-specific soft coral
richness (number of genera per site), turbidity
(visibility, in metres), and sediment (rated on a 4-point
scale) as a function of depth and shelf position. Values
are means, error bars represent 1 standard error.
Orange line, filled squares ϭ inner-shelf; green line,
filled triangles ϭ mid-shelf; and blue line, open
circles ϭ outer-shelf reefs.
© 2001 by CRC Press LLC
144 Oceanographic Processes of Coral Reefs
FIGURE 15 Mean levels of soft coral cover, hard
coral cover, and the ratio between soft coral cover and
total coral cover (hard corals plus soft corals) as a
function of depth and shelf position. Values are means,
error bars represent 1 standard error. Orange line,
filled squares ϭ inner-shelf; green line, filled
triangles ϭ mid-shelf; and blue line, open
circles ϭ outer-shelf reefs.
ANIMATION 1 Model of response in soft coral
richness (number of genera per reef; right panel) to
progressively decreasing water clarity (left panel). The
green dot indicates the location of Low Isles.
© 2001 by CRC Press LLC