ATOLL RESEARCH BULLETIN

NO. 556

COMPLEX ENVIRONMENTAL PATTERNS AND
HOLOCENE SEA-LEVEL CHANGES CONTROLLING REEF HISTORIES

ALONG NORTHEASTERN ST. CROIX, USVI

BY

IAN G. MACINTYRE, MARGUERITE A. TOSCANO,
AND JOYCE LUNDBERG

ISSUED BY

NATIONAL MUSEUM OF NATURAL HISTORY
SMITHSONIAN INSTITUTION
WASHINGTON, D.C., U.S.A.
NOVEMBER 2008

Figure 1 . Index map of northeastern St. Croix showing Long Reef and the Outer Shelf-Edge Reef Complex (gray
line), Buck Island, and the Inner Bank-Barrier Reef including Tague Reef, Sand Cay/Candlelight Reef (SC/CR), and
Boiler Bay.

1 7°45.5' W

COMPLEX ENVIRONMENTAL PATTERNS AND
HOLOCENE SEA-LEVEL CHANGES CONTROLLING REEF HISTORIES

ALONG NORTHEASTERN ST. CROIX, USVI

BY

IAN G. MACINTYRE, 1 MARGUERITE A. TOSCANO, 1
AND JOYCE LUNDBERG 2

ABSTRACT

The progressive timing of reef initiation along northeastern St Croix, and varying
histories of reef growth, provide insights into the effects of Holocene paleoenvironmental
and sea-level changes in the region, and into the importance of accurate and
comprehensive sea-level reconstruction as context for understanding local variations
in reef history. Two core transects provide new information on the Holocene history of
related reef systems off the northeastern coast of St. Croix ™ the outer shelf-edge reef
complex ranging from northern Lang Bank through Buck Island Bar and ending at Long
Reef, and the inner bank-barrier reef off Tague Bay. The outer shelf-edge reef system
records progressively shallower and thinner marginal reef initiation and development
above a westward-shallowing Pleistocene substrate; this surface rises from -15 m at Buck
Island Bar to -6 m MSL at Long Reef. The core transect through the inner bank-barrier
reef at Tague Bay reveals a 10-m thick accumulation of mAcropora palmata- dominated
reef framework on a well developed 12-15 m deep erosional terrace.

A sea-level analysis for northeastern St. Croix indicates three periods in which
different areas were the foci of reef initiation and development. The timing of initiation
of reef systems along these reef trends as sea level rose during the Holocene follows
the slope of the Pleistocene surface as it rises to the west, with the oldest (easternmost)
reefs forming on the deepest areas of the antecedent topography. The westernmost reefs
were initiated later as sea level reached the slightly higher elevations of the antecedent
surfaces in these areas. The timing of Holocene reefs at these two sites is a direct result
of the interaction of sea-level rise, antecedent topography, shelf erosion and evolving
environmental conditions affecting coral health. The unique reef history at Buck Island
Bar represents a particular case whose relationship to the Caribbean Holocene sea-level
record supports paleoenvironmental interpretations potentially impacting all of Lang
Bank to the east.

department of Paleobiology, National Museum of Natural History, Smithsonian Institution, Washington DC
20005

department of Geography and Environmental Studies, Carleton University, Ottawa, K1S 5B6, Ontario,
Canada

Manuscript received 12 June 2008; revised 11 September 2008.

2

INTRODUCTION

Reconstructing the history of reef growth is a multidisciplinary process requiring
detailed understanding of local oceanographic and climatic conditions, and the
relationship of regional sea-level history, local topography, erosion and other controls,
to reef initiation and maintenance. Given this complexity, the geologic history of reef
development may be temporally and spatially limited at any single reef site, and not
always simply predictable from the regional sea-level curve. We report an interesting
Caribbean case study from St. Croix, where the development of Holocene reefs in
response to rising post-glacial sea level was dependent on variations in geologic and
oceanographic factors along the northeastern coast. Reef structure and age were studied
from the stratigraphy revealed by two core transects (14 cores in total, 5 not previously
published), and from 14 C and U-Th dating of fossil coral samples (63 dates total, 10 of
them new). With the availability of new core and age data and an additional reef transect
site, a comprehensive regional synthesis of reef history, including multiple reef trends, is
herein accomplished. The Holocene sea-level reconstruction for the Caribbean (Toscano
and Macintyre, 2003) provides a well-constrained context for environmental changes
and the relationship of sea level and antecedent topography to reef development over
the Holocene in St. Croix. The results of the present study, which highlight significant
local variation due to to sea-level change, antecedent topography and the response
of individual species to evolving local conditions, have general implications for the
reconstruction of reef history and sea level using core data.

The insular shelf off the northeast coast of St. Croix exhibits a variety of
Holocene sediment and reef accumulation patterns. Two separate linear reef trends have
been previously mapped and cored (Adey et al., 1977; Burke et al., 1989; Macintyre and
Adey, 1990; Hubbard, 1991; Hubbard et al., 2005), describing reef development around
the eastern end of St Croix. The outer shelf-edge reef complex trends westward from
northeastern Tang Bank on its eastern end, through Buck Island Bar, and ending at Tong
Reef (offshore of Christiansted). Tang Bank, a major feature of the shelf edge all around
eastern St Croix (Adey et al., 1977) forms the northeastern end of the outer shelf-edge
reef system (Fig. 1). Tang Bank started reef building along its southwestern arm during
the early Holocene (-10,000 yrs BP; Adey et al., 1977), and began building a thick
accumulation of Acropora palmata framework on its northeastern end at approximately
7,700 cal BP (Hubbard et al., 2005). The depth to the Pleistocene surface is not known
but is estimated to be below -15 m mean sea level(MST). To the west, Buck Island Bar
(Fig. 1) has >10 m of accumulation of mainly A. palmata and sand from one core along
the shelf edge (Hubbard, 1991; Hubbard et al., 2005). The reef forming Buck Island Bar
was established on an elevated Pleistocene ridge at —15 m MST prior to 7,700 years ago,
but ceased active framework construction approximately at -1,200 cal BP, as indicated by
a near-surface dated A. palmata of 1,180 cal BP at a depth of about 5 m below present sea
level (Hubbard et al., 2005).

Macintyre and Adey (1990) drilled a core on Buck Island Bar just shoreward of
the core of Hubbard et al. (2005) and recovered mostly massive corals and well-lithihed
pavement limestone, a facies pattern signifying slow accumulation under exposed,

3

high-energy conditions. In addition, the surface of this reef has remained below -4.5 m,
never reaching sea level in elevation, during the last 4000 years. Adey and Macintyre
(1990) concluded that any A. palmata accumulations at this site were likely to have been
frequently damaged and transported shoreward by heavy seas or severe storms, thus not
allowing this reef to keep up with sea level.

Hubbard et al. (2005) also indicated that while A. palmata was becoming well
established as the primary reef framework builder on the outer shelf edge (Buck Island
Bar and Lang Bank), by 7,500 cal BP a predominance of massive head corals was
just becoming established across inner shelf areas. They also reported reductions in A.
palmata on the outer shelf edge by 7000 cal BP, which became absent from Buck Island
as well as possibly less plentiful throughout the Caribbean by 6000 cal BP. By 5,000 cal
BP A. palmata was again making a major contribution to the framework of outer shelf-
edge reefs as well as the bank-barrier reefs of Buck Island, which eventually caught up
with sea level. Tague Reef on the inner shelf was still head-coral dominated (Hubbard et
al., 2005).

The up to 10-m thick sections of the inner- shelf Tague Bay reef system are
related to the elevation of the Pleistocene substrate in the area. This antecedent surface
dips to the east, allowing early initiation of reef framework construction at Tague Bay
by 7030 cal BP, following the reefs at Buck Island Bar (7700 cal BP; Hubbard et al.,
2005). Candlelight Reef, an isolated westward extension of the Tague Bay inner reef
system, was established as a fringing reef on the Pleistocene terrace underlying Sand Cay,
a shore-parallel clastic bar. Burke et al. (1989) show a shore-parallel cross section of
Tague Reef, along which several cores intersect the Pleistocene surface at depths of —9
m (Sand Cay/Candlelight Reef- SC/CR; Fig. 1), —11 to -13 m (Tague Reef; Fig. 1) and
-14 m (Romney Point - RP; Fig. 3). At the eastern end of Tague Bay, Adey (1975) cored
the algal ridges in Boiler Bay (Fig. 1) to determine timing of initiation on top of ridges
of the Caledonia Formation as well as on a “bench of Caledonia boulders” (page 7) i.e.,
colluvium from the Caledonia Formation. This surface occurred along the W-E sloping
trend from -21 m, rising to benches at -5 m MSL (Adey 1975, Fig. 7). Algal ridges
formed after sea-level rise eroded the colluvium, allowing establishment of A. palmata
colonies on the lobes and benches of lag conglomerate, followed by algal ridges and
boilers.

Long Reef, the westernmost extension of the outer shelf-edge reef complex,
resting on the landward edge of the pre-Holocene surface, is the essential (and previously
missing) end-member of the geologic history of this reef system. We present for the first
time a 5-core transect drilled across Long Reef. We compare this northwestern limit of
the shelf-edge reef system with its eastern extension as far as Buck Island Bar. We also
provide new stratigraphic and age data from a core transect across the Tague Bay reef
complex to better document its Holocene history and response to local conditions and
regional sea-level rise (SLR) over the past 11,000 yrs. Eight of these cores across the reef
complex are presented in this study, with an earlier core (16) from Burke et al. (1989). A
diagrammatic sketch of this transect was included in Burke et al. (1989).

4

METHODS

A diver-operated hydraulic drill (Macintyre, 1975) was used to collect 54 mm
diameter cores along a transect across the Tague Bay reef system (Fig. 2) during two
held trips in May and July/August 1976. Ten cores (eight used in this study, Fig. 3) were
collected along a transect extending from the West Indies Laboratory pier, across the back-
reef lagoon, and the Tague Bay bank-barrier reef to the bottom of the outer slope at a depth
of 14.9 m (Burke et al., 1989). Core 16 (Burke et al., 1989) is added to this cross section.

In November 1978, five cores were drilled across the northwestern limit of the
shelf-edge reef system at Long Reef (Fig. 4; 5). These cores extended from the back-reef
lagoon (Core 5, Fig. 5) across the reef crest (Cores 1 and 2, Fig. 5), to two cores drilled
adjacent to each other on a slight spur (Core 4, Fig. 5) and groove (Core 3, Fig. 5).

We herein document all previously published radiocarbon dates (Burke et al.,

1989; Macintyre and Adey, 1990; Hubbard, 1991; Hubbard et al., 2005), and add ten
new dates, including four Thermal Ionization Mass Spectrometric U-Th dated samples,
to the St. Croix database (Table 1). Radiocarbon dates from the Tague Bay transect were
analyzed in the mid-late 1970s by Robert Stuckenrath of the Smithsonian Radiocarbon
Dating Laboratory. The dates for the Long Reef transect were analyzed in the 1970s
by Beta Analytic Inc., Coral Gables, Florida. These original radiocarbon dates were
not corrected for 5 13 C pDB and did not incorporate an oceanic reservoir correction, hence
some were published as basic 14 C dates based on the Libby half life (5568 years).
Conventional radiocarbon ages, which are now routinely reported, have been corrected
for isotope fractionation by normalizing 5 13 C to 0%o pDB for corals. For these samples we
have calculated 5 13 C pDB values using the Calib spreadsheet dl3ccorr.xls (linked from the
online Calib manual; http://calib.qub.ac.uk/calib), assuming the original measurement
was a 14 C/ 12 C ratio. The 5 13 C correction spreadsheet also calculates the corresponding
conventional radiocarbon date for each sample. Next, all coral radiocarbon dates were
calibrated using the Calib 5.1.0 Beta program (Stuiver and Reimer, 1993; http://calib.qub.
ac.uk/calib; on-line version Stuiver et al., 2005), the marine calibration dataset (marine
04) and a time-dependent global ocean reservoir correction (-400 years; CALIB Manual
version 4.1, Stuiver and Reimer, 1993). The difference in age between the local ocean
reservoir and modeled values (AR) was set at -5±20 (Darden Hood, Beta Analytic, pers.
comm.; Stuiver and Braziunas, 1993).

Thermal Ionization Mass Spectrometric (TIMS) U-Th dating was done in the
Isotope Geochemistry and Geochronology Research Facility, Carleton University,

Ottawa, Ontario, following standard techniques (e.g., Ivanovich et al., 1992). Samples
were ultrasonically cleaned, ignited for 5 hours at 875° C to remove organics, dissolved
in HN0 3 and spiked with 233 U- 236 U- 229 Th tracer. U and Th were co-precipitated with iron
hydroxide, and purified twice on anion exchange columns (Dowex AG1-X 200-400 mesh).
Measurement of U and Th isotopic ratios was done on the Triton TIMS, in peak jumping
mode using secondary electron multiplier with retarding quadrupole filter. Ages were
calculated using half lives from Cheng et al. (2000).

5

Figure 2. Aerial photograph looking east along Tague Bay bank-barrier reef. Arrow
indicates drill site.

Figure 3. Index map showing the general distribution of the bottom communities
and the location of the nine core sites (the numbered solid circles) across Tague Bay
bank- barrier reef.

Table 1. Compilation of all radiocarbon (59 samples) and TIMS (4 samples) age data for Buck Island Bar (BB or BIB, HSX =
haystacks), Tong Reef (TR), Buck Island (BI), Tague Bay, Tague Reef, Sand Cay Reef, and Sand Cay/Candlelight Reef (TB, TR,
SCR, SCC), northeastern St. Croix. The 14C data are from: 1- this study; 2 - Burke et al. (1985); 3 -Macintyre and Adey (1990); 4
- Hubbard et al. (2005).

6

cal BP or

TIMS U-ThO

7770

7670

4763

3970

2943

2005

1792

1180

4427

4052

3596

2007

m

6P£

7510

7175

6775

6180

5665

5270

5190

4570

4495

4305

4210

3990

3365

3320

2840

2725

Age

Error

o

oo

06

50

50

50

70

45

70

70

09

120

50

50

50

120

06

o

oo

o

oo

70

o

oo

70

100

06

120

o

oo

o

oo

09

70

70

70

Conventional
Age (yrs BP)

7360

7270

4552

3962

3152

2400

2192

1630

4297

4027

3657

2377

1107

LL9

7070

6680

6330

5760

5320

4930

4860

4440

4370

4220

4150

4010

3500

3440

3090

2950

Error

o

oo

06

50

50

50

70

<n

70

70

09

120

50

50

50

i

i

i

14 C date

6950

6860

4145

3555

2745

1990

1785

1220

3890

3620

3250

1970

700

270

i

i

Depth
(m MSL)

-12.75

-13.7

-12.39

-9.32

o

oo

<n

oo

-6.02

-5.05

oo

o

'O

-4.17

-3.08

-4.43

-2.02

-5.52

-12.5

-8.2

-5.9

-10.5

-14.25

-4.9

-7.15

-8.5

oo

-14.25

-5.45

-2.35

-3.90

-16.05

<r>

oo

r-C

Coral

Species

A. palmata

A. palmata

A. palmata

P. astreoides

M. annularis

A. palmata

Dipl or ia sp.

A. palmata

A. palmata

D. clivosa

A. palmata

M. annularis

D. strigosa

A. palmata

M. faveolata

M. faveolata

M. faveolata

M. faveolata

M. faveolata

A. palmata

A. palmata

A. palmata

M. annularis

M. faveolata

S. siderea

A. palmata

D. labyrinthiformis

A. palmata

M. faveolata

M. faveolata

Site Name
(reference)

Buck Island Bar 4

Buck Island Bar 4

Buck Island Bar 3

Buck Island Bar 3

Buck Island Bar 3

Buck Island Bar 4

Buck Island Bar 3

Buck Island Bar 4

oa

M-H

(D

Pi

tJ)

C,

o

H-)

0*

Dh

CD

CD

toD

C

o

Dh

CD

CD

6D

C

O

h-)

9*

Dh

CD

CD

td)

C

o

H-}

o*

Dh

CD

CD

bD

O

h-J

9*

Dh

CD

CD

toD

O

Buck Island 4

Buck Island 4

Buck Island 4

Buck Island 4

Buck Island 4

Buck Island 4

Buck Island 4

Buck Island 4

Buck Island 4

-i-

~a

c,

o3

00

M

(D

3

CQ

Buck Island 4

Buck Island 4

Buck Island 4

Buck Island 4

Buck Island 4

Buck Island 4

Sample ID

BB1-23

BB1-26

BIB4H1C5 1978

BIB3H1C3 1978

BIB2H1C2 1978

BB1-11

BIB2H1C1 1978

BB1-2

LRH4 C2

LRH2 C3

LRH2 C2

LRH1 C2

LRH2C1

LRH4C1

BI1-52

BI4-39

BI1-40

BI4-21

BI3-49

BI2-50

BI4-14

BI1-24

BI3-37

BI2-43

BI5-62

BI 1-13

BI4-02

BI1-02

BI7-29

BI2-27

Table 1, Con’td. 14 C Date represents the original, uncorrected basic radiocarbon date. Conventional Age is corrected for 5 13 C.
Cal BP is the calendar age calibrated from marine calibration data and corrected for the oceanic reservoir effect. Cal BP ages are
curve intercepts with probability ranges. Here we provide the curve intercept dates. Under Sample ID * denotes a replicate date.
U-Th dates are denoted by 0 and error range.

7

2305

2225

1840

1780

1740

1555

1345

1075

1045

970

in

©>

oo

o

o

in

oo

n

in

©>

n

o

in

Cl

091

7052

6329

5235

oo

-H

oo

oo

m

o

3779

3541

0 2295±4

2090

1870

0 1873±5

0 1661±3

1449

1432

1082

774

708

(N

in

o

oo

Ti

©>

o

oo

70

70

70

09

09

70

70

60

60

09

09

50

60

o

oo

94

o

oo

o

oo

in

c-

o

oo

in

i

09

o

oo

in

©>

50

90

o

oo

2620

2755

2250

2310

2175

2020

1830

1540

1510

1450

1310

o

in

c-

099

o

ci

v©

590

540

6542

5897

4932

!

3822

3627

1

2442

2257

1

1

1902

1862

1522

1210

1127

c-

Cl

in

i

i

!

i

i

i

i

!

i

|

!

i

|

!

|

!

i

i

o

oo

in

oo

o

oo

i

o

oo

in

c-

!

o

oo

in

©>

i

|

09

o

oo

•n

©>

o

•n

o

oo

o

oo

i

i

!

!

i

i

|

!

i

i

i

i

i

6135

5490

4525

|

3415

3220

i

2035

1850

i

|

1495

1455

1115

803

720

120

-12.25

Ti

OO

cy

-4.6

-4.6

-3.35

-8.35

-2.05

-4.65

-6.2

-2.9

-7.9

f-c

-7.9

-5.75

-13.1

-0.5

-10.4

-7.2

-8.62

-12.3

o

oo

-7.98

-7.75

-3.0

-2.7

oo

-y

-5.5

-3.0

-2.0

-7.9

-1.0

-1.2

-1.5

M. faveolata

M. faveolata

A. pal mat a

A. pal mat a

A. palmata

M. faveolata

A. palmata

A. palmata

A. palmata

A. palmata

A. palmata

M. faveolata

A. palmata

A. palmata

M. faveolata

D. clivosa

A. palmata

A. palmata

A. palmata

M annularis

Dipl or ia sp.

A. palmata

Dipl or ia sp.

M annularis

A. palmata

A. palmata

M. annularis

A. palmata

M. annularis

A. palmata

A. palmata

A. palmata

A. palmata

Buck Island 4

Buck Island 4

Buck Island 4

T3

ft

03

C/O

M

CJ

pq

Buck Island 4

ft

03

C/O

1— H

M

o

f3

PQ

Buck Island 4

Buck Island 4

Buck Island 4

Buck Island 4

-a

ft

03

C/O

M

M

o

f3

pq

"T

ft

o3

CO

M

M

o

f3

PQ

Buck Island 4

Buck Island 4

ft

o3

CO

M

o

3

CQ

— r
X3
C
o3

CO

M

o

PQ

Tague Bay Romney Pt 1

Tague Bay Romney Pt 1

PQ

<D

60

o3

H

Tague Reef 2

<N

PQ

<u

to

o3

H

>

PQ

<D

3

60

03

H

Q-1

M— 1
<D
<D

<u

3

60

03

H

Tague Bay 2

North Shore Reef 1

Tague Reef 2

Tague Reef 2

Sand Cay/Candlelight Reef 2

North Shore Reef 2

PQ

=3

6J)

o3

H

Tague Bay Romney Pt 2

Tague Bay Romney Pt 2

Tague Bay 2

BI7-20

BI2-27*

BI2-05

BI2-05*

BI2-01

BI5-40

BI2-16

BI5-22

BI5-24

BI5-12

HSX-1

BI7-7

HSX-2

BI7-2

BI6-6

BI3-05

SCR-3DC17

SCR-2C17

TB H8 C2b

H7 C2-4

TB Core 15

ZD 8H ai

TRH7 C2-1

TB Core 15

TB Core 13

TRH8C1-5

TRH6C1-6

TB Core 1 1

TB Core 13

ZD 6H ax

TB Core 17

SCC-3C19

TB Core 16

8

Figure 4. Aerial photograph looking east across Long Reef (shelf-edge reef).

Caribbean Sea

Shallow-water
mixed corals

AA Acropora palmata

Deeper-water
mixed corals

Sand

Pavement

Christiansted

St. Croix, USVI

600 m

Figure 5. Index map showing the general distributiion of the bottom communities and the location of the
core sites (the numbered solid circles) across Long Reef.

9

RESULTS

Drill Site Descriptions

Descriptions of the two drill site settings were made in the mid 1970s. Drilling
was completed prior to the major loss of Acropora palmata and Acropora cervicornis to
White Band Disease in this area in the late 1970s and early 1980s (Gladfelter, 1982) and the
regional mass mortality of Diadema antillarum in the early 1980s (Lessios et ah, 1984). Site
descriptions start at the western end of the older outer shelf-edge reef complex (Long Reef),
then the inner bank-barrier reef (Tague Reef).

Long Reef - Outer Shelf-Edge Reef The shallow back-reef zone (~1 m) consists of
a crustose coralline algal encrusted coral-rubble flat with patches of Thalassia- covered sand
and mounds of Porites porites. Other scattered corals on this rubble included A. palmata ,
Diploria clivosa , Diploria strigosa , and Montastraea annularis. This coral-rubble bottom
became more lithified (Fig. 6) on approaching the reef crest, with both A. palmata and
Porites astreoides becoming more dominant and with the addition of Millepora complanata.
At the time of our field work D. antillarum were very abundant and coral colonies and
fragments of Caledonia Formation stood out in relief from the surrounding urchin-eroded
substrate (Fig. 7).

Figure 6. Dense interlocking and lithified coral rubble reef flat close to reef crest at Core 2 site, Long Reef.

10

Figure 7. Bioerosion of reef-flat pavement by Diadema antillarum has left Caledonia Formation rubble
standing out in relief, Long Reef.

The A. palmata-domimtQd reef crest sloped gently seaward to form a low-relief
(~1 m) spur and groove pattern (Fig. 8). The spurs had coral cover consisting of A. palmata,
M. annularis , D. strigosa , D. clivosa , Millepora spp., M. cavernosa , Siderastrea siderea ,
Siderastrea radians , with some Agaricia agaricites , P. porites , Dichocoenia stokesi ,

Mussa angulosa , Meandrina meandrites , and Z). labyrinthiformis. The zoanthid Palythoa
caribaeorum was also very abundant. The adjacent grooves had a much denser and smooth
substrate (Fig. 9) with scattered A. palmata and small colonies of D. strigosa , Dichocoenia
stokesi , M. cavernosa , and Stephanocoenia michelinii.

This spur and groove gave way seaward to a smooth pavement with increasing
coverage of octocorals at about 8 m (Fig. 10). This community changed to thickets of A
cervicornis at 12 to 14 m. The slope then became covered with a few large colonies of M.
annularis and scattered colonies of S. siderea , M. cavernosa , A. agaricites , F! porites , Z).
strigosa , Z). clivosa , Z). labyrinthiformis , Z). stokesi , Colpophyllia natans , and Dendrogyra
cylindrus. The overall impression of the bottom community in this area was that there was
very little Holocene framework accumulating. There was extensive bioerosion, with coral
colonies mostly small and scattered over a hardground (Fig. 10). There was no indication of
the buttresses and pinnacles with downward slumping massive coral blocks that had been
reported on the outer reef slopes west of our site at Salt River (Hubbard et al., 1986) and Cane
Bay (Hubbard et al., 1990).

11

Figure 8. Shallow fore-reef spur and groove
zone, Long Reef. Spur covered dominantly
by Montastraea annularis and Millepora
complanata in contrast to the almost bare
groove. Depth 5 m.

Figure 9. Close-up photo of Core 3 in a groove along the shallow fore reef of Long Reef. Note the
smooth surface with no significant encrusting organisms. Coring revealed that this is a Pleistocene
limestone surface. Depth 6 m.

12

Figure 10. Octocorals and scattered coral heads on a hardground with very little sediment cover on the
fore-reef slope of Long Reef. Depth 8 m.

Tague Bay - Inner Bank-Barrier Reef. Starting at a lagoon depth of about 6 m,
the bottom was dominated by the seagrass Thalassia testudinum along with the calcareous
green algae Halimeda spp. and Penicillus sp. (Burke et ah, 1989). The back-reef area
consisted of coral patches, including P. porites , M. annularis , D. strigosa , and C. natans
with scattered octocorals, and surrounded by sand (Fig. 11). This graded upward to a dense
A. palmata community at the reef crest (Fig. 12), which also included D. strigosa , M.
complanata and P. astreoides (Dahl et al., 1974). This reef crest is about 50 m wide and falls
off to a seaward slope covered with a variety of corals including D. strigosa , M. annularis ,

S. siderea , Agaricia sp., A. cervicornis , M. cavernosa , D. labyrinthiformis , C. natans ,
Meandrina sp., Mycetophyllia sp., P. furcata , and M. angulosa (Dahl et al., 1974). This
slope leveled off to a sandy bottom at a depth of 15 m.

Core Transects

Long Reef Transect (westward end of outer shelf-edge reef trend). Four cores from
this shelf-edge reef contain no more than approximately 4.5 meters of reef accumulation
above the Pleistocene erosional surface at —6 to -7 m MST (Fig. 13). The oldest
radiocarbon date of 4427 cal BP (at -6 m MST) on A. palmata in the outer spur indicates
that reef initiation in this area probably occurred less than 5,000 years ago. Core 2, drilled
immediately leeward of the reef crest, yielded a head coral date of 4052 cal BP at -4 m
MST, and a date of 3,596 cal BP on A. palmata at -3 m MST. Much of the material from
the two reef-flat cores (Cores 1 and 5, Fig. 13) consisted of leeward-transported coral/

13

Figure 11. Drilling Core 7 in the shallow back
reef of Tague Bay reef. Note the coral patches
with surrounding sand.

Figure 12. The reef crest of Tague Bay reef showing the dense coral framework that includes Acropora
palmata, Diploria strigosa, and Millepora complanata.

14

reef debris. The most striking feature of this cross section is the exposure of Pleistocene
substrate at -6 m MSL in the low-relief fore-reef grooves (Core 3, Fig. 13). The pavement
areas seaward of the spur and groove (Fig. 12) are probably exposed Pleistocene limestone.

Tagne Bay Reef Transect (inner-shelf reef). The bank-barrier reef at Tague Bay was
established on a 100 m-wide terrace at a depth of about -12 m (Fig. 14). Tague reef has
many of the internal facies characteristics of a typical Caribbean A. palmata coral reef
(Macintyre and Glynn, 1976). Its central section is dominated by A. palmata framework,
while the shallow wave-washed section is more lithihed by submarine cement (Macintyre,
1977). Massive coral colonies, particularly M. annularis and D. strigosa , are concentrated at
the base and back-reef zone, while lagoonal sediments limit reef framework construction to
the leeward. Burke et al. (1989) obtained only one radiocarbon date at the reef crest (152 cal
BP, Core 16); additional dates were reported leeward (Core 8) and seaward (Core 9) of the
crest. We report four new high-precision Thermal Ionization Mass-Spectrometric (TIMS)
U-Th dates in back-reef Cores 6 (1661 ± 3 yrs at - 5.5 m MST, M. annularis ), 7 (3889 ±

8 yrs at -12.3 m MST, M. annularis ; 2295 ± 4 yrs at -7.75 m MST, Diploria sp.), and 8
(1873 ± 5 yrs at -4.8 m MST, A. palmata). We have no radiocarbon dates at the base of this
reef (~ -12 m under the reef crest; -15 m under the fore-reef); the oldest available date on
A. palmata is 5235 cal BP at a depth of- 8.62 m MST. This date/depth is stratigraphically
consistent with a date of 7060 cal BP on A. palmata at -10.4 m MST in a core 40 meters to
the east (Core 17 at Sand Cay Reef; Burke et al., 1989). We therefore conclude that Tague
Reef framework was established at approximately 7,000 years ago (when sea level was
—7.5 m MST), building upward and expanding seaward, with the leeward Cores 7 and 6
documenting the back-reef facies. The youngest dates in these back-reef cores indicate the
rapid accumulation of back-reef talus after the reef crest approached sea level ~ 2000 yrs
ago.

DISCUSSION

The outer shelf-edge and inner shelf-core transects document two separate but roughly
coeval coral reef systems whose histories were controlled by antecedent topography and
STR along the northeastern coast of St. Croix. The Tong Reef transect represents the
western end of the outer shelf-edge reef complex which formed progressively from east to
west on a westward-shallowing pre-Holocene surface (Hubbard et al., 2005). The Tague
Bay transect of the inner shelf reef complex occurs inside the outer shelf-edge reef trend
but is correlative with the portion of the outer reef trend that lies directly to the north (Buck
Island Bar), both temporally and in elevation range. Buck Island Reef initiated in between
these two trends on a “broad antecedent bench” from -13 to -16 m MST around Buck
Island (Hubbard et al., 2005), but dated A. palmata from the upper sections indicate the reef
framework is generally contemporaneous with that of Tong Reef.

In his study of the algal ridges near Tague Reef at Boiler Bay (Fig. 1), Adey
(1975) emphasized the “importance of pre-existing shelf level in determining the form
and developmental stage of ridge-reef systems.” Along each reef trend, the timing of reef

15

Horizontal Distance (m)

Figure 13. Cross section of Long Reef along the core transect showing the core data and location of radiocarbon
dates.

0 100 200 300 400

Figure 14. Cross section of Tague Bay reef along the core transect, showing the core data and the location of
radiocarbon dates.

16

initiation, thickness of accretion and duration of accumulation were directly related to the
elevation of the antecedent topography, local substrate character, rate of SLR and marine
environmental factors affecting coral growth (e.g. initial turbidity followed by improving
water quality over time vs. deepening water and decreasing light levels) at each site as
sea level rose over the shelf Macintyre (1988) introduced the concept of Holocene coral
reefs keeping pace with SLR, indicating that the thickness of accumulation in any area is
controlled by variations in the pre-existing shelf topography, but is initially inhibited by
shelf flooding, soil erosion and subsequent turbidity. Mineralogic evidence in support of the
shelf soil erosion hypothesis has been presented by Adey et al. (1977) along the southern
extension of the shelf-edge reef system. As the Holocene transgression proceeded, turbidity
would have eventually subsided and reefs formed shoreward in areas shallow enough to
support A. palmata reef crest framework close to sea level. A sea-level based general history
of northeastern St Croix reefs can be constructed from these data. Lhe western Atlantic sea-
level (WASL) curve of Loscano and Macintyre (2003) is used as a reference standard in this
area (see also Hubbard et al., 2005).

Lime-depth plots of all calibrated radiocarbon dates from northeastern St Croix reefs
against the western Atlantic sea-level (WASL) curve (Loscano and Macintyre, 2003; Fig.
15A-D) give the overall history of the progressive east to west flooding of the westward-
shallowing pre-Holocene erosional surfaces, followed by reef accumulations at these sites.
Starting -8,000 years ago in this area, sea level rose and the antecedent surface was initially
flooded and eroded to remove sediment and soils. Reef accumulation began on the deeper
(-16 m) shelf-edge reefs of Lang Bank and Buck Island Bar, then at Buck Island above the
-15 m Pleistocene surface there. A. palmata on Buck Island Bar never caught up with sea
level over its cored history (Fig. 15A).

Lhe inner-shelf reef at Fague Bay initiated on a pre-transgression surface ranging
from -12.5 to -15.5 m MSL. Although it appears from Figure 15D that reef-crest A. palmata
re-established a facies on Fague Bay reef above -6 m MSL after a prolonged hiatus, we
recognize that the reef crest facies critical to such an interpretation was only minimally
cored and not dated, leaving significant periods of the geologic history of Fague Reef
undocumented. At -5000 years ago, as sea level continued to rise, the higher pre-Holocene
surfaces westward to Long Reef were flooded, and a later episode of shallow water reef
building occurred at Long Reef that appears to have kept pace with SLR to the present.

Outer Shelf-Edge Reef Complex

Lhe progression of reef initiation along the outer shelf-edge reef trend follows
the depth (decreasing to the west from Lang Bank towards Long Reef) of the Pleistocene
surface (Hubbard et al., 2005), in context of Caribbean SLR in the region (Toscano and
Macintyre 2003; Fig. 15A-D) intersecting the exposed shelf areas beginning about 8,000
years ago. Adey et al. (1977) studied the relict outer shelf-edge reef complex around the
southeastern end of St Croix and noted that these reefs, although dominated by A. palmata ,
did not keep pace with early Holocene rates of SLR of 8 mm/yr (based on the sea-level
curve of Neumann, 1971 and later revisions). They interpreted the demise of the shelf-edge
reef system as coinciding with the flooding of the insular shelf and resulting turbidity from

17

« 8

o

CO

5

.CO 3

55 w

5:

o

+

O

• +

o

o

(0

so

<0

-c

to

V.

(0

c

c

&

CQ

CD

3

9»

,<o

I 1 I 1 I 1 I 1 I 1 I 1 I 1 I 1 I 1 I 1

+ •

• +

+ +

~co

0)

. 8-
<0

• Q

+ + +

CO

•

0

CD

LL

5

8 ,

t3

CD

1

0)

§ 8

.CO 3

55 co

! 1 I ' I 1 I 1 I 1 I 1 I 1 I 1 I ' I 1 I 1

CM

CD CO

CM

CD

CO O
t- CM

CM

CM

1SIAI IAI ifldea

o

o

o

o

o

o

o

CO

o

o

o

CO

o

o

o

Tt

o

o

o

CM

o

o

o

o

o

o

o

CO

o

o

o

CO

o

o

o

Tt

o

o

o

CM

o

Age Cal BP

Figure 15. Radiocarbon- and TIMS- dated coral samples from the northeastern coast of St Croix plotted against the western Atlantic sea-level curve (Toscano
and Macintyre 2003). This time-depth plot documents the progressive east-west shallowing of the antecedent surface and the various stages of development
of each site. Buck Island Bar samples never attain depths of less than 4-5 m below sea level. There are currently no published age data for Lang Bank.

18

the erosion of shelf soils. Non-carbonate mineralogic evidence supporting this hypothesis
from X-ray diffraction of reef sediments in cement crusts indicated the occurrence of clay
minerals and other siliclastic components coinciding with the disappearance of A. palmata
as a reef-framework builder (Adey et af, 1977).

In contrast to the southeastern shelf-edge reefs, two emergent northeastern shelf-
edge complex reef communities at Buck Island Bar and its western end at Long Reef
appeared to have kept pace with SLR. Adey et al. (1977) therefore interpreted these sites as
being located in areas isolated from the insular shelf or adjacent to very narrow shelf areas
and hence not significantly affected by shelf erosion turbidity “at 7000-8000 years BP.”
Lhese sites were thus expected to have maintained “continuous Holocene section[s] when
cored” (p. 20).

A sequential history of the northern outer shelf-edge reef system begins -8000
years ago when the A. palmata community presumably flourished on the northern flank of
Lang Bank (Fig. 1), while the adjacent inner shelf was initially flooded by rising sea level
(Hubbard, 1991). Lhe thickness of the Holocene reef cover on the northern flank of Lang
Bank is as yet unreported but may be similar to the 10.4 m reef section dominated by A.
palmata and sand intervals (Core BB-1; Hubbard, 1991; Hubbard et al., 2005; Fig. 16) on
Buck Island Bar, its western extension (Fig. 1). Buck Island Bar initiated by -7,700 cal BP
(Hubbard et al., 2005) above the -15 m to -16 m MSL deep Pleistocene surface as a shallow
A. palmata reef (two samples from Hubbard et al., 2005; Fig. 15A), followed by a -2,500
year coral gap (Fig. 15A; an artifact of limited coring?). Lhis gap (also noted by Hubbard
et al., 2005) appears to be real as the reef shows no significant framework accumulation of
A. palmata at this site. If so, this gap may indeed have been caused by local shelf flooding
and turbidity, or from another environmental cause. At about 4,000 years ago the reef record
resumed (until -1,200 cal BP) in a deeper water setting (Fig. 15A). Macintyre and Adey
(1990) had previously drilled Core BB-2 slightly leeward of BB-1 (Fig. 16), from which
they demonstrated that while this reef community tracked rising sea levels for the last 4,000
years (Fig. 15 A), it consistently remained below sea level by -5 meters (the depth of the
drill site). Macintyre and Adey (1990) particularly noted the number of heavily cemented
pavement limestone intervals and massive corals recovered in BB-2 (Fig. 16). Lhe heavily
cemented “pavement limestone” sections indicated long periods of exposure of the sea floor
to allow lithification to take place (Macintyre, 1977; Macintyre and Marshall, 1988). In
addition, they noted the deeper “anastomosing thickets of A. palmata on Buck Island Bar
as considerably more fragile than their robust shallow- water counterparts” (p. 6). Lhus they
interpreted the paleoenvironment at core BB-2 as one of frequent heavy seas breaking and
transporting fragile A. palmata from the site, leaving only the massive coral heads that grew
on the pavement around the A. palmata. Hubbard (1991) downplayed the magnitude and
destructive influence of waves and storms on A. palmata around St. Croix, and suggested
that the coral heads and cemented pavement sections in core BB-2 were typical of a
more protected back reef environment in comparison to the more exposed windward A.
palmata dominated core site BB-1 (Fig. 16). However, Hubbard's core BB-1 consisted
predominantly of sand intervals, suggesting a lower energy fore-reef setting where sand
could accumulate.

19

Buck Island Bar

m MSL

BB-1

Buck Island Bar

m MSL

BB-2

Figure 16. Core data from two cores drilled into Buck Island Bar (BB1- Hubbard 1991; Hubbard et al., 2005)
and (BB2 - Macintyre and Adey, 1990) showing the location of radiocarbon dates (after Hubbard, 1991;
Hubbard et al, 2005).

20

The question remained, however, as to why Buck Island Bar’s A. palmata reef
community lagged behind SLR, as indicated by the youngest radiocarbon date of 1180
cal BP at the top of Hubbard’s core BB-1 (-5 m MSL; Fig. 16). Figure 15A combines
all coral dates from Buck Island Bar (Hubbard et al., 2005; Macintyre and Adey, 1990).
Hubbard (1991) suggested that Buck Island Bar initiated before the oldest A. palmata date
of 7770 cal BP (date from Hubbard et al., 2005) “during a period of very rapid sea-level
rise” (Hubbard et al., 1991; p. 26) that presumably outpaced the accumulation rate of A.
palmata. Although the deepest A. palmata samples at the base of core BB 1 were not dated,
the two oldest dates of 7770 and 7670 cal BP, at present depths of -12.75 and -13.7 m MSL
respectively, appear to have grown in shallow water of no more than 3 m (Fig. 15 A) and
date to the oldest segment of the WASL curve when SLR was 5.2 mm/yr (Toscano and
Macintyre, 2003; Fig. 15 A), a rate slower than the growth rate capacity of A. palmata.

These dates are followed, after the -2,500 year gap, by a sample of A. palmata that grew
in -8 m water depths (close to their maximum) at 4763 cal BP, at which time the rate of
SLR had decreased markedly to 1.47 mm/yr (Toscano and Macintyre, 2003; Fig. 15A). All
subsequent A. palmata and head corals appear to have established as a deeper- water (10 to
5 m water depths) coral community on Buck Island Bar. It appears that whatever caused
the 2,500 year gap in the coral record prevented reef accumulation while water depths
simultaneously increased. When conditions again favored coral growth, water depth had
increased sufficiently to cause attenuation of sunlight (required for coral/algal symbiosis)
and thus reduce A. palmata s ability to grow rapidly and catch up to sea level for the next
4000 years. We also speculate that in deeper water these A. palmata likely utilized a thin,
upraised/outstretched branch growth habit to collect sufficient sunlight (as observed on site
by Macintyre and Adey, 1990). Thinner coral branches would have been more susceptible
to destruction by strong bottom currents generated by frequent heavy wave action on this
exposed shelf edge. Such ongoing conditions are hypothesized to have contributed to
limiting active reef accumulation at the shelf edge.

Interestingly, Hubbard (1991) reported that the northern reef communities on Lang
Bank ceased active accumulation about 5,000 years ago. The Virgin Islands C&GS Chart
905 shows that the crest of the northern flank of Lang Bank has present water depths of
about 9 to 11 meters. The WASL curve (Toscano and Macintyre, 2003; Fig. 15) indicates
that 5,000 years ago sea level was at -4 m MSL, thus the A. palmata communities were
growing in five to seven meter depths, or close to their maximum depth range, and not
accreting reef framework as part of a reef crest keeping pace with SLR. Again, deeper
water may have prevented “reef keep up” due to attenuation of light penetrating to these
depths. In addition, the possibility remains that the upward growth of a fragile, deep water
form of A. palmata was restricted or pruned by high energy conditions or frequent heavy
seas.

On the western end of the outer shelf-edge reef system, Long Reef (Fig. 13) exhibits
no more than 4.5 m accumulation of Holocene reef on top of a ~6-m higher, westward
shallowing Pleistocene surface. The higher elevation of the Pleistocene surface at the
western end of the outer shelf-edge reef (unknown to Macintyre and Adey, 1990), resulted
in -2500 yrs delay in the timing of reef initiation (at -4,500 cal BP) until sea level rose
above that surface. Framework began accumulating on the fore-reef spur at -6 m MSL
(Fig 13). While no dates at the Pleistocene/Holocene unconformity are available for Core

21

2 just behind the reef crest (Fig. 13), stratigraphically higher ages in this core suggest that
an estimate of >5,000 years ago would account for sufficiently high sea level (of -2.5 m,
Fig. 15B) over the ~6 m MSL Pleistocene surface to support coral growth in a back-reef
environment. Figure 15B illustrates the belated history of reef development in context of
SLR at Long Reef. Lhe few data are insufficient to determine how well the A. palmata
facies kept pace with SLR over the 5,000 years of its existence; however the reef as a whole
(including the head corals) appears to have tracked sea level consistently over that time.

Buck Island Reef, in its intermediate location between the outer shelf-edge reef
at Buck Island Bar and the inner shelf reef at Lague Bay, appears to be contemporaneous
with early reef initiation at Lague Reef as well as later reef development at Long Reef (Fig.
15C, D, B). Lhe Pleistocene surface under Buck Island Reef is slightly deeper (-15 to -18
m) than that underlying Lague Reef (-12.5 to -15.5 m), hence the earlier initiation of reef
growth at this more seaward location. With the exception of one deep (-14 m) A. palmata
date (Fig. 15C), head corals dominate the reef facies at Buck Island Reef until -5,000 cal
BP. Lhe depth of this A. palmata sample is inconsistent with the much shallower depths
of two other A. palmata of similar age. Hubbard et al.’s (2005) corelog indicates an A.
palmata and rubble interval at this depth below a head-coral facies. All cores from Buck
Island Reef exhibit sand, rubble, voids and head corals in the lower few meters (Hubbard et
al., 1991; 2005). Lhis sample may indicate a deeper water form of A. palmata in the midst
of a head coral facies contemporaneous with shallower reef crest A. palmata at this site. It
is also possible that this sample was detrital, or represents a drilling cave-in fragment based
on its age in comparison to more elevated samples.

Inner Bank-Barrier Reef Complex

Lhe core transect across the bank-barrier reef off Lague Bay (Fig. 14) shows the
optimum development of this reef system with 10 m of Holocene reef framework. Lhis
thickness, according to Burke et al. (1989), is related to the eastward dipping, deeper level
of the Pleistocene limestone substrate (-12 m MSL compared to - 6 m MSL underneath
Long Reef), providing more space for thicker and earlier development of the reef. Sea level
would have attained this surface at -8,500 years BP, or at least 3,500 years earlier than at
Long Reef. Burke et al. (1989) also postulated that the deeper antecedent surface would
have allowed for earlier improvement of water conditions at this site following the flooding
of the insular shelf and subsequent erosion of the sediment cover.

Figure 15D documents the history of Lague Reef in relation to SLR and the -12
to -13 m depth range of the Pleistocene surface under the reef. Although Lague Reef
is part of the inner shelf reef trend, the depth of the Pleistocene surface there is similar
to that underlying Buck Island Bar (-15 to -16 m). Lhese reefs are therefore roughly
contemporaneous despite their separate, parallel reef trends and histories. Lhey differ in
that Buck Island Bar (Fig. 15 A) initiated almost 1000 years earlier, may have experienced
a hiatus in A. palmata framework accumulation of -2,500 years (based on only two
available cores), then reestablished in a deeper environment wherein the reef could not
catch up or keep up with SLR. Lhe more protected inner shelf-reef trend in Lague Bay was
able to track and even attain sea-level elevations over most of its time frame of existence,
although according to Hubbard et al. (2005) Lague Reef and most of the inner shelf were

22

head coral-dominated before A. palmata framework took hold. Burke et al. (1989)
suggested that staggered dates of reef initiation along the length of the reef reflected
Tague Reef’s beginning as a series of isolated patch reefs. Figure 15D (from our reef-
normal cross section) suggests a hiatus in accumulation of A. palmata framework on
Tague Reef from -6,500 years ago to its apparent reestablishment just after -2,000 years
ago. This apparent hiatus occurred over a time of slow STR of 1.47 mm/yr (Toscano and
Macintyre, 2003), well below the regional accumulation rate of A. palmata framework
of 2.6 mm/yr (James and Macintyre, 1985). The ultimate reality of this apparent hiatus
cannot be ascertained owing to the lack of a dated reef-crest core, which might provide
the paleoenvironmentally-specific samples needed to fill this gap. In addition, only the
youngest and most elevateds, palmata from Tague Reef from Core 16 of Burke et al.
(1989) has been radiocarbon dated.

CONCLUSIONS

Two coral-reef core transects provide further insight into the Holocene history of
two parallel reef complexes of northeastern St. Croix. These new data further enhance
previous coring, radiometric dating and geologic interpretations for the area. Reefs
along the northeastern coast of St. Croix initiated sequentially from east to west along
a westward-shallowing pre-Holocene surface as rising sea level breached, eroded and
flooded these shelf areas in turn.

The eastern portions of the outer shelf-edge reef system have been previously
studied (Lang Bank and Buck Island Bar). Our new transect across the Long Reef shelf-
edge reef system (Fig. 13) represents the westernmost portion of the outer shelf-edge
reef system extending from Buck Island Bar and Lang Bank. The westward-shallowing
pre-transgression surface reaches a high point of -6 m MSL at Long Reef, resulting in a
thinner (< 4.5 m) and younger Holocene cover compared to the thick (-10 m) Holocene
section at Buck Island Bar. However, even the thicker eastern sections do not present long
continuous records of Holocene reef accumulation to present sea level. The Lang Bank
reefs stopped actively accumulating -5,000 years ago according to Hubbard et al. (2005)
in water depths of 5 to 7 meters. Buck Island Bar, where cored A. palmata remained
at depths of -4.5 meters over the past 8,000 years and never caught up with sea level,
stopped forming reef framework about 1,200 years ago when water depth was -5 m. One
core into Buck Island Bar offers evidence, in the form of leeward crest hard pavements,
of potentially destructive high-energy conditions. It appears that the deeper depths for A.
palmata growth resulted in this coral forming thinner, flatter, more fragile branches which
could have been easily broken and transported during periods of severe wave action,
resulting in a lack of reef accumulation.

The transect across the inner bank-barrier reef system off Tague Bay, by contrast,
illustrates the optimum development of this reef with an accumulation of 10 m of
Holocene deposits. The successful formation of this reef at this site appears to be the
combination of accommodation space due to the depth of substrate and the possibility

23

that water conditions in this area improved rapidly following the flooding of the shelf and
erosion of the sediment cover (Burke et al., 1989).

Implications of this study for reconstruction of reef history and sea level using core
data indicate that single reef localities (or cores) are unlikely to provide a comprehensive
record of local shelf evolution or sea-level rise. The spatial and temporal variations of a
reef complex (e.g., the outer shelf-edge reef complex) indicate the variety of site-specific
physical and environmental factors controlling reef location and affecting reef growth. In
addition, the biases inherent in interpreting minimal core data (such as at Buck Island Bar)
may result in incomplete or inaccurate reef history reconstructions, and insufficient data
to identify, contrain and understand true gaps in the record of A. palmata. The regional
Caribbean sea-level synthesis provides the needed context for understanding varying
depths, age ranges and histories of the reef settings along the northeastern coast of St. Croix,
given a sufficient number of core transect localities along the reef system. More data from
northeastern St Croix, particularly from Buck Island Bar and eastern Tang Bank, will help
to complete the history presented here.

ACKNOWLEDGEMENTS

Macintyre received valuable help with coring of Tague Bay reef in 1976 from W.H.
Adey, R.C. Shipp, D.K. Hubbard, and the late R.F. Dill. In the later 1978 coring operation
of Long Reef off Christiansted, C.V.G. Phipps, R.C. Shipp, M. Dominic, J. Eherts, and
the late R.F. Dill assisted with the drilling. A special thanks to W.H. Adey for the bottorm
community maps that were used to show core-hole locations. Walter H Adey and Denis
K. Hubbard provided important assistance in reviewing the manuscript. This is Ottawa-
Carleton Geoscience Centre, Isotope Geochemistry and Geochronology Research Centre
contribution No. 51.

REFERENCES

Adey, W.H., I.G. Macintyre, R. Stuckenrath, and R.F. Dill

1977. Relict barrier reef system off St. Croix: Its implications with respect to late
Cenozoic coral reef development in the western Atlantic. Proceedings of the
Third International Coral Reef Symposium 2: 15-21.

Burke, R.B., W.H. Adey, and I.G. Macintyre

1989. Overview of the Holocene history architecture and structural components, Tague
Bay, St. Croix, U.S.V.I. Pages 105-110 in D.K. Hubbard (ed.), Terrestrial and
Marine Geology of St. Croix, U.S. Virgin Islands. West Indies Laboratory, Special
Publication 8.

Cheng, H, Edwards, L R, Hoff, J, Gallup, C D, Richards, D A, and Asmerom, Y.

2000. The half-lives of uranium-234 and thorium-230. Chemical Geology, Vol.169: 17-
33.

24

Dahl, A.L., I.G. Macintyre, and A. Antonius

1974. A comparative survey of coral reef research sites. Pages 37-120 in Sachet,

M.H., and A.L. Dahl (eds.). Comparative Investigations of Tropical Reef
Ecosystems: Background for an integrated Coral Reef Program. Atoll Research
Bulletin 172.

Gerhard, L.C.

1981. Origin and evolution of Candlelight Reef -Sand Cay system, St.Croix. Atoll
Research Bulletin 242: 1-22.

Gladfelter, W.B.

1982. White band disease in Acropora palmata : Implications for the structure and
function of shallow reefs. Bulletin of Marine Science 32:639-643.

Hubbard, D.K.

1991. Geologic development of Buck Island Reef National Monument. A Report to the
National Park Service 36p.

Hubbard, D.K., R.B. Burke, and I.P. Gill

1986. Styles of reef accretion along a steep shelf-edge reef, St. Croix, U.S. Virgin
Islands. Journal of Sedimentary Petrology 56: 848-861.

Hubbard, D. K., A.I. Miller, and D. Scaturo

1990. Production and cycling of calcium carbonate in a shelf-edge reef system (St.
Croix, U.S. Virgin Islands): Applications to the nature of reef systems in the
fossil record. Journal of Sedimentary Petrology 60: 335-360.

Hubbard, D.K., H. Zankl, I.V. Heerden, and I.P. Gill

2005. Holocene reef development along the northeastern St. Croix shelf, Buck Island,
U.S. Virgin Islands. Journal of Sedimentary Research 75: 97-113.

Ivanovich, M, Latham, A G, and Ku, T-I.

1992. Uranium-series disequilibrium applications in geochronology. 62-89 in
Ivanovich, M and Harmon, R S (Editors), Uranium- series Disequilibrium:
Applications to Earth, Marine and Environmental Sciences. 2nd edition. [Oxford:
Clarendon Press.] 910pp

Lessios, H.A., D.R. Robinson, and J.D. Cubit

1984. Spread of Diadema mass morality through the Caribean. Science 226: 335-337.

Lighty, R.G., I.G. Macintyre, and R. Stuckenrath

1982. Acropora palmata reef framework: A reliable indicator of sea level in the western
Atlantic for the past 10,000 years. Coral Reefs 1: 125-130.

Macintyre, I.G.

1975. A diver operated hydraulic drill for coring submerged substrates. Atoll Research
Bulletin 185: 21-25.

Macintyre, I.G.

1977. Distribution of submarine cements in a modern Caribbean fringing reef, Galeta
Point, Panama. Journal of Sedimentary Petrology 47: 503-516.

25

Mac intyre, I.G.

1988. Modern coral reefs of western Atlantic: A new geological perspective.
American Association of Petroleum Geologists Bulletin 72: 1360-1369.

Mac intyre, I.G., and W.H. Adey

1990. Buck Island Bar, St. Croix, U.S.V.I.: A reef that cannot catch up with sea level.
Atoll Research Bulletin 336: 1-7.

Macintyre, I.G., and P.W. Glynn

1976. Evolution of modern Caribbean fringing reef, Galeta Point, Panama. American
Association of Petroleum Geologists Bulletin 60: 1054-1072.

Macintyre, I.G., and J.F. Marshall

1988. Submarine lithiheation in coral reefs: Some facts and misconceptions.

Proceedings of the Sixth International Coral Reef Symposium, Townsville,
Australia 1: 263-272.

Toscano, M.A., and I.G. Macintyre

2003. Corrected western Atlantic sea-level curve for the last 1 1,000 years based on
calibrated 14 C dates from Acropora palmata framework and intertidal
mangrove peat. Coral Reefs 22:257-270.