SY
+. *
seathan
ee NV
freee ‘ vn an asa re Ran eert scat ongeimntete Wa wears
uh Raced Catt
Ria ore a aa ee
Sein ige lanes
Rei are Me
Sein anak mR
ara EME a St selacsmt ner
erate ent ee
See) Prete a hatord ©
sea 4 our
‘utes
apa a
ae a
mate
oe ‘ th ; cher PEP EP TOR aaah
aio AE haha eas iaidbabeta eed tees pita
' | ’ St z th = if ny Ah ey i > bat ¢ F A : sh x + i %, a n t lay
1 fe mae
”
i
' : i eos { & saa * : f why By ' Ves ¢ t 9
f =) tp: 1 8 Tr ' = } ; 4
; } ; 5 f i te A ’ ips » : j ‘
Issued by NATIONAL MUSEUM OF NATURAL HISTORY
SMITHSONIAN INSTITUTION WASHINGTON, D.C. U.S.A. APRIL 1996
ATOLL RESEARCH BULLETIN NOS. 435-442
RESEARCH BULLETIN
ATOLL RESEARCH BULLETIN
NOS. 435-442
NO. 435.
NO.
NO.
NO.
NO.
NO.
NO.
NO.
436.
437.
438.
439.
440.
441.
442.
MORPHOLOGY AND MARINE HABITATS OF TWO SOUTHWESTERN CARIBBEAN ATOLLS: ALBUQUERQUE AND COURTOWN
BY JUAN M. DIAZ, JUAN A. SANCHEZ, SVEN ZEA, AND
JAIME GARZON-FERREIRA
CORAL FAUNA OF TAIPING ISLAND (ITU ABA ISLAND) IN THE SPRATLYS OF THE SOUTH CHINA SEA BY CHANG-FENG DAI AND TUNG-YUNG FAN
FIRST OBSERVATIONS ON THE FISH COMMUNITIES OF FRINGING REEFS IN THE REGION OF MAUMERE (FLORES- INDONESIA)
BY MICHEL KULBICKI
GROUPER DENSITY AND DIVERSITY AT TWO SITES IN THE REPUBLIC OF MALDIVES BY ROBERT D. SLUKA AND NORMAN REICHENBACH
EFFECT OF TYPHOONS ON THE LIZARD COMMUNITY OF A SHELF ATOLL BY MICHAEL JAMES MCCOID
FLOWERING AND FRUITING IN THE FLORA OF HERON ISLAND, GREAT BARRIER REEF, AUSTRALIA BY R.W. ROGERS
NAMU ATOLL REVISITED: A FOLLOW-UP STUDY OF 25 YEARS OF RESOURCE USE BY NANCY J. POLLOCK
CRUSTACEA DECAPODA OF FRENCH POLYNESIA (ASTACIDEA, PALINURIDEA, ANOMURA, BRACHYURA) BY JOSEPH POUPIN
ISSUED BY NATIONAL MUSEUM OF NATURAL HISTORY SMITHSONIAN INSTITUTION WASHINGTON, D.C., U.S.A. APRIL 1996
ACKNOWLEDGMENT
The Atoll Research Bulletin is issued by the Smithsonian Institution to provide an outlet for information on the biota of tropical islands and reefs and on the environment that supports the biota. The Bulletin is supported by the National Museum of Natural History and is produced by the Smithsonian Press. This issue is partly financed and distributed with funds from Atoll Research Bulletin readers and authors.
The Bulletin was founded in 1951 and the first 117 numbers were issued by the Pacific Science Board, National Academy of Sciences, with financial support from the Office of Naval Research. Its pages were devoted largely to reports resulting from the Pacific Science Board's Coral Atoll Program.
All statements made in papers published in the Atoll Research Bulletin are the sole responsibility of the authors and do not necessarily represent the views of the Smithsonian nor of the editors of the Bulletin.
Articles submitted for publication in the Atoll Research Bulletin should be original papers in a format similar to that found in recent issues of the Bulletin. First drafts of manuscripts should be typewritten double spaced and can be sent to any of the editors. After the manuscript has been reviewed and accepted, the author will be provided with a page format with which to prepare a single-spaced camera-ready copy of the manuscript.
COORDINATING EDITOR
Ian G. Macintyre National Museum of Natural History MRC-125 ASSISTANTS Smithsonian Institution Kasandra D. Brockington Washington, D.C. 20560
William T. Boykins, Jr. Theodore E. Gram
EDITORIAL BOARD
Stephen D. Cairns (MRC-163) National Museum of Natural History Brian F. Kensley (MRC-163) (Insert appropriate MRC code)
Mark M. Littler (MRC-166) Smithsonian Institution
Wayne N. Mathis (MRC-169) Washington, D.C. 20560
Victor G. Springer (MRC-159)
Joshua I. Tracey, Jr. (MRC-137)
Warren L. Wagner (MRC-166)
Roger B. Clapp National Museum of Natural History
National Biological Survey, MRC-111 Smithsonian Institution Washington, D.C. 20560
David R. Stoddart Department of Geography 501 Earth Sciences Building University of California Berkeley, CA 94720
Bernard M. Salvat Ecole Pratique des Hautes Etudes Labo. Biologie Marine et Malacologie
Université de Perpignan 66025 Perpignan Cedex, France
PUBLICATIONS MANAGER
A. Alan Burchell Smithsonian Institution Press
ATOLL RESEARCH BULLETIN
NO. 435
MORPHOLOGY AND MARINE HABITATS OF TWO SOUTHWESTERN
CARIBBEAN ATOLLS: ALBUQUERQUE AND COURTOWN
BY
JUAN M. DIAZ, JUAN A. SANCHEZ, SVEN ZEA, AND JAIME GARZON-FERREIRA
‘ Pr EM MISUNIATN = ¥ \
Ql
ISSUED BY NATIONAL MUSEUM OF NATURAL HISTORY SMITHSONIAN INSTITUTION WASHINGTON, D.C., U.S.A. APRIL 1996
RMON MMOQ AA NM
STAND NANT DAT RDNA BIA DTT
Ny S| . \ S \ N S N NN N N \ h N S| ‘| N N N hy RK N S N 's N N ny S ‘| N ‘ Ss N S 'S N S \ N N N Ss N N NY N NN Ss N N N N S| N N N yy N N N \ N S| N N Ss NN N N N N N N 's N N N N NY N \ N \ NN N N
AMD DE
MMMM MLL
QUITASUENO BANK PROVIDENCIA Is.
SONI
Honduras
ALBUQUERQUE ATOLL
Nicaragua
COLOMBIA
ZL
SSLSS EA.
LA GLIDOV ISAS IETS SE
Z
LFS.
Z
EIS.
FHV
GALS LS
PLIST.
LPS.
LIDS.
Gil He
VA
LED Vids
f the Colombian
h the location o
2.
wit
bbean Sea
ri ines are iSO
1. Map of the Southwestern Ca
igure
F
In m
baths (
lands, atolls and banks. Contour li
1S
MORPHOLOGY AND MARINE HABITATS OF TWO SOUTHWESTERN CARIBBEAN ATOLLS: ALBUQUERQUE AND COURTOWN BY JUAN M. DIAZ}, JUAN A. SANCHEZ , SVEN ZEA!” and JAIME GARZON-FERREIRA'
ABSTRACT
Albuquerque and Courtown are two small, uninhabited oceanic atolls, located in the southwestern Caribbean Sea, belonging to the San Andrés and Providencia archipelago, Colombia. These atolls have a volcanic basement and are surrounded by deep water. Based on photo-interpretation of geomorphological and ecological features as well as on data collected during field work, the gross morphology, marine bottom habitats and reef structures of both atolls are described down to a depth of 50 m. Distributions of morphological and bottom habitat units are presented in thematic maps showing the overall zonational patterns in the two atolls.
Morphological and ecological zonations in both atolls are primarily controlled by both wave exposure in a windward-leeward gradient and depth. The presence of an ample windward fore-reef terrace, a well developed windward barrier reef with spur-and-groove system, an extensive lagoonal terrace with sudden transition to the lagoon basin, and profuse development of ribbon and anastomosing patch reefs in the lagoon are characteristics common to both atolls. As in other Caribbean atolls, the outer slope in Albuquerque and Courtown is outlined by a sandy step or bench at 35 to 45 m depth. Significant differences between the two atolls exist in the degree of development and structure of leeward peripheral reefs, as well as in the amplitude of the leeward fore-reef terrace. At Albuquerque, peripheral reefs grow on a shallow flat and enclose the lagoon along a wide semicircle, whereas at Courtown such reefs have in part developed algal ridge-like structures and are unevenly distributed, leading to an open lagoon to the east. The broad leeward terrace in Albuquerque contrasts markedly with the rapidly dipping leeward slope towards the outer shelf margin in Courtown.
Accumulations of sand and rubble have led to the formation of cays and small islands on the lagoonal terrace in both atolls, but also on leeward peripheral reefs in
Instituto de Investigaciones Marinas y Costeras, INVEMAR, Apartado 1016, Santa Marta, Colombia
* Universidad Nacional de Colombia (Departamento de Biologia)
Manuscript received 31 March 1995; revised 21 Novermber 1995
Courtown, some of which have experienced remarkable changes in their size and shape in the last 25 years.
Biological composition and structure of reefs in both atolls show a great resemblance to one another and to the better-known reef complexes around the nearby islands of San Andrés and Providencia. Although no urban development exists in these atolls, recent decline of living coral and over-exploitation of marine resources were evident.
INTRODUCTION
There are about 425 atolls worldwide and only 15 of them are located in the Atlantic, of which four are part of the San Andrés and Providencia archipelago in the Southwestern Caribbean Sea (Milliman, 1973; Geister, 1983). This archipelago comprises a series of islands, atolls and coral shoals running in SSE-NNE direction, parallel to the Nicaraguan Rise for more than 500 km. It is separated from the Central American continental shelf by the San Andrés Trough (Fig. 1). The southernmost reefs of the archipelago, Albuquerque and Courtown (the latter are also called Bolivar Cays) are two small atolls lying about 200 km east of the Nicaraguan coast. Although geographically closer to Central America than to the South American continent, the archipelago has belonged to the Republic of Colombia since 1822.
The accurate date of human discovery of these atolls is uncertain but their locations were well known to the Spanish sailors of the 16th century and were probably occasionally visited by Miskito Indians from the Central American coast, who came for fishing and turtling (Parsons, 1956). None of the tiny sand cays on the atolls has sufficient land to warrant permanent settlement, but one of them on each atoll serves presently as a military post for the Colombian navy, and they are visited regularly by fishermen and tourists in chartered yachts from nearby San Andrés.
Briefly mentioned by Darwin (1842) in his interpretation of Caribbean reef structures and their origin, the reefs of Albuquerque and Courtown have since received little scientific attention in comparison with those around the nearby islands of San Andrés and Old Providence (Geister, 1969, 1973,1975,1992; Kocurko, 1977; Marquez, 1987; Diaz et al., 1995) and other West Atlantic and Caribbean atolls such as Hogsty Reef (Milliman, 1967), Alacran Reef (Kornicker and Boyd, 1962; Bonet, 1967), Chinchorro (Jordan and Martin, 1987) and those off Belize (Stoddart, 1962; James and Ginsburg, 1979: Riitzler and Mcintyre, 1982; Gischler, 1994). Albuquerque and Courtown were briefly visited by the Fifth George Vanderbilt Expedition in 1941. Published observations include reports on the birds (Bond and DeSchauensee, 1944), fishes (Fowler, 1944) and crustaceans (Coventry, 1944). The R/V GERDA, of the University of Miami, stopped in May 1966 for few days at Albuquerque and Courtown and conducted observations on the
ecology, morphology and oceanography of the atolls. From this visit, Milliman and Supko (1968) made preliminary conclusions on the geological origin, and Milliman (1969) described the general characteristics of the reefs and commented on hydrography. Further oceanographic findings from the waters surrounding the atolls have been recorded during research cruises by the Colombian navy (Gonzalez, 1988; Téllez et al., 1988). Aspects of the terrestrial environment and fauna were more recently discussed by Chirivi (1988). However, very little is known about the distribution of marine bottom habitats and the zonation of the reefs constituting these atolls.
Therefore, the purpose of this paper is to give the first detailed systematic description of the gross morphology and the marine habitats of Albuquerque and Courtown atolls, with emphasis on the reef structures.
REGIONAL SETTING
Albuquerque and Courtown are the southernmost reef complexes of the San Andrés and Providencia archipelago. Albuquerque (12° 10' N and 81° 51' W) is located 37 km south of San Andrés Island and about 190 km east of the Nicaraguan coast. It is nearly circular in shape, about 5.5 km E-W and 4.5 km N-S. Two small islands, North Cay and South Cay, rise up to 2 m above mean sea level behind the seaward barrier reef and are separated from each other by a 250 m shallow channel.
Courtown (12° 24' N and 81° 28' W) lies 30 km southeastward from San Andrés and 47 km northeast of Albuquerque. It is kidney shaped, about 3.5 km E-W and 6.5 km SSE-NNW (Fig. 3). Although this atoll presently bears two cays (East Cay and West or Bolivar Cay) and a tiny sand spit, their size, shape and number seem to be quite variable in the course of time, as can be easily inferred from an earlier description and map of the atoll by Milliman (1969).
Toward the north and eastern sides of both atolls, an almost continuous barrier reef is well developed, whereas the leeward peripheral reefs are absent or ill defined and are separated by wide gaps and channels.
Both atoll foundations rise from the surrounding sea floor more than 1000 m deep, and apparently have a volcanic basement. Unequivocal evidence for the volcanic origin of these atolls and nearby islands comes from the magnetic anomalies detected at San Andrés Island and Courtown, one volcanic pebble dredged from Albuquerque basement (Milliman and Supko, 1969), as well as the volcanic rocks of Providencia (Geister, 1992) and the Corn Islands (McBirney and Williams, 1965). Further aspects of the geological origin of the archipelago are discussed by Geister (1992: p. 56-58)
Available meteorological data recorded from nearby San Andrés between 1959 and 1986 (Diaz et al., 1995) are used here, as there are no recorded observations from
either atoll. The mean annual air temperature is 27.4°C, with a 1°C range in monthly values. The annual rainfall measured at San Andrés is about 1900 mm, of which over 80% falls between June and November. Winds are trades, from the ENE, with a mean annual intensity of 6.1 m/s and mean monthly variations between 4.5 m/s (May, September- October) and 6.6 m/s (December-January, July). Sporadic storms occur mostly in the second half of the year, with westerlies or northwesterlies attaining speeds over 20 m/s.
Albuquerque and Courtown lie in the Caribbean hurricane belt. Hurricanes were recorded in 1818, 1876, 1877, 1906, 1940, 1961, 1971 and 1988 (cf. Barriga et al., 1969: 23; Geister, 1992: 7; Diaz et al., 1995: 112). The latter, 'Joan', on October 20-22 1988, passed westwards 90 km south of San Andrés (about 50 km south of Albuquerque); its eye attained a diameter of about 35 km and the wind reached speeds over 210 km/h (Geister, 1992).
The Caribbean Current reaches Albuquerque and Courtown from the NE with speeds of 0.5-1 m/s and passes over the atoll shelf in a SW to W direction, being highly affected by the irregular bottom topography of the shallow-water zones. Waves are generated by the trade winds and approach the atolls from the NE to E, the effective fetch extending for nearly 2,000 km over almost the entire width of the Caribbean Sea. Hence, the considerable amplitude and height of waves breaking on the barrier reef along the windward side of the atolls.
The sea surface temperature averages 27.5°C, with mean monthly values ranging between 26.8 (February-March) and 30.2°C (September-October). Surface salinity fluctuates between 34.0 and 36.39/00 (Gonzalez, 1988). Tides on the atolls are mixed with a strong diurnal component. Tidal ranges between 0.3 and 0.6 m are recorded from nearby San Andrés (Geister, 1975).
METHODS
A preliminary photo interpretation of geomorphological and ecological features of both atolls was done on panchromatic total coverage air photography taken in 1971 and 1984 by the Colombian Geographical Institute (Instituto Geografico 'Agustin Codazzi') approximately 1:22,500, 1:23,000 and 1:30,000, which was then used as basis for field sampling. Preliminary morphological and habitat distribution maps at 1:20,000 scale were drawn combining reef and lagoon photo-patterns defined on the basis of tone, texture and location, as well as bottom topography inferred from bathymetric charts 1:20,000 COL- 203 (Albuquerque) and COL-204 (Courtown). Further detail of the spur-and-grove system of the barrier reefs and lagoonal patch reefs was obtained from oblique aerial colour slides taken on September 29, 1994 from a chartered aircraft at altitudes of 200 to 500 m.
During a cruise to the atolls in May-June 1994 aboard the R/V ANCON of the Instituto de Investigaciones Marinas y Costeras (Santa Marta, Colombia), 8 days (May 20-27) were spent at Courtown and 12 (May 28 - June 8) at Albuquerque. A total of 23 (Courtown) and 25 (Albuquerque) observation and sample sites were visited (Figs. 2-3). Location of sample sites included several examples of each of the photo-pattern units, and their exact geographical placement was carried out with an accuracy of 20 m with the aid of a portable Geographic Positioning System (GPS) instrument. SCUBA was used for depths over about 6 m SCUBA was used, otherwise observations were made while skin diving or walking for shallower areas. Observations of bottom types, depth, direction of currents, dimensions and distribution patterns of the reef structures, as well as species composition of dominant biota were recorded on acrylic data sheets. Complementary depth profiles were recorded with the ship's echosounder (28 khz).
Final thematic maps at 1:20,000 (bathymetry, geomorphology, bottom habitats, wave exposure) were entered via a digitizing table into a geographic information system (GIS-ILWIS) for storage, processing and further analysis. Morphology and _ habitat classification and terminology vary considerably between authors, and the terms used here to define morphological units and reef zones follow those of Geister (1975, 1977, 1983). Marine habitats are named, where possible, after the substrate dominating macrobiota or substrate features, as was done by Duyl (1985) for the reef environments of the Netherland Antilles.
RESULTS
Both atolls have the same basic morphological features (Figs. 2 and 3) and, with minor differences, the same marine environments (Figs. 4 to 7). To save space, a general description of each of the morphological units is given below with comments on the bottom habitats found there (map units on Figs. 6 and 7) and, where necessary, on the pecularities of each atoll. Table 1 includes a brief description of the habitats (map units) and allows cross referencing to morphological units. Figures 4 and 5 are representative profiles of the atolls and show the morphological features and bottom habitat distribution along a windward-leeward (right to left) gradient.
FORE-REEF TERRACE AND OUTER SLOPE
The windward margin of the atolls is characterized by the presence of a gently dipping terrace, descending at low angle (from 6 to 9 degrees) to -24 to -30 m (somewhat deeper in Albuquerque than in Courtown), where a topographical break gives way to a subvertical slope below -30 m. The break marks the transition to the outer slope of the atoll shelf. From a depth of 4-8 m seaward of the barrier reef, to about -15 m, this flat, calcareous platform is, with the exception of scattered gorgonians (Pseudopterogorgia sp.) and large sheets of excavating sponges (Cliona aprica and C. caribbea), mostly devoid of sessile organisms and sediments (‘bare calcareous hard bottom' unit). Low relief calcareous ridge-like structures, with a parallel layout similar to the spur and groove
system of the barrier reef (see below), are found along the entire width of the terrace and are more conspicuous on the northeastern section of the atolls. Shallow furrows between these low ridges are filled with coarse sediments and rubble below 18 m. Toward the outer margin of the fore reef terrace, faunal richness and diversity increase gradually, at first especially with brown algae (Sargassum sp., Stypopodium sp.), green algae (Halimeda spp.), massive scleractinians (Diploria spp., Porites astreoides, Siderastrea siderea) and many branching octocorals (Pseudopterogorgia spp., Pterogorgia citrina, Eunicea spp., Plexaurella spp.) (‘Gorgonaceans on hard bottom’ unit). Below 18 m more and more hemispherical scleractinians (Montastraea spp., Colpophyllia natans, and others) and sponges come into sight, as well as coarse sediments that accumulate in shallow hollows. Although coral heads often attain considerable size, they are mostly solitarily, tens of meters apart. In contrast to this, a narrow belt along the transition zone to the outer slope (24 to 30 m) exhibits a well developed coral community, and the calcareous platform appears therefore almost totally covered by corals (Vontastraea spp., Colpophyllia natans, Agaricia agaricites, Dichocoenia stokesii, Stephanocoenia intersepta, among others) , algae (Lobophora sp., Halimeda spp.), sponges and octocorals (‘mixed corals’ unit).
The windward outer slope was visited only in Courtown, but its morphology seems to be similar in both atolls, as could be inferred from the recorded bathymetric profiles. The outer slope dips gently (ca. 40-50°) to a sand step beginning at -30 to -35 m. Since this step can be easily recognized on the aerial photographs as a narrow, light grey band along the windward margin of the atolls, thus it seems to be covered by high- reflectance sediments. Below this step, the outer slope decreases subvertically to -400 m and then at lower angle to depths beyond 1,000 m.
WINDWARD BARRIER REEF
The barrier reef does not completely encircle the atolls, but extends only along the inner shelf from the NNW, N, NE, E, and SE almost continuously for about 5.6 km at Albuquerque and 7.5 km at Courtown. The continuous reef segments are 50-250 m across, being formed by more or less coherent ridges rising from the upper margin of the fore-reef terrace at 5 to 6 m to a reef flat near low tide level.
The barrier reef is normally deeply penetrated by surge channels oriented perpendicular to the reef front, forming a typical spur-and-groove system which is easily recognizable on the aerial photographs. Also scattered coral pinnacles rise in some places from about -4 m, just windward of the surf zone, often breaking the surface. The spurs rise 0.5 to 2 m above the adjacent grooves, the latter being 1 to 5 m or more wide and often exhibiting anastomosing bifurcations (Plate 1). At Courtown, the barrier reef is indented at two places, giving the atoll its distinctive kidney shape. Here, the reef crest becomes discontinuous, and a well developed buttress-groove system appears instead (Plate 2), creating a transition zone 300 to 500 m wide between the fore-reef terrace and the lagoonal terrace in its lee. The 2-3 m depth surge channels in this area allow small boats to pass the barrier during calm days. At Albuquerque, the barrier reef is virtually continuous, but on its NE margin a few unusually wide grooves interrupt the reef flat for
10 to 20 m, permitting some waves to pass undisturbed into the lagoonal terrace. At this place, a second, discontinuous barrier reef, located 100 to 200 m behind the former and nearly parallel to it, generates a displaced surf zone clearly observable from the air.
The main framework builder in the windward reef flat is the hydrocoral Millepora complanata, which is commonly associated with incrustations of coralline algae. Millepora and the zoanthid Palythoa sp. overgrow the shallowest zone of the barrier reef flat and the upper surfaces of the spurs (‘Millepora-Palythoa' unit), the high surf splashing and washing permanently the emergent colonies. In the buttress-groove area in Courtown, as well as in the second barrier at the NE margin of Albuquerque, Palythoa is generally replaced by Porites porites (growing usually within the Millepora colonies) and crustose forms of Porites astreoides and Diploria clivosa, which overgrow with Millepora the upper parts of the buttresses and the reef flat (Willepora-P. porites' unit). The upright sides of the spurs and buttresses are encrusted with Diploria spp., Porites astreoides and Agaricia agaricites, often assuming a flat form. The hydrocoral Stylaster roseus, the green alga Halimeda, as well as coralline red algae (Porolithon sp.) are also common elements in this zone. Large (up to 2-3 m in diameter) sheet-like excavating sponges (Cliona aprica, C. caribbea) may be fairly common at the sides and bottom of grooves.
Leeward of the reef crest, following the ‘Millepora-Palythoa’ unit, cushion-like colonies of Porites porites as well as massive P. astreoides and Diploria strigosa occur at some places among small ridges of Millepora and calcareous boulders (‘Millepora-P. porites’ unit). The displaced rear barrier reef on the NE side of Albuquerque consists likewise of extensive ridges with Millepora complanata and Porites porites rising from - 1.5 to -2.5 m. In some places, like in the NE barrier of Albuquerque and the SE section of Courtown, the coral growth on the rear reef zone extends for about 250 m. There, the end of the barrier reef is marked lagoowards by patchy thickets of Acropora palmata, accompanied by small colonies of Diploria strigosa, Montastraea spp. and occasionally also by cushion shaped colonies of Porites porites (‘Diploria-A. palmata’ unit).
In Courtown, the southernmost portion of the barrier reef becomes discontinuous after it bends westward. Numerous pinnacles, constituted mostly by a framework of Millepora at their upper parts, rise in this area from -4 to -5 m reaching up usually to a few centimeters below the surface (Plate 3). At the base of the pinnacles are massive colonies of Diploria spp. commonly more than 2 m in diameter, small thickets of Acropora cervicornis and branching octocorals. The pinnacles are generally arranged in groups, separated by anastomosing sandy channels, with a characteristic wave-induced pattern of ripple marks. Coral rubble (mostly of Acropora cervicornis) accumulates at the sides of the channels.
LAGOONAL TERRACE
The leeward margin of the reef flat leads down to the lagoonal terrace usually with an abrupt, 0.6 to 1.5 m high, steep slope. The lagoonal terrace is a flat platform attaining a width of 200 to 900 m and increasing in depth from 1 to 3 m towards its inner margin. The lagoonal terrace is one of the most discernible features from the air due to its pale
hue. Close to the rear reef, the terrace is covered by rubble (‘hard bottom and rubble' unit), which is gradually replaced lagoonwards by gravel and coarse sand (‘sand and rubble’ unit). The rubble zones are usually arranged in elongated layers perpendicular to the barrier reef, apparently related to the grooves and depressions of the reef crest. The innermost rubble areas on the terrace are overgrown by green (Halimeda, Padina), brown (Dictyota, Turbinaria) and red algae (Amphiroa, Neogoniolithon), as well as scattered encrusting scleractinians (P. astreoides, Siderastrea)('‘rubble with algae’ unit). Some portions of the sandy bottom, particularly in Albuquerque, are sparsely colonized by green algae (Penicillus, Rhipocephalus, Udotea), where juvenile individuals of the gastropod Strombus gigas are fairly common. The lagoonal terrace normally terminates on its lee with a steep 'sand cliff, leading down into the lagoon basin with slopes up to 40°. It represents an accretionary fore-set of fine-grained sediments transported from the reef area to the leeward margin of the terrace
The two cays existing in Albuquerque (North Cay and South Cay, Plate 4), as well as East Cay in Courtown (Plates 5 and 6), are sand and rubble accumulations on the lagoonal terrace. Coconut palms, Ficus trees, Scaevola bushes and Tournefortia shrubs are the dominant vegetation. North Cay, at Albuquerque, serves today as military post for the Colombian navy. Several bands of beachrock, paralleling the windward shoreline of these cays, extend eastward on the lagoonal terrace for about 15 (both cays in Albuquerque, Plate 7) to 70 m (East Cay in Courtown), suggesting the location of previous shorelines and thus a lagoonward migration of the cays. The two cays at Albuquerque are presently very close to the leeward margin of the lagoonal terrace. Where submerged beachrock is not covered by rubble and sand, it is mostly overgrown by encrusting coralline and green algae (Halimeda, Rhipocephalus) that contain dense populations of boring sea urchins (Echinometra lucunter). The only sea grasses on the atolls occur on the sheltered leeward side of North Cay in Albuquerque and East Cay in Courtown, where they cover the shallow sandy bottom of the terrace (‘sea grass' unit) The dominant grasses in Courtown are Syringodium and Halodule, whereas Thalassia is more abundant in Albuquerque. The edible urchin, 7ripneustes ventricosus, is abundant in these grass meadows.
LAGOON WITH PATCH REEFS
The depth of the lagoonal basin is as much as 18 m (in Albuquerque, see below) but generally it varies between 8 and 10 m. Where corals and coral reefs are lacking, the lagoon floor is covered by white calcareous sediments, the coarser fractions of which consist mostly of fragments of coral, molluscs, foraminifera, coralline algae and Halimeda, and rubble. Numerous burrows, mouds and faecal pellets throughout the deeper parts of the lagoon evidence an active bioturbation of the bottom (‘bioturbated sediments’ unit). Green algae (Rhipocephalus, Udotea, Halimeda) grow sparsely around coralline areas forming small patches, where one or more individuals of the Queen Conch, Strombus gigas, as well as patchy aggregations of garden eels (7aeniconger sp.) are occasionally found.
A significant portion of the lagoon is occupied by coral reefs, which are highly variable in shape and size, as well as in the dominant scleractinian species, depending mainly on the depth and wave exposure. Reefs occur as solitary mounds and miniatolls, or as ribbon and anastomosing patch reefs. In order to simplify the nomenclature, we divided the patch reefs found inside the lagoon into three main types (map units, see Table 1), according to the dominant scleractinian species: a) emergent to very shallow 'Diploria-A. palmata’ reefs dominated at their summit by Diploria strigosa and Acropora palmata, b) 2-5 m deep 'A.cervicornis' reefs dominated by thickets of Acropora cervicornis, and c) 4- 16 m deep '‘Montastraea spp'.-reefs dominated by one or more species of the Montastraea annularis species complex (see Weil and Knowlton, 1994).
At Courtown, lagoon depths vary between 7 and 15 m. Patch reefs cover about 30% of the lagoon floor. In the northern half of the lagoon, where the average depth is about 10 m, a dense net of anastomosing reefs (Montastraea spp.) covers nearly 50% of the bottom. Most of them are low-lying, rising no more than 4 m above the bottom (Plate 8), but some are nearly emergent and form a wave-breaking zone of thickets of A. palmata. The relative coverage of living scleractinians composing these reefs ranges between 10 to 50% from one patch to another. In many places, heads of Montastraea annularis are extent overgrown by filamentous and brown algae (mainly Lobophora variegata), and scattered thickets of Acropora cervicornis are up to 90% devoid of living tissue. Although the bottom in the central and southern portions of the lagoon is predominantly covered by sand, solitary mounds and scattered coral heads are common. In some places of the central area, large aggregations of single coral heads and small thickets of A. cervicornis occur (at present largely dead), forming diffuse, non-cohesive reef communities. The lagoon is rather open to the E and NE, lacking a well defined sill.
Nearly 25% of the lagoon floor at Albuquerque is covered by patch reefs. The lagoon exhibits two distinctly depth levels, which are easily recognized from the air because of their different blue hues (Plate 9). A first level, with an average almost constant depth of 9 m, takes up the N and E parts of the lagoon and about 65% of its whole area. The second depth level averages about 15 m and takes up the leeward half of the lagoon to the W and S. Both levels are separated by a meandering ribbon reef of '‘Montastraea spp.', which wanders for nearly 6 km, attains 10 to 30 m in width and rises up to -4 m. On the upper lagoon level there are also several nearly circular shaped miniature atolls which break the surface. These reefs are of type Montastraea spp. at their base but show a typical zonation to the 'Diploria-A.palmata' type towards the summit. Anastomosing patch reefs (Montastraea spp.), with the same basic structure as those at Courtown, are found in the northeastern and southeastern parts of this lagoon level. The deeper level is more sparsely covered by reefs. These are mostly low-lying, isolated patch reefs of the ‘Vontastraea spp' type. The depth of the lagoon diminishes leewards to about -5 m or less and the bioturbated sediments of the bottom give way to a gravel-rubble zone, representing the lagoon sill and the transition zone to the western terrace.
According to our observations, lagoonal currents are completely wind-driven and perceptible over the entire water column. Although some differences in direction and
10
intensity were noticed from one location to another, average current velocities of about 2.5 m/min were estimated on the surface at an almost constant wind intensity of 3 m/s in Courtown, and of about 3.5 m/min (wind velocity: 6.5 m/s) in Albuquerque. Considering the rather small size as well as the shallow and open nature of the lagoons, the residence time of lagoonal water masses are thus apparently short, probably not exceeding 24-36 hours.
LEEWARD PERIPHERAL REEFS
Leeward peripheral reefs are poorly developed in both atolls. In Courtown, the absence of such reefs for more than 2 km results in a widely open lagoon to the west. The northernmost portion of the barrier reef becomes interrupted after it curves southwestward semi-enclosing the northern part of the lagoonal terrace. Southwards, detached reef flats rise from 5 to 7 m depth and break the surface in irregular intervals of 50 to 400 m for about 1.4 km, building the northern peripheral reefs. Wave refraction around the north end of the atoll results in colliding surf from both the NE and the NW. Similarly, beginning at the southwestern tip of the atoll, a series of detached reefs and Shoals semi-enclose the southern third of the lagoon. Some of these reefs are partly emergent at low tide and most of them are almost completely coated by calcareous red algae (Porolithon sp., 'coralline algae! unit), resembling the algal ridges characteristic of Pacific atolls. The algal crust usually exhibits numerous bores caused by chitons (Choneplax lata) similar to the systems described elsewhere in the Caribbean (Littler ez al., 1995). Scattered colonies of Diploria strigosa and Millepora encrust the reef flat, whereas on the subvertical to overhanging walls Dendrogyra cylindrus, Agaricia agaricites, branching octocorals (Plexaura sp., Pseudoplexaura sp.) and bunches of Halimeda are common. Wave turbulence, swift currents and the presence of an intricate system of caves in the northern and southernmost peripheral reefs in Courtown create a bizarre and attractive environment. Sand and rubble accumulations over the larger leeward peripheral shoals at Courtown led to the formation of one island (formerly two, see discussion) and a small sand spit. The island serves today as military post for the Colombian navy (Cayo Bolivar).
At Albuquerque, leeward peripheral reefs grow on a shallow, wide sand flat, which represents the lagoon sill. A series of small, low lying reefs enclose the lagoon basin along a wide semicircle between the northwestern tip of the barrier reef and the southern margin of the lagoon. Two navigable channels on the NW and SW breach the flat into the lagoon basin. The peripheral reefs are constituted mainly by large thickets of Acropora palmata, as well as isolated heads of Diploria strigosa and Porites astreoides. Crustose coralline algae (Porolithon sp.), coating large areas of the coral framework, are also major constituents of these reefs. Octocorals and dense beds of brown algae (Dictyota) extensively cover the reef flat bottom. In some places, the scleractinians are dead and overgrown by Dictyota or encrusted by coralline algae. The patch reefs in the southwestern edge, at both sides of the navigable channel, are particularly affected. Here, large thickets of A. palmata were found broken and even overthrown. Large amounts of coral debris were dispersed around the reef flat, including fragments of A. cervicornis, at present an uncommon species in Albuquerque's reefs. This perturbation may have been
11
caused by hurricane ‘Joan’, whose eye passed westward in October 1988 only a few kilometers south of Albuquerque, with winds of more than 200 km/h, which produced very abrasive swells from the south.
LEEWARD TERRACE AND OUTER SLOPE
In the leeward margin of Courtown Atoll there is not a well defined fore-reef terrace. A slope descends in a distance of no more than 200 to 300 m from the shallow reef flat or the lagoon sill to 17 to 20 m, giving rise suddenly to a subvertical sand slope or to a vertical cliff with locally overhanging ledges. The sand-covered slope of the terrace acts as sedimentary ramp, across which reef detritus falls to greater depths. In the northern and central sections, extensive but somewhat diffuse coral carpets cover as much of the bottom, forming elongated low buttresses in an E-W direction and alternating with rather broad sandy channels. Much of the coral (ca. 75% of the bottom) is at present dead and overgrown by fleshy brown algae (Lobophora, Dictyota), whereas living scleractinians cover no more than 10% of the bottom. Although in the southern half of the terrace coral carpets are scantier and have a patchy distribution, they are better developed and form a distinct hardground on the sandy slope, showing a coverage of nearly 70% of living tissue (‘scattered corals’ unit).
At the outer edge of the terrace, the angle of the sandy slope increases to nearly 45°, whereas the reef slope drastically changes to a near vertical wall at about -15 m. Species richness and abundance of scleractinians are very high on the outer margin of the terrace, where massive Montastraea annularis, M. franksi, M. cavernosa and Colpophyllia natans form especially in the southern part, large dome-like structures rising up to 3 m above the bottom (‘mixed corals’ unit). Between these structures usually run ‘sand rivers', which continue as sand falls on overhanging locations along the drop-off. Apart from scattered, small plate-like agariciids such as Agaricia undata, the vertical cliff is mostly devoid of corals and the only organisms attached to the rather smooth substratum are large tube-like and ramose sponges (Agelas conifera, Aplysina spp., Totrochota birotulata), antipatharians and clumps of Halimeda. At the southern locality visited, the cliff remains vertical to about -45 m, where a slanting sand-covered step, about 40 m wide, lines the outer slope of the atoll shelf. In this area, the sand-covered bench deepens at an angle of nearly 30° to about -55 m, where a steep slope continues to greater depths. The loose sand on the slope is composed of Halimeda with accessory shell and coral grains. Large plate-like corals (probably Agaricia and Montastraea) and antipatharians could be observed from above growing along the outer margin of the sandy slope. At another locality, situated in the central section, the drop-off is subvertical to -28 m and is mostly covered by plate-like scleractinians (Agaricia, Montastraea), the sand step is much wider and dips at a lower angle. It seems probable that such a sandy step does occur along the entire leeward margin of the atoll, although the indicative lighter photo-pattern is not always visible on the aerial photographs, possibly due to its variable slope angle and width.
In contrasting to Courtown, the leeward fore-reef terrace at Albuquerque is broader, extending for 1 to 1.6 km, and reaching depths greater than 30 m. It is an
12
extensive, gently dipping platform, descending at a low angle (4 to 7 degrees) to about 15 m and then gradually steeper to nearly 40 m, where the subvertical drop-off of the outer slope begins. The bottom in the upper portions of the terrace is mostly covered by ripple marked sand and rubble, although the calcareous hardground appears at certain locations as elongated buttresses, about 1.5 m high, being thus sparcely overgrown by brown algae (Stypopodium, Dictyota) and branching octocorals. Scleractinians are very scarce to depths of about 12 to 15 m (living coral coverage: 5-20%), but their abundance and species richness increase gradually with a simultaneous increase of the slope angle. At the two localities visited, the outer margin of the terrace is marked respectively at -18 and -27 m by a subvertical escarpment, densely covered by plate-like and pagoda-like scleractinians (Agaricia spp., Montastraea franksi), sponges and antipatharians, which descend to nearly -35 m and give way to the accustomed sand step. Such a sand-covered bench or step at 40-45 m depth was recorded on bathymetric profiles at other places of the leeward outer margin of Albuquerque (Fig. 8), and can be distinguished on aerial photographs as a lighter narrow band, outlining almost the entire outer slope around the atoll shelf.
DISCUSSION
Rather than by its origin (e.g. Darwin, 1842), an atoll is defined by its geomorphic features (Milliman, 1967, 1973; Geister, 1983). Hence, Albuquerque and Courtown may be called atolls. When Milliman (1969) first described the gross morphology and environmental features of the southwestern Caribbean atolls, he was impressed by their close climatologic, oceanographic and geologic resemblance to many Pacific atolls: surrounded by deep water, little seasonal change, appreciable windward fetch, and a Millepora-Palythoa zone that emerges at low tide, resembling somehow the leeward portions of the algal ridge found in Pacific reefs. Besides this, the atolls belonging to the archipelago of San Andrés and Providencia are supposedly the only ones in the Caribbean atolls with a volcanic basement (cf. Milliman and Supko, 1968; Geister, 1992).
The atolls of Albuquerque and Courtown share with nearby San Andrés Island and other reef areas of the archipelago, the geological foundations upon which they rest and a similar set of environmental conditions. Other Atlantic atolls, such as the ones found off Belize and the Yucatan Peninsula (Lighthouse Reef, Glover's Reef, Turneffe Islands, Chinchorro Bank, see Stoddart, 1962; James and Ginsburg, 1979; Jordan and Martin, 1987), in the Gulf of Mexico (Alacran Reef, Kornicker and Boyd., 1962) and the Bahamas (Hogsty Reef, Milliman, 1967) show indeed some analogies with Albuquerque and Courtown in their basic morphology, but they have different geological histories.
The presence of an extensive windward fore-reef terrace in Albuquerque and Courtown is a characteristic common to most Caribbean atolls. As in the Belizean atolls, the outer margin of the fore-reef terrace is defined by a sudden change of the slope angle at about -20 to -25 m, where the nearly vertical cliff of the outer slope begins. The fore- reef terrace or seaward bank is likely one of the essential morphological differences between Caribbean and Pacific atolls. In the latter, the exposed reef margin margin is the
site of most active coral growth, leading to the development of characteristic shelf-edge reefs.(cf. Wiens, 1962). The existence of a sandy step or bench at -35 to -45 m, that outlines the outer slope of the atoll shelf, is also a common feature of the Belizean atolls (cf. James and Ginsburg, 1979). This step, called by some authors the '-40 m Terrace’, is a widespread characteristic of Caribbean reefs. It occurs also in the Bahamas (Zankl and Schroeder, 1972), Jamaica (Goreau and Land, 1974), Curagao (Focke, 1978), San Andrés (Geister, 1975), Providencia (Geister, 1992) and other Caribbean islands.
The present morphology of the outer margin in Caribbean reefs has been interpreted in relation to the fluctuations of sea level in the last 80,000 years. As did James and Ginsburg (1979) for the Belizean reefs, and Geister (1975, 1992) for the fore-reef terraces of San Andrés and Providencia, respectively, we may assume that the outer margin of Albuquerque and Courtown, indicated by the '-20 m Terrace’, corresponds to a truncation of the former marginal reef area that occurred before the last interglacial (Sangamon, about 125,000-80,000 years b.p.). In the period between Sangamon and 10,000 years b.p. sea level was not constantly low (about -120 m under present sea level). At least three high stands of sea level took place during that time, reaching to nearly -25 to -40 m below present sea level (Bowen, 1988). The coincidence of the sandy bench in present morphology at -35 to -40 m around both atolls, as well as the occurrence of a deep intertidal notch at this level on the vertical cliff (at least at the visited locality in Courtown), led us to explain this topography as a truncation of the emerging shelf margin during a Pleistocene sea-level stand at about -40 m that may be regarded primarily as an erosional feature. Unlike other Caribbean reefs, such as those off Belize (James and Ginsburg, 1979) and Jamaica (Goreau and Land, 1974), where this feature is now subdued by overgrowing modern facies, no significant accretion to the reef margin seems to have occurred during the Holocene rise of the sea level either in Albuquerque or in Courtown, or in the reefs surrounding San Andrés (cf. Geister, 1975) and Providencia (cf. Geister, 1992), where a truncation of the outer margin at -35 to -40 m and an intertidal notch are very distinctive.
The uppermost part of the reef front in both atolls shows a well developed spur- and-groove system, similar to other reef complexes in the western Caribbean, such as those off Belize, Yucatan, San Andrés and Providencia (cf. Stoddart, 1962; James and Ginsburg, 1979; Jordan and Martin, 1987; Geister, 1975, 1992). In some localities, such as the northeastern barrier of Courtown, where the relief between the spurs and grooves often attains more than 3 m, and the grooves penetrate deeply into the reef flat, they are apparently cut into Pleistocene rock, indicating an essentially erosional origin of this system. It acts as an effective baffle for the immense energy expended by incoming surf (Roberts, 1974; Geister, 1982). In other parts of the Caribbean, where the effective windward fetch and the energy of the incoming surf are not as great, the spur-and-groove system may owe much of its relief to differential rates of scleractinian growth (cf. Goreau, 1959). The presence of an extensive lagoonal terrace between the reef crest and the lagoon basin on the windward side, as well as its abrupt transition into the lagoon in the form of a 'sand-cliff, are also characteristics common to most oceanic reefs with a considerable windward fetch, due to active movement of debris associated with the
14
prevailing northeasterly winds and waves. The presence of seagrasses on the lagoonal terrace is conditioned by shelter created on the leeward side of the cays and islands.
The depth of the lagoon floor in Albuquerque and Courtown is not very different from most Caribbean atolls, whose average lagoon depth ranges between 10 and 15 m (Milliman, 1973). A singular feature is the existence of two well defined lagoon depth- levels in Albuquerque. It is likely a consequence of the barrier effect of the 'Montastraea spp.'-ribbon reef, which restricts leeward transport of bottom sediments to fill the lagoon basin evenly. The occurrence of anastomosing and ribbon patch reefs covering unusually large portions of the lagoon floor seems to be a rather common feature of oceanic reef complexes in the Caribbean, such as Serrana Bank (Milliman 1969) and Alacran Reef (Kornicker and Boyd, 1962). The NE portion of the lagoon bottom in Providencia Island exhibits also several coalescing patch reefs (J.M. Diaz, J.A. Sanchez and S. Zea, pers. obs., Sept. 1994). It seems likely that the greatest development of anastomosing patch reefs is attained always on the windward side of the lagoon.
Contrasting with Pacific atolls, the absence or poor development of leeward peripheral reefs is a characteristic common to most Caribbean atolls (Milliman, 1973). However, Chinchorro Bank and some of the Belizean atolls exhibit a discontinuous leeward reef crest which almost completely encloses the lagoon. Residence time of lagoonal water may hence undergo a notable prolongation in these atolls. Coincidentally, the abundance and development of lagoonal patch reefs in these atolls is apparently much reduced in comparison to Albuquerque and Courtown (cf. Stoddart, 1962; Jordan and Martin, 1987). It seems probable that the residence time of lagoonal water plays an important part in the luxuriance and relative bottom coverage of patch reefs in Caribbean atolls along with other physical factors, such as substrate availability and depth.
As stated by Milliman (1969), it seems probable that leeward peripheral reefs in Albuquerque have originated from coalescing patch reefs. On the aerial photographs, most peripheral reefs and the rubble zones surrounding them are arranged in a meander- like fashion. Former ribbon and cellular reefs on the leeward lagoon margin have apparently been damaged again and again by storms and hurricanes, leaving only the most resistant frameworks of Acropora palmata and coralline algae, which built such peripheral reefs. In Courtown, leeward peripheral reefs have developed only in the NW and SW parts of the atoll, where the windward barrier reef bends southwestward at its northern end and northwestward at its southern end. They are heavily exposed to colliding surf from both the NW and the NE (or SW and NW) and are formed mainly by a framework of coralline algae (Porolithon sp., Titanoderma spp., Lithophyllum sp.) comparable to that of algal ridges. Although algal ridges had been thought characteristic of the Indo- Pacific region until recently (Frost and Weiss, 1975), the southernmost leeward peripheral reef in Courtown, with its emergent crest, represents in fact a true algal ridge, such as those described recently elsewhere in the Caribbean (Glynn, 1973; Adey, 1975; Adey and Burke, 1976). This feature was apparently overlooked by Milliman (1969), who refers to it as a ‘small rocky spit, composed of massive coral debris’. Although not so well developed, similar structures also have been recognized adjoining the NW end of the
barrier reef in nearby San Andrés by Geister (1975). Interesting discussions concerning the existence and development of Caribbean algal ridges are found in Adey and Burke (1976), Stoddart (1977) and Littler et al (1995).
At present, two cays exist in Albuquerque, both lying on the lagoonal terrace. Their position, size, and shape have not changed significantly in the last 25 years, except that North Cay has currently a more rounded shape than in the map of Milliman (1969) and on the aerial photograph taken 1971. On the 1984-photograph it exhibits approximately the current shape and size. The western and southern shores of this cay have been dammed with piles of Strombus shells by the marines of the Colombian navy. On the other hand, islands and cays in Courtown experienced remarkable changes in number, size and shape since that time, and it seems likely that further changes are even now taking place. Milliman (1969) mentioned four small cays, a sand spit and a rocky spit. Sand Cay and East Cay lay close together on the lagoonal terrace and have currently coalesced in an arrow-shaped island (about 800 m long), which seems to grow further to the NW by accretion of sand and rubble (Plates 5 and 6). Of the formerly two cays sitting on leeward peripheral reefs, Middle Cay was the only one visited by Milliman, who noticed the presence of Yournefortia and Scaevola bushes and even some native fishermen living on it. This cay might have disappeared between 1966 and 1971, since no trace of it can be seen in the aerial photographs taken in August 1971. On the contrary, West Cay (currently called Cayo Bolivar and serving as military post) and the sand spit have experienced little change. The shallow bottom (1-2 m depth), where Middle Cay lies, is currently covered with rubble and coral debris. It is not known if the.disappearance of the cay was a slow erosional process that took place within five years or a rapid loss produced by a forceful weather event. The latter seems less probable, since the only hurricane recorded between 1966 and 1971 affecting this area, 'Irene' in 1971, had only trivial consequences in nearby San Andres (IGAC, 1986).
Although a detailed checklist of scleractinians from Albuquerque and Courtown has not yet been published, our survey indicates no noteworthy differences in species composition and structure between the reefs of both atolls. It can be stated however that the reefs in both atolls show a highly diverse fauna of about 40 species, not significantly diverging from those known from neighbouring San Andrés and Providencia, where 44 and 43 species have been respectively recorded (Geister, 1975; 1992). The distribution pattern of reef framework associations in both atolls, at least in shallow-water to about 15 m, is highly controlled by wave-energy and corresponds well to the ‘wave zones’ model postulated by Geister (1977). With the exception of a ‘Porites zone’, each of the most important reef framework associations recognized in the Caribbean Sea were found in Albuquerque and Courtown. Only the names employed by Geister (1977) for his 'Melobesiae-zone' has been modified to designate the 'Coralline algae! unit (including the algal ridges) in our maps.
In spite of a generally similar distribution pattern of reef framework associations, there are some qualitative differences between Albuquerque and Courtown. Neither 'A.cervicornis' reefs nor a '‘coralline algae’ (or algal ridges) unit occur in Albuquerque.
Due to the interrupted windward reef crest in Courtown (ie.,discontinuity of the ‘Millepora-Palythoa” unit), medium-energy waves can penetrate in some places into respective rear reef and lagoonal areas, leading to a better development of 'Diploria- A.palmata" and 'A.cervicornis' reefs in this atoll. Contrary to Albuquerque, Courtown lacks a gently dipping and extensive leeward terrace, which represents a highly abrasive environment during storms and hurricans coming usually from the SW. This is seemingly the main reason for a much reduced ‘scattered corals' unit and the lack of a ‘bare calcareous hard bottom’ unit there.
Although detailed information about the current conditions of reef health in these atolls will be presented and discussed elsewhere, some preliminary statements can be made here. In the description of habitats presented above, we mentioned several signs that are indicative of some degradation of the coral reef environment in both atolls. Proliferation of algae overgrowing scleractinian colonies, low proportions of living coral cover at several sites, abundance of heaps of skeletons of recently dead scleractinians (i.e. Acropora spp.), as well as a noticeable depletion of commercial organisms, such as queen conchs (Strombus gigas), lobsters (Panilurus spp.), snappers (Lutjanidae), groupers (Serranidae) and turtles, are the most evident signs of degradation. Although no human development exists in the atolls, they have been visited for many years by San Andrean and Providencian natives for fish and turtles. In contrast to the condition in 1944, when Fowler reported abundant fish and lobsters, by 1966 the populations of these resources seemed to be low at Courtown, possibly a result of the increasingly fishing pressure caused by overexploitation of Strombus, lobster and fish stocks at San Andrés (Wells, 1988). The health condition of reefs in these atolls is even at several sites not significantly different from those around the densely populated San Andrés island (cf. Diaz ez. al., 1995), indicating that, besides local human factors (i.e. sand mining, siltation, pollution) and local natural agents (i.e. hurricanes), recent coral mortality is highly associated rather to a generalized phenomenon of coral decline occurring in the Caribbean from beginning of the 1980's (Hallock ez al., 1993; Ginsburg, 1994). Overfishing has also been recently recognized as an indirect agent of coral mortality (Hughes, 1994).
ACKNOWLEDGMENTS
The authors express their gratitude to Luz S. Mejia, Guillermo Diaz (INVEMAR, Santa Marta) and the crew of the R/V 'Ancon' for assistance in the field surveys. We extend our appretiation to Dr. Jorn Geister (University of Bern, Switzerland) for his helpful discussions and encouragement to make possible the flight over the atolls, as well as friendly loan of the photos included in Plates 2, 4, 5 and 9. We thank Martha Prada for her friendly hospitality at San Andrés. For their help in map digitizing and improvement of computer drawings we are indebted to the students P. Sierra, J.A. Pulido and N. Ardila. This study has been funded by the Instituto Colombiano de Ciencia y Tecnologia (COLCIENCIAS, Grant No. 2105-09-023-93), the Instituto de Investigaciones Marinas y Costeras (INVEMAR, Santa Marta) and the Universidad Nacional de Colombia.
REFERENCES
ADEY, W.H., 1975. The algal ridges and coral reefs of St. Croix: their structure and Holocene development. Atoll Res. Bull., 187: 1-67.
ADEY, W.H. and R. BURKE, 1976. Holocene bioherms (algal ridges and bank-barrier reefs) of the eastern Caribbean. Bull. Geol. Soc. Amer., 87: 95-109.
BARRIGA, E., J. HERNANDEZ, I. JARAMILLO, R. JARAMILLO, L.E. MORA, P. PINTO and P. RUIZ., 1969. La Isla de San Andrés: contribuciones al conocimiento de su ecologia, flora, fauna y pesca. Univ. Nacional de Colombia, Direccion de Divulgacion Cultural, Bogota, 152 p., 4 + 31 figs, 6 tab.
BOND, J. and R. DOSCHAUENSEE, 1944. Results of the Fifth George Vanderbilt Expedition (1941): The birds. Monog. Acad. Nat. Sci. Philad., 6: 7-57.
BONET, F., 1967. Biogeologia subsuperficial del arrecife Alacranes, Yucatan. Bol. Inst. Geol. Univ. Nac. Auton. Mexico, 80: 1-192.
BOWEN, D.Q., 1988. Quaternary Geology: a Stratigraphy Framework for Multidisciplinary Work. Pergamon Press, Oxford, 234 pp.
CHIRIVI, H., 1988. Fauna tetrapoda y algunos aspectos ecologicos de los cayos del archipiélago de San Andrés y Providencia. Trianea, 2: 277-337.
COVENTRY, G.A., 1944. Results of the Fifth George Vanderbilt Expedition (1941). The Crustacea. Monog. Acad. Nat. Sci. Philadelphia, 6: 531-544.
DARWIN, C., 1842. The Structure and Distribution of Coral Reefs. (Reprinted 1962 by University of California Press, 214 pp.).
DIAZ, J.M., J. GARZON-FERREIRA and S. ZEA, 1995. Los arrecifes coralinos de la isla de San Andrés, Caribe colombiano: estado actual y perspectivas para su conservacion. Acad. Col. Cienc. Exact. Fis. Nat., Col. Jorge Alvarez Lleras, No. Tem Ol:
DUYL, L.F. Van, 1985. Atlas of the Living Reefs of Curacao and Bonaire (Netherlands Antilles). Uitg. Natuurwetenshappelijke Studiekring Suriname & Ned. Antilles, Utrecht, 117, 37pp. + 41 maps.
FOCKE, J.W., 1978. Holocene development of coral fringing reefs, leeward off Curagao and Bonaire (Netherlands Antilles). Mar. Geol., 28: M31-M41.
FOWLER, H.W., 1944. Results of the Fifth George Vanderbilt Expedition (1941): The fishes. Monog. Acad. Nat. Sci. Philad., 6: 57-529.
FROST, S.H. and M.P. WEISS, 1975. Caribbean reefs reviewed. Geotimes, 20(2): 2-23.
GEISTER, J., 1973. Los arrecifes de la Isla de San Andrés (Mar Caribe, Colombia). Mitt. Inst. Colombo-Aleman Invest. Cient., 7: 211-228
GEISTER, J., 1975. Riffbau und geologische Entwicklungsgeschichte der Insel San Andrés (westliches Karibisches meer, Kolumbien). Stuttgarter Beitr. Naturk., Ser.B (Geol. & Palaont.), 15: 1-203.
GEISTER, J., 1977. The influence of wave exposure on the ecological zonation of Caribbean coral reefs. Proc. Third Int. Coral reef Symp. (Miami), 1: 23-29.
GEISTER, J., 1983. Holozine westindische Korallenriffe: Geomorphologie, Okologie und Fazies. Facies, 9: 173-284.
GEISTER, J., 1992. Modern reef development and Cenozoic evolution of an oceanic island/reef complex: Isla de Providencia (Western Caribbean Sea). Facies, 27: 1- 70.
GINSBURG, R. (compiler), 1994. Proceedings of the Colloquium on Global Aspects of Coral Reefs: Health, Hazards and History. Rosenstiel School of Marine and Atmospheric Sciences, University of Miami, Miami, 420 pp.
GISCHLER, E., 1994. Sedimentation on three Caribbean atolls: Glover’s Reef, Lighthouse Reef and Turneffe Island, Belize. Facies, 31: 243-254.
GLYNN, P.W., 1973. Aspects of the ecology of coral reefs in the western Atlantic region. In O.A. Jones and R. Endean (Ed.): Biology and Ecology of Coral Reefs, vol.2 (Biology 1), Academic Press, New York-London, pp. 271-324.
GONZALEZ, E., 1988. Oceanografia fisica descriptiva del archipiélago de San Andrés y Providencia con base en el analisis de los cruceros Océano IV a IX. Bol Cient. CIOH, 7: 73-100.
GOREAU, T.F., 1959. The ecology of Jamaican coral reefs. 1. Species composition and zonation. Ecology, 40: 67-89.
GOREAU, T.F. and L.S. LAND, 1974. Fore-reef morphology and depositional processes, north Jamaica. Spec. Publ. Soc. Econ. Paleontol. Miner., Tulsa, 18: 77-89.
HALLOCK, P., F.E. MULLER-KARGER and J. HALLAS, 1993. Coral reef decline. Nat. Geogr. Res. & Explor., 9(3): 358-378.
19
HUGHES, T.P., 1994. Catastrophes, phase shifts, and large-scale degradation of a Caribbean coral reef. Science, 265: 1547-1551.
IGAC, 1986. San Andrés y Providencia, aspectos geograficos. Instituto Geografico ‘Agustin Codazzi', Bogota, 156 pp.
JAMES, N.P. and R.N. GINSBURG, 1979. The Seaward Margin of Belize Barrier and Atoll Reefs. Special Publ. No.3, Int. Assoc. Sedimentologists. Blackwell Scientific Publications, Oxford, 191 pp.
JORDAN, E. and E. MARTIN, 1987. Chinchorro: morphology and composition of a Caribbean atoll. Atoll Res. Bull., 310: 1-20, 9 figs.
KOKURKO, M.J., 1977. Preliminary survey of modern marine environments of San Andrés Island, Colombia. Tulane Stud. Geol. Paleont., 13(3): 111-134.
KORNICKER, L.S. and D.W. BOYD, 1962. Shallow-water geology and environments of Alacran Reef complex, Campeche Bank, Mexico. Bull. Amer. Assoc. Petrol. Geol., 64: 640-673.
LITTLER, M.M., D.S. LITTLER and P.R. TAYLOR, 1995. Selective herbivore increases biomass of its prey: a chiton-coralline reef-building association. Ecology, 76(5): 1666-1681.
MARQUEZ, G., 1987. Las Islas de Providencia y Santa Catalina. Ecologia regional. Fondo FEN Colombia-Univ. Nacional de Colombia, Bogota, 110 pp.
McBIRNEY, A.R. and H. WILLIAMS, 1965. Volcanic history of Nicaragua. Publ. Geol. Sci. Univ. California, 55: 1-73.
MILLIMAN, J.D., 1967. The geomorphology and history of Hogsty Reef, a Bahamian atoll. Bull. Mar. Sci., 17(3): 519-543.
MILLIMAN, J.D., 1969. Four southwestern Caribbean atolls: Courtown Cays, Albuquerque Cays, Roncador Bank and Serrana Bank. Atoll. Res. Bull., 129: 1- 22, 10 Figs.
MILLIMAN, J.D., 1973. Caribbean coral reefs. Jn O.A. Jones and R. Endean (Ed.), Biology and Ecology of Coral Reefs, vol. 1 (Geology). Academic Press, New York-London, pp. 1-50.
MILLIMAN, J.D and P.R. SUPKO, 1968. On the geology of San Andrés Island, western Caribbean. Geol. Mijn., 47: 102-105.
20
PARSONS, J.J., 1956. San Andrés and Providencia. English speaking islands in the western Caribbean. Publ. Geogr., Univ. California, 12: 1-84.
ROBERTS, H.H., 1974. Variability of reefs in regard to wave power around an island. Procc. 2th. Internat. Coral Reef Symp., 2: 497-512.
RUTZLER, K and I.G. MACINTYRE, 1982. (Ed.): The Atlantic barrier reef ecosystem at Carrie Bow Cay, Belize. I, Structure and communities. Smith. Contrib. Mar. Sci., 12: 1-539.
STODDART, D.R., 1962. Three Caribbean atolls: Turneffe Islands, Lighthouse Reef and Glover's Reef, British Honduras. Atoll Res. Bull., 87: 1-151.
STODDART, D.R., 1977. Structure and ecology of Caribbean coral reefs. FAO Fish. Rep., 200: 427-448.
TELLEZ, C., G. MARQUEZ and F. CASTILLO, 1988. Fitoplancton y ecologia en el archipiélago de San Andrés y Providencia: Crucero Océano VI en el Caribe colombiano. Bol. Cient. CIOH, 8: 3-26.
WELLS, S.M., 1988. Coral reefs of the world. Vol 1: Atlantic and Eastern Pacific. UNEP/IUCN, Gland, Switzerland, 373 pp.
WEIL, E. and N. KNOWLTON, 1994. A multi-character analysis of the Caribbean coral Montastraea annularis (Ellis & Solander, 1786) and its two sibling species, M. faveolata (Ellis & Solander, 1786) and M. franksi (Gregory, 1895). Bull. Mar. Sei, SS) rls IES»
WIENS, H.J., 1962. Atoll environment and ecology. Yale Univ. Press, New haven and London, 532 pp.
ZANKL, H. and J.H. SCHROEDER, 1972. Interaction of genetic processes in Holocene reefs off North Eleuthera Island, Bahamas. Geol. Rundschau, 61(2): 520-541.
21
Table 1. Marine habitats (Map units) of Albuquerque and Courtown atolls with their corresponding geomorphological units and absolute and relative area.
Habitat unit
‘pioturbated
sand’
‘rubble with algae’
‘sand and
rubble’
‘A.cervicornis’
‘scattered corals’
‘rubble on hard bottom’
‘Gorgonaceans on hard bottom’
‘mixed corals’
‘Diploria-A. palmata’
Geomorphol.
units
(depth)
Lagoon (6-18m)
Lagoonal terrace (1-2m)
All zones
Lagoon (3-5m)
Leeward terrace (15-30m)
Laggonal terrace (1-2m)
Fore-reef terrace (15-30m)
Fore-reef and Leeward terraces (25-37m)
Lagoon and
Lagoonal terrace (0.5-3 m)
Brief description
Calcareous sand (Halimeda, coral, shells) with many burrows and mounds (Arenicola, Callianasa).
Coral debris with rodoliths formed by coralline algae mostly overgrown by brown algae.
Bare coarse to medium sand with scattered coral rubble and algal rodoliths.
Patch reefs dominated by thickets of Acropora cervicornis, scattered coral heads (Siderastrea, Montastraea) and plexaurid octocorals.
Scattered massive and hemispheric scleractinians (Siderastrea, Colpophyliia, Diplona, Montastraea), gorgonaceans and fleshy brown algae.
Rather barren bottom with coral debris sometimes encrusted with coralline algae.
Rather flat bottom with luxuriant growth of
gorgonaceans (Pseudopterogorgia spp., Pterogorgia, Plexaurella, Eunicea, Munceopsis, etc), scattered massive
scleractinians, many fleshy algae and large sponges.
Diverse scleractinians (Montastraea franksi, Diploria, Colpophyllia, Pontes, Mycetophyliia, etc.) gorgonaceans, sponges and Halimeda.
Moderate wave exposed reefs dominated in the shallow zones by thickets of Acropora palmata, massive Diplona stngosa and encrusting Porites astreoides.
Area (Ha) Courtown Albuquerque 631.8 841.2 (12.8%) (11.7%)
127 122.4 (2.5) (1.7) 1270 237 (25.5) (33.2) 28.2 - (0.6) 226.6 357.8 (4.6) (5)
' 361.3 260.2 (7.3) (3.6) 910 327.4 (10.3) (4.5) 134.1 180.2 (2.7) (2.5) ]
29.6 107.6 (0.6) (1.5)
22
Table 1. continued.
ae iad eos Brief description Area (Ha) uni units (depth) Courtown Albuquerque ‘ : Patchy seagrass meadows with Thalassia, sealgiiass goonal Halodule and/or Syringodium growing on 35.5 3.8 aay sandy bottom. (0.7) (<0.1) -om ‘ , Emerging sand and rubble accumulations, land (cays) Fee mostly vegetated with shrubs (Scaevola, 9.2 78 eared Tournefortia), coconut palms or Ficus trees. (0.2) (0.1) eewarl peripheral reefs ‘Millepora- Barrier reef Highly Wave-exposed reefs dominated by 195 195.4 ; 0-3 m) Millepora complanata and Palythoa sp., (4) (2.7 Palythoa ( mostly accompained by crustose coralline 1) algae. ‘Millepora- Barrier reef ear surf zone of the barrier reef. Millepora 107 59.2 pila 0-3 m) complanata, Portes ponites, P.astreoides (2.2) (0.8) P.porites ( and Diploria strigosa. ; : ‘Montastraea Lagoon Ribbon and anastomosing patch reefs 325 1 304.1 ‘ 5-15 m) dominated by massive Montastraea (6.5) (4.2) Spp. ( annulans and M. faveolata, brown algae : : (Lobophora-Dictyota) and some octocorals. bare Fore-reet's War otioctigiemanidalytitcl) noggiiS.6 trafi224 calcareous hard terrace algae, scattered sea fans, brown algae. (8.4) (17.2) bottom (3-15m) Heavily excavated by sheet-like sponges (Cliona spp.). ‘outer slope’ Fore-reef Vertical to subvertical drop-off of the atoll 522.4 806.2 shelf. Sedimentary ramp or subvertical 10.4 11.2 and calcareous wall (covered or not with plate-like (10.4) (11.2) Leeward scleractinians, sponges and antipatharians). terraces (>35m) ‘ : ’ Wave-exposed reefs, almost completely z SEU I US cele Leeward covered by encrusting algae (Porolithon) - - Peete building algal-ridge-like emerging crests. (0.5) reets (0-5m) eee 4953.3 7172.5
a
23
BB RNK EEA ACS
@) es
CE HOS A ASKS —S YA Y SP
YS
LE PAWN ae
Lagoon with patch reefs Windward barrier reef ca Cays
oa Fore-reef terrace and outer slope
KA Leeward terrace and outer slope
¢| Lagoonal terrace
Leeward peripherical reefs
Sample sites
Figure 2, Geomorphological units and visited stations at Albuquerque Atoll. Straight lines mark the location of the schematic profiles of Fig. 4.
24
81° 28'W
12° 28’N
Lagoon with patch reefs | Ninaward barrier reef ow Cays
[+ Fore reef terrace and outer slope
RY Leeward terrace and outer slope
0 500 1000 1500 m
Figure 3. Geomorphological units and visited stations at Courtown Atoll. Straight lines mark the location of schematic profiles of Fig. 5.
A. OUTER LEEWARD TERRACE PR LAGOON LAGOONAL TERRACE] BR | fendez Ores 0 1 2 3
T4 15 16 7 km PEN Fk Ce eRe od pe ta eer a pr ere erence REE level depth = MO7TLLG aed SYP? seattered corals — mixed OTa 60 m L— B. OUTER LEEWARD TERRACE lP.r| LAGOON Sty’ | LAGOONAL TERRACE B.R| F.R.TERRACE OUTER (0) 1 2 3 4 5 6 7 8 9 10km se siti on ear ee ee depth bb with a ip Montastraea spp. gorgonian on hard bottom re < QO mixed corals 60 mL_—
Figure 4. West-East schematic profiles (straight lines in Fig. 2), showing the different
geomorphological and habitat units of Albuquerque Atoll. P.R- peripheral reefs, B.R- barrier reef, F.R.- fore-reef.
25
26
§ 3
> © [ste |
ee | LAGOON LAGOONAL Terrace] 8.«| Seb 3)
i 2 4km seq depth Millepora—Palythoa Le coralline algae rubble with algae i gorgonians on.hard bottom ool of Montastraea spp. eee scattered corals mixed corals 50 m B. Lev [eri ¢ said GOON MMA Lee
0 Fr 19 aie 4 5 km
On == eye Oe SO ee ie Nim oe depth
Millepora—Palythoa
= cervicornis
= algae or grass meadows
= coralline algae gorgonians on hard bottom il Montastraea spp.
= scattered corals
mixed corals 50) tm ==
Figure 5. West-East schematic profiles (straight lines in Fig. 3), showing the different
geomorphological and habitat units of Courtown Atoll. P.R- peripheral reefs, B.R- barrier reef.
77)
= Outer slope Z ward bottom with rubble Fra Scattered corals Sea grasses E24 Rubble with algae
—] Gorgonaceans on hard bottom Ej Diploria—A.palmata 12°07’N = lap) w 2
2] Bare calcareous hard bottom Montastraea spp. 4 Millepora—Palythoa
| Bioturbated sediments ai Millepora—P.porites EC] Sand and rubble
Figure 6. Distribution of bottom habitats and reef types at Albuquerque Atoll (for brief description of map units see Table 1).
28
Diploria—A.palmata Bioturbated sediments
= Millepora—Palythoa A.cervicornis
fj Gorgonaceans on hard bottom Eel Millepora—P.porites
(2) Bare calcareous hard bottom Scattered corals
[_] Sand and rubble
E2 Montastraea spp. @4 Coralline algae
[] Sea grasses
—] Outer slope
22 Mixed corals
Rubble with algae
Hard bottom with rubble
500 1000 1500 m
0
Figure 7. Distribution of bottom habitats and reef-types at Courtown Atoll (for brief
description of map units see Table 1).
Figure 8. Echosounder bathymetric profile of the leeward terrace and outer slope at Albuquerque atoll. Note the presence of a truncation (sandy bench or step) at about -40 m depth on the outer slope.
Plate 1. The spur-and-groove system of the windward barrier reef. The spurs are overgrown on the top by Millepora complanata and by Porites spp. and crustose coralline algae on the sides, whereas the narrow groove is filled with sand (Courtown, 22 May, 1994).
we
30
Plate 2. Oblique aerial view to the N, showing the buttress-groove system on the central portion of the windward barrier reef at Courtown atoll (Sept. 29, 1994).
Plate 3. Rounded pinnacle (left) and narrow pillar formed by Millepora spp. at the SW section of Courtown atoll, where the barrier reef becomes discontinuous (27 May, 1994).
31
Plate 4. Oblique aerial view to the W of Albuquerque atoll showing the two cays lying close to the leeward margin of the lagoonal terrace (Sept. 29, 1994),
Plate 5. Oblique aerial view to the NW of Courtown atoll. The arrow-shaped island in the center right is East Cay, which currently is connected with Sand Cay by a sand bar. Sand
Cay grows seemingly further to the NW by recent gradual addition of sand (Sept. 29, 1994).
32
Plate 6. East Cay, Courtown atoll, looking SE along the sand bar which at present connects this cay with Sand Cay (21 May, 1994).
Plate 7. East shore of South Cay, Albuquerque atoll. Note the conspicuous band of beachrock parallelling the shore line ( June 6, 1994)
33
Plate 8. Lagoonal patch reef in the upper depth level at Albuquerque atoll, made up mostly by Montastraea annularis and M. faveolata (June 7, 1994).
Plate 9. Oblique aerial view to the SE of Albuquerque atoll. Note the two different hues of the lagoon basin denoting the two depth-levels of the lagoon floor (Sept. 29, 1994).
)
now tga ‘A ots i ‘en iil
) aimbogeal, MM bas ‘svn voruncnnahs yd
avid sal! OA jlote
ie) #8 ie iShabt ut eu inet
ATOLL RESEARCH BULLETIN
NO. 436
CORAL FAUNA OF TAIPING ISLAND (ITU ABA ISLAND) IN THE SPRATLYS
OF THE SOUTH CHINA SEA
BY
CHANG-FENG DAI AND TUNG-YUNG FAN
ISSUED BY NATIONAL MUSEUM OF NATURAL HISTORY SMITHSONIAN INSTITUTION WASHINGTON, D.C., U.S.A. APRIL 1996
115°E Taiwan
¢ Pratas I.
Hainan
Paraceles I. a o
rh) %2¢ (a ‘ 1S)
15°N
Vietnam
SOUTH CHINA SEA
~ cA ~
7
2 : ~
, os XN
Taiping I. \ y Palawan ping fe] :
114°21E
Taiping Island
E
Fig. 1. Locations of the survey sites (A-G) at Taiping Island in the Spratlys of the South China Sea.
CORAL FAUNA OF TAIPING ISLAND (ITU ABA ISLAND) IN THE SPRATLYS OF THE SOUTH CHINA SEA BY
CHANG-FENG DAI AND TUNG-YUNG FAN
ABSTRACT
The coral fauna of the Taiping Island (Itu Aba Island) in the Spratlys of the South China Sea was surveyed on April 19-23, 1994. A total of 163 species of scleractinians in 15 families and 56 genera; 15 species of alcyonaceans in three families and five genera; and six species of gorgonaceans in four families and five genera were recorded. The coral communities of the Taiping Island were dominated by scleractinian corals with high species diversity and coral cover found on the lower reef flat at depths between 1 and 3 m. Alcyonaceans and gorgonaceans are mainly distributed on the reef slopes at depths below 15 m. Wide reef flats and reef terraces exist on the east and west sides of the island indicating that the reef development is better in these areas. Species diversity of coral communities was the highest on the east side and the lowest on the west side of the island. The depauperate coral fauna on the west side is possibly related to the strong SW monsoon during summer and autumn. In comparison with other tropical coral reefs, species diversity and abundance of coral communities of Taiping Island are relatively low. Dead coral skeletons and debris were widely spread on the reefs below 3 m deep and only small colonies were found. These facts indicate that coral communities of Taiping Island may have been heavily damaged by natural catastrophes or artificial destruction during the last decade. The possible destruction forces are typhoon disturbances and sea warming events.
INTRODUCTION
The South China Sea, situated between the Indian and Pacific Oceans, has an historical importance in politics, economics, military affairs and transportation (Gomez, 1994). As the South China Sea is surrounded by continental Asia and many islands, it is generally recognized as the major marginal sea in Asia. Major islands in the South China Sea such as Tungsha Island (Pratas Island), Xisha Islands (Paracel Shoals) and Nansha Islands (Spratly Islands) are reef islands. Most reef islands are atolls or emergent islands, which are mainly composed of coral debris and sand. The emergent islands constitute only a small portion of the reefs; the major parts are underwater reefs, shoals and banks.
Institute of Oceanography, National Taiwan University, P.O. Box 23-13, Taipei, Taiwan, R.0.C.
Manuscript received 8 September 1995; revised 8 March 1996
The Spratly Islands, consisting of some 600 coral reefs and associated structures scattered across an area north of Sabah and southern Palawan stretching for more than 500 km, are a group of atolls, islets, and sea mounts in the South China Sea. The structures which protrude above the sea surface at high tide include at least 26 islands and seven exposed rocks (McManus, 1992). Taiping Island, or Itu Aba Island, is one of the major islands in the Spratly Islands.
The Indo-Pacific region, which includes the Spratlys, is characterized by a high diversity of marine organisms. Among reef building corals, for example, the region in which the Spratlys reside includes at least 70 genera (Veron, 1986, 1993). Inthe coral reef ecosystem alone, more than 400 species of corals (Veron and Hodgson, 1989), 1500 species of reef fishes and 200 species of algae are found in this area (McManus, 1994). The exact number of all marine species in the South China Sea is difficult to estimate given the inadequate state of taxonomy, but the total number of species to be found at all depths in the Spratlys certainly ranges to the tens of thousands (McManus, 1992).
The marine ecosystem of the South China Sea can be assumed to be dependent on the Spratlys, at varying levels, for sources of larvae of renewable resources. Due to prevailing monsoonal currents, the Spratly reefs may serve as sources of larvae that could recruit to the disturbed coral reefs in the South China Sea (McManus, 1994). The semi- enclosed nature of the South China Sea and hydrodynamic patterns prevailing in the area could explain this linkage of coastal ecosystems in terms of nutrient level and fauna. It is very likely that the Spratly Islands and similar groups of uninhabited reefs serve as a mechanism for stabilizing the supply of young fish and invertebrates to these areas. This becomes increasingly important wherein coastal populations of adult fish decline, as appears to be the case in many coastal reefs of the Philippines and elsewhere. The dispersal of larvae from the Spratlys possibly contribute to the coral reef fishery in the region. The contribution of coral reef fishery to the national fish production of countries bordering the South China Sea varies between 5-60% (McManus, 1994). Thus, the Spratlys could be considered as a “saving bank” where commercially important fish and invertebrates are saved from overharvest and supply a constant flow of larvae to areas of depletion.
Coral reefs are widely distributed in shallow water areas in the South China Sea. The high spatial heterogeneity and productivity of coral reefs provide not only various habitats for marine organisms but also feeding and nursery grounds for fishery resources such as fish, shells, crustaceans and cephalopods. Flourishing coral reefs also constitute beautiful underwater scenery that are valuable resources for the development of touristic industry. As corals play a key role in marine ecosystems of the South China Sea, a better understanding of the coral fauna in this area is necessary for conservation and management of the marine resources in the future.
Several scientific expeditions in the South China Sea over the last 50 years have provided oceanographic information and taxonomic listing of marine organisms, mainly fishes. Although corals are widely distributed in the South China Sea, the coral fauna of
3
this area is poorly documented because of its remoteness and difficulty of access. Bassett-Smith (1890) first described corals from Tizard Bank. Ma (1937) studied the growth rates of scleractinian corals from Tungsha Island (Pratas Island). In recent years, a few expeditions have been conducted to investigate the fauna and flora of the South China Sea (Yang et al., 1975; Zou, 1978a, b; Fang et al., 1990). These studies have provided valuable information for a preliminary understanding of the coral fauna of this area. However, in comparison with the vast area of the South China Sea, these studies have only covered a very restricted area. Studies on the coral fauna in other areas are thus necessary.
We sought to provide baseline information for resource conservation and exploitation of Taiping Island (or Itu Aba Island). The objectives of this work were to survey and to describe the distribution of coral reefs and reef topography, to provide an inventory of coral species and their estimated relative abundance, and to identify special coral biotopes.
STUDY SITE AND METHOD
Seven sites around Taiping Island (Fig. 1) were surveyed on April 19-23, 1994. Taiping Island (10°23'N, 114°22'E), located on the northwest side of Tizard Bank, is one of the major islands on the west side of the Spratly Islands (Nan-sha Islands). The island, with an area of 0.49 km’, is about 1300 m long and 350 m wide (Fig. 3). The climate is tropical oceanic. The average water temperature is about 28-29°C. The island is influenced by seasonal monsoons. The northeast monsoon blows from October to March, the southwest monsoon from May to October. The current flows southeast during the former and east or north during the latter (UNEP/ITUCN, 1988).
Coral reefs were surveyed by snorkeling and scuba diving. Reef topography, coral Species, community types and estimated coral cover were recorded. The relative abundance of each coral species was estimated according to the number of colonies encountered during each survey as common with more than 50 colonies, occasional with about 10-50 colonies, or rare with less than 10 colonies. Underwater camera and video were used to record photographs of coral colonies and reef topography. Coral species were identified in the field. Whenever confronted with an uncertain species identification, a piece of coral skeleton was detached and brought to the laboratory for further identification. The identification of species was based on Veron and Pichon (1980, 1982), Veron and Wallace (1984), Veron et al. (1977), Veron (1986), Dai (1989), Hoeksema and Dai (1991), and Dai and Lin (1992).
RESULTS AND DISCUSSION Description of Reef Topography and Coral Community
Site A is located on the south side of the island. The substrate of the upper reef flat at 1-2 m depth is covered with sand and seagrasses. On the lower reef flat at 2-4 m depth, there are abundant massive and stoutly branching colonies of Porites, Acropora and Pocillopora spp. Below the reef flat at depths between 4 and 15 m, there is a steep slope; only a few foliaceous Montipora and branching Acropora colonies were found on the surface of the slope. At depths between 15 and 21 m, it is a reef terrace. The substrate is flat and composed of coral debris with some ridges and grooves (Fig. 2a). The coral cover is less than 5%; only a few small colonies are scattered on the substrate. The species diversity was quite high, more than 67 species were recorded. The most abundant species at this site is the octocorallian, /sis sp. (bamboo coral). They can form large colonies of 1 m long and in dense assemblages at some locations. Scleractinians found here are mainly species of Montipora, Favia, Favites, Goniastrea and Cyphastrea. They typically exist as small colonies with a diameter less than 10 cm. The widespread coral debris covering the substrate was mainly Acropora and Pocillopora skeletons indicating that there were flourishing branching coral communities in the past. The scarcity of coral species and scattered small colonies indicate that the community might have been destroyed recently and that recovery is slow.
Site B is located on the southeast of the island. Reef topography is similar to Site A. There is a reef flat about 50 m wide at depths between 0 and 4 m. Living coral cover on the reef flat exceeds 50%, but a trend of decrease toward the west is evident. Scleractinian corals of about 120 species were found. Species commonly occurring on the reef flat were stoutly branched colonies of Pocillopora damicornis, P. verrucosa, P. eydouxi, Acropora monticulosa and A. gemmifera (Fig. 4). Colonies of A. digitifera, A, palmera, Favia speciosa, Leptoria phrygia, Platygyra lamellina and the hydrocoral, Millepora platyphylla were also commonly found on the reef flat. These species generally form large colonies with diameters greater than 1 m. Corals existing on the flat are mainly massive, encrusting and stoutly branched forms. The colony morphology of corals of this area indicates that the reef flat is exposed to strong wave action. Below the reef flat on the seaward side between 5 and 18 mis a steep slope on which coral cover was less than 5%; only a few coral colonies were found to grow on the surface of the slope. A few solitary corals of Fungia spp. and several large colonies of the blue coral, Heliopora coerulea, were found on the sandy grooves. Below 18 m the bottom is sandy and no coral was found.
Site C is situated on the west of the island. It is characterized by a wide reef flat that extends westward to over 500 m from shore with depths about 3-8 m (Fig. 2b). On the surface of the flat, there are low reef ridges alternating with shallow grooves running in the NE-SW direction. Currents of this area are generally strong especially during flood and ebb tides. This area is also exposed to strong waves during the summer monsoon. The
5
substrate on the upper reef flat was characterized by a dense seagrass bed. The lower reef flat was covered with dead coral skeletons; some of them were clearly identifiable based on skeletal features. Few small colonies were found and the coral cover was less than 2%. These phenomena indicate that the coral communities might have been destroyed during the past decade. Some small soft coral colonies such as Sarcophyton spp. and Lobophytum spp. were scattered on the substrate (Fig. 5); few attained a diameter of 50 cm.
Site D is located on the northwest side of the island. The reef flat has a width about 100 m and stretches from 1 to 6 m deep (Fig. 2c). Coral communities on the reef flat can be divided into two zones. In the upper zone between | and 3 m deep, coral cover is higher than 50%. Species common in this zone are Favia, Favites, Goniastrea, Coeloseris mayeri and Pavona spp. Some large colonies with diameters greater than 1 m were found. In the lower zone between 3 and 6 m deep, coral diversity is low and coral cover is less than 10%. The reef surface is covered with dead coral skeletons and algae. Below 6 m, there is a steep drop-off, descending at a nearly perpendicular angle to a depth about 60-80 m. On the wall of this drop-off, there are colonies of Dendronephthya spp., Junceella fragilis and Isis sp. Scleractinians were rare; only few small colonies of foliaceous corals were found to grow on the slope. The coral cover is less than 5%. However, sponges, bryozoans and other sessile invertebrates are abundant.
Site E is situated on the northeast side of the island. Reef topography and coral fauna of this site are similar to those of Site D. On the upper zone of the reef flat, the coral cover was higher than 50% and approximately 100 scleractinian species were found. Among the most abundant species are Pocillopora verrucosa, P. eydouxi, Acropora digitifera, Heliopora coerulea, and Millepora platyphylla (Fig. 6). Species of Montipora, Porites, Favia, Favites and Goniastrea are also common in this zone; most of them are massive, encrusting or stoutly branched forms, with colony sizes often less than 30 cm in diameter. At the lower zone between 3 and 6 m deep, the substrate 1s covered mainly by dead coral skeletons and green algae, Caulerpa spp. The coral cover is less than 5% in this zone. There is a steep drop-off below 6 m; many large gorgonian and antipatharian colonies were found overhanging on the slope. Sponges, bryozoans, crinoids and other groups of marine invertebrates are abundant, which comprise a rich benthic fauna and colorful scenery (Fig. 7). Below 35 m the bottom 1s sandy and no coral was found.
Site F is located on the east side of the island. The reef flat is wider in the north where it extends seaward to approximately 500 m from shore but becomes narrower to the south (Fig. 2d). Dense coral cover (>50%) and high species diversity were found on the upper part of the reef flat at depths between 1 and 3 m. More than 100 scleractinian Species were recorded, most of them were small colonies. Species commonly present in this area are Pocillopora damicornis, P. verrucosa, Acropora digitifera, Cyphastrea chalcidicum and Favites abdita. Coral cover and species diversity are low on the lower part of the reef flat. Less than 5% of the substrate was covered by corals and only few small colonies were found. Below 6 m there is a steep drop-off that extends to about 30 m and reaches the sandy bottom. The most peculiar organisms on the surface of the slope are
6
many colorful soft corals, Dendronephthya spp. hanging on the wall. Other corals are rare and scattered. Below 35 m the bottom is sandy and no corals were found.
Site G is located on a reef ridge on the southeast of the island. The reef ridge is separated from the island by a trough approximately 20 m deep (Fig. 2e). The surface of the ridge is smooth and about 7 m deep. More than 70 species of scleractinian corals were found on the top of the ridge, mainly species of Acropora, Favia, Favites, Goniastrea, and Fungia. The coral cover is about 30-40%. Many colonies of solitary corals such as Fungia cyclolites, F. costulata, F. tenuis, F. fungites, F. scutaria and Herpolitha limax were found on the sandy grooves. The edge of the reef ridge is about 8 m deep. Below 8 m there is a steep slope down to approximately 37 m. There are several 7ubastraea micranthus colonies growing on the upper part of the slope. The lower part of the slope between 20 and 37 m deep is covered by thick patches of Dendronephthya colonies (Fig. 8). These colorful soft corals, when fully extended, form a gorgeous underwater "flower wall”. The slope reaches the sandy bottom at 37 m.
Coral Fauna
A total of 163 species in 15 families and 56 genera of scleractinians; 15 species in three families and five genera of alcyonaceans; and six species in four families and five genera of gorgonaceans were recorded during this survey (Table 1). The results showed that coral communities of the Taiping Island are dominated by scleractinian corals with high species diversity and abundant coral cover found on the reef flat between 1 and 3 m deep. Alcyonaceans and gorgonaceans are relatively rare and their distributions are limited to reef slopes at depths below 15 m. Although the coral fauna varied slightly among the surveyed sites, species compositions of the coral communities are similar and can be regarded as typical of tropical reef communities. The abundance of small coral colonies indicates that coral communities are in their early stages of succession (Grigg, 1983). As early succession communities generally have high species diversity (Connell, 1978), this conditions may also relate to the high diversity of coral communities at Taiping Island.
In comparison with the known coral fauna of other reefs in the South China Sea, the number of scleractinian species recorded during this study exceeds those of Tungsha Island (Pratas Island, 101 species; Dai et al., 1995) and Xisha Islands (Paracel Shoals, 127 species; Zou and Chen, 1983). In general, the species composition of the coral fauna among these islands is similar. Biogeographically, these coral fauna belong to the Indo- Pacific province. Because Taiping Island is situated at a lower latitude and closer to the area of highest coral diversity, it is natural that its coral fauna is more diverse than those of other reefs in the South China Sea. According to the biogeographical location of Taiping Island, this island is expected to have more than 70 genera and 400 species of scleractinians (Veron, 1993). However, during our brief survey to the island, only 51 genera and 163 species were recorded (Table 1). Further intensive surveys of adjacent islands may reveal more species.
7
The coral reef of Taiping Island is a typical oceanic reef. It has a wide, shallow reef flat and a steep drop-off on the edge of the flat. The reef flat is a site of intensive coral calcification that forms the reef framework. The substructure of this region is invariably composed of large, massive, interlocking colonies of hermatypic corals cemented by calcareous algae. The drop-off borders the reef framework and generally descends to depths below 30 or 60 m. At the base of the drop-off there are abundant coral debris and accumulation of sediment. These facts indicate that physical and biological destruction of the reefs is relatively high and debris produced through these processes are transported to a deeper zone at which accumulation occurs.
The development of reefs on the southwest and northeast sides of Taiping Island is better than that of other areas. On both sides there are wide reef flats extending beyond 500 m from shore which basically conform to the shape of the island. Such a pattern of reef development is likely related to the water flow of the reef as both sides are located in the path of tidal current entering and leaving Tizard Bank. Reef growth is usually better where there is strong water flow (Stoddart, 1969; Goreau and Goreau, 1973) because this flow brings food and raw materials at the same time that it removes sediments and waste products. In terms of species diversity, coral communities on the east, southeast and northeast sides of the island are higher than in other areas. The depauperate coral fauna on the west and southwest sides are possibly related to the strong SW monsoon during summer and fall. Zou et al. (1978) reported that coral communities of Xisha Islands (Paracel Shoals) were well developed on the northeast side and poorly developed on the southwest side of the islands and that such distribution patterns are likely related to local flow patterns. Due to the influence of the prevailing SW monsoon during summer, such distribution patterns of coral communities are likely common in the South China Sea.
The tropical reef environment of Taiping Island implies that its coral fauna is rich and the reef is highly developed. However, in comparison with other tropical Indo-Pacific coral reefs, the species diversity and abundance of coral communities at Taiping Island are relatively low. Dead coral skeletons were widely spread on the reef surface below 3 m and only small coral colonies were found. These facts indicate that the coral communities of Taiping Island have suffered severe damage during the last decade. The cause of such extensive coral death is uncertain. Many natural and anthropogenic stresses on coral reefs have been reported (see reviews by Brown and Howard, 1985; Grigg and Dollar, 1990). According to the current status of the reef environment, the possible disturbances are likely include artificial destructions, pollution, storms, predation of Acanthaster planci, and El Nifio events.
Artificial destructions including blast fishing and underwater bombardment may have caused heavy destruction in certain areas. The presence of idle troops at Taiping Island is also of concern because they may engage in environmental damaging activities such as shooting and fishing with explosives. Substantial damage may also come from occasional parties of blast fishers and coral-smashing muroami fishers from the Philippines and Vietnam (McManus, 1992).
The possibility of oil pollution is also of concern because the Spratlys lie near to major shipping lines for oil and nuclear waste. Oil and nuclear waste could be released in the event of a tanker accident in these reef-studded waters (McManus, 1992). However, we found no substantial record or evidence of these pollutants.
The tropical position of Taiping Island places it within the area of frequent typhoon disturbances. The typhoon-generated waves and storm surges may erode reef crest corals and sediments down to about 20 m depth (Stoddart, 1985; Scoffin, 1993). The recognition of past storm disturbances may rely on several features such as the deposits of coral debris, the assemblages of corals and other reef biota, the reef framework structure, and the existence of reef flat storm deposits (Stoddart, 1971; Scoffin, 1993). During this survey, widespread coral debris were found to accumulate as talus at the foot of the fore-reef slope, on submarine terraces and in grooves on the reef front. In addition, on the shallow reef flat there are mainly massive, encrusting or stout branching corals that are basically wave-resistant forms. These facts indicate that typhoon disturbances are possibly the major destructive forces that have caused severe damage to the coral communities of Taiping Island.
The population outbreak of the crown-of-thorn starfish, Acanthaster planci, has been recognized as the most potent biotic disturbance affecting coral communities on many Indo-Pacific reefs (Endean and Cameron, 1990). However, on reefs where marked destruction of hard-coral cover was not apparent, A. planci was either not observed or found at very low populations densities. Since we did not find any individual of A. planci during this survey, it was unlikely that the crown-of-thorn starfish was the major destructive force to the coral communities of Taiping Island.
Global sea warming associated with El Nifio events has caused widespread coral bleaching in the Caribbean and the Pacific (Glynn, 1984, 1988; Williams and Bunkly- Williams, 1990; Gleason, 1993). The ecological consequences of bleaching events include widespread mortality with resultant decreases in coral cover, changes in species composition, reduced growth rates and reproductive output of corals (Szmant and Gassman, 1990; Gleason, 1993). Mortality rates in bleaching events have ranged from zero (Hoeksema, 1991) to very severe (50-98%) as on the eastern Pacific during the 1982-83 El Nifio event (Glynn, 1988). This severe event also had other associated secondary disturbances following coral mortality such as a subsequent increase in number of grazers and bioerosion rates (Glynn, 1988). Whether the widespread mortality of corals at Taiping Island is related to the El Niio-Southern Oscillation (ENSO) events need to be studied. Analysis of the environmental record in coral skeletons and marine environmental data are thus needed to answer this question.
In conclusion, the coral fauna of Taiping Island is dominated by scleractinian corals, distributed mainly on the shallow reef flat at depths of 1-3 m on the east, south and north sides of the island at which flourishing coral communities were found. Few
9
gorgonaceans and alcyonacean species were found mainly on deeper reef slopes. Coral cover and species diversity of Taiping Island are relatively low in comparison with other tropical Pacific coral reefs indicating that the coral communities of Taiping Island may have been destroyed by artificial or natural disturbances. Since flourishing of coral communities and reef-building activities are the basis of sustained development of this island, we propose that reef conservation and protection are urgent and should be enforced immediately by reducing artificial destruction and pollution to the reefs. In addition, the changes of reef environment and biotic communities should be monitored. On a broader scale, the Spratly Reefs, including Taiping Island, are ecologically important, with abundant and relatively unexploited resources and where endangered species still abound. The Spratlys may also serve as a pool of larvae for fishes and other marine organisms that recruit to depleted fringing reefs and coastal habitats of the South China Sea. For these reasons, it is worthwhile to conserve the ecosystem and genetic diversity of the Spratlys by establishing a marine park in the Spratlys as proposed by McManus (1992).
ACKNOWLEDGEMENTS
We are grateful to Dr. L.-S. Fang, National Museum/Aquarium of Marine Biology for his support and to Mr. D.-S. Chen for his assistance with field work. Special thanks are due to the captain and crew of the Fishing Training Ship No. 2, Deep Sea Fishing Training Center, Council of Agriculture. This study was supported by a grant from the Council of Agriculture, Executive Yuan, R. O. C. (83-S.T.-2.15-F.-13).
REFERENCES
Bassett-Smith, P.W. 1890. Report on the corals from Tizard and Macclesfield Banks. Ann. Mag. Nat. Hist. 6(6):353-374, 443-458.
Brown, B.E. and L.S. Howard. 1985. Assessing the effects of “stress” on reef corals. Adv. Mar. Biol. 22:1-63.
Connell, J.H. 1978. Diversity in tropical rain forests and coral reefs. Science 199:1302- 1B 10;
Dai, C.-F. 1989. Scleractinia of Tarwan. I. Families Astrocoentidae and Pocilloporidae. Acta Oceanographica Tatwanica 22:83-101.
Dai, C.-F., T.-Y. Fan and C.-S. Wu. 1995. Coral fauna of Tungsha Tao (Pratas Islands). Acta Oceanographica Taiwanica 34:1-16.
Dai, C.-F. and C.-H. Lin. 1992. Scleractinia of Taiwan. III. Family Agariciidae. Acta Oceanographica Taiwanica 28:80-101.
Endean, R. and A.M. Cameron. 1990. Acanthaster planci population outbreaks. In: Dubinsky, Z. (ed.), Coral reefs, Ecosystems of the world 25, p. 419-437, Elsevier, Science Publ. Co., Amsterdam, The Netherlands.
10
Fang, L.-S., K.-T. Shao and L. Severinghaus. 1990. Report on the marine ecological resources of Tungsha Island. Fishery management Bureau, Kao-Hsiung City Government. 61 pp. (in Chinese).
Gleason, M.G. 1993. Effects of disturbance on coral communities: bleaching in Moorea, French Polynesia. Coral Reefs 12: 193-201.
Glynn, P.W. 1984. Widespread coral mortality and the 1982-83 El Nifio warming event. Environ. Conserv. 11: 133-146.
Glynn, P.W. 1988. El Nifio-Southern Oscillation 1982-83: nearshore population, community and ecosystem responses. Ann. Rev. Ecol. Syst. 19:309-345.
Gomez, E.D. 1994. The South China Sea: Conservation area or war zone? Mar. Pollut. Bull. 28: 132.
Goreau, T.F. and N.I. Goreau. 1973. The ecology of Jamaican coral reefs. II. Geomorphology, zonation, and sedimentary phases. Bull. Mar. Sci. 23: 399-464.
Grigg, R.W. 1983. Community structure, succession and development of coral reefs in Hawaii. Mar. Ecol. Prog. Ser. 11: 1-14.
Grigg, R.W. and S.J. Dollar. 1990. Natural and anthropogenic disturbance on coral reefs. In: Dubinsky, Z. (ed.), Coral Reefs, Ecosystems of the World 25, p. 439-452, Elsevier, Science Publ. Co., Amsterdam, The Netherlands.
Hoeksema, B. 1991. Control of bleaching in mushroom coral populations (Scleractinia: Fungiidae) in the Java Sea: stress tolerance and interference by life history strategy. Mar. Ecol. Prog. Ser. 74: 225-237.
Hoeksema, B. and C.-F. Dai. 1991. Scleractinia of Tarwan. II. Family Fungiidae (including a new species). Bull. Inst. Zool. Academia Sinica 30:201-226.
Ma, T.Y.H. 1937. On the growth of reef corals and its relation to sea water temperature. Mem. Nat. Inst. Acad. Sinica Zool. 1:1-226.
McManus, J.W. 1992. The Spratly Islands: a marine park alternative. ICLARM 15(3): 4-8.
McManus, J.W. 1994. The Spratly Islands: a marine park? Ambio 23(3): 181-186.
Scoffin, T.P. 1993. The geological effects of hurricanes on coral reefs and the interpretation of storm deposits. Coral Reefs 12: 203-221.
Stoddart, D.R. 1969. Ecology and morphology of recent coral reefs. Biol. Rev. 44: 433-498.
Stoddart, D.R. 1971. Coral reefs and islands and catastrophic storms. In: Steers, J. E. (ed.), Applied coastal geomorphology. Macmillan, London, p. 155-197.
Stoddart, D.R. 1985. Hurricane effects on coral reefs: conclusion. Proc. Sth Int. Coral Reef Symp. 3: 349-350.
Szmant, A.M. and N.J. Gassman. 1990. The effects of prolonged “bleaching” on the tissue biomass and reproduction of the reef coral Montastrea annularis. Coral Reefs 8: 217-224.
UNEP/IUCN. 1988. Coral reefs of the world. Volume 3: Central and western Pacific. UNEP Regional Seas Directories and Bibliographies. IUCN, Gland, Switzerland and Cambridge, U.K./UNEP. Nairobi, Kenya. xlix + 329 pp., 30 maps.
Veron, J.E.N. 1986. Corals of Australia and the Indo-Pacific. Angus & Robertson, Sydney, Australia, 644 pp.
11
Veron, J.E.N. 1993. A biogeographic database of hermatypic corals. Australian Institute of Marine Science Monograph Series Vol. 10, 433 p.
Veron, J.E.N. and G. Hodgson. 1989. Annotated checklist of the hermatypic corals of the Philippines. Pac. Sci. 43: 234-287.
Veron, J.E.N. and M. Pichon. 1980. Scleractinia of Eastern Australia. III. Families Agariciidae, Siderastreidae, Fungiidae, Oculinidae, Merulinidae, Mussidae, Pectiniidae, Caryophylliidae, Dendrophyllidae. Aust. Inst. Mar. Sci. Monogr., Vol. 4, 422 pp.
Veron, J.E.N. and M. Pichon. 1982. Scleractinia of Eastern Australia. IV. Family Poritidae. Aust. Inst. Mar. Sci. Monogr., Vol., 5, 159 pp.
Veron, J. E. N., M. Pichon and M. Wijsman-Best. 1977. Scleractinia of Eastern Australia. II. Families Faviidae, Trachyphyllidae. Aust. Inst. Mar. Sci. Monogr., Vol. 3, 233 Pp.
Veron, J.E.N. and C.C. Wallace. 1984. Scleractinia of Eastern Australia. V. Family Acroporidae. Aust. Inst. Mar. Sci. Monogr., Vol. 6, 485 pp.
Williams, E.H.Jr. and L. Bunkley-Williams. 1990. The world-wide coral reef bleaching cycle and related sources of coral mortality. Atoll Res. Bull. 335: 1-71.
Yang, R.-T., Y.-M. Chiang and J.-C. Chen. 1975. A report of the expedition to Tungsha reefs. Inst. Oceanogr., Nat’! Taiwan Univ., Special Publ. 8, 33 pp. (in Chinese)
Zou, R.-L. 1978a. Studies on the corals of the Xisha Islands, Guangdong Province, China. III. An illustrated catalogue of scleractinian, Hydrocorallian, Helioporina and Tubiporina. Report on Marine Biological Survey of the Xisha and Zhongsha Islands. p. 91-124. Scientific Publishing Society, Beijing, China. (in Chinese)
Zou. R.-L. 1978b. A preliminary analysis on the community structure of the hermatypic corals of the Xisha Islands, Guangdong Province, China. Report on Marine Biological Survey of the Xisha and Zhongsha Islands. p. 125-132. Scientific Publishing Society, Beying, China. (in Chinese)
Zou, R.-L. and Y.-Z. Chen. 1983. Preliminary study on the geographical distribution of shallow-water scleractinian corals from China. Nanhai Studia Marina Sinica 4:89- 96. (in Chinese)
12
Table 1. Distribution and relative abundance of shallow water corals at seven study sites (A-F) of Taiping Island. Relative abundance, +: rare, ++: occasionally, +++: common.
species / Site A B (C D E F G SUBCLASS ZOANTHARIA ORDER SCLERACTINIA Family ASTROCOENIIDAE Stylocoeniella armata + 3 + + + S. guentheri + 3 + + Family THAMNASTERIIDAE Psammocora profundacella ar ase + + ++ ++ + P. digitata 3° Te + + P. contigua 2 35 25 + + Family SIDERASTREIDAE Pseudosiderastrea tayami = + Coscinarea columna + + ~ C. exesa + Famliy POCILLOPORIDAE Pocillopora damicornis =r + ar a7 +++ ++ P. eydouxi 3 sete ++ + + P. meandrina + 3 ae + 4 + P. verrucosa oF states + ++ <FIaF +++ ae P. woodjonesi + Seriatopora caliendrum 4° + S. hystrix + 35 + 36 + + Stylophora pistillata 3 + + e 4 Palauastrea ramosa + Family ACROPORIDAE Acropora humilis absp 3 + +++ + a A. gemmifera +++ + + + + + A. monticulosa +4++ +++ + + A. digitifera ar +++ ct ate +++ H+ + A. robusta ++ + +f + A. palmerae ++ + oF + + A. nobilis ++ + ++ + A. grandis + + A. microphthalma + ain + + A. aspera + +: A. millepora + A. tenuis a stot + A. cytherea + A. hyacinthus + + + A. nana + + A. cerealis et: + A, nasuta +
A. valida qe + + ++ +
A. lutkeni 4 ae
A. divaricata a ie
A. florida +
A. sp 1 4
A. sp 2 su
Astreopora myriophthalma 4
A. listeri +
A. gracilis 4 +f + + + + +
Montipora monasteriata + ae + + +4 ae a
M. turgescens + 4 cfs a fh
M. undata +5 + + + a
M. verrucosa =H; qr + + ++ ++ a
M. danae 42
M. foveolata af
M. venosa + ++ + +
M. digitata oT +
M. grisea +
M. informis +P + + er iF +
M. foliosa + a
M. aequituberculata + + + + Family AGARICIIDAE
Pavona clavus +
P. explanulata ++ + ++ a
P. varians + ++ ++ a
P. venosa + ++ + + +- ++ 4+
Gardineroseris planulata 4p + + +
Leptoseris mycetoseroides + + JL
L. explanata + +r
Coeloseris mayeri + + + + a
Pachyseris rugosa qe ++ ++ ++ - wee
P. speciosa + ++ + ated a dine Family FUNGIIDAE
Fungia (Cycloseris) cyclolites cata ++
F. (C.) fragilis ns a
F. (C.) costulata + + + ats
F. (C/) tenuis + + a ffl
F. (C.) vaughani + ae =e
F. (Verrillofungia) repanda af + + + + +
F. (V.) concinna ae
F, (Danafungia) horrida 4
F’, (D.) scuposa i ay
F.. (Fungia) fungites ae + te spo
F. (Wellsofungia) granulosa a; te 1
F. (Pleuractis) gravis oe + + + a at
F. (P.) paumotensis at
F. (Lobactis) scutaria aF + + He nee i
species / Site A B (& D E F G Ctenactis echinata + + ae ne C. crassa x 24: Herpolitha limax 3 2s + + + Polyphyllia talpina + + Sandalolitha robusta + + + - +. Heliofungia actiniformis + Family PORITIDAE Alveopora verrilliana Efe a A. spongiosa ss oy Goniopora minor + 4: G. columna ie aL G. stuchburyi a + Porites (Porites) solida ++ + + ++ + + P. (P.) lichen + oF + P. (P.) lobata + ++ + ++ ++ ne P. (P.) lutea + ++ + ++ + 48 P. (P.) cylindrica + + + + zt P. (P.) nigrescens + + a ae se P. (P.) annae + + P. (Synaraea) rus + ee + + Family FAVIIDAE Cyphastrea chalcidicum + +4++ + ++ +++ Fete C. microphthalma + + + + + C. serailia + ++ + + ++ + Caulastrea furcata + - Diploastrea heliopora + + + + Echinopora lamellosa + + + + + as E. gemmacea + Favia favus sme + + oo = = F. pallida + ++ + ++ ++ + a F. rotumana + F. speciosa + +++ a5 +++ ++ ++ + F. stelligera + + - + + F. laxa + a Favites abdita + +++ + ++ ++ shekss ae F. chinensis + + + + F.. complanata + F. flexuosa ++ + + ++ + + F. russelli at ay be F. pentagona + + + ++ + + F. halicora + Barabattoia amicorum + a Montastrea valenciennesi or + M. curta + ++ + + qa eet + M. magnistellata + + x Goniastrea edwardsi + + + + ra G. aspera + +
G. pectinata ++ ++ + + G. retiformis FF + + + = Leptoria phrygia EF + ARP ote ser + Platygyra pini staat + tte ay + P. lamellina steht aaah ++ + ++ P. daedalea state + + ++ + + P. sinensis ape +r atte tp + Plesiastrea versipora ats ar +P Leptastrea purpurea ats a + L. pruinosa +f L. transversa at +P + Family OCULINIDAE Galaxea fascicularis 2 ++ + 4 ++ ar “++ G. astreata +f alanis F + ++ ar + Family MERULINIDAE Merulina ampliata +P as af + + Scapophyllia cylindrica ate + Hydnophora exesa + ser + + ++ ++ ++ H. microconos qe ++ HP a Family PECTINIIDAE Echinophyllia aspera ct + + 4 + + E. echinata + + oF + Oxypora lacera + + ete + O. glabra 4 Mycedium elephantotus + Pectinia lactuca + 4p + P. paeonia + anata ++ + + Family MUSSIDAE Blastomussa merleti +r Cynarina lacrymalis ote Scolymia cf. vitiensis + + + Acanthastrea echinata + ++ + 4 =r + + A. hillai + + Lobophyllia hemprichii ale L. corymbosa + +r + Symphyllia recta ate a a S. radians + + S. agaricia + + Family CARYOPHYLLIIDAE Euphyllia (E.) glabrescens a0 at Family DENDROPHYLLIIDAE Turbinaria mesenterina oF
T. reniformis + +
Tubastraea aurea + T. micranthus ae He rs
SUBCLASS OCTOCORALLIA ORDER STOLONIFERA Family TUBIPORIDAE Tubipora musica 35 ++ 35 + ++ - ef
ORDER COENOTHECALIA Family HELIOPORIDAE Heliopora coerulea + sears ats + ao ++ ++
ORDER ALCYONARIA Family Alcyoniidae Sarcophyton ehrenbergi S. trocheliophorum S. glaucum S. sp. Lobophytum sarcophytoides L. mirabile Sinularia exilis ++ S. gibberosa *e S. numerosa i S. sp. 1 Sa Spe2 +
+++ 4+ 44+
++++ +4 + + +
Family Nephtheidae Dendronephthya sp. 1 + + ++ D. sp. 2 = + D. sp. 3 zt ae ef:
Family Xentidae Xenia sp. + a
ORDER GORGONACEA Family Isididae Isis sp. qe ae i + + + -
Family Melithaeidae Melithaea ochracea + + + ~
Family Subergorgidae Subergorgia sp. ia + + ++ S. sp. + ~
Family Ellisellidae Ellisella robusta + i Junceella juncea + + +
CLASS HYDROZOA ORDER MILLEPORINA Family MILLEPORIDAE Millepora platyphylla aF Tee I ++ +++ +++ + M. tenera + + ee fe M. intricata +++ ae mn M. tuberosa +
Total No. of species 67 121 88 103 106 107 86
18
500 1000 b (o) 30 a) 500 1000 E < c d 2 2 : c?) a 40 40 0) 200 0) 500
wii Sea grass
xn Coral debris
««v Branching coral
a n Massive coral
= Soft coral Gorgonians
cv Alcyonaceans 0 500 700
Distance from shore (m)
Fig. 2. Reef profiles and distribution of benthic organisms at the study sites. a: Site A, b: Site C, c: Site D, d: Site F, e: Site G.
Fig. 3. Taiping Island, or Itu Aba Island, is a reef island about 1300 m long and 350 m wide.
Fig. 4. Coral community on the reef flat at Site B is dominated by stoutly branched colonies of Acropora spp.
19
20
Fig. 5. Some small soft coral colonies of Sinularia sp. scattered on the substrate at Site C.
Yar OF : “od ery Ae Fig. 6. Coral community on the reef flat at Site E is dominated by stoutly branched colonies such as Pocillopora eydouxi.
Fig. 8. Colonies of Dendronephthya sp. on the reef slope at Site G on the southeast of the island.
21
ry ile
1 was
bh ala AN eodbiine at as waters on NT ag eee” agi ; : : ) .
ATOLL RESEARCH BULLETIN
NO. 437
FIRST OBSERVATIONS ON THE FISH COMMUNITIES OF FRINGING REEFS
IN THE REGION OF MAUMERE (FLORES - INDONESIA)
BY
MICHEL KULBICKI
ISSUED BY NATIONAL MUSEUM OF NATURAL HISTORY SMITHSONIAN INSTITUTION WASHINGTON, D.C., U.S.A.
“ir a, asa HO RALIAS HIOTA |
| a ct ’ Al
OIE Oe OTT UMS Bem VAT ‘a0 SORTA VINE Ta | : eae ; i (VORA autos DAM AO nanan auT ma :
et
AON A SEM
{9 cyan ; Won
VROT DART A TO MUM Senet mt MOUTUTIVEMT AAIMOCTIME. 0
a) Bee | hada
FIRST OBSERVATIONS ON THE FISH COMMUNITIES OF FRINGING REEFS IN THE REGION OF MAUMERE (FLORES - INDONESIA). BY,
MICHEL KULBICKI
ABSTRACT
Total fish counts were made along 6 transects on fringing reefs in the region of Maumere (Flores - Indonesia). This represents the first description of fringing reef communities in this area of the Pacific. A total of 255 species, distributed among 36 families, were recorded. The major families were the Pomacentridae, Labridae, Serranidae, Acanthuridae and Chaetodontidae. The number of species per station was high (96 species) compared to similar counts for fringing reefs in New Caledonia. Density was 7.2 fish/m? and biomass was 187 g/m?. The average weight of fish was low (21.7 g), with the Pomacentridae comprising 68% of the density. Large fish (over 40 cm) were scarce, possibly due to fishing pressure. The major contributors to the biomass were Scaridae, Caesionidae, Acanthuridae and Pomacentridae. Carnivores had the highest number of species followed by zooplanktivores and microalgae feeders. Most of the density consisted of planktivores and microalgae feeders, whereas biomass was dominated by microalgae feeders, zooplantivores and macroinvertebrate feeders. Small species with short life spans constituted most of the density. The trophic structure and distribution of life-history strategies were very similar to observations made on the fringing reefs of mainland New Caledonia, but were different from those of fringing reefs of two isolated islands (Ouvea Atoll and Chesterfield Island). There was a relationship between the number of dominant species and diversity. Structure of the fringing reef fish communities was mainly linked to habitat type, in particular, terrestrial runoffs could be a major factor.
INTRODUCTION
The reef fish fauna of Indonesia is one of the most diverse in the world, with over 2000 species. The Flores islands are at the eastern end of the Indonesian archipelago and are likely to support a species diversity lower than the larger islands further west such as Java, Sumatra or Borneo (species diversity decreases eastwards in the Pacific, and smaller islands tend to have fewer species than large ones). Other than a recent checklist (Kuiter and Allen, unpublished), very little is known of the reef fish communities of Flores. There is no account of the abundance, biomass, size distribution, trophic structure and the life history strategies of the major reef fish species in that region. The first objective of this article is to present a set of data relating to these subjects that were obtained in the Maumere region in 1993.
The second objective of this article is to compare the species rich region of Flores with a less diversified region (New Caledonia). Several questions come to mind when studying
ORSTOM - B.P. A5 Nouméa New Caledonia
Manuscript received 4 June 1995; revised 1 February 1996
i)
e702 i i ee A en Rg PO dS
TIMOR SEA
Figure | : map of the Maumere region. The 4 stations are indicated by a & on the map inset. The numbers on the inset correspond to the transects.
3
communities found in a species rich area. For a given habitat, are there more species per unit area than in a less diverse region with similar habitat? Are there more "dominant species " (species making more than 2% of the density or the biomass) than in a less diverse region? Is the trophic structure or the distribution of the life-history strategies different from those observed on fish communities from a similar habitat but a different region? One of the major problems in answering such questions is to develop comparable sets of data. In the present case, the data from Flores were collected using the same methods as those used for a large set of data collected in New Caledonia (Kulbicki et al, 1994a).
MATERIAL AND METHODS
During the Pre-Indo-Pacific Fish Conference in Maumere (November 1993), the author had the opportunity to visit 4 fringing reefs and to perform 6 transects (Figure 1). The start of each transect was chosen at random on the reefs and the transects were laid in the direction of the slope. The transects were 50 m long. All fish, except the cryptic species (most Muraenidae, Ophichtydae, Syngnathidae, Gobiidae, Blenntidae, Synodontidae, Scorpaenidae, Antenariidae) and juveniles (newly recruited fish, usually less than 5 cm, but may be as small as 3 cm, i.e. Chromis viridis), were counted. For each record, the species name, number of fish observed, size of fish and distance of fish to the transect were noted. The size of the fish were noted in 1 cm classes for fish less than 10 cm, in 2 cm classes for fish between 10 and 30 cm, in 5 cm classes for fish between 30 and 50 cm and in 10 cm classes for fish more than 50 cm. The distances of the fish to the transect were estimated in 1 m classes for fish less than 5 m from the transect, and in 2 m classes for greater distances. Fish beyond 12 m from the transect were not counted. The diver covered each transect only once. The average time per transect was 90 min. Densities were calculated according to the method given by Burnham et al (1980) and Buckland et al. (1993). Fish weights were estimated from length-weight equations (Kulbicki et al., 1994a). Biomasses were estimated using these fish weights and the same method as for densities.
The diet of each fish species was either taken from the data used by Kulbicki et al. (1994a) or from information in FISHBASE (Froese et al., 1992). Species with no direct information available were assigned the same diet as the closest species within the same genus or family for which dietary information was available. The food items are divided into 9 categories: fish, macroinvertebrates, microinvertebrates, zooplankton, other plankton, macroalgae, microalgae, coral, detritus. The diet of each species is distributed among these 9 food categories. The percentage of each of these food items is taken into account when calculating the contribution of a given species to a trophic category. For instance, if species A eats 50% fish and 50% microalgae, and if this species has a density of 0.1 fish/m?, the contribution of species A to piscivory will be of 0.1 x 0.50 =0.05 fish /m?.
Each fish species was classified within one of the 6 life-history strategy classes defined in table 1 (see Kulbicki 1992 for a discussion on this classification). For most species the classification is given by Kulbicki et al. (1994a). For the remaining species, data from FISHBASE (Froese et al., 1992) was used to assign the species to a given class. For a number of species the information available was absent or too scant for a classification. In such a case, I used the same classification as for the closest species within the genus or the family.
Each transect was divided into five sections of 10 m each. On each section the cover of each of the substrate categories (see Kulbicki et al., 1993 for details of the method) given in Table 2 was noted (the total for each section being 100%) for a 5 m wide strip. Algae and coral cover were noted in the same manner.
RESULTS
The stations (Table 2) were between 3 and 7 m deep with a minimum depth of 1 m and a maximum of 12 m. The substrate was characterised by a large proportion of rubble (debris, small and large boulders) and a small coverage of sand, which was either muddy or coarse, no fine sand being found. Rock formations were usually from eroded reefs and not of volcanic origin, as found on land. Macroalgae were very scarce. Coral and alcyonarians were present in significant amounts at only one station.
A total of 255 fish species, distributed among 36 families, were recorded (Appendix 1). The families with more than 5 species accounted for 77 % of the total species seen (Table 3), and only 6 families (Serranidae, Chaetodontidae, Pomacentridae, Labridae, Scaridae and Acanthuridae) had more than 10 species. The number of species per transect (95.7 species), density (7.1 fish /m?) and biomass (187 g/ m?) were high (Table 4), but average weights were small (21.7 g) due to the dominance of Pomacentridae in the counts. Pomacentridae accounted for 16% of the diversity, 68% of the density and 9.5% of the biomass. One species, Pomacentrus coelestis, formed 48.7% of the total density and four other Pomacentridae (Chromis amboinensis, Chromis xanthura, Neopomacentrus azysron, Pomacentrus amboinensis) were among the 10 most important contributors to density. The other important species with respect to diversity and density were in the Labridae, but no particular species in this family dominated in density. Most species had a low number of individuals in the counts, even the planktivorous Labridae, which are usually found in schools elsewhere in the Pacific. The major contributors to biomass were the Scaridae and the Caesionidae. Most of the biomass for the Scaridae was made of juveniles, which cannot be easily identified underwater, but two species, Scarus fasciatus and S.quoyi, formed one-third of the Scaridae biomass. The Caesionidae, which are all schooling species, were dominated by Pterocaesio tile and Pterocaesio chrysozona. One of the major contributors to biomass was Pomacentrus coelestis, a very small fish (3 g average weight), but which was present in extremely high densities.
The trophic structure can be considered in species numbers, density or biomass (Table 5). Most species were carnivores (23.2% macrocarnivroes, 14.2% microcarnivores, 11.9 % piscivores), zooplanktivores and microherbivores represented respectively 21.7 and 20.5% of the species. Density was dominated by zooplanktivores (59.9%), followed by microherbivores (17.2%). The other trophic categories had little importance with respect to density. Three categories dominated biomass: microherbivores (34.9%), zooplanktivores (29.9%) and macrocarnivores (19.3%). Coral and detritus feeders were low in all respects. The low numbers for "other planktivores" are normal for reefs in the tropical Pacific. Macroherbivores were not an important group. As is usually the case in the Pacific, this group exhibits little diversity and low densities, but the large size of macroherbivores makes this category, at times, a significant contributor to the biomass. In Flores, these fish were small in size, most likely because of fishing pressure.
The distribution of the life-history strategies was dominated by the abundance of short- lived species (classes | and 2) (Table 6). Short-lived species were also the most diverse; however, species with an average life span (classes 3 and 4) were also represented by large number of species. Biomass was evenly distributed between short and average life-span species.
There were major differences in the distribution of the life-history strategies among trophic categories (Figure 2). In particular, zooplanktivores were essentially short-lived species
5
whereas, the long living species were mainly macrocarnivores and piscivores. Microherbivores were split between many small, short-lived, species which dominated the density of this group, and a few large longer-lived species (Scaridae, Acanthuridae), which made up most of the biomass.
The average size of the commercially important species (essentially Serranidae, Lethrinidae, Lutjanidae, Scaridae, Acanthuridae) indicates that there are very few large fish (Appendix 1). In particular, not a single species with more than 10 individuals sighted, had an average size > 40 cm. The size frequencies for the most abundant commercial species are given on Figure 3. Most Serranidae were juveniles or small species. The Lethrinidae, Caesionidae and Scaridae were small in size (sizes at least 30% less than average reproductive size). This could be due to fishing pressure, but the high densities observed indicate that other factors could possibly be involved.
DISCUSSION
The data set presented here are minimal and one should be cautious in generalizing these results to a large area. In the absence of other comparable data from the Flores Islands or even Indonesia, it is difficult to assess how representative are these results. In particular, it is noteworthy that the stations were sampled in a leeward zone and that on the windward side of the island the morphology of the reefs is very different, and it is likely that the reef fish communities there would be different also. However, data from New Caledonia (Kulbicki et al. , 1994a) indicate that even in a wide zone, reef fish communities from the same type of reef habitat share much in common in species richness, density, biomass and structure.
The substrate found on the stations is typical of many fringing reefs in the region. Indeed, in many cases terrestrial runoffs bring very fine sediment, and wave action induces the formation of rubble and coarse sediment. The very low algae and coral cover is not unusual either, especially in turbid areas.
It is difficult to compare the total number of species with other areas, because the sampling effort was low. However, this number (255) is higher than observations made on fringing reefs in Hawaii, 81 - 187 species (Hayes et al. , 1982) or French Polynesia, 80 species (Galzin, 1985), which have been sampled much more thoroughly. These numbers are comparable to the highest diversities found in New Caledonia, 168 - 252 species, but with a much larger sampling effort (Kulbicki, 1992). The number of species /station is a better indicator, if the stations are sampled in a similar manner. The only data (Table 7) that have been collected according to the same methods are from Kulbicki et al. (1989, 1994a). The species richness observed in Flores is higher than in any of the New Caledonian areas. It is estimated that there are 1140 reef and lagoon fish species in the Maumere area (Kuiter and Allen, unpublished), whereas there are 940 species in the SW lagoon of New Caledonia (Rivaton et al. 1989), with 550 species in the Chesterfield Islands (Kulbicki et al. , 1994b) and 630 in Ouvéa (Kulbicki et al, 1994a). The families that are best represented in Flores exhibit considerable species diversity in most parts of the tropical Pacific, but some families that contain many species elsewhere (Apogonidae, Holocentridae, Scaridae, Acanthuridae) (Thresher, 1991) did not exhibit similar diversity in our observations.
The densities observed in Flores are very high, especially for fringing reefs. Such densities have not been recorded in this type of environment in the tropical Pacific (Kulbicki, 1991). However, most of this density is due to only one species, Pomacentrus coelestis, a
6
planktivore. Large densities of planktivores are common on reefs (Kulbicki et al. , 1994a), and these species are usually short lived and experience large temporal variations. The other components of the density in Flores are usually found on fringing reefs in the Pacific, in particular, the Acanthuridae, Pomacentridae and small Labridae. This is confirmed by the few published studies on fringing reefs in the Pacific that give a detailed account of the contribution of the various species to density. In Hawaii (Hayes et al., 1982), the dominant species were two Acanthuridae (A.nigrofuscus, Ctenochaetus striatus), followed by small Labridae (Thalassoma duperrey, Gomphosus varius), the Pomacentridae being the third major component of the Hawaiian reef communities. In French Polynesia, Galzin (1985) also found a majority of Ctenochaetus striatus on the fringing reefs, the second most abundant species being another herbivore, the Pomacentridae Stegastes nigricans. In New Caledonia, the composition of the density varied from one zone to another. In Ouvéa (Kulbicki et al., 1994a) the most abundant fish were Acanthurus nigrofuscus and Stegastes nigricans, followed by three planktivorous Pomacentridae (Pomacentrus coelestis, Chromis chrysura, Chrysiptera cyanea). In the Chesterfield islands (Kulbicki et al., 1989) the most abundant species were Mulloides flavolineatus, juvenile Scaridae, Acanthurus nigrofuscus, Ctenochaetus striatus, three species of Caesio and three Pomacentridae, all herbivores (Pomacentrus molluccensis, Stegastes nigricans, Pomacentrus vaiuli). On the main island of New Caledonia (Kulbicki, unpubl.data), the major contributor to density were planktivorous Caesionidae (Pterocaesio diagramma, P.tile), several Pomacentridae (the two major ones being Chromis viridis and Dascyllus aruanus, which are mainly planktivores), Acanthurus nigrofuscus, small Labridae (Thalassoma lunare, T.lutescens) and juvenile Scaridae.
The biomass (187 g/m?) found in the Flores is high for fringing reefs. In Hawaii Brock et al. (1979) found 106 g/m?, on the GBR (inshore reefs) Williams and Hatcher found 92 g/m?; the results for New Caledonia are given in table 7. The distribution of the biomass can be compared only to the studies from New Caledonia. There, the major contributors varied greatly from one zone to another. In Ouvéa (Kulbicki et al., 1994a) the top three species in terms of biomass were herbivores (Hipposcarus longiceps, Acanthurus blochii, Acanthurus xanthopterus); in the Chesterfield Islands (Kulbicki et al., 1989) the top species were two herbivores (Kyphosus vaigiensis, Naso unicornis) and a carnivore (Mulloides flavolineatus); and on the mainland the main species were planktivores (Pterocaesio tile, P.diagramma) and herbivores (Acanthurus nigrofuscus, Scaridae spp.). The similarity between Flores and New Caledonia is the presence of Acanthuridae and Scaridae as major contributors to the biomass. The differences are in the species involved, with larger species in New Caledonia than in the Flores Islands.
The comparison of some length frequencies (Figure 3) between Flores and New Caledonia show that there is usually no difference in the size range. However, no small Siganus doliatus were observed in Flores, which could be due to the season, small Siganus doliatus (less than 15 cm) being found mainly during the dry season in New Caledonia. Monotaxis grandocculis did not exceed 22 cm in Flores, whereas this species was found to reach 38 cm in New Caledonia, with the largest sizes found on the barrier reef.
It is often assumed that the number of species contributing in an important manner (major species; more than 2% in the present case) to the density or biomass decreases as diversity increases (Richards, 1952 and Whittaker, 1964 in McIntosh, 1967; Spight, 1977; Wahington, 1984). The relationship is not clearcut, because it is often not specified which diversity is taken into account: the observed diversity (number of species in the sample) or the potential diversity (number of species in the region). The correlation between density and biomass for major species exists both for the observed diversity and the potential diversity, but is not as good for the latter
gi
(Table 8 and Figure 4). This result suggests that highly diverse communities have lower numbers of dominant species. In other words, one would expect the resources to be better shared and utilised in these communities that in less diverse ones. Analysis of the trophic structure and of distribution of the life-history strategies will in part answer this question.
It is difficult to compare the trophic structure found in Flores with most of the findings in the literature, because the methods were very different from one study to another (Kulbicki, 1991). The data from New Caledonia were collected and analysed with the same methods used in the present study and are, therefore, comparable (Figure 5). The distribution of species among trophic categories (Figure 5a) is very similar in all 4 studies. However, Flores had more zooplankton feeding species than the fringing reefs of New Caledonia. In density (Figure 5b) and biomass (Figure 5c) the results from Flores and mainland New Caledonia are almost identical. The latter two islands differ from Chesterfield and Ouvea, both of which are offshore islands, in having larger numbers of zooplanktivores, lower abundances of microherbivores and carnivores, and larger biomasses of zooplanktivores. This larger importance of zooplanktivores in the Flores and mainland New Caledonia could be linked with high terrestrial runoffs (these islands have similar land masses -10 000 and 20 000 km? - and average rainfall - 1500 to 2000 mm/ year). There are also trends common to all four studies. In particular, coral feeders form 2-7% of the species but account for very little in density or biomass. Detritus feeders and "other planktivores" are never an important component of the trophic structure, whereas they form between 10 and 15% of the abundance or weight for the coastal (mangroves and estuaries) areas in New Caledonia (Thollot, 1992). Fringing reefs and coastal areas are often adjacent in New Caledonia, thus indicating that the trophic structure is greatly influenced by the substrate.
Very few studies on reef fishes have treated life-history strategies (Kulbicki, 1991; Kulbicki et al., 1992, 1994a) or assimilated structures (ecological categories x size classes) (Harmelin-Vivien, 1989). Kulbicki (1992), based on original data, compared life-history strategies from several types of reefs across the Pacific using the same classification. The data of the present study can be compared with data processed in the same way for fringing reefs in New Caledonia (Figure 6).
The distribution of species among life-history strategies is almost identical for all reefs (Figure 6a). This result could be expected from the findings of Kulbicki (1992), who demonstrated that within the Western Pacific there were little differences in this structure at the species level. Flores and mainland New Caledonia also have very similar structures in terms of density and biomass (Figures 6b, c). In particular, they differ from the fringing reefs of the islands of Ouvea and Chesterfield by having more class-1 species, which have the fastest turnover. Conversely, Flores and mainland New Caledonia have a low proportion of biomass represented by long living fishes (classes 5 and 6) which are important on the Ouvea and Chesterfield islands. This suggests that in Flores the fish communities of the fringing reefs should be more sensitive to short term variations than they would be on isolated islands such as Ouvea or the Chesterfield. This is logical since most of these class 1 and 2 fish feed mainly on zooplankton and microalgae, which are variable food sources, depending on primary production and mineral inputs.
Our findings indicate, therefore, that the functioning of the fringing-reef fish community of Flores is very similar to what is observed on mainland New Caledonia where ecological conditions are similar. Conversely, fringing reef fish communities from isolated islands of New Caledonia, despite their similar species composition, have different structures. Diversity alone does not account for the major differences in the structure of these fish communities.
ACKNOWLEDGEMENTS
The author wishes to thank the following persons and organisations: Prof. Dr. Kasijan Romimohtarto and the organizing committee of the Pre Indo-Pacific Fish Conference workshop held in Maumere (November 20-25, 1993), R.Kuiter, Dr.G.Allen, G.Moutham, P.Dalzell and the two anymous reviewers.
LITERATURE CITED
Brock R.E., Lewis C. et Wass R.C. 1979 Stability and structure of a fish community on a coral patch reef - Marine Biology 54: 281-292
Buckland S.T., Anderson D.R., Burnham K.P., Laake J.L. 1993 Distance sampling, estimating abundance of biological populations. Chapman & Hall London 446p.
Burnham K., Anderson D.R., Laake J.L. 1980 Estimation of density from line transect sampling of biological populations. Wildlife Monographs 72: 202p.
Froese R., Palomares MLD, Pauly D. 1992 Draft user's manual of Fishbase software 7 - International Center for Living Aquatic Resources Management- Manila Philippines 56 p.
Galzin R. 1985 Ecologie des poissons récifaux de Polynésie Francaise Thése Doctorat Université de Montpellier: 195 p.
Harmelin-Vivien M. 1989 Reef fish community structure: an Indo-pacific comparison. in Ecological studies - Vertebrates in complex tropical systems (Harmelin-Vivien M., Bourliére F. eds) Springer Verlag N.Y. 69: 21-60
Hayes T., Hourigan T., Jazwinski S., Johnson S., Parrish J., Walsh D. 1982 The coastal resources, fisheries and fishery ecology of Puako, West Hawaii - Hawaii Cooperative Fishery Research Unit Technical Report 82-1: 159 + Annexes
Kulbicki M. 1991 Present knowledge of the structure of coral reef fish assemblages in the Pacific - in Coastal resources and systems of the pacific basin: investigation and steps toward a protective management - UNEP Regional Seas Report and Studies : 147: 31-53
Kulbicki M. 1992 Distribution of the major life-history strategies of coral reef fishes across the Pacific. Proc. 7th Intern. Coral Reef Symp. - Guam 1992 : 918-929
Kulbicki M., Doherty P., Randall J.E., Bargibant G., Menou J-L., Mou-Tham G., Tirard P. 1989 - La campagne Corail 1 du N.O. Coriolis aux iles Chesterfield (du 5 aoit - 4 sept. 1988) : données préliminaire sur les peuplements ichtyologiques ORSTOM Nouméa. Rapp. Sci. Tech. Sci. Mer Biol. Mar. 57 : 88 p.
Kulbicki M., Thollot P., Wantiez L. 1992 Life history strategies of fish assemblages from reef, soft bottom and mangroves from New Caledonia. Seventh Intern Coral Reef Congress - Guam June 1992 abstract
Kulbicki M., Dupont S., Dupouy C., Bargibant G., Hamel P., Menou J.L., Mou Tham G., Tirard P. 1993 Caractéristiques physiques du lagon d'Ouvéa - in Evaluation des ressources en poissons du lagon d'Ouvéa: 2éme partie: l'environnement physique: sédimentologie, substrat et courants - Convention Sciences de la Mer ORSTOM Nouméa 10: 47-150
Kulbicki M., G. Bargibant, Menou J.L., Mou Tham G., P.Thollot, L. Wantiez, Williams J.T. 1994a Evaluations des ressources en poissons du lagon d'Ouvéa. in Evaluation des ressources en poissons du lagon d'Ouvéa: 3éme partie: les poissons; Convention Sciences de la Mer ORSTOM Nouméa 11: 448 p.
Kulbicki M., Randall J.E., Rivaton J. 1994b Checklist of the fish from the Chesterfield islands. Micronesica - 27 (1/2): 1-43
Kuiter R., Allen G. submitted Fishes of Maumere Bay, Flores Indonesia - Tropical Diversity Indonesian Journal
McIntosh R.P. 1967 An index of diversity and the relation of certain concepts to diversity - Ecology 48 (3) : 392 - 404
Rivaton J., Fourmanoir P., Bourret P., Kulbicki M. 1989 - Catalogue des poissons de Nouvelle- Calédonie. Catalogues Sciences de la Mer, ORSTOM Nouméa 2: 170 p.
Spight T.M. 1977 Diversity of shallow water gastropod communities on temperate and tropical beaches - The American Naturalist 111 (982): 1077-1097
Thollot P. 1992 - Les poissons de mangrove du lagon sud-ouest de Nouvelle-Calédonie - écologie des peuplements, relations avec les communautés ichtyologiques cotiéres. Ph.D. Thesis University of Aix-Marseille II (France), 406 p.
Thresher R.E. 1991 Geographic variability in the ecology of coral reef fishes : evidence, evolution, and possible implications - in The ecology of fishes on coral reefs (P.Sale ed.) Academic Press Inc. New York 754 p.
Washington H.G. 1984 Diversity, biotic and similarity indices. A review with special relevance to aquatic ecosystems - Water Research 18 (6): 653-694
Williams D.McB., Hatcher A. 1983 Structure of fish communities on outer slopes of inshore, mid-shelf and outer shelf reefs of the Great Barrier Reef - Marine Ecology Progress Series 10: 239-250
10
Piscivores Macrocarnivores
Life-history strategy classes Life-history strategy classes Microcarnivores Zooplanktivores
70 + Oce2p
eo Bic2B
50 +
40 +
x
30
20 +
10 +
(0) 1 2 3 4 5 6
Life-history strategy classes Life-history strategy classes
Microherbivores
Life-history strategy classes
Figure 2: distribution of trophic categories according to life-history strategies. D: density; B: Biomass; Pi: piscivores; C1: macroinvertebrate feeders; C2: microinvertebrate feeders; Zoo.: zooplanktivores; Mi.: microalgae feeders
Pterocaesio tile
number
16 Size (cm)
Scolopsis bilineatus
o
Ynumber =
a
= Ss Sette Sea zz
Size (cm)
Ctenochaetus striatus
11
Monotaxis grandocculis
© Flores N=320 BI NC N=3825
Scarus fasclatus
O Flores N= 29 O Flores N= 23 BANC N= 1421 FNC N=295
Size (cm)
Figure 3: size distribution of the most abundant commercial species (NC:
Caledonia)
Y=10.13 -0.023X r1r?=0.60
12 i= 2 10 o E =, 8 ore 06 ea ow iS o a
o Nn BR DD
100 Number of species in sample
200 300
Size (cm)
data for New
InY=6.95 -0.75InX 17=0.69
2.50 = In%B
2.00 5 |In%D 1.50 1.00
0.50
In % major species
0.00 6.00
400
7.50 In Number of potential species
6.50 7.00 8.00
Figure 4: correlation between number of species ("major species") contributing to more than 2% of density (%D) or biomass (%B) and number of species in sample, or number of reef species known in region. Data from Table 8. Note that for second figure a log scale is used.
WA
_] Flores Sp. ES Ouvea Sp. [J Chest. Sp. NC Sp.
% species number
a) b) ” 7) Oo £ 2 fe} 2 : N . =N % EN =f i oo 5 3) S o <= io} Ss N 5 s O fo) C)
Figure 5: comparison of trophic structure (a: species, b: density, c: biomass) of fringing reefs Flores with New Caledonia: Ouvéa (Kulbicki et al., 1994a), Chesterfield islands (Kulbicki et al., 1989), main island (NC) (Kulbicki (1991). Pi: piscivores; Cl: macrocarnivores; C2: microcarnivores; Zoo: zooplankton feeders; Other P.: other plankton feeders; MaH.: macroalgae feeders; MiH.: microalgae feeders; Cor.: coral feeders; De.: detritus feeders
13
Flores S. ES Ouvea S. Chest. S. NC S.
Life-history strategy classes
a) L] Flores D. ES OuveaD. [J Chest.D. N NCD. 70 7 60 | > 50 @ 40 ® So 307 32 20 10 (0) ce ss CoE a 1 2 3 4 5 6 Life-history strategy classes b) Flores B. 3 OuveaB. [:] Chest. B. NC B. Life-history strategy classes c)
Figure 6: comparison of life-history strategy classes in Flores and New Caledonia. Key same as Figure 5.
14
Table 1: definition of the 6 life-history strategy classes used for defining structure. Life length can be considered as life expectancy (LSO after recruitment)
Class Size Reproduction Behavior Growth Mortality Life length
1 Small to Very early in life Most species Very fast High 0.5 to 3 medium __—~ Very high gonado-somatic school years < 30cm index or reproductive Simple sexual effort behavior 2 Small to 1-3 years old at first Often schools, Rapid initially Medium 3 to7 years medium reproduction may be < 30cm High gonado-somatic territorial index Sexual behavior may be complex 3 Medium to 2-3 years old at first Often schools, Rapid initially Medium 3 to7 years large reproduction seldom or through > 30 cm High gonado-somatic territorial life index Simple sexual behavior 4 Small to Late in life Seldom Slow after Low 7 to 12 medium - Usually > 50 % maximum schools first years < 30cm size at first reproduction Often reproduction Medium gonado-somatic territorial _— initial growth index often fast 5 Medium to Late in life Seldom Slow after Low 7-12 years large Usually > 60% maximum schools first > 30cm size at first reproduction Often reproduction usually Low gonado-somatic index __ territorial Often rapid >50cm initial growth 6 Large to Very late in life Almost never Veryslow Verylow > 12 years very large Usually > 60% maximum schools especially > 50cm size at first reproduction except for after usually > 1m Often ovoviviparous reproduction reproduction
Low gonado-somatic index
15)
Table 2: composition of substrate. Depths in m. All other numbers are percentages.
STATION NUMBER
1 2 3 4 5 6 Total SUBSTRATE Sand - muddy 12 6 8} Sand - fine Sand - coarse WZ, 5) 5 17 11 8 10 Gravel and Debris 3 7 10 24 7 36 16 Small boulder 3 3 2 10 SY 16 14 Large boulder 23 3 4 7a) 18 34 22 Rock 47 41 71 28 3 5 33 Beachrock 8 8 3 TOTAL 100 100 100 100 100 100 100 ORGANISMS Algae 5) 1 Coral 13 <1 <1 D, Alcyonarians 15 DEPTH RANGE 3/9 2/9 2/4 2/10 1/12 7/9 1/12
Table 3: major fish families and their contribution to total diversity and comparison with New Caledonia (NC)
Family Number of %total Species in | Family Number of %total Species in species species common with NC
Serranidae : Labridae :
Caesionidae 7 Da, 5 Scaridae 15 5.9 13 Mullidae 8 3a 7 Acanthuridae 16 6.3 £5 Chaetodontidae 15) D9) 13 Siganidae
Pomacanthidae 7 De 5 Balistidae
Pomacentridae 49 19.2 42 Total 197 77 170
16
Table 4: density (fish/m?) and biomass (g/m?) of the major families and species.
FAMILIES DENSITY BIOMASS SERRANIDAE 0.099 6.36 Pseudanthias squamipinnis 0.047 0.17 Cephalopholis urodeta 0.013 0.99 Epinephelus fasciatus 0.010 1.10 LUTJANIDAE 0.021 3.95 Lutjanus decussatus 0.015 2S LETHRINIDAE 0.025 Sle Lethrinus harak 0.006 E92 Monotaxis grandocculis 0.012 1.85 NEMIPTERIDAE 0.040 4.28 Scolopsis bilineatus 0.021 1.67 MULLIDAE 0.042 9.48 Parupeneus indicus 0.003 4.53 Parupeneus trifasciatus 0.021 1.10 CHAETODONTIDAE 0.049 1.67 POMACANTHIDAE 0.044 29, POMACENTRIDAE 4.954 18.4 Chromis amboinensis 0.163 0.64 Chromis xanthura 0.226 0.23 Neopomacentrus azysron 0.139 0.48 Pomacentrus amboinensis 0.074 0.31 Pomacentrus brachialis 0.103 0.63 Pomacentrus coelestis 3.468 10.4 LABRIDAE 0.374 7.86 Cirrhilabrus cyanopleura 0.027 0.11 Cirrilabrus sp. 0.027 0.06 Halichoeres melanurus 0.056 0.29 Novaculichthys taeniourus 0.004 1.09 Thalassoma amblycephalum 0.048 0.23 SCARIDAE 0.106 33:1 Scarus spp. juvenile 0.052 13.0 Scarus fasciatus 0.016 5.56 Scarus quoyi 0.014 6.94 ACANTHURIDAE 0.132 18.4 Acanthurus leucocheilus 0.033 2.50 Ctenochaetus striatus 0.059 5.60 Naso hexacanthus 0.008 2.14 SIGANIDAE 0.023 4.85 BALISTIDAE 0.065 4.84
TOTAL fas 187
Table 5 : trophic structure. All numbers are percentages.
CATEGORY DIVERSITY DENSITY BIOMASS Piscivores 11.9 Mp 8.4 Macrocarnivores DD 43 19.3 Microcarnivores 14.2 6.5 3.8 Zooplanktivores Dey Se) 29.9 Other planktivores 0.1 0.1 0.1 Macroherbivores 2 0.1 0.8 Microherbivores 20.5 ee 34.9 Coral feeders Soe! 0.5 0.9 Detritus feeders 2.0 9.2 D0)
Table 6: distribution of the life-history strategies. All numbers are percentages. Classes refer to the classification given in table 2.
LIFE-HISTORY STRATEGY DIVERSITY DENSITY BIOMASS CLASS 1 10.0 61.6 8.2 D, 39.8 eS Sled! 3 16.1 5.8 36.4 4 Zk 3.8 3} 5) 10.0 1S 10.0 6 2.8 0.1 Med
Table 7: species richness (species /transect), density (fish/m?), biomass (g/m?) from fringing reefs in New Caledonia (SW lagoon, Chesterfield and Ouvéa)(Kulbicki, 1991; Kulbicki et al., 1989, 1994a).
REGION SPECIES RICHNESS DENSITY BIOMASS Chesterfield 64 QBS) 90/200 Ouvéa 85 2.4 340
SW Lagoon 55 2.2/5.8 61/155
18
Table 8: number of species (N) contributing to more than 2% of density or biomass for Flores and other fringing reefs in the Pacific. Sampled species: number of species sampled. Potential species: number of reef species known in the area; %N: percentage of N in the number of species recorded during the survey.
1: Kulbicki unpublished; 2: Kulbicki et al., 1994a; 3: Kulbicki et al. 1989; 4: Galzin, 1985; Hayes etaliy1982
Region N density %Ndensity Nbiomass %Nbiomass Sampled Potential Land are species species (km?) Flores 6 3) 10 3.9 255 1140 ~10 000 New Caledonia (1) 10 29 11 3.2 348 940 20 000 Ouvéa (2) 14 ies 8 4.3 152 630 130 Chesterfield (3) 14 10.8 10 7.8 130 550 10 Moorea (4) 6 I>) 80 630 130
Hawaii (5) 9 4.8 187 460 =500
19
Appendix 1: list of species observed. St: number of stations where species was observed; N: total
number of individuals seen; Sch.: average size of schools; Size: average size in cm
NAME
Taeniura lymma
Plotosus lineatus
Saurida gracilis
Synodus variegatus Synodus dermatogennis Synodus spp.
Sargocentron caudimaculatum Aulostomus chinensis Pterois antennata
Pterois volitans Pseudanthias squamipinnis Pseudanthias tuka Anyperodon leucogrammicus Cephalopholis argus Cephalopholis cyanostigma Cephalopholis leopardus Cephalopholis microprion Cephalopholis miniata Cephalopholis sexmaculatus Cephalopholis spiloparea Cephalopholis urodeta Epinephelus cyanopodus Epinephelus fasciatus Epinephelus hexagonatus Epinephelus merra
Variola louti
Variola albomarginata Pseudochromis exquisitus Pseudochromis paccagnellae Apogon fraenatus
Apogon nigrofasciatus Cheilodipterus lineatus Malacanthus latovittatus Carangidae spp.
Caranx para
Caranx tille
Caranx spp.
Gnathanodon speciosus Lutjanus decussatus Lutjanus fulvus
Lutjanus rivulatus Lutjanus vittus
Macolor niger
Caesio cuning
Caesio lunaris
Pterocaesio chrysozona Caesio xanthonota
2
NPN NY HH KH KN DH WK KP KN KN KH RK KN WK K DN WK KK WWNnN DN WH OK NH RK We Ke eS
Nn i=)
KS NOK DK KK WH eS
100
— WD We WD HAwW NY NM VY
— lon
Se UMW pe HP OW HN NY OY
18 232) 17.3 15.2 W267)
NAME
Pterocaesio diagramma Pterocaesio teres Pterocaesio tile Plectorhinchus picus Lethrinus olivaceus Lethrinus harak Lethrinus rubrioperculatus Monotaxis grandoculis Pentapodus caninus Scolopsis affinis Scolopsis bilineatus Scolopsis lineatus Scolopsis margaretifer Mulloides flavolineatus Parupeneus barberinus Parupeneus bifasciatus Parupeneus cyclostomus Parupeneus indicus Parupeneus macronema Parupeneus trifasciatus Upeneus tragula
Platax orbicularis Chaetodon adiergastos Chaetodon baronessa Chaetodon citrinellus Chaetodon kleinii Chaetodon lineolatus Chaetodon lunula Chaetodon melannotus Chaetodon ornatissimus Chaetodon pelewensis Chaetodon rafflesi Chaetodon trifascialis Chaetodon trifasciatus Chaetodon vagabundus Chaetodon xanthurus Heniochus varius Centropyge bicolor Centropyge tibicen Centropyge vrolicki Genicanthus lamarcki Pomacanthus imperator Pomacanthus xanthomethopon Pygoplites diacanthus Abudefduf saxatilis Acanthochromis polyacanthus Amblyglyphidodon aureus
St
KBWNWNNK DHE NPE UWHWHE NEP HE NEP KP PEP SPE NHK DPEPRWRKE WHE DRE UEP NHK EP wD eG
= S| ne
i) i)
—
ins DNDN WOOWWN KH WD
NO nA ©
20
NAME Amblyglyphidodon curacao
Amblyglyphidodon leucogaster
Amphiprion clarkii Amphiprion melanopus Amphiprion perideraion Chromis amboinensis Chromis atripectoralis Chromis atripes
Chromis viridis
Chromis chrysura Chromis flavicauda Chromis flavomaculata Chromis margaritifer Chromis retrofasciata Chromis vanderbilti Chromis spp.
Chromis xanthura Chromis weberi Chrysiptera rex Chrysiptera rollandi Chrysiptera talboti Dascyllus aruanus Dascyllus melanurus Dascyllus reticulatus Dascyllus trimaculatus Discistodus melanotus Neopomacentrus azysron Neopomacentrus nemurus Neopomacentrus violascens Paraglyphidodo nigroris Neoglyphidodon crossi Plectroglyphidodon dicki Plectroglyphidon lacrymatus Pomacentrus alexanderae Pomacentrus amboinensis Pomacentrus bankanensis Pomacentrus brachialis Pomacentrus coelestis Pomacentrus lepidogenys Pomacentrus philippinus Pomacentrus reidi Pomacentrus simsiang Pomacentrus sp. Pomacentrus taeniometopon Pomacentrus vaiuli Cirrhitichtys falco Paracirrhites forsteri Sphyraena barracuda Sphyraena japonica Anampses caeruleopuncta Bodianus mesothorax
2) road
BWOrrP Ke HB PWWNYK WN DWW WH PP KP WWE NH WNKH KH UNDUADWK KEP NNN KH WHEN KEN HK HK NH WwW
NAME
Cheilinus celebicus Cheilinus chlorourus Cheilinus diagrammus Cheilinus fasciatus Cheilinus trilobatus Choerodon anchorago Cirrhilabrus exquisitus Cirrhilabrus cyanopleura Cirrhilabrus sp.
Coris gaimard
Coris schroederi Diproctacanthus xanthurus Epibulus insidiator Gomphosus varius Halichoeres argus Halichoeres chrysus Halichoeres hortulanus Halichoeres melanurus Halichoeres miniatus Halichoeres prosopeion Halichoeres podostigma Halichoeres nebulosus Halichoeres scapularis Hemigymnus fasciatus Hemigymnus melapterus Hologymnosus annulatus Hologymnosus doliatus Labrichthys unilineatus Labroides bicolor Labroides dimidiatus Macropharyngod meleagris Macropharygodo ornatus Novaculichthys taeniourus Pseudocheilinu evanidus Pseudocheilinu hexataenia Pseudocheilinu octotaenia Pseudodax mollucanus Stethojulis bandanensis Stetholulis interrupta Stethojulis strigiventer Stethojulis trilineata Thalassoma amblycephalum Thalassoma hardwicke Thalassoma janseni Thalassoma lunare Scarus spp.
Cetoscarus bicolor Scarus bleekeri
Scarus altipinnis
Scarus dimidiatus
Scarus flavipectoralis
~ oo
NK KE DK AADANNWNNRKP KF KF WK NWN DPRK KP PEN RP RP UN WNnNNK PK NWWN WD KN WHY
RWW We PNY W
oo Wo
—_ (Gey ey ee ee
Nn os)
10.4
— (oe) —_— ~~) — 00
— ee eo ee dO oe, oe, a
eet et — ee Won fp ~TIN
1.3
NAME
Scarus fasciatus Scarus forsteni
Scarus microrhinos Scarus niger
Scarus oviceps
Scarus psittacus Scarus quoyt
Scarus prosognathos Scarus sordidus Parapercis clathrata Parapercis cylindrica Parapercis multiplicata Parapercis tetracantha Ecsenius bandanus Ecsenius bicolor Ecsenius midas
Plagiotremus rhinorhynchos
Amblygobius rainfordi Istigobius decoratus Ptereleotris evides Ptereleotris heteroptera Valenciennea strigatus Acanthurus mata Acanthurus fowleri Acanthurus dussumieri Acanthurus nigricans Acanthurus blochii Acanthurus lineatus Acanthurus nigrofuscus Acanthurus leucocheilus Acanthurus olivaceus Acanthurus pyroferus Ctenochaetus binotatus Ctenochaetus striatus Naso hexacanthus Naso lituratus Paracanthurus hepatus Zebrasoma scopas Siganus argenteus Siganus canaliculatus Siganus corallinus Siganus doliatus Siganus puellus Siganus vulpinus Zanclus cornutus Rastrelliger kanagurta Amanses scopas Aluterus scriptus Balistapus undulatus Balistoides viridescens Melichthys vidua
n oo
BEA HBB BPWHAN WR DHY HE UANN HK HNN KKH KH eK Ke PN HEP Hanne aernweyr Oe WHE WDA |
23
NAME St Odonus niger
Pervagor melanocephalus Rhinecanthus verrucosus Sufflamen bursa Sufflamen chrysopterus Arothron meleagris Arothron nigropunctatus
WN NY Fe W
Canthigaster solandri
N
65
Sch. 16.2
aN
Size 13.7
$4 nies on vga)
ninnabayg: oma rvyiny ava
te a wise Te MAA
t Be ag 54 senor aR, ys vivaess thea sgh ahh Pee janis ran as
rete ea) preg re ee wears Bs
lanes ay, Soyk ce apilinnd
SORA 9
A a
r Ney od t ey "
Mi, 1 ae
Re nih
a ue, eee |
ane ah Aled in ads ass Ruf. fl S: !
Sey
f 4 aly, Aiea atgeanee M j
ert |
if ili Bad rabbi
ot.” ays ny an GY Cpe s eee
ae a a
oh
4 é
a %
% i ‘ th ‘a ey i 1 ; i wi &
,
: |
j i
}
A ;
’ New “ ty a
om, {2 t |
4 Beh or) y By a ub f deeb !
WA A uta rierver ath
Woe fae white
—
| 0 ig tal knee
i
= ss iP
sr > es
—
= ae
—_—_— — = - &
— = =
a oe oe -
“ Jy a5
ti aber a J vin mab
apse tuntniycanid yee se : al Fo j
nleyag tat .
ATOLL RESEARCH BULLETIN
NO. 438
GROUPER DENSITY AND DIVERSITY AT TWO SITES IN THE REPUBLIC
OF MALDIVES
BY
ROBERT D. SLUKA AND NORMAN REICHENBACH
ISSUED BY NATIONAL MUSEUM OF NATURAL HISTORY SMITHSONIAN INSTITUTION WASHINGTON, D.C., U.S.A. APRIL 1996
Tp) a) > — (am) =] << = io GaN noe otal?
GROUPER DENSITY AND DIVERSITY AT TWO SITES
IN THE REPUBLIC OF MALDIVES
BY
ROBERT D. SLUKA! AND NORM REICHENBACH?
ABSTRACT
The density and diversity of shallow-water groupers at Gaagandu, North Male Atoll and Olhugiri, Thaa Atoll, Republic of Maldives was enumerated using visual transects. Four different habitat types were surveyed: reef lagoon, reef crest, reef slope, and a well- developed lagoonal reef. Twenty-two species in seven genera were recorded. Median densities ranged from 7 to 23 grouper 240 m*. At Gaagandu Island, the reef slope was repeatedly sampled using 20-m belt transects to estimate the efficiency and accuracy of the sampling methodology. Fifteen transects were necessary to estimate the median density of all species within 10% of the reference value and to develop a species list containing 80% of the total number of species observed. The species observed varied in their degree of site attachment. Those species which were most closely tied to their habitat exhibited clumped spatial distributions while those species which ’roamed’ over large areas had random spatial distributions. The number of transects necessary to adequately characterize the median density of a species was related to the degree of clumping in its spatial distribution.
INTRODUCTION
Groupers are an important fishery resource throughout the world and are important predators in coral reef ecosystems. Approximately 30 grouper species occur in the Republic of Maldives. Maldivians prefer to eat tuna and have not developed extensive reef fish fisheries (Anderson et al. 1992). Total reef fish catch is approximately 3000 tons per year (Anderson et al. 1992). At present, we are aware of only two operations exploiting groupers, one of which has had little effect on the grouper population (Sluka unpublished data). A market has developed exporting groupers to other southeast Asian countries and to supply many of the resorts located around North Male Atoll. It is therefore likely that reef fish, especially groupers, will come under increasing exploitation in the near future in the Republic of Maldives. Differences in catch
1 University of Miami, Department of Biology, P.O. Box 249118, Coral Gables, Florida 33124 USA
2 The Oceanographic Society of Maldives, Male, Republic of Maldives
Manuscript received 19 July 1994; revised 19 November 1994
2
composition during exploratory fishing were found between a southern atoll (Laamu) and more northern atolls (Alifu and Shaviyani) (Anderson et al. 1992). Shepherd et al. (1992) reported that the abundance and biomass of all species combined was lower on reef flats that were mined than on unmined reef flats. However, the abundance and biomass of fish on slopes adjacent to mined flats was greater than on slopes adjacent to unmined flats. ‘Four grouper species were among the 20 fish which showed the most dissimilarity between these slopes. Cephalopholis miniata and Variola louti had higher biomass on slopes adjacent to mined flats, while Plectropomus pessuliferus and Gracila albomarginata had higher biomass on slopes adjacent to unmined reef flats.
The difficulties in using visual survey methods such as transects has been reviewed by other authors (De Martini and Roberts 1982; Bortone et al. 1986; Sanderson and Solonsky 1986; Greene and Alevizon 1989). Various techniques for solving problems such as transect width (Sale and Sharp 1983), transect length (Fowler 1987), duration of the survey (St. John et al. 1990), and sample size (Sale and Douglas 1981) have been developed. However, these studies usually involved sampling the whole community and in many cases were specifically directed towards sampling patch reefs. Methodologies for surveying serranids were examined by the Great Barrier Reef Marine Park Authority (1979) and Craik (1981) for the Great Barrier Reef region. Groupers are relatively sedentary and site attached. Survey methods must take into account their cryptic behavior and the likelihood of having a patchy or clumped dispersion pattern. This clumped dispersion could lead to misleading results if only a few samples are collected. The number of samples necessary to accurately assess population density will depend on the degree of clumping in their dispersion pattern.
The density and diversity of groupers was studied at two sites in the Republic of Maldives and related to habitat preferences of the different species. The sample size necessary to accurately estimate the density and diversity of groupers in a specific area was examined using visual belt transects.
METHODS
Habitat characterization: The atolls were divided into three habitat zones: 1) lagoon, 2) reef crest, and 3) reef slope. The habitat was characterized by recording the coverage class of dominant substrate (sand, sand-mud, rubble, and hard reef) and lifeforms (seagrass, algae, sponges, octocorals, and hard coral). Substrata and lifeform information were collected by visually estimating the coverage in a belt of 1 m* quadrats. Coverage was scored in the following categories: 1) < 10%, 2) 10 - 30%, 3) 30 - 70%, and 4) > 70%. In order to convert to cm’ the midpoints of each coverage class were summed for each quadrat and averaged.
Visual surveys: Prior to observation, the observer was trained to accurately estimate length using models of fish with a known size-frequency distribution (Bell et al. 1985). Visual surveys were conducted similarly to GBRMPA (1979). A 20-m transect was placed in a haphazard fashion along a particular depth gradient (parallel to shore). An
3
area 6 m out from one side of the transect was intensively searched for all grouper species and then the diver searched the other side in a similar fashion. The number and size of all groupers observed were recorded. Groupers were placed in one of five size categories: <5 cm, 5-15 cm, 15-25 cm, 25-35 cm, and >35 cm. The depth and time of each survey were recorded. All of the habitat zones had similar sampling effort except the reef slope at Gaagandu, which was more intensively surveyed. A distance of approximately 300 m along the reef slope from 6 to 20 m depth was repeatedly sampled in order to assess the number of transects necessary for reliable estimates of density and diversity. Species identifications were made using Heemstra and Randall (1984), Randall (1992), Randall and Heemstra (1991), and Allen and Steene (1987). When information on species identification differed between sources, Randall and Heemstra (1991) was used. Species presence/absense data was collected at Chicken Island, near Gaagandu, for comparison.
Statistical analysis: Descriptive statistics, histograms, correlations and other calculations were performed using Microsoft Excel® software. The frequency distributions of numbers of groupers observed per transect (240 m”) exhibited various degrees of skewing to the right (Figure la, b). Because of the skewed distributions, medians were considered to characterize the densities better than means. Performance curves based on cumulative medians and species-sample curves were used to determine the number of transect replicates needed to obtain adequate density and diversity estimates for groupers observed in the 48 transects from the slope area (Brower et al. 1990). Medians were compared statistically using a Chi-square procedure (Zar 1984).
For species with median density estimates greater than zero, performance curves were calculated. The performance curves calculated were considered to stabilize when all subsequent cumulative medians fell between the 40th and 60th percentiles calculated from the entire set of 48 transects. The least number of transects required to stabilize the performance curve was considered the number of replicates required for a reliable density estimate. This process was repeated 20 times, with the order of the 48 transects entering the cumulative median calculation being randomized each time. Medians were then calculated from the 20 estimates of replicates required to obtain a reliable density estimate. The median estimates for required replicates were then correlated with the species dispersion pattern using Morisita’s Index of Dispersion (I,) (Brower et al. 1990).
For the density and diversity of all species combined on the reef slope, performance curves and species-sample curves were calculated. The number of replicates required for a stable density estimate was determined in a fashion similar to that noted for individual species except for the criteria used to determine performance curve stability. Instead of using one level for determining stability, i.e. the 40th and 60th percentiles, several levels were evaluated. These levels included 20% of the median (30 and 70 percentiles), 15% (35 and 65 percentiles), 10% (40 and 60 percentiles), and 5% (45 and 55 percentiles). If the median estimated from all 48 transects is considered to be the reference median density, then these different levels for assessing performance curve stability would indicate the accuracy of the median estimated from a given number of transects. The number of replicates based upon the species-sample curves were also
4
48).
is urodeta (n
48) and (b)
Figure 1: Frequency distribution of number of grouper observed per transect for (a) = Cephalopholis
Aethaloperca rogaa (n
a)
Asuenbel4
11S SMS IS
10
A. rogaa (#/transect)
b)
Prrrrrs
OOO II
RrEDLCEetirLeceeeenereLereeLieeLereeLee eles
PETRELELEMERILEL EERE LULLEL ELE LE LeELeerery
OOK OOK HO HHH III A MHI
ANIPIPIID IDI L ILI SL LL LIN IID LLLP LLDPE
DO SESASNAASSA XMAS
Asuenbel4
C. urodeta (#/transect)
5
assessed at various levels of percent of species observed. The levels included >70%, >80%, =90%, and 100%. This process was repeated 20 times, randomizing the order of the transects each time. Medians were then calculated from the 20 estimates of replicates required for each level of percentage of species observed.
RESULTS
Habitat characterization: Gaagandu Island is located inside the main atoll ring of North Male Atoll. The northern and western sides of the island are surrounded by a lagoon approximately 50 m wide and approximately 2 m deep at high tide. The lagoon was primarily rubble with very small areas of sand (Figure 2a). The rubble areas of the lagoon were covered by turfing algae, had no soft coral or sponges, and very little hard coral (Figure 2b). The reef crest consisted of large, eroded coral heads covered by algal turf. The crest had only slightly higher hard coral cover than the lagoon and had very low coverage of sponges and soft coral. From the crest, the reef sloped down steeply to a sand flat at 30 m depth. The reef slope appeared to be divided into areas of high vertical relief separated by ’landslides’ of rubble with sand. The reef slope had the highest percentage cover of hard coral (approximately 30%) and low numbers of sponges and soft coral. The southwestern portion of the island had a well-developed reef consisting of a huge bed of Acropora sp. interspersed by massive coral colonies. This reef is designated as reef 1 for further analyses. The depth ranged from 1-10 m at reef 1 and no substrate/lifeform data was taken at this site.
Olhugiri island is located on the northern edge of the outer ring of Thaa Atoll, approximately 2.35 N latitude, 73.05 E longitude. The lagoon of the atoll stretches approximately 50 m in each direction around the island. The northern side of the island is open to the sea and has a reef crest which slopes steeply down to 50 m where the slope becomes much gentler. The western portion of the island is lagoonal connecting to another island without any deep passages. The inner side of the island has a reef crest which slopes gently to about 10 m into a sand flat. The eastern portion of the island has a channel about 10 m in depth which allows passage of water into the atoll. The outer and inner reef crests were sampled for grouper density and diversity No quantitative habitat data was collected at Olhugiri.
Density _and_ diversity of groupers: There was no correlation between any species abundance, nor total abundance, with depth (minimum, maximum, or mean) or time of day (p> 0.05) along the reef slope. There was a significant difference in the median number of grouper observed per transect between sites (X? = 44.84, df = 4, p < 0.001, Table 1). The slope at Gaagandu had the highest median density with 23 grouper observed per transect. Excluding the slope data, the other sites had no significant differences in the median number of grouper observed per transect (X? = 4.74, df = 3, p > 0.05). The lagoon at Gaagandu had a median density of 5 and the lagoon at Olhugiri 16. These two sites were not included in the density comparisons due to the low sample size (2 and 4 transects, respectively).
6
Figure 2: Substrata (a) and Lifeform (b) coverage of the site at Gaagandu Island, North Male Atoll. Open bars represent the slope area (n=100 1 m? quadrats), solid bars represent the reef crest (n=100), and striped bars represent the reef lagoon (n=40). (a) S = sand, RB = rubble, and HR = hard reef. (b) AT = algae, SP = sponge, SC = octocoral, and HC = hard coral.
a)
Percent Cover
Substrata
b)
Percent Cover
\ \ N N \ \ \ \ \ NN: NY
Lifeforms
qj
Table 1: Median, maximum, and minimum number of grouper observed per 240 m? transect within each zone at the two island sites.
GAAGANDU OLHUGIRI INNER OUTER CREST SLOPE REEF 1 CREST CREST MEDIAN 7 7a) 2 10 JUGS) MAXIMUM 13 50 18 24 15 MINIMUM 3 11 4 3 7
The lagoon at Gaagandu was characterized by low diversity (4 species). There were 7 species observed on the reef crest, dominated by Cephalopholis argus and C. urodeta (Table 2). Reef 1 was dominated by C. argus and Epinephelus merra. The slope had the highest diversity with 17 species (also the largest sample size). Cephalopholis miniata, C. leopardus, C. urodeta, E. spilotoceps, and C. argus dominated numerically in decreasing order of importance. Along the slope the densities of ’roving’ species, such as G. albomarginata, Variola louti, and Plectropomus spp., were probably underestimated; these species were frequently observed swimming along the reef slope, but outside transect boundaries. Overall, the species of Cephalopholis tended to dominate numerically with many Epinephelus spp. being rarely observed. The Epinephelus groupers commonly observed (E. spilotoceps, E. merra, and E. macrospilos) were similarly colored, a white to cream background with brown spots or hexagonal markings.
The inner reef crest of Olhugiri had 16 species present and the outer reef crest 15. The dominant species on both reefs was C. argus, with a median number per transect of 7 inside and 6 outside (Table 2). C. leopardus and E. spilotoceps were the second most abundant species on the inner crest, whereas C. urodeta was second most abundant on the outer slope.
Length-frequency distribution: The majority of grouper observed in the lagoons at Gaagandu and Olhugiri were small (5-15 cm Total Length (TL)). No groupers were observed over 25 cm TL. The reef crest and slope had similar size - distributions (X? = 7.07, df = 3, p > 0.05). The < 5 cm and 5-15 cm categories were combined due to an expected value < 1 (Everitt 1992). The majority of grouper observed were 5-25 cm TL. On the slope the smaller grouper (5-15 cm) were dominated numerically by Cephalopholis leopardus and C. urodeta. The largest fish observed on the slope (> 35 cm) were Anyperodon luecogrammicus, Aetheloperca rogaa, C. argus, Variola louti, E. polyphekadian, and C. miniata. Fish observed were mostly less than 50 cm TL. Fish greater than 50 cm were mostly V. louti and P. laevis. The larger grouper observed on the reef crest (25-35 cm) were C. argus. Reef 1 had similar numbers of fish in the 5-15 cm, 15-25 cm, and 25-35 cm categories when compared to the other sites at Gaagandu (X? = 0.43, df = 2, p > 0.05). Reef 1 had a larger percentage contribution of the >
8
Table 2: Median and maximum number of groupers observed per transect (median, maximum) for Gaagandu slope (GS), Gaagandu crest (GC), Gaagandu lagoon (GL), Gaagandu reef 1 (GR), Olhugiri inside crest (OI), Olhugiri outside crest (OO). The minimum number observed per transect was zero except * = 3, + = 1, and # = 2. % = species observed outside boundaries of transects
SPECIES GS GC” Gl. (GEKea Or OO
Number of transects 48 11 2 12 13 6 Aethaloperca rogaa 156 --- --- = 0,1 ee Anyperodon luecogrammicus 1,4 0,2 --- 12 0,2 --- Cephalopholis argus 3,11 Bhi vatgaas 5: naeOs te (OF C. leopardus 4.5,16 --- --- 0,1 1,6 Ua C. miniata 5: aes) ele -<aw yO ae C. sexmaculata 0,2 = a ae aes ae C. spiloparea 0,4 _ al ns is ae C. urodeta De OMe LO pe aa (0) pane 72 S500 Epinephelus caeruleopunctatus 0,1 0,1 oo 0,1 0,1 Om E. fasciatus = aes % ee es a E. fuscoguttatus 0,1 ee — owe ie ae E. macrospilos 0,1 | | I = --- --- E. merra --- O'S ..2 B95; S18. 5169 (O22 = E. ongus 0,1 --- O51 .051 oo --- E. polyphekadian 0,2 --- --- 0,1 = 0,1 E. spilotoceps 314 03 -- 0,3 1,4 0.4 E. tauvina --- = =e ese 0,1 id Gracila albomarginata 0,2 --- a = 0,1 ect Plectropomus areolata --- --- = 0,2 0,1 0,1 P. laevis 0,1 --- --- --- 0,1 ---
P. pessuliferous = AY aes £5 en at
Variola louti 0,2 = —_ ? cic mr
9
35 cm category than the other sites at Gaagandu. These larger grouper were mainly C. argus with a few A. luecogrammicus.
There was a significant difference in the length-frequency distributions of groupers on the inner and outer crests at Olhugiri island (X? = 12.51, df = 4, p < 0.05). Many small (< 5cm) C. leopardus were observed on the inner crest, whereas only 1 < 5 cm C. urodeta was observed on the outer crest. There were more smaller (5-15 cm) grouper and fewer larger (25-35 cm) grouper on the inner crest than would be expected if the two size-frequency distributions were similar. Alternatively, there were fewer smaller (5-15 cm) grouper and more larger (25-35 cm) grouper on the outer crest than would be expected.
Similarity index: The similarity in species composition was compared using Jaccard’s coefficient, which is based on species presence/absence data (Table 3). The reef slope at Gaagandu was most similar to reef 1 and Chicken Island (53%). The rest of the sites at Gaagandu were less than 50% similar, with the lagoon the least similar to the reef slope and reef 1. The Olhugiri reef crests were most similar to each other (82%).
Sample number: Seven species in the slope area had median densities greater than zero (Table 2). The median number of transects necessary for a reliable density estimate ranged from 2 to 16 (Table 4). The number of transects needed was related to the degree the species exhibited a clumped distribution as indicated by their I, values (r = 0.73, p = 0.06). Two species, Anyperodon leucogrammicus and Aethelaperca rogaa, had I, values which were not significant or nearly so; this indicated their dispersion patterns were not significantly different from random.
These species required only a few transects to determine their density. In contrast,
the other 5 species showed various degrees of clumping and required more transects to reliably estimate their densities (Table 4).
For all species combined, the number of transects needed for an accurate survey ranged from 7 to 37 depending upon the level of accuracy desired for the median density and the percent of the species observed (Figure 3). Increasing the number of transects from 7 to approximately 15 provided a large increase in the accuracy of the median density estimate and percent of species observed. The accuracy of the estimate of median density increased from 20% to approximately 10% of the reference median density, while the percent of species observed increased from 70% to over 80%. Further increases in the number of transects provided more moderate increases in the accuracy of the median density estimate and percent of the species observed.
10
Table 3: Similarity matrix of Jaccard’s coefficient comparing the presence - absence of species among survey sites.
SURVEY SITE 1 2 3 4 5 6 7 8 1. Gaagandu slope 1.00 2. Gaagandu crest 0.33 1.00
3. Gaagandu lagoon 0.11 0.38 1.00
4. Gaagandu Reef 1 0.53 0.46 0.14 1.00
5. Chicken Island 0.53 0.46 0.00 0.36 1.00 6. Olhugiri inside Os7e" 0°39" OC. O565 70°47) =£-00 crest 7. Olhugiri outside 0.68 0.29 0.06 0.59 0.60 0.82 1.00 crest 8. Olhugiri lagoon Only O38" Or53. O53) "Orte Ol25° Oro eee
Table 4: Morisita’s index of dispersion (I,) in relation to the median number of transects necessary for a reliable density estimate for the 7 most common species of grouper observed in the slope zone at Gaagandu Island, North Male Atoll. Chi-square test Statistics and associated probability levels indicate whether or not the species’ dispersion pattern was significantly different from a random distribution.
MEDIAN NO. SPECIES TRANSECTS I, Ss P Anyperodon Z 1.31 64.5 0.045 luecogrammicus Aethaloperca rogaa 3 bel5 57.9 0.133 Cephalopholis u 1.44 119.0 < 0.001 spilotoceps C. argus 11 1.47 111.4 < 0.001 C. miniata ihe. 1.34 136.9 < 0.001 C. leopardis 15) 1.38 141.7 < 0.001
C. urodeta 16 1.91 220.0 < 0.001
11
Figure 3: Number of transects needed to obtain a desired level of accuracy in estimating the median number of groupers per unit area (dashed line) and percent of all species observed (solid line) on the slope at Gaagandu (reference values are 23 for median density and 17 for total number of species observed).
100 20 18 95 9 16 = rye a re) 14 s $ 3 @ 85 12 < C¢p) c - $ S 10 6 8 80 o 8 75 6 70 4 0 5 10 15 20 25 30 35 40
Number of Transects
DISCUSSION
Habitat can be viewed on a number of different scales. The density and distribution of groupers were related to within and among zone differences in habitat type. First, at the macro-scale, there were clear differences in the density and diversity of groupers at Gaagandu Island between the lagoon, crest, reef 1, and the slope. The slope had a higher sampling effort so that rarer species were more likely to be observed. These different zones vary in the amount of refuge available for groupers. The lagoon and crest had little relief. The lagoon at Gaagandu has been mined for coral (M. Haleem pers. com.). The lagoon at Olhugiri has not been mined extensively and still has large coral heads. The density at the lagoon at Olhugiri exceeded all sites except the slope at Gaagandu. This indicates that the lagoon at Gaagandu probably supported a much higher density of groupers prior to mining. Reef 1 had high relief, but consisted mainly of dense thickets of Acropora sp., which might have limited their use by certain species as the interstices were probably too small for movement and hiding (the dense thickets most likely inhibited the efforts of the surveyor as well). Harmelin-Vivien (1977) found that spur and groove reefs at 6-18 meters depth had more species of fish and a higher biomass than the deeper sloping platform.
12
Within the different zones the species were associated, to varying degrees, with specific features. Some species had very little association with structural features of the zone such as the species of Plectropomus, Variola louti, and Gracila albomarginata. These species were observed to freely roam large areas generally > 15 meters deep. Variola louti was not observed in caves or hiding in the Society Islands, but swam off the bottom (Randall and Brock 1960). Gracila albomarginata was observed frequently in shallow water 5-10 m, however, Randall and Heemstra (1991) reported that this species was more abundant in depths greater than 15 m. This species tended to swim along the slope and did not appear to hide when frightened, but swam away, as is consistant with Randall and Heemstra’s (1991) observations. Smith-Vaniz et al. (1988) also indicated that this species was an active swimmer, not resting on the reef substratum. Plectropomus areolata appeared more substrate attached; the younger ones were observed swimming among the Acropora thickets on reef 1. The species of Plectropomus feed mainly on fishes and tend to be less sedentary than most groupers (Randall and Hoese 1986). Aethaloperca rogaa tended to be intermediate between these free-roaming species and the more substrate attached species. Individuals tended to swim about freely, but would often hide under coral heads and ledges when approached. They did not traverse long distances as did the previously mentioned species, but would remain near a large coral structure in the water column.
The reef slope contained areas with high coral relief, in between which occurred *landslides’ of coral rubble and sand. Stoddart (1966) documented these same features of Maldivian reefs. These rubble patches were frequently inhabited by small Cephalopholis urodeta and, especially, Epinephelus spilotoceps. The latter species was usually observed on the edge of these rubble patches near high coral relief rather than out in the open. Epinephelus merra was abundant in the lagoons of the islands and at reef 1. This species is similar to E. spilotoceps, being a demersal carnivore living under ledges near the bottom of coral mounds and rubble (Hiatt and Strasburg 1960). E. merra is typically found in shallow water on patch reefs in lagoons and bays (Heemstra and Randall 1993). Many C. urodeta observed had a coloration with the posterior 1/3 to 1/2 of the fish black. Species descriptions of this fish indicate that the Indian Ocean variety has only a dark caudal fin, but that in "dark habitats" in the Comoros Islands it was uniformly brown (Randall and Heemstra 1991). Small specimens (< 10 cm) of C. urodeta were observed in shallow water that appeared uniformly black or with a red head region and black body posteriorly. Most of the individuals of this species conformed to the species description in Randall and Heemstra (1991), however many followed this pattern of more extensive black coloration on the posterior 1/3 to 1/2 of the body and the soft dorsal and anal fins. Cephalopholis urodeta is strongly demersal and rarely ventures away from shelter (Hiatt and Strasburg 1960). The most site attached of the slope species was C. leopardus. It was always seen within patches of coral with closely set *finger’ arrangements. When approached it would dart into the coral head. Anyperodon luecogrammicus was often seen in pairs. Cephalopholis sexmaculata was observed only in Caves as iS consistent with the observations of Randall and Ben-Tuvia (1983).
13
C. argus tended to have a higher density at shallower depths and dominated the diversity on the reef crest. This species is one of the most common food fishes (Randall et al 1985), and is generally one of the most abundant piscivores at most locations thoughout the Indo-Pacific (Randall and Ben-Tuvia 1983). It is more common on exposed rather than protected reefs (Randall and Brock 1960) and prefers depths of 1-10 m (Heemstra and Randall 1993). Shpigel and Fishelson (1989) found this species on the shallow reef table and reef wall in the Gulf of Elat. Harmelin-Vivien (1977) observed C. argus at depths of 6-18 m on spur and groove reefs and 18-25 m on the lower sloping platform at Tulear. Cephalopholis miniata is abundant in deep lagoons and dominates coral knolls that are isolated at depths of 17-33 m (Randall and Brock 1960). At one knoll off the slope at Gaagandu at 30 m depth, this species was the most numerous of the groupers observed. The grouper species observed on the reef crest and lagoon were in close association with structural features such as overhangs and crevices (with the exception of C. argus, which roamed about freely while darting into cover when approached). The species observed in the lagoon were all similarly colored (brown spots or hexagons on a light background) and tended to blend into the background of algal covered rubble. Hiatt and Strasburg (1960) found E. macrospilos under large coral heads and rock ledges, seldom far from cover. Our observations on this species in the lagoon at Gaagandu support their findings. Epinephelus fasciatus was observed in the lagoon closely associated with shelter. Fishelson (1977) observed this species near rocks in the lagoon of the Gulf of Eilat (Aqaba) as well as in the fore reef.
The number of transects required to adequately characterize grouper density and diversity is dependent upon the dispersion patterns and the desired levels of precision, accuracy, and percent of the species observed in the community. A single visit to a reef is not likely to record all species present, especially cryptic ones (Sale and Douglas 1981). An analysis similar to that conducted here could be done on a preliminary set of transects in order to determine the number of transects required. The number of transects should be determined not only by the dispersion patterns of the species of interest, but also by logistical constraints on effort. Collecting a large sample might increase accuracy minimally and use time that could be applied to other sites (Bros and Cowell 1987). In addition, if only species densities are required, the level of effort devoted to a particular species could be tailored to the degree to which a species is clumped. Only a few transects would be required to characterize the density of a randomly dispersed species, while a species which is clumped would require more transects.
The groupers observed in this study appeared to have specific habitat requirements or preferences. The dispersion of the groupers throughout the site is probably related to the dispersion of their preferred habitat. Cephalopholis leopardus is strongly substrate attached and its distribution was significantly clumped (Table 4). The clumped distribution of the species is likely due to a clumped distribution of its preferred habitat. Thirteen transects would be needed to adequately characterize the density of this species whereas a species such as Aethaloperca rogaa which had a random distribution (Table 4), would need only 3 transects. A. rogaa is a species which is not strongly substrate attached. However, our data on Anyperodon luecogrammicus does not follow this pattern as it was randomly dispered, but appears to be strongly substrate attached. A
14
more detailed investigation of its habitat might reveal that it is a generalist in its association with the substrate.
ACKNOWLEDGEMENTS
We gratefully acknowledge the help of Mohamed Haleem, Omar Maniku, Ahmed Shakeel, and Steve Holloway. Without their contributions this research could not have been accomplished. We also thank the men of Gaagandu and Olhugiri Islands for helping with the research and providing a great living environment. The manuscript was significantly improved by two anonymous reviewers. This project was sponsored by the Oceanographic Society of Maldives.
LITERATURE CITED
Allen, G.R. and R.C. Steene. 1987. Reef fishes of the Indian Ocean. T.F.H. Publications, New Jersey. 240 pp.
Anderson, R.C., Z. Waheed, M. Rasheed, and A. Arif. 1992. Reef fish resources survey in the Maldives - Phase II. Bay of Bengal Program BOBP/WP/80, Madras, India.
Bell, J.D., G.J.S. Craik, D.A. Pollard, and B.C. Russel. 1985. Estimating length- frequency distributions of large reef fish underwater. Coral Reefs 4:41-44.
Bortone, S.A., R.W. Hastings, and J.L. Oglesby. 1986. Quantification of reef fish assemblages: a comparison of several in situ methods. Northeast Gulf Science 8:1-22.
Bros, W.E. and Cowell, B.C. 1987. A technique for optimizing sample size (replication). J. Exp. Mar. Biol. Ecol. 114:63-71.
Brower, J., J. Zar, and C. von Ende. 1990. Field and Laboratory Methods for General Ecology. Wm. C. Brown Publishers, Dubuque, IA, 237pp.
Craik, G.J.S. 1981. Underwater survey of coral trout Plectropomus leopardus (Serranidae) populations in the Capricorn section of the Great Barrier Reef Marine Park. Proc. 4th Int. Coral Reef Symp. 1:53-58.
De Martini, E.E. and D. Roberts. 1982. An empirical test of biases in the rapid visual technique for species-time censuses of reef fish assemblages. Mar. Biol. 70:129-134.
Everitt, B.S. 1992. The Analysis of Contingency Tables, Second Edition. Chapman & Hall, New York. 164pp.
Fishelson, L. 1977. Sociobiology of feeding behavior of coral fish along the coral reef of the Gulf of Elat (= Gulf of Aqaba), Red Sea. Isr. J. Zool. 26:114-134.
15
Fowler, A.J. 1987. The development of sampling strategies for population studies of coral reef fishes: a case study. Coral Reefs 6: 49-58.
Great Barrier Reef Marine Park Authority (GBRMPA). 1979. Great Barrier Reef Marine Park Authority workshop on reef fish assessment and monitoring. Workshop Series No. 2 GBRMPA, Townsville, Australia. 64pp.
Greene, L.E. and W.S. Alevizon. 1989. Comparative accuracies of visual assessment methods for coral reef fishes. Bull. Mar. Sci. 44:899-912.
Harmelin-Vivien, M.L. 1977. Ecological distribution of fishes on the outer slope of Tulear reef (Madagascar). Proc. Int. Coral Reef Symp. 3rd 1:289-295.
Heemstra, P. and J.E. Randall. 1984. Serranidae. In: Fischer, W. (Ed.), FAO Species Identification Sheets for Fishery Purposes, Western Central Atlantic (fishing area 31). Vol. 4,5. FAO, Rome, Italy.
Heemstra, P.C. and J.E. Randall. 1993. FAO Species Catalogue. Vol. 16. Groupers of the world (Family Serranidae, subfamily Epinephelinae). An annotated and illustrated catalogue of the grouper, rockcod, hind, coral grouper and lyretail species known to date. FAO Fisheries Synopsis No. 125, Vol.16. Rome, FAO. 382pp.
Hiatt, R.W. and D.W. Strasburg. 1960. Ecological relationships of the fish fauna on coral reefs of the Marshall Islands. Ecol. Monogr. 30:65-127.
Randall, J.E. 1992. Diver’s guide to fishes of Maldives. Immel Publishing, London. 193 pp.
Randall, J.E., M.L. Bauchot, and A. Ben-Tuvia. 1985. Cephalopholis argus Schneider, 1801 and Cephalopholis sexmaculata (Ruppell, 1830) (Ostiechthyes, Serranidae: Proposed conservation by suppression of Bodianus guttatus Bloch, 1790, Anthius argus Bloch, 1792 and Serranus zanana Valenciennes, 1828 Z.N.(S.)2470). Bull. Zool. Nom. Vol 42 pt.4:374-378.
Randall, J.E. and A. Ben-Tuvia. 1983. A review of the groupers (Pisces: Serranidae: Epinephilinae) of the Red Sea, with description of a new species of Cephalopholis. Bull. Mar. Sci. 33:373-426.
Randall, J.E. and V.E. Brock. 1960. Observations on the ecology of epinepheline and lutjanid fishes of the Society islands, with emphasis on food habits. Trans. Am. Fish. Soc. 89:9-16.
Randall, J.E. and P. Heemstra. 1991. Revision of Indo-Pacific groupers (Perciformes: Serranidae: Epinephelinae), with descriptions of five new species. Indo-Pacific Fishes 20: 1-332.
16
Randall, J.E. and D.F. Hoese. 1986. Revision of the groupers of the Indo-Pacific Genus Plectropomous (Perciformes: Serranidae). Indo-Pacific Fishes 13:1-31.
Sale, P.F. and W.A. Douglas. 1981. Precision and accuracy of visual census technique for fish assemblages on coral patch reefs. Env. Biol. Fishes 6:333-339.
Sale, P.F. and B.J. Sharp. 1983. Correction for bias in visual transect censuses of coral reef fishes. Coral Reefs 2:37-42.
Sanderson, S.L. and A.C. Solonsky. 1986. Comparison of a rapid visual and a strip transect technique for censusing reef fish assemblages. Bull. Mar. Sci. 39:119-129.
Shepherd, A.R.D., R.M. Warwick, K.R. Clark, and B.E. Brown. 1992. An analysis of fish community responses to coral mining in the Maldives. Env. Biol. Fishes 33:367-380.
Shpigel, M. and L. Fishelson. 1989. Habitat partioning between species of the genus Cephalopholis (Pisces, Serranidae) accross the fringing reef of the Gulf of Aquaba (Red Sea). Mar. Ecol. Prog. Ser. 58:17-22.
Smith-Vaniz, W.F., G.D. Johnson, and J.E. Randall. 1988. Redescription of Gracila albomarginata (Fowler and Bean) and Cephalopholis polleni (Bleeker) with comments on the generic limits of selected Indo-Pacific groupers (Pisces: Serranidae: Epinephelinae). Proc. Acad. Nat. Sci. Philad. 140(2):1-23.
St. John, J., G-.R. Russ, and W. Gladstone. 1990. Accuracy and bias of visual estimates of numbers, size structure and biomass of a coral reef fish. Mar. Ecol. Prog. Ser. 64:253-262.
Stoddart, D.R. 1966. Reef studies at Addu Atoll, Maldive Islands. Atoll Res. Bull. 116. 122pp.
Zar, J.H. 1984. Biostatistical Analysis, 2nd Edition. Prentice-Hall, Inc., Englewood Cliffs, N.J. 718 pp.
ATOLL RESEARCH BULLETIN
NO. 439
EFFECT OF TYPHOONS ON THE LIZARD COMMUNITY OF A
SHELF ATOLL
BY
MICHAEL JAMES MCCOID
ISSUED BY NATIONAL MUSEUM OF NATURAL HISTORY SMITHSONIAN INSTITUTION WASHINGTON, D.C., U.S.A. APRIL 1996
Cocos I.
~ 4 Km is es ae |
Figure |. Map of the Mariana archipelago with the location of the study site.
EFFECT OF TYPHOONS ON THE LIZARD COMMUNITY OF A SHELF ATOLL BY MICHAEL JAMES MCCOID!.2
ABSTRACT
Two major typhoons hit the southern Mariana Islands within an 11 month span and provided a- unique, unplanned opportunity to investigate storm influences on the herpetofauna of an atoll. Habitat specialists (Emoia atrocostata and Cryptoblepharus poecilopleurus) endured the largest population declines because of habitat destruction. All other species, particularly scincids, suffered less drastic population declines. The highest population declines for all species occurred on the developed (resort) end of the island, suggesting that removal and restructuring of typhoon-adapted vegetation allowed complete overwash and local extirpations. Cumulative effects of typhoons suggest a resilience to storm influences by atoll-dwelling reptiles.
The Mariana Islands comprise an archipelago of volcanic origin oriented north-south roughly equidistant between New Guinea and Japan. There are 15 major islands, with the northernmost (Farallon de Pajaros = Uracas) located at approximately 20°N, 145°E and the southernmost (Guam) at 139N, 145°E (Fig. 1). Two km south of Guam, situated on the southern portion of a coral lagoon, is Cocos (Dano) Island. This atoll has a maximum elevation of 2 m and is approximately 100 m by 2 km. As of September 1992, forest vegetation on Cocos Island was dominated by Cocos nucifera (Coconut Palm), Hernandia sonora (no common name) and Casuarina equisetifolia (Australian Pine). Understory vegetation in the forest was dominated by Carica papaya (Papaya) with ground cover dominated by unidentified grasses and Ipomoea pes-caprae (Railroad Vine). Bordering the surf / tidal splash zone on the windward side of the atoll were dense thickets of Pemphis acidula (no common name). Vegetation on the developed (resort) northeastern 1/3 end of the atoll was dominated by C. equisetifolia, C. nucifera, and ornamental trees and shrubs. An historical record detailing vegetation was provided by Neubauer and Neubauer (1981). The atoll has undergone substantial changes during the past half-century including the development of a coconut plantation prior to WWII, construction of a U. S. military installation (formerly occupying approximately 1/4 of the island), two resorts (occupying a total of 1/2 the island; the present resort occupies only 1/3 the atoll), and at least three typhoons since 1949 that overwashed the island (Neubauer and Neubauer, 1981; per. obs.). Only about 1/3 of the island remains as atoll forest, albeit regenerated.
The climate of the southern Marianas is tropical with annual diurnal temperatures
ranging between 22° and 31°C (Anon., 1990). Rainfall is seasonal (Anon., 1990) with
1Division of Aquatic and Wildlife Resources, P. O. Box 2950, Agana, Guam 96910, USA. ;
2Present Address: Caesar Kleberg Wildlife Research Institute, Texas A&M University, Kingsville, Texas 78363, USA.
Manuscript received 19 July 1994; revised 25 April 1995
Z
most occurring between June and December. Typhoons in the western Pacific are common and have been recorded on Guam in most months of the year (Myers, 1991). The typhoon season on Guam is between June and December.
Information on the effects of typhoons on the fauna of atolls is minimal; Jackson (1967) reported that insects and vertebrates persist despite catastrophic impacts and that lizards "somehow have found sufficient protection". Damage to and recovery of vegetation is better documented, with estimates of as long as ten years for a marked recovery (Wiens, 1962). In this unplanned study, I document changes in the herpetofauna of Cocos Island after the cumulative effects of two major typhoons.
While typhoons are a yearly event in the Mariana Islands, two storms of severe magnitude recently hit Guam within an 11 month span. Typhoon Russ hit Guam in December 1990 and Typhoon Yuri in November 1991. Minimum sustained wind speeds to attain classification as typhoons are the same as hurricanes (>74 MPH = 119 KPH) but these storms had sustained wind speeds recorded at 175 MPH (281 KPH). Along southeastern exposures (including Cocos Island), the direction that typhoons usually approach Guam, maximum estimated wave heights were 9 m. Damage caused by high winds and waves, in both typhoons, were substantial on Guam and catastrophic on Cocos Island. Typhoon Russ totally overwashed the atoll, defoliated all broadleaf vegetation, and downed an unknown, but large number of trees, particularly C. equisetifolia along the windward side of the island. Typhoon Yuri inflicted similar damage including loss of a substantial portion of the remaining C. equisetifolia on the windward side of the atoll. An estimated 40-60% of C. equisetifolia on Cocos Island were cumulatively lost during the typhoons. Another cumulative overt vegetation change observed was the virtual elimination of the P. acidula thickets bordering the high energy zone on the windward side of the atoll. An estimated 95% of the thickets were destroyed by Typhoon Yuri. Between typhoons, dominant forest vegetation releafed, seeded, and a dense understory of C. papaya and C. nucifera developed. Also during this period, the remaining P. acidula thickets releafed. Due to Typhoon Yuri, the papaya and coconut palm understory was destroyed and tremendous amounts of debris from the resort were strewn throughout the forest. The dominant understory vegetation that emerged after the second storm was /. pes-caprae.
The herpetofauna of the Mariana Islands has been characterized as depauperate (Rodda, et al. 1991) consisting of a pre-western contact terrestrial reptile fauna of 13 species (McCoid, 1993). Ten of these species occur on Cocos Island (Gehyra mutilata, G. oceanica, Lepidodactylus lugubris, Perochirus ateles, Cryptoblepharus poecilopleurus, Emoia cyanura, E. caeruleocauda, E. atrocostata, E. slevini, and Varanus indicus) and an additional two species (Hemidactylus frenatus and Carlia cf. fusca), both introduced to the Marianas (McCoid, 1993), are established on Cocos Island. At present, Cocos Island possesses the most diverse reptile fauna (12 species) of any island in the Mariana archipelago. Declines in the herpetofauna of the Mariana Islands were discussed by Rodda, et al. (1991) but most species formerly found on Guam still occur on Cocos Island. Although there are no native amphibians on the Mariana Islands, Bufo marinus is established on Guam and Cocos Island.
The pre-typhoon reptile fauna on Cocos Island was not uniformly distributed in all habitats. The gekkonids G. oceanica, H. frenatus, and P. ateles, were found in both developed and forested areas (McCoid and Hensley, 1994), but differences in densities between these habitats were not investigated. Gehyra mutilata and L. lugubris, however, were far more common in the relatively undisturbed forested areas; | encountered only two L. lugubris in the resort area during nocturnal surveys and no G. mutilata (G. Rodda, pers. com., recorded these species in the forest). Scincids were also not evenly distributed in all habitats. Carlia cf. fusca, perhaps introduced as recently as the late 1980's to Cocos Island (T. Fritts, pers. com.) was found only at a boat landing and public park on the western end
8
of Cocos and at the resort on the eastern end on the island in early 1989. By mid-1990, the species was observed in intervening habitats on Cocos Island. By early 1991 (see below), the species was abundant in all areas. Cryptoblepharus poecilopleurus was most conspicuous on the windward (east) side of the island where it commonly occurred on tree trunks in C. equisetifolia groves (Hensley and McCoid, 1994). Generally, any tree with a trunk diameter > 2.5 cm had at least one resident C. poecilopleurus. Emoia cyanura was found in both resort and forest areas but was associated with sunlit, open habitat. Expansive areas of dense undergrowth harbored few individuals. Emoia caeruleocauda favored heavily shaded areas and was common in the forest and resort, but was occasionally found in open areas. Emoia atrocostata was restricted to the high energy P. acidula zone (total habitat 4 ha) on the windward side of Cocos Island. Emoia slevini only occurred in forest (total habitat 9 ha) (McCoid, et al. 1995).
Qualitative surveys of the herpetofauna of Cocos Island were initiated in April 1989 and initially consisted of nocturnal surveys for gekkonids, diurnal surveys for arboreal scincids (both time-constrained surveys), and diurnal surveys for terrestrial scincids using rubberbands. Time-constrained surveys (N = 5, between April 1989 and December 1991) for C. poecilopleurus were limited from 15 to 30 min during which all lizards seen while walking through C. equisetifolia groves were recorded. Time-constrained surveys for gekkonids were conducted on the resort and lasted between 1.5 and 2 h during which all lizards encountered along a predetermined route were either collected or recorded. In September 1990, sticky traps (see Rodda, et al. 1993), which provide a mechanism to estimate relative abundance, were first employed to sample terrestrial reptile faunas in forested, resort, and beach areas of Cocos Island. Traps (10-80) were placed at five m intervals and checked every 15 min at which time any lizards captured were removed. Generally, sticky trapping spanned the time between 0700 and 1200 h. Rubber-banding was only rarely employed after September 1990. After the December 1990 typhoon, nocturnal surveys were discontinued (see below) and only arboreal diurnal and sticky trapping survey techniques were used.
Pre-typhoon Russ herpetological surveys of gekkonids in the resort yielded a qualitative estimated community structure (expressed as percentage of total number of lizards) of P. ateles (4.5 %), G. oceanica ( 6.2 %), L. lugubris (0.6 %), and H. frenatus (88.8%) (N = 315 lizards in 30 person-hours survey effort). Unfortunately, the survey route for gekkonids was completely destroyed by the cumulative effects of both typhoons. This was exacerbated by the clean-up efforts of the resort corporation in which remaining debris was removed. Thus, no comparable post-typhoon data could be generated.
Surveys immediately after Typhoon Yuri yielded no lizards of any species on the approximately 1/3 of the island occupied by the resort. This portion of the island was subjected to the most intense vegetation / structural loss from typhoons. Although gekkonids were common in the resort prior to the typhoons, population densities of gekkonids in the relatively unsurveyed forest sections of Cocos Island are unknown; I can only assume that a sizable fraction of the gekkonids on Cocos Island were lost because of typhoons. Post-Typhoon Yuri diurnal surveys in forest areas targeting gekkonids revealed the persistence of all previously recorded species on Cocos Island.
Pre-typhoon sticky trapping surveys for E. atrocostata yielded a Catch-Per-Unit-Effort (CPUE) of 0.304 lizards/trap hr (N = 51 lizards, trap hrs = 168). Trap-hours are defined as one trap set for one hr = one trap hr. CPUE's are the number of lizards captured/trap hr. Post-typhoon surveys yielded a CPUE of 0.022 (N = 2 lizards, trap hrs = 90). This is a decline of an order of magnitude in catch rates and suggests that the population on Cocos Island declined by over 90% due to cumulative typhoon effects.
The remaining Emoia species (cyanura, caeruleocauda, and slevini) and C. cf. fusca can be discussed as a group as no changes in ranking of species collected (see below) in the forest area were noted after or between typhoons. These four species were initially
4
sampled in forest using rubber-banding in early 1989 through late 1990 and sticky trapping in September 1990. Initial levels of efforts were low (total trap hrs = 22) or not quantifiable (rubber-banding). Numbers of lizards collected, ranked in terms of most to least abundant, indicated that C. cf. fusca was the most common followed by E. caeruleocauda, E. cyanura, and E. slevini. All sticky trapping surveys in the forest after December 1990 (N = 5) were conducted along the same transects and yielded the same ranking in abundance as above. Trapping (N = 1400 trap hrs) was conducted in January, June, October, and December 1991, and September 1992. Two surveys (January 1991 and December 1991) were conducted within two weeks after typhoons. Percentage composition for each of the species (grand total = 365 skinks) in the five forest surveys ranged between 57.6 and 68.9 for C. cf. fusca, 20.7 and 30.3 for E. caeruleocauda, 2.6 and 12.9 for E. cyanura, and 0.0 and 2.6 for E. slevini. Changes in percentage compositions between surveys were tested using a R X C test of independence with a
William's correction and were not significantly different (X7cale,12,.05 = 9.197). This suggests that responses of individual species to typhoon effects were not statistically different. Similarly, CPUE's for all surveys were within the same order of magnitude (range 0.171 - 0.475) indicating that the cumulative effects of the typhoons did not dramatically decrease catch-rates of forest-dwelling scincids. Since at least 1/3 of the island was devoid of any lizards after Typhoon Yuri (see above), it is safe to assume that total population declines were greater for E. cyanura, E. caeruleocauda, and C. cf. fusca than for E. slevini, which occurred only in forest.
Numbers of C. poecilopleurus were gauged by sightings per min (range 0.33 - 1.1). These sighting data, including both pre- and post-typhoon observations, are within the same order of magnitude suggesting that typhoon effects were minimal on survivorship of C. poecilopleurus. importantly though, post-typhoon observations were made on existing trees and since sighting rates after typhoons did not increase on these trees, perhaps indicating emigration of surviving lizards from felled trees to existing trees, it is assumed that if a tree was lost during a typhoon, the resident lizards were also lost.
The ability of a herpetofauna to persist on an atoll after substantial environmental perturbations are also highlighted by observations on two species not directly surveyed in this report. Varanus indicus, although found on Guam, was probably introduced to Cocos Island in the late 1980's (pers. obs.) and managed to persist through two major typhoons. By December 1991, in addition to a number (3 - 5) of 200 to 450 mm snout-vent length (SVL) lizards, a small (ca. 100 mm SVL) individual had been observed on Cocos Island. These observations suggest that successful reproduction had occurred and monitor lizards had survived the typhoons. Bufo marinus was probably introduced to Cocos Island in 1989 and successful reproduction (large numbers of tadpoles in rain pools) was observed in September 1989. In September 1992, after both typhoons, two adult (ca. 830 mm SVL) B. marinus were observed in a freshwater pool.
Observations of the herpetofauna on Cocos Island after typhoons suggest a resilience to environmental perturbations. Terrestrial forest-dwelling scincid populations appeared to persist relatively unscathed despite substantial typhoon impacts. Habitat specialists (E. atrocostata and C. poecilopleurus) were more susceptible to population declines due to habitat destruction. All gekkonid species also persisted after the substantial effects of the typhoons. Besides C. poecilopleurus and E. atrocostata, the largest localized population declines of other species are associated with the developed (resort) section of the atoll. This may be related horticultural / architectural practices that restructure typhoon adapted vegetation allowing complete overwash and loss of most structures, soil, and sand during severe storms.
Considering the absence of all lizards on the resort 1/3 of the atoll, a substantial fraction of the lizard population was lost because of the cumulative effects of typhoons. Habitat
5
specialists E. atrocostata and C. poecilopleurus probably suffered much greater population declines, which is related to susceptibility of these habitats to typhoon damage. Despite that, the data suggest that relatively undisturbed atolls will tend to retain herpetofaunal components despite substantial typhoon influences.
ACKNOWLEDGMENTS
Assistance in the field was provided by Rebecca Hensley, Robert Cruz, and Earl Campbell III. Gordon Rodda and Thomas Fritts generously provided unpublished data. Rebecca Hensley, Gordon Rodda, Thomas Fritts, and Kevin de Queiroz commented on a version of the manuscript. Portions of this study were funded by the Endangered Species Conservation Program, Project E-4 (to Guam) and by the U. S. Department of the Interior, National Biological Survey.
LITERATURE CITED
ANONYMOUS. 1990. Local climatological data. Annual summary with comparative data. Guam, Pacific. NOAA Natl. Clim. Data Center, Asheville, NC. 8 p.
HENSLEY, R. A. and M. J. MCCOID. 1994. Cryptoblepharus poecilopleurus (Snake- eyed Skink). Activity. Herpetol. Rev. 25:121.
JACKSON, W. B. 1967. Productivity in high and low islands, with special emphasis to rodent populations. Micronesica 3:5-15.
MCCOID, M. J. 1993. The 'new' herpetofauna of Guam, Mariana Islands. Herpetol. Rev. 24: 16-17.
.and R. A. HENSLEY. 1994. Distribution and abundance of Perochirus ateles (Gekkonidae) in the Mariana Islands. Herpetol. Rev. 25: 97-98.
,G. H. RODDA, and T. H. FRITTS. 1995. Distribution and abundance of Emoia slevini (Scincidae) in the Mariana Islands. Herpetol. Rev. 26: in press.
MYERS, R. F. 1991. Micronesian Reef Fishes. 2nd ed. Coral Graphics, Barrigada, Guam. 298 p.
NEUBAUER, C. P. AND D. R. NEUBAUER. 1981. The vegetation of Cocos Island (Mariana Islands). In L. Raulerson (ed.). Plant biogeography of Guam. Univ. Guam Mar. Lab. Tech. Rep. 69. pp. 23-39.
RODDA, G. H, T. H. FRITTS, AND J. D. REICHEL. 1991. The distributional patterns of reptiles and amphibians on the Mariana Islands. Micronesica 24: 195- 210.
, M. J. MCCOID, AND T. H. FRITTS. 1993. Adhesive trapping II. Herpetol. Rev. 24:99-100.
WEINS, H. J. 1962. Atoll Environment and Ecology. Yale Univ. Press, New Haven. 2p.
~
crebigas enh tad feanangea eee tes om Madeacdienveg actoohitbn Sages kei csinhagt ren un nd pang th eg EB ioc chee tdi Baan aelgl dant) winwhid teaunit oLtoinin efxivi2 Remiineb i wan tons val igtabiingan| woo! rey Nt nila nO ine (oviulovsadideep dom miall odvtdy least, woudanl,. Ucates Anat 4 LO aye OTe he mmpanaieeg edb wegbabgrdad | ip ry! yi mjunes ft iit ab Pe WR Nes . Wi a ny rey aT Ye ye fy after cerita UD CN Oe) Wirth «ave WMA IFOA IO: insects and yombdogd Lie ma
PART EMR AN, ais 1 ORE | . as ra) wih aoe sucha ten vi jaar: ee ee a pate) pe We ri acy” yal Ley at Hy yon hoe beste: ‘dow bles dump 4 Jeane Ie ‘sant ag ‘Lob ive Vela nine wih i cea. wis abbot yalitinodh, 1 ifeieety aoa bolt er tek acne eb A | ay PR ine pbb Sh rovointen Yi, Neo hart cosy yi pane ae Decne ashi 1 ork Vote 2B omits (og Aioanmnmeledt iad i venir Ley ed sai ih Lea ine yee wl wel bua Perinae Fe nb Shs paiut Vena Dh coca AS OR A MSTA Re I arnt He bey Pres 4 BYimpadinc qx:
WY Ulan eg ara C eli a ee a Hay ART SEH ice Ce ia fy i & ve A per A pst De as oie eres ne DB re EM a Meeihin Genny | 4 Re i, petal rei Min thea ai te ven aw vei
i in yi yi & : ay ec ii . : iperny ui mali ingle eg A a Sis ra sen Fl hs mye
nee ea yr Aa sy Wet ing
ab ANVid ae » a We a it aa ot nate i: “a ie
Witte Fi Weibe eet iF thle obi Ayes
vty a itiatt user Las
eH mi ie oihicin le 4 Ai )} La i a On Gaabent rth, POL. ka) is yo yi iy Meee cach pol etal ea Aes ieoe Peal qa i) aan 5 ‘| t cy Ae id ‘ye ate abit ii nate sie te ii ii Ww ie ty its if eh 1 Mt ih | LY: iy
AAA Beh vi een aan ihe | ork taceuy ¥i vic uy
ae
i* pale ! ns Ve
fi
deh aes inept i MO eae } sycmneni VA PANES ioe om bh min iss igi hi an ye re aft ms fan ny oy J vadwet «Agia detenh a Fok Ny ail rH, oft bralel ee abi r.
a
le ili : a) ) a tye ty i rit DP Maplbahina oanals ed ee pik} OL on SH i ba dala}: ) Hoe he ba A oy id sii ‘POE tit ih ns boii ail! any og) rier CPO thaw | a vf aa cits ila Louvre abd ity Lobo ane vial yr wae? 4 mayen i, Vathop nsec 1M dia aus iit aloe Daa ; He ¥ vr reek pgekes wenn acy by
(
' -
iw
LP an hit, are bc wer aut’ (em. BO cio can ‘ Re. eae “i. a UMA © a TAG
ey daha mn ee Mad dad Cohdeanueniaten i wd civithaneloconehinal
jive baa p ti wut a eee NM A eget cc Wes} sree) aise
i my Me We a MAR oe Une Piet «yas wren’ Ie
Tyee Hinata tt ub al P| pW tea ia Vahudieh. ay) Viper Med ji tae
25 bated veo mena pom i Selene ted anak ste wr nfo lasiagtiom twirl aot mon ss , med ec Miami mie Fk tle Lorast boaathe od Lat
| ier are atvaniahead whee abagalenpmadl Gieaet seek ak ee ail
Central skubieg tevep gems ty vr mala eal RRR Bt b SUVA Meal ATA AWD nh Magia
PVA SC a aD: Deh a i renin Maines nf MAL reasty
ce beled » aut Ts tie re (acetate eh Shik iA) | nut | “atuinet wl Me ee ee
ATOLL RESEARCH BULLETIN
NO. 440
FLOWERING AND FRUITING IN THE FLORA OF HERON ISLAND,
GREAT BARRIER REEF, AUSTRALIA
BY
R.W. ROGERS
ISSUED BY NATIONAL MUSEUM OF NATURAL HISTORY SMITHSONIAN INSTITUTION WASHINGTON, D.C., U.S.A. APRIL 1996
yi \ Genie? See WITAIIUG HORAGEAS JI01 ay co a
Abi KONTO AMOI TIT M1 OATTIUNT GA OVIAWORE
iaiasebern 0 AMLAAR TARA
SHADOW
Ya (ae (OTA JARUTAM FO MUU AAMONT AM VOITUTTP SYR FART Ee en al, OT OMIRA AT | ove) JADA
FLOWERING AND FRUITING IN THE FLORA OF HERON ISLAND, GREAT BARRIER REEF, AUSTRALIA.
BY
R.W.ROGERS'
ABSTRACT
The plant species in flower and fruit on Heron Island, a sandy cay on the Great Barrier Reef, Australia, were observed at intervals of three months for three and a half years. At no time were less than 20 nor more than 36 of the 49 species monitored found to be in flower, nor were less than 20 nor more than 41 of the 50 species monitored found to bear fruit. Despite a strongly seasonal climate there was not a strong seasonal pattern evident in the number of species in flower or fruit, although some species were themselves strongly seasonal. A principal components analysis of all flowering records, however, demonstrated a seasonal polarity with March and September representing the two extremes. Fleshy fruited species, important for frugiverous birds such as silvereyes, bore fruit throughout the year.
INTRODUCTION
Temporal patterns in flowering and fruiting are significant attributes of vegetation, for these are attributes subject to selection as are any others. Variation in seasonal flowering patterns has proved to be significant in understanding of heathlands in Australia (Specht et a/. 1981) and Europe (Woolhouse & Kwolek 1981), both in terms of ecophysiology, and in terms of the evolutionary derivation of the floras. The availability of flowers and fruit is manifestly important to those animals which depend on fruit, seed and nectar as food resources, and an interaction between plant phenology and the birds responsible for seed dispersal has been postulated (Herrera 1986).
There