Chapter Seventeen
INTRODUCTION
Coral cays are generally referred to as low islands constructed entirely from the biogenic materials of the reef itself (Hopley 1982), and are a common physiographic and ecologic zone of many reefs throughout the archipelago (figs. 17.1 and 17.2). The purely biogenic origin of coral cay sediments from the underlying reef, and the absence of continental rocks, clearly distinguishes them from volcanic islands as well as the high islands of the Sunda Shelf and other regions of the archipelago. The composition of beach sand alone is, however, not an appropriate criterion to distinguish between the two types of islands (fig. 17.3).
The vast majority of the 17,508 islands of the archipelago are most likely coral cays or low islands. Their contribution to the total land area of the archipelago is, however, quite small (c. 3%). Soegiarto and Polunin (1981) estimated that, since the 12 major islands of the archipelago constitute 97% of the total landmass, the average size of the smaller islands is about 4 km2 (note that they based their estimate on the assumption that there are 13,000 islands in the archipelago). The exact number of islands, and therefore the length of the coastline, is not known since different definitions of an island give different numbers. Development and implementation of a formal classification system for Indonesian coral-reef-associated islands would greatly clarify the issue. In our discussion of coral cays we refer only to those coral reef features which are situated above the Lowest Low Mean Water or the supralittoral zone affected only by sea splash and spray.
The nature of the underlying foundation from which the reef has originated should be of no consequence to the classification. Thus, volcanic, sedimentary or old Pleistocene limestone foundations capped by thick Holocene reef deposits may share the same type of coral cay (e.g., unvegetated sand cay). The condition that must be met, however, is that the pre-Holocene volcanic, sedimentary or biogenic foundations do not contribute any material whatsoever to the structure of the existing islands, which stand upon the Holocene reefs.
FORMATION OF CORAL CAYS
Among the best-known, if not the most-studied, coral cays in Indonesia are the 120, or more, islands of Kepulauan Seribu. Their geomorphology and mode of formation have been studied and examined in great detail since the early 1920s. The formation of coral cays may be considered as an evolutionary process driven by biogenic and physio-chemical processes. Note that this is not analogous to the suggestion by umbgrove (1929a,c) that variations in reef island morphology are indicative of an evolutionary sequence (i.e., age). It is obvious that the evolution of a vegetated coral island (cay) begins with the formation of an embryonic sand (or shingle) cay. If the embryonic cay subsequently stabilizes, it may develop further into one of a variety of coral cay types. The final form of a cay, if it is possible for a cay to achieve one, is determined by a complex interaction of biotic and abiotic processes, among which reef-top geometry, hydrodynamics, climate and biogenic carbonate production must rate at the top.
Figure 17.1. Coral cays (i.e., coral islands) are a ubiquitous physiographic feature of Indonesian reefs. A) A series of inshore reefs of the Spermonde Barrier Reef illustrating various stages of reef development, from submerged reef (lower left), intertidal reef (lowest left), unvegetated sand cay, and vegetated sand cay. B) Vegetated mixed sand and shingle cay in the lagoon of the Spermonde Barrier Reef.
Photos courtesy' of M. B. Best, National Museum of Natural History, Leiden.
Figure 17.2. Coral cays are not restricted to shelf patch reefs, but develop on many oceanic reefs that have reached sea level. Pulau Raja is a 3-km-long coral cay formed on top of a large reef at the southwest rim of Taka Bone Rate Atoll, the largest atoll in Indonesia.
Photo by Tomas and Anmarie Tomascik.
Figure 17.3. Composition of beach sediments of small offshore islands should not be used as a criterion to distinguish between high and low islands. A) The geologic nature of many islands is clearly apparent from their beach deposits. East coast of Central Sulawesi. B) White coral beaches and lush tropical vegetation of many small high islands often hide their volcanic origins. C) On the other hand, white sandy beaches are the characteristic features of true coral cays.
Photos by Tomas and Anmarie Tomascik.
Sea-Level Influences
It seems obvious that past sea-level fluctuations must have had considerable influence on the development of coral islands. However, as Kuenen (1933a) pointed out: "…the importance of the negative movements (falling sea levels) to the present aspect of the reefs has been greatly underrated,…", and, indeed, this aspect is still seldom mentioned when discussing formation of coral cays. Both Gardiner (1931) and Sewell (1932) were strong advocates of falling sea levels as a prerequisite for coral cay formation. They believed that without the relative fall in sea levels, formation of coral cays was not possible. Kuenen (1933a), however, pointed out that most raised (whether by eustatic or tectonic forces) reefs will eventually be attacked by the weathering and erosional forces as in the case of raised islands. His observations of numerous raised and rapidly abrading coral cays throughout the archipelago (e.g., Jakarta Bay, Kepulauan Seribu, Postiljon and Paternoster Islands, etc.) led him to conclude that, in many instances, falling sea levels will eventually lead to the destruction of the former land formation, even if, initially, a new cay develops. His views of recent fall of sea level have been clearly proven to be correct, at least for Kepulauan Seribu, where the relative sea levels have dropped 1-2 m during the past 4500 years (Park et al. 1992). Nonetheless, Kuenen (1933a) points out that in the sheltered regions of the archipelago many cays occur: "…which were built since the present relative levels were reached on reefs that were not laid dry, but were still covered by living corals. Obviously, however, very many and probably most islands have been formed as a result of the emergence of their flats". This view is supported by recent radiocarbon dates of corals at Pulau Putri Besar taken from 1 m above present sea level. The cores seem to indicate that the cay post-dates the underlying conglomerate platform which was deposited between 8000-4500 yrs B.P., and that the cay has developed in the last 4000 years during falling sea levels. The various coral islands that abound throughout the archipelago are, therefore, products of local and regional environmental settings at one point along an evolutionary time continuum of their reef foundations. Accordingly, one must expect continual changes in their size, shape and form over time, a fact of nature that is still not fully appreciated by coastal developers (fig. 17.4).
Coral Cay Framework
The formation of coral cays depends firstly on the availability of reefal sediments. Coral cay sediments are biogenic, originating from the skeletal material of numerous calcifying plants and animals which build and live on the reef. Sediments are formed from their skeletons either after their death, through subsequent physical and biochemical breakdown, or through the action of various destructive processes (biotic and abiotic) that attack the skeletons of still-living reef-builders. Boring or grazing reef organisms discussed earlier are important components of this process.
According to Hopley (1982), sediment size on reefs is distinctly bimodal, which results in the formation of windward shingle or leeward sand cays. Sediment distribution patterns of Indonesian reefs with cays are not available. On the Great Barrier Reef the most common sediment size range is 0-1.5 φ, and sorting is less than 1 φ (Hopley 1982). Similar to Hopley's (1982) observations, beach sands on all cays visited throughout the archipelago are the coarsest sediments on the reef, especially on the windward side of the cays. In addition to the purely reefal carbonates, non- carbonate allochthonous sediments may also be present. On many shelf reefs siliciclastics originating from the nearby continental landmasses may constitute a significant fraction of the total sediments. For example, the carbonate sediments of Pulau Rambut only a few kilometres off mainland Java are heavily contaminated with siliciclastics.
Figure 17.4. The dynamic nature of coral cays is clearly illustrated in this time sequence for Nyamuk Besar in Jakarta Bay. The placement of concrete blocks on the windward (northwest) side of the reef had a significant impact on the size and shape of the coral cay. The reef upon which Nyamuk Besar stands is now a functionally dead coral system.
From Stoddart 1986.
In addition to the purely reefal sediments, some cays located close to the main land of large islands (e.g., Java), or active volcanoes (e.g., Banda), may have non- carbonate sediments incorporated into the cays. Not surprisingly, pumice is quite common on coral cays that are in close proximity to active, volcanoes (e.g., Lampung Bay, near Krakatau). However, Kuenen (1933a) also found large quantities of pumice on Pulau Pelokang (Sapoka Atoll) in the west Flores Sea, far removed from any active volcanoes. It has been demonstrated that marine and terrestrial organisms are able to disperse widely via floating "rafts" made of various materials, such as trees, seaweed, floating coral and volcanic pumice (Woodjones 1912; Jackson 1986; Jokiel 1984,1990; DeVantier 1992). It is therefore probable that some of the earliest sand-cay colonizers are "hitchhiking" species (DeVantier 1992). With rapid development of coastal areas, and the use of oceans as waste-dumps, the amount of "flotsam" originating from anthropogenic activities has increased dramatically within the past few decades. These days, it is not uncommon to find floating shoes, cans, plastic water bottles, etc., in the middle of the Flores Sea and other inland seas. These modern-day "rafts" are frequently overgrown with numerous organisms, including corals (Jokiel 1984).
Environmental Factors Involved
Prevailing winds, waves and currents generated by the reversing monsoons are key physical factors of the Indonesian marine climate, influencing the structure and function of marine ecosystems including the development of coral cays. The monsoonal winds and the resultant wave and current patterns constitute an important atmospheric-oceanic coupling which has a significant influence on the geomorphology and stability of the thousands of coral cays within the archipelago (Molengraaff 1922,1929; Verwey 1931b; Kuenen 1931,1933a; umbgrove 1931,1947; Verstappen 1954). Without the influence of wind-generated waves, the evolution of vegetated coral cays cannot proceed.
On the Togian reefs, however, not even sand islands occur, so that not a single trace of the action of wind or waves can be noted.—UMBGROVE 1947
Waves play a key role in the formation of coral cays, since it is their sweeping action that facilitates the movement and concentration of resuspended reefal sediments. The energy of the waves is also an important factor in coral cay formation, since it is the wave energy that primarily transports the reefal sediments. The energy of a wave varies with the square of the wave height (i.e., size of the waves). Thus, the higher the amplitude, or the bigger the wave, the more total energy it contains. Doubling of wave amplitude will quadruple the total energy of the wave. The size of waves is governed by the fetch (i.e., the unobstructed distance of sea over which the wind is blowing), actual wind velocity, duration and constancy of direction. When waves break on the reef, their total energy is transferred into kinetic energy which transports reefal sediments. However, for a coral cay to develop, convergence of waves must take place so that sediments are deposited in one particular area of the reef throughout the tidal cycle. Wave refraction is governed by the shape, size and bathymetry of the reef as well as the tidal range and direction of wave approach. It is therefore not surprising that coral cays are most often present on oval-shaped reefs, since this geometry concentrates the pattern of wave refraction into one area of the reef flat (fig. 17.5). Since the incoming waves' from the windward side have more energy, sediment deposition will frequently occur at the leeward side of the reef.
Numerous patch reefs are without sand cays, because their shape allows waves to pass over the reef without refraction (bending). Since these waves are not concentrated (focused), sediments are usually deposited evenly over the reef flat. On these reefs we frequently find large windward rubble areas that grade into sand sheets to the lee of the reef (Hopley 1982). The size of the reef is not an important factor. Small reefs in the north sector of Kepulauan Seribu have large cays that cover almost the entire reef flat (e.g., Pulau Cina), while some very large reefs at the south sector of the complex have only small cays (e.g., Pulau Semadaun). Large reef size may, in fact be detrimental to cay formation. Attenuation of wave energy, and therefore the competency of waves to move and carry large-size sediments, across the reef flat of large reefs will vary from day to day with weather (i.e., energy) conditions (Hopley 1982). As a result, sediments may be moved and dispersed more evenly over the reef flat without being concentrated in one particular area. Figure 17.6 illustrates the potential of water velocities caused by various marine processes to move different-sized sediments during the monsoonal year.
Figure 17.5. Oblique aerial photograph of Pulau Panjang, west outer rim of Taka Bone Rate Atoll, Flores Sea. Note the sanded reef flat on the north (upper left) and south (lower right) points of the island. Sand chutes through which sediments move out of the reef flat are visible at the far side of the reef (upper left).
Photo by Tomas and Anmarie Tomascik.
Figure 17.6. Potential capacity of different velocity currents to transport reefal sediments. Heavy dashed lines represent peak monsoonal currents. Thin lines are wave maximum horizontal velocities for non-breaking waves. Lines 1 and 2 are average and maximum Southeast Monsoon waves. Lines 3 and 4 are Northwest Monsoon waves.
From Brown 1991.
Coral Cay Stabilization
During the early stages of development, most sand cays are rather unstable systems, frequently changing their position on the reef flat. Some cays have been measured to migrate tens of metres in one day. In many tropical regions affected by the Trade Winds, where wind directions are rather unidirectional, the migration of sand cays over the reef flats are not as pronounced as they are under the influence of bidirectional monsoonal systems. Once sand cays reach a certain mass or size, the movements become much less pronounced and eventually the cays stabilize. Provided that environmental conditions are appropriate, for not all embryonic sand cays will develop further, the formation of a vegetated island may take a few years to decades. Stabilization of sand cays is greatly enhanced by the colonization of plants, whose seeds arrive by atmospheric transport (cyclones, storms, etc.), water currents or on flotsam, or carried by birds that may use the sand cays for nesting. In some areas where there are large concentrations of sea birds, formation of vegetated cays may be slowed due to the negative effects of phosphatic guano (i.e., bird droppings) (Hopley 1982). Another important process that begins as soon as plants and birds are established on the new sand cay is the formation of cay sandstone, which further stabilizes the base of the cay. In the humid tropics, the abundant rainfall rapidly leaches phosphates from the guano and transports it deeper into the cay sediments, where cementation slowly takes place (Kuenen 1933a; Hopley 1982). Kuenen (1933a) recognized two types of cay sandstones in die archipelago. One type is softer than beach rock and the other is harder, however, both are horizontally bedded and located just above the high water mark. Kuenen (1933a) speculated that the hard type is formed from the soft type through evaporation of seawater and hardened by percolating rain water. Further hardening is attributed to the fresh water in the interior of the cay (i.e., freshwater lens). However, he also observed that cay sandstone was not found on all islands, and was particularly rare on the cays of the Spermonde Shelf.
Figure 17.7. A) Seaward-dipping slabs of Recent beach rock are common features of many sand cays throughout the archipelago. Beach rock is often found along the weather side of cays, where it is frequently exposed. P. Neijalanka, Banda Sea. B) However, beach rock is not present or visible on all beaches. Where tidal amplitudes are high (>2 m), frequent movement of beach sand may prevent formation of beach rock. On many green turtle (Chelonia mydas) nesting beaches, turtles frequently disturb the beach sediments. Pangumbahan, West Java.
On the beach face (intertidal and lower supratidal) of many mature sand cays, especially in areas of high wave exposure, exposed beach rock is often present (fig. 17.7A). This feature is a product of cementation and has an important function in beach stabilization. Beach rock has been defined as a sedimentary formation of seaward-sloping flagstones indurated at the intertidal level by a calcareous cement, consisting initially of aragonite or calcite (Stoddart and Cann 1965; Guilcher 1988). However, beach rock may be absent from beaches whose sediments are frequently disturbed by nesting turtles. An excellent example is Pulau Sangalaki (East Kalimantan), where 30-45 turtles dig their nests every night of the year (fig. 17.7B).
One of the most striking features of many tropical coasts, especially of coral coasts, is the frequent conglomeration of the sand to a fairly hard rock laying between tide levels. Darwin had already noted its existence on Keeling Atoll.—KUENEN 1933A
Kuenen (1933a), who referred to beach rock as "beach sandstone", observed that with one exception it always showed a dip of about 10°, generally corresponding with the angle of the beach face. Cay sandstone and beach rock can be easily distinguished by the angle of layers. While beach rock is characterized by seaward-dipping layers, cay sandstone is horizontal. Beach rock is frequently found in a few layers and varies considerably in hardness. Hopley (1982) reviewed the various theories offered to explain the formation of this feature, but they can be generally characterized as being products of three processes, organic (microbial and/or algal action), inorganic from fresh water, and inorganic from seawater. Some workers have dismissed organic processes as insignificant due to the low organic content in the beach sands (Milliman 1974; Davies and Kinsey 1973). Kuenen (1933a) was not able to offer any new explanation for the formation of "beach sandstone" on the numerous cays visited during the Snellius Expedition, but concluded that fresh water was not a necessary agent, and that evaporation of seawater was one of the cementing agents due to its restriction to the intertidal. Others suggest that beach rock forms by cementation of subsurface beach deposits through the action of calcium carbonate cements precipitated from the ground water of the cay (Hopley 1982). Since ground water in cays is mostly of marine origin, the mineralogy of the beach cements is mainly aragonite, and occasionally high-magnesium calcite (Inden and Moore 1983). The exact mechanism for the formation of beach rock is still being debated.
Ecological Succession
Subsequent to the arrival of the first pioneering plant species, environmental conditions on the cay continue to change, and a simple succession takes place (Hopley 1982). How early pioneering plants and organisms arrive at some of the most remote oceanic islands has perplexed many early naturalists, arid numerous theories have been proposed (Guppy 1890; Wood-Jones 1912; Renvoize 1979; Stoddart and Steers 1977). The major environmental factors that influence the early colonization process, and subsequent growth, are, among others, degree of wave and current action, tides, salinity, temperature and the composition and nature of the sediments. Early colonization processes have not been studied in detail in Indonesia, however, a considerable amount of literature exists for sand cay vegetation from other parts of the world (e.g., Sachet et al. 1983; Stoddart et al. 1979; Lee 1984).
The size of the initial 'embryonic' cays varies, but the colonization process by plants usually begins with the seemingly ubiquitous pes-caprae group (table 17.1). These hardy "crawling" strand plants, as well as many other groups, have a number of means at their disposal to get their diaspores to new areas (e.g., wind, ocean currents, birds, bats, etc.). On many small cays the first arrivals are salt-tolerant (halophytes) and drought-resistant species of which the beach morning glory (Ipomoea pes-caprae) is by far the most common (fig. 17.8). The diversity and distribution of these plants is not known, however, Forbes (1885) described 20 species of Ipomoea from Timor. This hardy species with characteristic bilobed leaves (thus the uncommon English name - goat's foot) colonizes unconsolidated areas within reach of high tides. The arrival of early colonizers able to tolerate harsh and unstable (i.e., ephemeral) environments begins the stabilization process, and subsequently allows other plants to take root in more landward areas of the cay, provided the cay keeps expanding. Together with other strand line species such as Thaurea involuta, Canavalia maritima, C. cathartica, Spinifex littoreus, S. sericeaus, and Vigna marina, these early arrivals begin to change the physical and chemical composition of the cay sediments.
Table 17.1. Common pioneering beach-front vascular plants of the pes-caprae community found on coral cays and beaches in the Indonesian Archipelago.
Figure 17.8. Small vegetated sand cay in the Salabanka Islands, east coast of Central Sulawesi. Strand vegetation belonging to Ipomoea pes-caprae is one of the early colonizers capable of tolerating wide fluctuations in salinity and low water content. The vegetation on top of the sand cay is predomi nantly J. pes-caprae and Vigna marina. The taller shrubs at centre are Scaevola sericea. The intertidal and subtidal sandy slopes are covered by dense mix seagrass meadows of Cymodocea serrulata, Thalassia hemprichii and Halophila oval is.
Photo by Tomas and Anmarie Tomascik.
Pioneering species have different tolerances to saltwater exposure following germination, which results in a loose, but discernible, zonation. Obviously, angle and width of the beach as well as the tide range are important environmental factors that determine early zonation patterns, even in the pioneering stage of community development. Sporobolus fertilis seems to be the most tolerant to saltwater exposure, and is frequently found in beach areas that are occasionally inundated with seawater. Just above, we find the creeping Vigna marina and a fast-growing Spinifex littoreus which produces spiky fruit clusters. Higher parts of new cays or established beaches, affected only by sea spray, are frequently dominated by the cay grass Thaurea involuta and Lepturus repens. On mature vegetated cays the early pioneers are often found in the most unstable environments (Hopley 1992). Note that this association is not exclusive to developing sand cays, but is common on most calcareous tropical beaches in moderate exposures as well as on sand dunes in high-exposure environments. For example, the sand dunes at Pangumbahan, on the south coast of West Java, have an Ipomoea pes-caprae-dominated community that includes Ischaemum muticum, Canavalia maritima and Cynodon dactylon. Pangumbahan is one of the most important green turtle (Chelonia mydas) nesting beaches in Indonesia. Unfortunately, years of unregulated egg harvesting and poor conservation planning has placed the green turtle population at risk.
The Post-Pioneer Stage
With the establishment of the pes-caprae pioneering community, environmental conditions become progressively more favorable for the survival of new plant species. The new plants that become established are the shrubs and trees commonly grouped into the Barringtonia community (table 17.2). These plants establish themselves characteristically behind the pes-caprae community. The majority of the larger plants that arrive have seeds that are well adapted for sea dispersal (e.g., buoyancy and dormancy), and are able to float in surface currents for many days, or months, without losing their germinating potential. Among the most widespread of these plants is Argusia, whose seeds are protected within a fruit that is surrounded by a thick cork-like outer skin. Seeds of numerous plant species get washed onto small cays as soon as they become fully exposed, however, many are considered as strays and die without reproducing. Other species are transients; they arrive, but subsequently become extinct either due to frequently changing environmental conditions or competition; this holds for plants and animals alike. Local extinctions are, however, much more common in the faunal groups.
As the cay matures, greater physical stability and higher nutrient content in the soil allows the survival of less tolerant plants, whose earlier arrivals perished under harsh environmental conditions. In many areas where sea birds are abundant, guano may significantly affect the successional dynamics. Mixed with decaying vegetation, guano not only adds significant amounts of nutrients (it is rich in phosphate) into the new cay system, but significantly alters the subsurface pH of cay sediments. Whitten et al. (1988) reported that the faces of the white-capped noddy, Anous minutus, which nests on Sangisangiang Island in the Flores Sea, contributes 250 and 1030 kg.ha-1.yr-1 of phosphate and nitrate, respectively. Changes in soil chemistry become increasingly more important as larger trees become established. Hopley (1982) has reported a decrease in pH from an initial 8.2 to 5.0 on cays with extensive Pisonia vegetation and sea birds. A similar example can be found in the Java Sea at Pulau Rambut, which is located just west of Jakarta Bay (4 km offshore). The island is an important waterbird sanctuary with 10 breeding species, as well as a roosting site for about 20,000 fruit bats Pteropus vampyrus (flying fox) (Silvius et al. 1987). The pH of the soil varies from 6.7 to 7.6, while organic content is 2.0% to 62.1% (Silvius et al. 1987). However, sea-bird guano is not an essential component for cay development, since many small vegetated cays in the Kepulauan Seribu are hardly ever visited by sea birds. Large sea-bird rookeries on coral cays, such as Sangisangiang Island and Pulau Rambut, are in fact a rather rare sight in the archipelago. Nesting sites are more common on small oceanic high islands, where they are generally unreachable by fishermen (fig. 17.9).
Table 17.2. Some of the common species of the Barringtonia community found on coral cays and beaches of the Indonesian Archipelago.
Figure 17.9. The 107-m-high Pulau Manukan (Suanggi) in the Banda Sea is an important nesting island for Sula leucogaster (brown booby), S. sula (redfooted booby) and Fregata minor (great frigatebird).
Photo by Tomas and Anmarie tomascik.
The ecological succession (both floral and faunal) that follows the pioneering stage of development is not universal by any means, as is clearly evident from the great variety of vegetated cays, each with a distinct floral and faunal composition. These differences are on both local and regional scales. At local scales (1-500 km2) habitat differences between cays may play the key role in determining species abundance and composition. Two adjacent cays (one exposed and one sheltered by the other) will differ, since, for example, shrubs such as Pemphis acidula and Suriana maritima prefer sheltered shores, while Scaevola taccada and Argusia argentea dominate the exposed shores (Williams 1994). On regional scales (> 500 km2) habitat diversity continues to be a factor, but may be subordinate to climatic differences. Climatic differences (i.e., rainfall) will have greater impact on community structure of shrubs and trees which differ in drought-tolerance. Nonetheless, the early strand flora is remarkably similar, especially on smaller cays (Whitehead and Jones 1969). Since the pioneering strand flora consists of relatively few halophytic and drought-resistant species of creepers, grasses, herbs and sedges, which are generally pan tropical, it is not all that surprising to see Ipomoea pes-caprae on a sand cay in Kepulauan Seribu, on the foraminiferan beach of Nusa Penida, Bali, on the north coast of Biak, Irian Jaya, or on the sand dunes of Pangumbahan facing the Indian Ocean. However, the distribution and composition of the pes-caprae community in the archipelago has not been investigated.
Once the pioneering community becomes established, the successional dynamics of the Barringtonia community and larger trees become dependent on environmental factors (e.g., rainfall), geographic location and the size of the cay. The transition from the pes-caprae community to the Barringtonia community can be startling (fig. 17.10). The major difference in floristic composition of cays is mainly in the shrub and tree groups, since each group has specific tolerances to saltwater exposure and drought. Vegetated sand cays in the wet regions of the archipelago (e.g., Berau Islands) are covered by a lush forest year-round. For example, the floristic makeup of Pulau Sangalaki, with monthly rainfalls between 100-300 mm, consists of Terminalia catappa (sea almond), Calophyllum inpphyllum (laurel wood), Barringtonia spp.(putat laut), Pisonia sp., Casuarina equisetifolia, Pandanus spp. and Erythrina variegata (coral tree). A two- to three-metre-wide belt of Hibiscus tiliaceus(sea hibiscus), with large bright yellow flowers, surrounds the island, just inside the Barringtonia fringe. In the dry regions, however, sand cays are dominated mainly by sedges and grasses (Poaceae), with cacti (Cactaceae) and Pandanus spp. often forming a solid wall around the small cays (fig. 17.10).
Note that the size of the cay itself does not exert a direct effect on the kind and number of species present. Size of the area defines the potential carrying capacity of the cay, by assuming that the larger the cay (area), the more diverse the habitats it supports. Smaller cays will have a lower carrying capacity than larger cays, and thus there should be a positive correlation between the area of a cay and its species number within a particular taxon. This is nothing new, since it is the fundamental thesis of the theory of island biogeography proposed by MacArthur and Wilson (1967). The relationship can be quantified by the area-species curve given by the equation S=CAZ, where S is the number of species of a given taxon found on the island, and A is the area of the island (MacArthur and Wilson 1967). Parameterz is the slope of the log-log plot, and varies depending on what values are being used for the area (MacArthur and Wilson 1967). Currently there is not sufficient information available in Indonesia to test whether sand-cay flora and fauna conform to the theory's prediction. Based on purely qualitative observations, smaller cays will have higher rates of extinction than larger cays. Williams (1994) has recently demonstrated that the area-species (flora) curve for Cocos (Keeling) Atoll islands is statistically very significant, and is expressed by the equation S=6.73a,0.28 where 'a' is in hectares (ha).
Indonesia offers a unique natural laboratory to study the distribution and successional dynamics of coral cay vegetation, since the archipelago is not a climatically homogeneous region. Different regions can be crudely divided into three climatic subzones with regards to rainfall, namely 1) wet all year-round (300-500 mm.month-1; e.g., Tarakan, Konolodale); 2) wet-dry seasons (10-600 mm.month-1, dry, wet, respectively; e.g., Luwuk, Manokwari); and 3) dry most of the year (0-100 mm.month-1; East Nusa Tenggara). Note that more elaborate classification systems are available based on dry/wet period ratios and vegetation types (Schmidt and Ferguson 1951; Whitmore 1984a,b). The classification system by Schmidt and Ferguson (1951) would be most useful to use, since it closely corresponds to major vegetational distribution patterns. Comparative studies of cay vegetation in these areas may shed some new light on this unique and important (yet neglected) group of coastal plants. Beach degradation on many cays in the Kepulauan Seribu is partly attributed to replacement of natural strand line vegetation by more visually pleasing garden variety plants. In fact, it seems doubtful whether many cays still exist with their natural vegetation intact.
Figure 17.10. Sand cay vegetation exhibits clear zonation from the seaward pescaprae community (A) to the shrub fringe often dominated by the so-called Barringtonia community (B). However, in the dry southeast regions of the archipelago, beach and sand cay vegetation is often dominated by Cactacaea (C).
Photos A and B by Tomas and Anmarie Tomascik; Photo C courtesy of A. Ibrahim, P30-LIPI, Ancol, Jakarta.
TYPES OF CORAL CAYS
The great variety and complexity of form of Great Barrier Reef cays results from the range of factors that affect island-building and that cannot be matched in any other single reef province.—HOPLEY 1982
Very little has actually been done with regards to classification of coral cays in Indonesia, even though Kuenen (1933a) laid an excellent foundation upon which to build. Stoddart and Steers (1977) and Stoddart et al. (1978) recognized seven classes of coral cays, ranging from unvegetated sand cays to vegetated cays on emerged limestone, and made a generalization that the greatest variety and morphological complexity of coral reef islands occurs on the Great Barrier Reef. Explaining the great geomorphological diversity of coral cays on the Great Barrier Reef ('Variation in reef-top ages, morphology, and sea-level history combined with the differences in energy conditions and tidal ranges produce this diversity"), Hopley (1982) could have been describing the Indonesian Archipelago.
In his classification of the Great Barrier Reef sand cays, Hopley (1982) recognized four basic classes: 1) Unvegetated solitary islands; 2) Vegetated solitary islands; 3) Multiple islands; and 4) Complex low wooded islands. Coral cays belonging to classes 1 and 2 were subdivided into the four types listed in table 17.3. In a preliminary attempt to determine the types of coral cays in the archipelago we will follow closely Hopley's (1982) classification system which also allows a preliminary first comparison with the Great Barrier Reef (table 17.3). However, before we discuss the various coral cays associated with offshore patch reefs, a mention should be made of a generally neglected class of coral islands, which for the lack of appropriate terminology we shall refer to as fringing reef islands.
Table 17.3. Classification scheme of reef islands of the Great Barrier Reef, Australia.
Figure 17.11. Many large fringing reefs in the archipelago have wide reef flats with unvegetated or vegetated islands. This long reef along the east coast of Pulau Tanajampea, Flores Sea, has a migrating sand cay at the southeast point (Tanjung Parumang) of the island (upper right). The origin of these formations has not been investigated.
Photo by Tomas and Anmarie Tomascik,
Fringing Reef Islands. The so-called fringing reef islands are found on the reef flats of many large fringing reefs, particularly in the eastern parts of the archipelago. Geomorphologically and structurally they are very similar to the well-known offshore sand cays (unvegetated or vegetated) associated with patch reefs, barrier reefs and atolls. Fringing reef islands have not been studied in detail, and their geologic origins remain obscured. However, on islands where wide fringing reefs run unbreached by embayments or rivers along much of the coastline, formation of sand cays has been observed (fig. 17.11). Strong longshore currents seem to be a requirement. The formation of reef flat sand bars suggests that many of the reef flat islands may in fact be products of Holocene reefs, and not features rooted in Pleistocene history. Note that fringing reef flat islands have been described earlier (e.g., Cockermouth Island, Great Barrier Reef), but these were composed of dune calcarenite of Pleistocene age (Hopley 1975).
Two island groups where fringing reef islands are very prominent are the Tanimbar and Aru Islands in the east Moluccas. The southeast point of Yamdena, including the 4-km-long and 1.5-km-wide Pulau Asutubu, is fringed by an extensive fringing reef whose intertidal reef flat, along the east coast, extends up to 1.8 km offshore. Located south of Olilit village and about 1.25 km offshore, just at the seaward margin of the intertidal reef flat are two oval-shaped vegetated coral islands of undetermined origins (fig 17.12). The fringing reef islands are also found on reefs of smaller continental islands. For example, the fringing reef island Pulau Nusmesa is located off the north coast of Pulau Wuliaru (c. 24 km long) in the middle of an intertidal reef flat that extends some 2.5 km offshore. Along the east side of the Nusmesa, which is 200 m long, is a 3-m-deep channel that may correspond to a pre-Holocene riverbed.
Figure 17.12. Fringing reef along the southeast point of Yamdena, Tanimbar Islands. The 20 small fringing reef islands located on the wide reef flat are of unknown origins. Depth in metres.
Figure 17.13. Kuenen's (1933a) interpretation of a small fringing reef island off the west coast of Wotap, Tanimbar Islands, Moluccas.
From Kuenen 1933a.
Kuenen (1933a) seems to have been the first to point out these interesting fringing reef features during his visit to the Tanimbar Islands. Figure 17.13 is Kuenen's interpretation of Pulau Itrankaililengat, a small fringing reef island off the west coast of Wo tap. The reef flat on which the small island stands is intertidal and dominated by branching corals. Whether the sand cay was vegetated or not, Kuenen (1933a) did not say, however, just south of the cay is a well-developed shingle rampart (tongue), another unusual feature of Indonesian fringing reefs. Shingle tongues have been observed on the patch reefs in Kepulauan Seribu, where they lay behind and perpendicular to the shingle rampart.
Along the east coast of the Am Islands, fringing reefs apparently extend 10-15 km offshore and may be part of an extensive reef complex that includes extensive sub tidal and intertidal reef flats dotted by numerous small cays (Schulz 1989). Seagrass beds are the dominant habitat, covering large areas of reef flat and lagoonal habitats between the various cays. The vast system of fringing reefs, patch reefs, coral cays, seagrass and algal meadowy makes the Aru Islands one of the most important green turtle (C. mydas), hawksbill (Eretmochelys imbricata) and dugong (Dugong dugon) feeding and breeding areas in the archipelago (Schulz 1989). The origin of the fringing reef islands is uncertain. Some may be bathymetric highs of pre-Holocene Sahul Shelf fringed by reefs. In this case they would be considered as continental islands joined with the mainland by an extensive fringing reef. Alternatively, they could be composed of Pleistocene dune calcarenite. The area clearly offers interesting research possibilities.
Unvegetated Solitary Islands
Linear Sand Cays. Unvegetated (linear) sand cays, commonly referred to as sand bars, are perhaps the most numerous of all islands in Indonesia, and are found from the supratidal to the intertidal (fig. 17.14). They are associated with all reef sizes and are locally referred to as gosongs. Gosong Sepa Kecil and Karang Pandang in the Kepulauan Seribu Complex are fine examples. Unvegetated sand cays tend to form on elongated narrow reefs where the opposing wave trains meet along the entire length of the reef, or in areas affected by reversing wind (Hopley 1982). Their abundance in the archipelago is therefore related to the bidirectionality of the monsoonal winds, which are highly conducive to their development. However, a characteristic feature of these small sand islands is their ephemeral and unstable character. Frequent migration of unvegetated sand cays in many parts of the archipelago (e.g., Kepulauan Seribu, Spermonde Archipelago, etc.) is related to the reversing influence of the monsoonal winds.
While unvegetated sand cays are real calcareous deserts above the high water mark, the intertidal and subtidal regions of the reef flat are highly productive systems. A large narrow and elongated sand cay parallels the coastline in Lebaleba Bay on the northwest coast of Lembata Island, East Nusa Tenggara. The sand cay is very unusual, since the sand covers the entire area of the reef flat. The characteristic reef flat substrate, consisting of coral rubble and hard pavement, is not present, and the sand has spilled over the reef crest. A shingle rampart and large coral rubble zone are absent. From a depth of about 1.5 m below low water, branching corals grow in profusion to a depth of 10 m or more. In fact, the sand cay resembles a sand bar, but since all the slopes are covered by actively growing coral, it may need a new definition. In general appearance, Lembata cay resembles Pankaja sand cay in the Spermonde Archipelago described by Kuenen (1933a) (fig. 17.15).
Figure 17.14. A) Unvegetated sand cay in Lembata Strait, East Nusa Tenggara. Only about 50 m2 of this large (600 m by 300 m) sand cay is supratidal. B) An intertidal sand cay in the Salabanka Islands at high tide. The top of this intertidal sand cay (centre left) is flooded only during spring high tides. The sanded reef flat supports mixed Thalassia hemprichii and Cymodocea rotundata seagrass community.
Photos by Tomas and Anmarie Tomascik.
The Pankaja cay is continually swept by high-velocity currents that flow from the Flores Sea (to the north) to the Indian Ocean (to the south) through the Boling Strait. The intertidal and subtidal areas support interesting assemblages of macrobenthos. Among the most unusual were pink-coloured, slightly translucent sea cucumbers belonging to the Family Synaptidae, that were observed feeding on detrital matter. They were found in shallow subtidal sand pools among seagrasses (Thalassia hemprichii). More unusual, however, was the high abundance of large (c. 5-10 cm) herbivorous sea hares, Dolabella auricularia (Aplysiidae), found in shallow intertidal pools with T. hemprichii, and larger-sized (10-15 cm) red-coloured, side-gilled carnivorous Pleurobranchus forskali (fig. 17.16). The genus Dolabella contains the world's largest sea slugs, some of which may be up to 2 kg live-weight, true sea slug giants. Perhaps due to their large size, Dolabella spp., unlike many other aplysiids (Aplysia), have not acquired the ability to swim, even though their evolutionary reduction of shell certainly offered them the possibility. Their greenish colour is an excellent camouflage, since it matches perfectly with the seagrass background colouration. In fact, they were first noticed only when one specimen was accidentally stepped upon, which prompted it to release copious quantities of deep purple ink, not unlike that of the octopi. The brightly coloured (red) Pleurobranchus, on the other hand, is an opistobranch gastropod (Notaspidea) with a small internal shell, which makes it a close relative to the true sea slugs, the Nudibranchia. It has been reported to feed on sponges, ascidians or coelenterates (Mather and Bennet 1993). The absence of these prey groups in the intertidal area suggests that it may be feeding on other organisms, or be a scavenger.
Figure 17.15. Sketch of Pankaja sand cay in Spermonde Archipelago, South Sulawesi, as described by Kuenen (1933a). Note the absence of shingle ram part and the double crescent shape of the sand bar. The sand cay has smoth ered coral on the east side of the reef. Depths in metres.
From Kuenen 1933a.
The sand cay is a valuable economic asset to a nearby Bajau village. At each low tide, the sand cay is visited by women and children who come in small canoes to collect the abundant bivalves and gastropods. The sand cay is thus an important gathering area, supporting the livelihood of six Bajau families. The abundance of filter-feeding bivalves as well as herbivorous and carnivorous gastropods is probably related to the continual supply of nutrient-rich water. High current velocities (>2.50 m.sec-1) associated with the southward current and bottom bathymetry of the 150-m-deep strait combine to create strong upwelling cells that seem to be ubiquitous features of the Boling Strait. Enhanced primary productivity caused by upwelling in large archipelagic straits is clearly demonstrated in colour plates 22.7 and 22.8. In the past (i.e., 2-4 years ago), the Boling Strait supported one of the most productive pearl-oyster grounds in the region (Pinctada maxima and P. margaritifera), however, overexploitation of adult shells for the pearl-farming industry has decimated the adult populations in the area and placed the former pearl-diving operators out of business. High planktonic productivity is also reflected in the great abundance of bivalves, dominated by Tapes literatus, Anadara antiquata, Asaphis violascens, Pinna muricata, Tellina sp., Fragum undeo, and Fimbria fimbriata. These bivalves are the staple protein of the Bajau community. Casual observations of 18 A. antiquata shells (3-5 cm wide) revealed that four specimens had a single small hole drilled into the ventral region of the left valve, just inside the pallial line (fig. 17.17A). Single drill holes were also observed on the spires (midway between apex and aperture) of numerous auger shells (Terebra crenulata), which were very abundant in the intratidal (fig. 17.17B). These carnivorous gastropods (equipped with poison apparatus similar to cone shells) feed mainly on worms and other intertidal invertebrates. The mystery of what animal had drilled the holes was solved when, among the gastropods collected by the Bajau women, we found numerous specimens of Natica spp. and Polinices spp. (Naticidae; sand snails or Moon shells). These shell-drilling gastropods are major bivalve and gastropod predators.
Figure 17.16. Pleurobranchus forskali (Pleurobranchidae) is found in a wide range of reef habitats and was very abundant on the sand cay shown in figure 17.14A.
Photo courtesy of R. Steene, Cairns.
Figure 17.17. Molluscan gastropod predators such as Naticidae (Moon shells) prey on other bivalve and gastropod molluscs. On productive intertidal sand flats, predation can be intense. A) Left valve of mature Anadara antiquata (5 cm posterior to anterior length) with a clean drill hole on the ventral portion of the valve. B) Even the carnivorous and poisonous auger shells (Terebra crenulata) fall prey to naticid gastropods that are able to drill through the shell of their prey.
Photos by Tomas and Anmarie Tomascik.
Linear Shingle Cays. A number of reefs in the Kepulauan Seribu Complex have only a shingle rampart visible, and fit into Hopley's (1982) linear unvegetated shingle cay category. However, in Kepulauan Seribu, shingle cays are also found that are non-linear. One example is a small reef just west of Pulau Belanda (fig. 17.18). The reef flat is subtidal, and the shingle rampart forms almost a half circle around the east margin of the reef flat. The shingle rampart is about 25-50 m from the seaward margin of the reef. These cay types will most likely be common in high-energy environments. Kuenen (1933a) also observed numerous shingle cays in the Spermonde Archipelago, but noted that they did not occur as regularly as in the Kepulauan Seribu Complex. Many can be classified as submerged shingle cays, since they become exposed only during low water. Shingle ramparts at the south extremity of the Spermonde Barrier Reef are apparently the highest (Kuenen 1933a). Shingle cays at Salisi and Sarapo are about 30 m long, and according to Kuenen (1933a), the reef flat of the two adjacent reefs is covered by fragile coral species that grow to sea level without being broken by the waves. Microatolls protected by shingle ramparts are a very common feature on many reef flats in this area.
Figure 17.18. P. Belanda is the easternmost reef of the Seribu Complex. It is classified as a vegetated mixed sand and shingle cay. Located to the west of the cay is an oval-shaped submerged reef with a semicircular shingle rampart, thus classified as a shingle cay.
From Cook 1985.
Compact Sand Cays. Compact unvegetated sand cays form where strong centripetal transport of sediments occurs (Hopley 1982). Numerous small oval-shaped sand cays in Kepulauan Seribu as well as Spermonde Archipelago fit into this class. In addition, umbgrove (1947) describes a "young island" called Gosong Opak (east of Opak Besar) in the Kepulauan Seribu which has a compact oval shape, but is also protected from the Southeast Monsoon by a well-developed shingle rampart along the east margin of the reef flat. Compact sand cays with a shingle rampart are not uncommon, and may be considered in a class of their own.
Hopley (1982) points out that compact cays make take on another shape as they migrate from one part of the reef flat to another, and this is clearly evident in both Kepulauan Seribu and the Spermonde Archipelago, where numerous sand cays are in various in-between shapes during their reef flat migration (i.e., linear). One explanation for this is the seasonal monsoonal reversal in wind direction as well as in wave energy. Wave regimes are distinctly different during the two monsoons, which facilitates sand migration. On more stable cays, the cay may stand intact, however, its shape will change, and distinct crescent-shape forms are common (fig. 17.19). Numerous unvegetated compact sand cays occur in the lagoon of the Spermonde Archipelago, however, their geomorphology has not been investigated in detail.
Figure 17.19. Taka Blukuran is located in the Madura Strait, south of Madura, and is fully exposed to the influence of the Northwest and Southeast Monsoonal winds. During the Southeast Monsoon, when the winds are blowing from the east, the crescentic sand patch is open to the west, whife during the Northwest Monsoon, when winds are from the west, the crescentic patch is open to the east.
Compact Shingle Cays. The compact, unvegetated shingle cays are characteristic of high-energy environments and may form on small reefs that are fully exposed to the monsoonal winds. Their presence in Indonesian waters is likely, especially along the southwest coast of Sumatra, where numerous offshore reefs are exposed to strong Indian Ocean swell. Kuenen (1933a) mentions the presence of numerous reefs with unvegetated and vegetated shingle ramparts on the Great Sunda Barrier Reef, but no surveys were conducted.
Solitary Vegetated Cays
Sand Cays. Solitary vegetated cays are the most visible and familiar of all coral cays. The main characteristic of vegetated cays is their relative stability in relation to the underlying reef. The size range of vegetated cays in Indonesia is wide, from less than 0.5 ha to over 15 km2. Small cays only a few hundred square metres in area abound in Kepulauan Seribu and are too numerous to mention, however, very little is actually known about the cay vegetation. Kuenen (1933a) noted that many cays in the Spermonde Archipelago were being destroyed by mining for coral blocks, and that many unvegetated linear sand cays disappeared from the charts in heavily mined areas. In contrast, the sand cays in Kepulauan Seribu are large relative to their reefs, but unfortunately more heavily exploited. Fishing with explosives (i.e., reef blasting) has caused a great deal of damage to the entire Kepulauan Seribu patch reef complex. Considering the increasing levels of pollution affecting the area, the damaged reefs may not be able to recover even if this illegal practice is stopped. Since the existence of coral cays is tied directly to the well-being of the reefs, the prognosis for many cays in Kepulauan Seribu is indeed very bleak.
Figure 17.20. Vegetated sand cays in Kepulauan Seribu are dynamic entities that undergo large-scale fluctuations in shape and size related to environmental forcing. Air Kecil was first mapped in about 1875 and underwent considerable changes, even though the island was a well-developed vegetated sand cay, before finally disappearing.
From Stoddart 1986.
Large coral cays occur in the Am Islands, and among the largest is Pulau Enu, with an area of c. 15.0 km2. Vegetation covers about 90% of the island. Pulau Enu is one of the most important green turtle nesting beaches in the world (Schulz 1987). Detailed phycological studies have not been conducted. The presence of a large turtle breeding population is indicative of the long stability of the island. Its immense size is difficult to explain. The island is located about 30 km off the southeast coast of Tarangan. The densely vegetated and uninhabited cay occupies 90% of the reef flat. White sandy beach skirts the entire island. The reef is situated on a flat shelf not more that 24 m deep. The relative stability and apparent longevity of such large cays as Pulau Enu is a result of an appropriate set of abiotic environmental conditions, as well as biological and chemical processes initiated and perpetuated by colonizing vegetation and associated micro- and macro-fauna. Beach rock may play an important role in stabilization of cays, however, on many cays it is not visible. On Pulau Enu, as well as on Pulau Sangalaki, frequent disturbance by nesting turtles prevents the formation of beach rock, yet this absence does not seem to be detrimental to the cay's long-term stability. It should be pointed out that stability is only in relative terms, and that many vegetated cays are highly dynamic entities known to undergo extensive changes over time in response to changing environmental conditions (fig. 17.20).
Figure 17.21. Sketch map of Pancallarang group located north of P. Putri, Kep. Seribu Complex. Hatched areas indicate location of cays. Note Gosong Pancallarang, a linear unvegetated cay south of Pancallarang Kecil.
From Cook 1985.
Vegetated sand cays vary in form as much as they vary in size. Even within a small complex such as Kepulauan Seribu, we find an amazing diversity of shapes (fig. 17.21). However, there is a persistent trend for an east-west elongation of cays, particularly in more exposed locations (fig. 17.22). Hopley (1982) points out that elongated cays result from two-directional wave convergence, which in Kepulauan Seribu is greatly enhanced by the bidirectional nature of the monsoonal system. Oval-shaped cays have a tendency to develop at the lee of larger reef platforms (fig. 17.22) where more centripetal sweep of waves occurs (Hopley 1982). The size and shape of cays is also indicative of their elevation above sea level. Large cays are 2 to 3 m above MHWS, with the oval-shaped cays having their central portion raised slightly above the surrounding area.
Most natural cays in the Kepulauan Seribu Complex are densely vegetated as a result of humid tropical climate (fig. 17.23). Rainfall is highly seasonal, with the Northwest Monsoon bringing in most of the rainfall, with an annual rainfall of about 1500 mm (Wyrtki 1957, 1961; Silvius et al. 1987). However, during exceptionally wet years, rainfall can be much higher. For example, Pulau Pari in the south sector of the reef complex received 2203 mm of rain in 1986 (Ongkosongo 1988). While rains are frequent throughout the Northwest Monsoon (December - March), the heaviest precipitation occurs mainly during December through February. The Southeast Monsoon (May through November) is the dry season during which evaporation exceeds precipitation (mid-June to mid-October), however, the Kepulauan Seribu islands still receive about 50 mm/month of precipitation (Wyrtki 1957; Ongkosongo 1988). Air temperatures in Kepulauan Seribu are very similar to those measured for Jakarta (Wyrtki 1957). During the cool Northwest Monsoon average air temperatures are about 26°C, while during the hot Southeast Monsoon average temperatures increase to about 28°C (maximum between April-May). However, daily temperature fluctuations can range from a low of 21 °C to a high of 36°C.
Figure 17.22. Sketch map of Pulau Putri and surroundings, north sector of Kep. Seribu Complex. Hatched areas indicate location of cays. Note narrow inter-reef channels.
From Cook 1985.
The moist humid climate is partly reflected in the natural vegetation (what is left of it) of the cays. Mature sand cays in Kepulauan Seribu, such as Pulau Pari, are vegetated by a number of plant species, the most common being Terminallia catappa, Pandanus spp., Casuarina equisetifolia, Denris trifoliata, Barringtonia sp., Ardisia humilis, Hibiscus tiliaceus, Calophyllum inophyllum and Desmodium umbellatum (Silvius et al. 1987). Many cays in Kepulauan Seribu have a mangrove fringe along parts of their coastline. At Pulau Pari, for example, well-developed mangrove covers 25% of the island, even though the diversity of the mangrove community is low. Rhizophora stylosa is among the most conspicuous mangroves in Kepulauan Seribu able to survive in a variety of environments that range from sheltered sandy beaches, to reef flats to highly exposed shingle ramparts. Mangroves that tend to dominate on Pulau Pari are Rhizophora stylosa, Sonneratia acida, S. alba, Bruguiera spp. and Aegicerassp. (Silvius et al. 1987).
Figure 17.23. The monsoonal climate of the west Java Sea. Mean monthly oceanographical (temperature, salinity, surface current velocity and direction) and climatological data (temperature, relative humidity, precipitation and evaporation, wind velocity and direction) for west Java Sea. Northwest Monsoon (rainy season) from December through March. Southeast Monsoon (dry season) May through November.
Figure from Brown 1991, modified after Wyrti 1961.
Mixed Sand and Shingle Cays. The sediments of vegetated mixed sand and shingle cays consist of shingle and sand. Hopley (1982) discussed the possible mechanism for their formation, which requires high-energy conditions at some point in their development. The shingle may originate from a shingle rampart that was invaded by sand from the lee side due to changes in environmental conditions. This scenario fits well with the bidirectional monsoonal wind system. Many islands in Kepulauan Seribu have a well-developed shingle rampart facing the weather side. While the reef complex lies outside the cyclone corridor, frequent high-intensity, short-duration storms during the Northwest Monsoon may generate sufficient force to generate shingle ramparts, while sand deposition continues during the calmer Southeast Monsoon. Stoddart and Steers (1977) point out that many atoll islands belong to this class. It is currently not known how many cays in Kepulauan Seribu would fit into this classification.
Shingle Cays. Vegetated shingle cays may be present on the Great Sunda Barrier Reef. Kuenen (1933a) observed that development of shingle ramparts on many of the reefs on the Great Sunda Barrier Reef were not obviously oriented with the direction of the prevailing wind, suggesting that other factors (e.g., currents) may play a role in their construction. Some islands seemed to have been completely surrounded by ramparts, with the interior of the cays vegetated by bushes and palm trees. Kuenen (1933a) observed: "The only other case of shingle ramparts with a vigorous vegetation I saw was on the Postiljon and Paternoster islands," and noted that in the western Flores Sea shingle ramparts are built in a different manner than they are in Jakarta Bay (umbgrove 1929c).
Mangrove Cays. Examples of mangrove islands that fit Hopley's (1982) classification are found in the Berau Islands. Pulau Panjang is located 15 km off East Kalimantan's mainland. It is situated on a large linear reef that is 30 km long and 8 km wide. Vast areas of the reef flat are subtidal, covered by about 0.5 m of water at low tide. The 4-km-long and 1.5-km-wide island is uninhabited, and densely vegetated by mangrove forest. The dominant seaward mangrove was Rhizophora apiculata. The Berau Islands in general are located in a relatively low-energy environment not affected by strong monsoonal winds. As a result, shingle ramparts are absent from all reefs in the area. Hopley (1982) pointed out that conditions necessary for development of mangrove islands are: 1) high reef tops; 2) low-energy conditions; and 3) relatively low tidal range. Environmental conditions in the Berau Islands satisfy criteria 1 and 2, however, tidal amplitude in this area is 2.8 m, which is considered high. Note that information is also lacking on the type of vegetation in the interior of the island.
Located about 25 km south of Pulau Panjang is Pulau Samama. The reef is situated 50 km east of the Berau River delta in a delta-front setting. The patch reef is about 4 km long and 2 km wide and the densely vegetated island occupies roughly 35% of the reef flat area. With a 2.8 m tidal range, most of the island is subtidal and vegetated by mangrove. The shingle rampart is absent and a large sand spit has formed at the southwest margin of the reef, about 500 m from the island. The reef flat is subtidal, covered by 0.3 m of water at low spring tides. At low tide most of the mangrove area was covered by about 0.2 m of water.
The dominant mangrove at the seaward margin of the island was Rhizophora spp. (possibly R. stylosa), with larger trees located in the interior of the island (e.g., Bruguiera spp., Sonneratia spp., Aegiceras spp. and Avicennia spp.). The island is a reported nesting site for frigatebirds, but none were seen, nor was any evidence of nesting observed. However, flocks of Fregata ariel were observed soaring daily high above the neighbouring islands. Of greater interest, however, was the sighting of Fregata andrewsi, which is known to nest only on Christmas Island in the Indian Ocean, just south of Java. This is obviously of considerable interest and additional surveys in the area should be conducted.
Interspersed among the mangrove, and deep into the forested area of the island, were large mixed stands of Enhalus acoroides and Thalassia hemprichii (fig. 17.24). These seagrass meadows were found in areas where the mangrove canopy was broken. However, E. acoroides was found adjacent to Rhizophora roots in a relatively shaded environment. Massive Porites spp. and Goniastrea spp. were abundant on the reef flat where they formed microatolls, but perhaps more interesting was their presence within the mangrove. Large monospecific stands of Goniastrea asperacovered large areas, up to 50 m2, and massive colonies of Goniastrea spp. formed medium-sized microatolls. Each microatoll invariably had a resident damselfish (Dischistodus pseudochrysopoecilus) that was maintaining an algal garden on the top of the microatoll (fig. 17.25).
Figure 17.24. Mixed Enhalus acoroides and Thalassia hemprichii meadow on Pulau Samama in a clearing about 100 m inside a mangrove forest. Coral is Goniastrea aspera;monarch damselfish (Dischistodus pseudochrysopoecilus) above.
Photo by Tomas and Anmarie Tomascik.
The presence of corals in a mangrove setting is possible due to high tidal range and strong currents that prevent sedimentation. The water clarity within the mangrove at high tide was similar to visibility on reefs in the area. Currents flowing across the reef flat and the mangrove had velocities of about 75-100 cm.sec-1. It was interesting to observe that the abundance of boring macroinvertebrates was considerably higher in the mangrove area than on the adjacent reef flat and reef slope. Localized enrichment effect within the mangrove may be one explanation.
Multiple Islands. Reefs with more than two islands are not a rare occurrence in the archipelago. According to Hopley (1982), this island class does not occur on the Great Barrier Reef in its strictest form. However, there are two examples from Kepulauan Seribu that do fit the definition. The 2.5-km-long Lancang Platform is located 10 km from mainland Java, and has three discrete vegetated sand cays. The largest cay is Pulau Lancang Besar (c. 0.5 km2), which occupies the west part of the platform. Pulau Lancang Kecil (c. 0.35 km2) occupies the southwest part of the reef flat, and the smallest cay, Pulau Gosonglancang (c. 0.06 km2), is located north of Pulau Lancang Kecil. Shingle ramparts are located along the northwest side of the platform. Because the islands have been inhabited for decades, the natural vegetation has been replaced by exotic (i.e., introduced) plants, and most of the area is cultivated.
The second example of multiple islands is the Pulau-Pulau Tidung Complex located 8 km north of Pulau Lancang. The Tidung Platform is 7 km long and has two shallow lagoons. The 2.5-km-long and c. 300-m-wide Pulau Pari is the largest of the cays. The vegetation of Pulau Pari was described earlier. In addition there are four other discrete sand cays, namely Pulau Tikus, Pulau Burung, Pulau Komgsi, Pulau Tengah and Pulau Tonggang. Tidung Reef is perhaps the most studied reef in Indonesia since the LIPI research laboratory is located on Pulau Pari. However, detailed geomorphological studies have not been conducted.
Figure 17.25. In small clearings deep inside the mangrove forest of P. Samama, massive Porites spp. and Goniopora aspera form wide microatolls. The monarch damselfish, Dischistodus pseudochrysopoecilus, is guarding its algal patch on top of the Porites sp. microatoll. Note the large density of Christmas tree worms, Spirobranchus giganteus corniculatus, and two giant clams Tridacna crocea. On P. Samama the abundance of boring macroinvertebrates is exceptionally high.
Photo by Tomas and Anmarie Tomascik.
Complex Low Wooded Islands.
On the inner reefs north of Cairns are a group of reef islands with a complexity unique to the Great Barrier Reef.—HOPLEY 1982
The islands being described "consist of a windward shingle island and leeward sand cay, with intervening mangrove in the lee of the shingle" (Hopley 1982). One of the characteristic features of this island class is the presence of mangroves. Low wooded islands can occupy up to 75% of the reef flat or more, but usually it's between 25%-50% (Stoddart et al. 1978). While we do not know whether or not there are low wooded islands in the archipelago, one island that is unique among the Jakarta Bay and Kepulauan Seribu cays is Pulau Rambut,mentioned earlier (fig. 17.26). Pulau Rambut is located just outside of Jakarta Bay, about 4 km from mainland Java. To the south of the island, the aggrading Java Shelf extends 3 km offshore as subtidal (0.5 m below low tide) and intertidal mud flats and sand bars. Pulau Rambut is separated from this shallow shelf area by an 18-m-deep (c. 750 m wide) channel. Figure 17.27 illustrates the rapid advance of north Java's coastline during the past 6000 years. The relatively deep channel between Java and Pulau Rambut will probably slow down the tombolo process (fig. 17.28). Formation of tombolos in the past has resulted in the incorporation of offshore patch reefs into the present-day coastal plain along the north coast of Java. Tanjung Priok and Kamal Reef are two examples (Ongkosongo 1988). The formation of the wide alluvial coastal plain that separates Muria volcano (northeast of Semarang, Central Java) from mainland Java may have also originated from the tombolo process. Tombolos occur when mud, sand or gravel bars, or barriers, connect an island with the mainland (fig. 17.29). This process may also connect two adjacent islands.
Figure 17.26. Pulau Rambut is a unique low wooded island in the southwest Java Sea.
From Stoddart 1986.
Figure 17.27. The north coast of West Java has advanced rapidly during the past 6000 years. The black areas in Java Sea represent present-day locations of active carbonate deposition. Vertical lines mark the Holocene strand plain deposited during the past 6000 years. The Pleisto-Holocene Java mainland is the siliciclastic shelf. There are many recent examples where pre-Holocene reefs have been incorporated into the coastal plains through the tombolo process.
From Brown 1992.
Figure 17.28. In the past, offshore patch reefs were joined with Java mainland through the formation of tombolos. Tombolos are sand or gravel bars or barriers that connect an island with the mainland. In this figure, a tombolo may be observed to be forming between mainland Java and Pulau Rambut (A) in the southwest Java Sea. Depths in metres (error ±1 m). Not to be used in navigation.
Figure 17.29. Schematic illustration of a hypothetical tombolo formation, and subsequent incorporation of a patch reef, such as Kamal Reef (northwest of Jakarta), into the alluvial plane.
After Ongkosongo 1988.
Pulau Rambut, with a total area of 56 ha, stands on a patch reef with an area of about 100 ha (Kartawinata and Walujo 1977). Mangroves account for 18 ha of the total area while the woodland covers the rest. The mangrove occupies the east and north sections of the island with a narrow fringe along the west coastline. The woodland occupies the western part of the island and is dominated by Sterculia foetida and Chisocheton pentandrus. The secondary growth is dominated by Scyphiphora hydrophyllaceae, Lumnitzera racemosa scrub and climbers such as Wedelia biflora and Merremia sp. (Cucurbitaceae). The maximum elevation of the island is 3 m above sea level, while the tidal amplitude is <1.0 m. The island is fringed along the east, north and west coasts by a well-developed shingle rampart located just seaward to the mangroves. A wide moat exists on the west side of the island, and numerous microatolls (Porites spp.) are found along the north coast. The north and northwest margins of the reef are fringed by a wide coral rubble zone. The shingle rampart consists of coral rubble that is thrown up during the Northwest Monsoon.
A large portion of the island has primary vegetation cover and is uninhabited. However, some cultivation was done in the interior of the island some years ago, and the area is now covered by a secondary shrubland (Silvius et al. 1987). Since the island is an important bird sanctuary with limited access, it has maintained most of its natural condition. The beach vegetation along the south and southwest coasts consists of Ipomoea pes-caprae community. The mangrove forest is not highly diverse (17 species) but is well developed (table 17.4). The dryland primary woodland is dominated by Sterculia foetida and Chisacheton pentadorus, while the secondary shrublands consist of Wedelia biflora and Merrennia sp. (Silvius et al. 1987). The Sterculina trees serve as a roosting place for thousands of flying foxes (Pteropus vampyrus).
The coral reef upon which the island stands has not been inventoried, however, the reef flats and surrounding shallow-water habitats are covered by a mixed sea-grass, community consisting of Enhalus acoroides and Thalassia hemprichii. Major seagrass meadows are found in the moats (fig. 17.26) on the west side of the reef, which is protected by a well-developed shingle rampart. Pulau Rambut has a significant biodiversity and conservation value. With 10 breeding waterbird species, the island has been classified as one of the most important breeding sites of Java (table 17.5). One of the most important breeding waterbirds is the milky stork (Mycteria cinerea), which, together with the roosting Leptoptilos javanicus (lesser adjutant), are listed in IUCN's Red List of Threatened Animals, and classified as 'vulnerable' (IUCN 1990). In addition to waterbirds, the island has a high avifaunal diversity (table 17.6).
Table 17.4. The mangrove community of Pulau Rambut, Java Sea.
The Future of Coral Cays
The numerous vegetated coral cays of the Kepulauan Seribu reef complex are the backbone of a multimillion-dollar tourism industry that is steadily expanding. Unfortunately, the patch reefs, and particularly the vegetated coral cays, are very misunderstood systems. The lack of basic understanding of the dynamic nature of coral cays has resulted in significant damage to numerous coral reefs throughout the archipelago. Few developers and planners realize that coral cays are an integral part of the reef system, and that they owe their very existence to the well-being of the reef upon which they are built. Coral cays are very dynamic entities, continually changing according to the environmental conditions. The lack of knowledge and basic understanding of these systems has led to highly destructive management practices. It is, therefore, not surprising that the degradation of these systems has intensified and accelerated in recent years (fig. 17.30).
Table 17.5. The waterbirds of Pulau Rambut, southwest Java Sea. '*' = breeding species. Species in bold-face are on IUCN's Red Data Book list.
In a misguided effort to protect an initially ill-advised development of numerous coral cays in Kepulauan Seribu, many developers have decimated the reef flats to obtain coral rock material for the construction of 'reef walls', which they proceeded to build around entire islands. These actions have resulted in reef flat instability, and, as a result, water turbidity in the reef complex has increased dramatically. Resuspended sediments from the reef flats are now affecting deeper reef slope coral communities through direct smothering as well as through reduction of light penetration. Specific long-term environmental and ecological consequences of these actions are difficult to predict, since current monitoring studies are restricted to too few areas for meaningful interpretation. What is abundantly clear, however, is that a considerable amount of funding will be required in the future to keep the natural processes at bay, since the natural coral reef breakwaters that have protected these islands for decades to centuries have been mostly destroyed or made functionally ineffective.
Table 17.6. The avifauna of Pulau Rambut, southwest Java Sea, excluding waterbirds.
Figure 17.30. The life history of Ubi Kecil, a sand cay in Jakarta Bay.
From Stoddart 1986.