Coral transplantation : A useful management tool or misguided meddling?

PII: Marine Pollution Bulletin Vol. 37, Nos. 8±12, pp. 474±487, 1998 Ó 1999 Elsevier Science Ltd. All rights reserved Printed in Great Britain 0025-326X/99 $ - see front matter S0025-326X(99)00145-9 Coral Transplantation: A Useful Management Tool or Misguided Meddling? ALASDAIR J. EDWARDS* and SUSAN CLARK Centre for Tropical Coastal Management Studies, Department of Marine Sciences and Coastal Management, University of Newcastle, Newcastle upon Tyne NE1 7RU, UK The primary objectives of coral transplantation are to improve reef `quality' in terms of live coral cover, biodiversity and topographic complexity. Stated reasons for transplanting corals have been to: (1) accelerate reef recovery after ship groundings, (2) replace corals killed by sewage, thermal e‚uents or other pollutants, (3) save coral communities or locally rare species threatened by pollution, land reclamation or pier construction, (4) accelerate recovery of reefs after damage by Crown-ofthorns star®sh or red tides, (5) aid recovery of reefs following dynamite ®shing or coral quarrying, (6) mitigate damage caused by tourists engaged in water-based recreational activities, and (7) enhance the attractiveness of underwater habitat in tourism areas. Whether coral transplantation is likely to be e€ective from a biological standpoint depends on, among other factors, the water quality, exposure, and degree of substrate consolidation of the receiving area. Whether it is necessary (apart from cases related to reason 3 above), depends primarily on whether the receiving area is failing to recruit naturally. The potential bene®ts and dis-bene®ts of coral transplantation are examined in the light of the results of research on both coral transplantation and recruitment with particular reference to a 4.5 year study in the Maldives. We suggest that in general, unless receiving areas are failing to recruit juvenile corals, natural recovery processes are likely to be sucient in the medium to long term and that transplantation should be viewed as a tool of last resort. We argue that there has been too much focus on transplanting fast-growing branching corals, which in general naturally recruit well but tend to survive transplantation and re-location relatively poorly, to create short-term increases in live coral cover, at the expense of slow-growing massive corals, which generally survive transplantation well but often recruit slowly. In those cases where transplantation is justi®ed, we advocate that a reversed stance, which focuses on early addition of slowly recruiting massive species to the recovering community, *Corresponding author. 474 rather than a short-term and sometimes short-lived increase in coral cover, may be more appropriate in many cases. Ó 1999 Elsevier Science Ltd. All rights reserved Introduction As pressures on coral reef resources have increased as a result of demographic changes so has degradation of the support systems (coral reef ecosystems) which produce those resources. Degradation of coral reefs results from human-induced impacts such as dredging, coral quarrying, sewage discharge, dynamite ®shing, chemical pollution, oil spills, ship groundings, tourist damage and run-o€ of sediment, fertilizer and pesticides as a result of changing land-use (e.g., Brown and Howard, 1985; Clark and Edwards, 1995; Salvat, 1987; Hatcher et al., 1989; Rinkevich, 1995). Thus, as burgeoning coastal communities have become increasingly dependent on a continuing supply of resources such as reef ®sh, molluscs, algae and crustaceans generated by coral reef and associated ecosystems so the functioning of those same ecosystems is being eroded by human activities. These anthropogenic pressures on reefs have been exacerbated in the 1980s and 1990s by several strong El Ni~ no Southern Oscillation events, which have been correlated with widespread warm water anomalies and associated ÔbleachingÕ and mortality of corals (Glynn, 1984, 1993; Glynn and DÕCroz, 1990; Wilkinson et al., 1999). Recognition of the value of coral reefs, the development of marine parks in coral reef areas and increased e€orts focused on reef management have resulted in widespread interest in reef rehabilitation using coral transplantation as an aid to management in areas where coral reefs have been degraded (e.g., Guzman, 1991; BowdenKerby, 1997; Hudson and Goodwin, 1997; Mu~ nozChagin, 1997; Oren and Benayahu, 1997; Lindahl, 1998). Indeed there is some danger that coral transplantation may be being oversold as a management tool. In the Maldive Islands of the central Indian Ocean building materials are scarce and traditionally coral rock Volume 37/Numbers 8±12/August±December 1998 has been quarried from shallow reef ¯ats for use in the construction industry. Similarly, coral rubble has been collected for use as aggregate and coral sand for making mortar and concrete. Since the 1970s demand for coral rock has been very high in the vicinity of the capital island Male where 26% of the country's 275,000 population live and where tourism and urbanization have developed extremely rapidly. Coral rock is quarried manually using crow-bars to break up the reef ¯at. It usually involves removal of the top half metre or so of the living reef, leaving behind a severely degraded wasteland of shifting rubble and sand o€ering minimal shelter to ®sh and other animals. Some reefs quarried over 25 years ago have shown virtually no recovery and this has been tentatively attributed to both a lack of suitable surfaces for settlement of coral larvae and smothering and abrasion of juvenile corals by the highly mobile sediment (Brown and Dunne, 1988). In a multifaceted study of the factors constraining recovery of these degraded reefs (Clark and Edwards, 1994, 1995, 1999; Edwards and Clark, 1992, 1993), we looked at ®rst, whether provision of stable surfaces in the form of concrete arti®cial reefs would be sucient to allow coral settlement and growth and secondly, whether transplantation of corals to such surfaces was justi®ed in terms of signi®cantly accelerating recovery. Detailed monitoring was carried out for four and a half years with subsequent sporadic visits. This paper examines the use of coral transplantation in coral reef rehabilitation in the light of the results of the Maldives study and other recent research on coral recruitment and transplantation. Although each reef rehabilitation project needs to be evaluated on its own merits, we attempt to provide some general guidelines on the use of transplantation based on this research. To provide a framework for discussion of reef rehabilitation using coral transplants, we ®rstly examine why coral transplantation has been carried out and highlight some key broad issues. Secondly, we explore in more detail what the potential bene®ts and dis-bene®ts of transplantation are with reference to recent research. Why Transplant Corals? Table 1 lists examples of transplantation studies and their ultimate aims. Many of these studies, like our work in the Maldives, were essentially feasibility studies to discover whether the aims listed might be achievable. From a management viewpoint two important points, which emerge from these studies, are that: (1) there is no point in transplanting colonies to areas where water quality is poor as they will tend to die, and (2) loss of transplants in high-energy environments tends to be high whatever methods are used to attach them. However, where water quality was good in relatively lowenergy environments, transplants tended to survive reasonably well. The aims listed in Table 1 are diverse but have some common elements. Firstly is the idea that natural recovery is inadequate in some way (e.g. too slow) and requires human assistance. This stems partly from human impatience ± ®ve to ten years to recover from a small localised disturbance to one to several decades for larger impacts (Alcala and Gomez, 1979; Connell, 1997; Curtis, 1985; Grigg and Maragos, 1974; Guzman, 1999; Maragos, 1974; Pearson, 1981; Shinn, 1976) seems a long time ± and partly from a political need to be seen to be doing something. The scienti®c case for using transplantation to speed recovery is often less than compelling but may be argued for areas where natural recruitment is poor (e.g., Guzman, 1991; but see later conclusions: Guzman, 1999) or where potential bene®ts clearly outweigh the dis-bene®ts (Table 2). Secondly is the wish to preserve coral colonies threatened by pollution, reclamation or other human activities. This is entirely laudable but we should be beware of setting a dangerous precedent. If politicians or other decision TABLE 1 Examples of studies of coral transplantation and the reasons why corals were transplanted. Site Reason for transplanting corals Philippines, Indonesia Aid reef recovery following dynamite ®shing Guam Guam Singapore, Cozumel Island, Florida Hawaii Replace corals killed by thermal e‚uent Save rare coral species threatened by pollution Relocate coral colonies (and other reef organisms) threatened by reclamation, pier construction, outfall repair, respectively Reintroduce species into an area previously polluted by sewage, dredging, etc. Accelerate reef recovery following ship groundings Florida, Cayman Islands Gulf of Aqaba Eilat Costa Rica Great Barrier Reef Enhance attractiveness of tourism area Rehabilitate tourist damaged reefs, create arti®cial reefs to relieve diving pressure Rehabilitate coral reefs severely impacted by 1982±1983 El Ni~ no warming and 1985 dino¯agellate blooms Accelerate recovery of reefs damaged by Crown-of-thorns star®sh References Auberson (1982), Yap et al. (1990, 1992), Yap and Gomez (1984), Fox et al. (1999) Birkeland et al. (1979) Plucer-Rosario and Randall (1987) Newman and Chuan (1994), Mu~ noz-Chagin (1997), Dodge et al. (1999) Maragos (1974), Maragos et al. (1985) Gittings et al. (1988), Hudson and Diaz (1988), Miller and Barimo (1999), Jaap (1999) Bouchon et al. (1981) Rinkevich (1995), Oren and Benayahu (1997) Guzm an (1991, 1993, 1999) Harriott and Fisk (1988b) 475 Marine Pollution Bulletin TABLE 2 Potential bene®ts and drawbacks of transplanting corals. Potential bene®ts Immediate increase in coral cover and diversity Increased recruitment of coral larvae as a result of presence of transplants Survival of locally rare and threatened coral species when primary habitat is destroyed Reintroduction of corals to areas which are larval supply limited or have very high post-settlement mortality Improved aesthetics of areas frequented by tourists Instant increase in rugosity and shelter for herbivores in bare areas makers think that corals can always be transplanted if they get in the way of development, then there is little incentive to resolve the management issues leading to the coral reefs being threatened in the ®rst place. Also the widespread emphasis on transplanting corals per se as `¯agship' species is misleading because they are but one component of a complex ecosystem whose other components thus tend to be ignored (but not always, e.g. Newman and Chuan, 1994; Mu~ noz-Chagin, 1997). Transplanting a few coral colonies does not mean you have transplanted a reef ecosystem or its associated goods and services. Bene®ts and Dis-bene®ts of Transplantation In this section we examine the potential bene®ts and drawbacks of transplantation methodologies (Table 2) in the light of research ®ndings. Potential adverse environmental impacts will be considered ®rst. Potential drawbacks (dis-bene®ts) 1. Loss of coral colonies from donor reef areas The corals to be transplanted have to come from somewhere. In general they are likely to be taken from adjacent undamaged or less damaged reef areas. These donor areas need to be suciently large and rich in coral colonies that they themselves will not be signi®cantly impacted by the removal of transplant material. Either whole colonies or fragments of colonies may be transplanted. Where fragments survive well, the ability to produce several viable colonies from one donor colony is clearly attractive, increasing potential bene®ts and decreasing the amount of damage to donor areas. However, the fact remains that there is collateral damage to the environment from transplantation and recovery from this damage may be slow. For example, Lindahl (1998) found that donor areas showed no recovery 2 years after collection of corals. Clearly this damage should be minimised and only a small proportion of available colonies in donor sites should be selected for transplantation and, where fragments are being transplanted, at least 50% of donor colonies should be left intact (see Harriott and Fisk (1988b), 476 Potential dis-bene®ts Loss of coral colonies from donor reef areas Higher mortality rates of transplanted corals Reduced growth rates of transplanted corals Loss of transplanted colonies from reef as a result of wave action (attachment failure) Reduced fecundity of transplanted colonies due to stress of transplantation Raised public expectations followed by disillusionment when transplants su€er high mortality Miller et al. (1993) and Lindahl (1998) for guidelines to reef rehabilitation methodologies). Where juvenile mortality is expected to be relatively low and substrata suitable for settlement are available, transplantation of gravid adult colonies may result in the greatest return for a given e€ort (cost) and loss of material from the donor area (Richmond and Hunter, 1990; Rinkevich, 1995). Recently, several workers have looked at ways to minimise or circumvent the problem of damage to donor reef areas. Rinkevich (1995) coined the term `gardening coral reefs' to describe the strategy of using (a) small colonies or fragments maricultured in nursery areas, or (b) spawned gametes and shed planula larvae which have been allowed to develop, settle and metamorphose in the laboratory, for transplantation into degraded areas. This approach is elaborated further for Stylophora pistillata planulae and autotomised fragments of the soft coral Dendronephthya hemprichi by Oren and Benayahu (1997) and for the culture of small coral fragments (down to 1±2 cm) by Franklin et al. (1998). Bowden-Kerby (1997) suggested a two-step `coral gardening' methodology for backreef and reef ¯at rehabilitation. Firstly, reef ¯at rubble areas are used as a nursery to culture unattached fragments into 25±50 cm3 colonies over 2±3 years. These colonies are then either used as a source of further fragments for transplantation or transplanted intact to create patches in backreef lagoons. The feasibility of this requires testing as the studies on which it was based followed survivorship over only 3 months. Reef ¯at rubble areas can be some of the most physically testing of environments and seem less than ideal as nursery areas. Raymundo et al. (1999) used planulae collected in laboratory aquaria from wild adult colonies of Pocillopora damicornis, allowed to settle and then reared for up to six months (to at least 10 mm diameter) to seed reefs in the Philippines. They recorded over 95% survival of transplants >10 mm diameter over 6 weeks. Szmant (1999) indicated that a similar approach should be feasible with the broadcast spawners Montastraea annularis (sensu lato) and Acropora palmata, with successful collection of spawn, and larval culture and settlement in the laboratory or ®eld now achieved. Sammarco et al. Volume 37/Numbers 8±12/August±December 1998 (1999) suggest a novel technique where larvae are cultured in laboratory tanks until fully developed and competent to settle, then seeded into the centre of eddies associated with target reefs. They consider that larvae could be retained in the eddies for 1±3 weeks promoting enhanced local settlement and possibly one to two orders of magnitude better regeneration rates than would be achieved by transplantation of juvenile or adult colonies, respectively. These studies indicate how collateral damage (loss of corals from donor areas) could be minimised and recruitment success enhanced by rearing in the laboratory or in protected nursery areas. The feasibility on a large scale and relative costs of these various approaches, which may be considerable, need to be evaluated. 2. Higher mortality rates of transplanted corals Even with careful handling, transplanted colonies tend to have higher mortality rates than undisturbed colonies (Plucer-Rosario and Randall, 1987; Yap et al., 1992) and thus the act of transplantation is putting coral colonies at risk. This risk may be small for some species, such as Pavona spp. and Heliopora coerulea, but may be signi®cantly higher for others, notably fast growing branching species, such as Acropora spp. and P. damicornis (Auberson, 1982; Plucer-Rosario and Randall, 1987; Yap et al., 1992). As noted by Clark and Edwards (1995), there tends to be a trade-o€ between growth rates of transplants (Fig. 1) and survivorship (Fig. 2). The use of fragments for transplantation is attractive for the reasons listed in the previous section. Although many coral species naturally reproduce by fragmentation (see Table 1 of Highsmith (1982) for details), studies indicate that, for some of the same species, survival of arti®cially transplanted fragments may be very poor (Auberson, 1982; Harriott and Fisk, 1988a,b). For example, Yap et al. (1992) recorded mortality rates for transplanted fragments of Acropora hyacinthus in the Philippines which were about 30 times those we recorded for transplanted whole colonies (average diameter 17 cm) in the Maldives. Highsmith (1982) proposed a general relationship between fragment size and survivorship in corals, with a continuum of strategies from production of many small fragments that survive relatively poorly to production of a few large fragments with high survivorship. However, although the size of fragments seems to be inversely related to mortality for some species, this does not appear true for all species. Highsmith et al. (1980) found size-dependent survivorship of hurricane generated A. palmata fragments (with high mortality of small fragments) and Harriott and Fisk (1988b) reported sizedependent survivorship of transplanted Acropora fragments (>30 cm, 10±30 cm and <10 cm) followed over 7 months. Similarly, Bowden-Kerby (1997) found signi®cantly greater survival of larger Acropora cervicornis fragments (8±12 cm and 15±22 cm) than smaller ones (3± 5 cm) on reef ¯at rubble areas, and that unattached 8±12 cm Acropora fragments transplanted to backreef sand areas all died whereas 95% of larger (>30 cm) colonies Fig. 1 Growth rates over 28 months of principal species of corals transplanted onto concrete mats on a high-energy Maldives reef ¯at estimated from repeated measurements of least and greatest diameters of colonies (Clark and Edwards, 1995). Colonies which showed negative or zero growth rates because of partial mortality, breakage or predation were excluded from the analysis. 477 Marine Pollution Bulletin Fig. 2 Mortalities after two years for nine coral species transplanted onto concrete mats on a high-energy reef ¯at in Maldives. Transplants represented a cross-section of the neighbouring unimpacted reef ¯at assemblage. n is the number of transplants of each species at the start of the study. Acroporid and pocilloporid transplants as a group were three times as likely to die as poritids and faviids (p < 0.01). survived. Smith and Hughes (1999) also found size-dependent survivorship of fragments of three species of Acropora, with the species that naturally produced the fewest fragments (A. hyacinthus) having the lowest survival (8% after 17 months). A. intermedia which produced relatively large fragments at an intermediate rate had the best survival (32% after 17 months). By contrast, Bruno (1998) found no overall signi®cant increase in survivorship with fragment size in Madracis mirabilis and no inter-speci®c relationship between fragment size and survivorship. Furthermore, Bowden-Kerby (1997) found no evidence of size-dependent mortality in Acropora prolifera fragments transplanted onto reef ¯at rubble. These studies indicate that mortality of fragments is highly site and species-speci®c and that relying on generalizations to guide transplantation is dangerous. The balance between increased mortality and the potential for generating more o€spring via fragmentation thus needs to be carefully evaluated for each site and species to be transplanted. 3. Reduced growth rates of transplanted corals The stress of transplantation may have less dramatic consequences than death. One such consequence may be reduced growth rates of transplanted colonies compared to undisturbed ones. Clark and Edwards (1995) found that average growth rates of transplanted A. hyacinthus, A. humilis and A. cytherea colonies were signi®cantly slower during the initial seven months after transplantation than thereafter (note large variance in growth rates for these species in Fig. 1). They also found that a 478 signi®cantly higher percentage of colonies showed negative growth (i.e. loss of living tissue) during the initial seven months after transplantation than thereafter. Yap and Gomez (1985) reported that growth rates of transplanted A. pulchra colonies were considerably less than those of undisturbed controls and Plucer-Rosario and Randall (1987) found that growth rates of transplants (for Pavona cactus, A. echinata, Leptoseris gardneri and Montipora pulcherrima) averaged 50±75% of those of controls. Franklin et al. (1998) noted that larger fragments had higher growth rates than smaller ones. By contrast, Yap et al. (1992) indicated signi®cantly increased growth rates of transplanted Pavona frondifera and P. damicornis compared to undisturbed controls. However, they measured growth as mean areal increments (cm2 moÿ1 ) and then compared percentage growth rates (based on previous colony area). With such an analysis, if the control colonies were on average larger than the transplants, for the same increase in diameter they would appear to be growing more slowly 1. These data therefore need to be treated with caution. On balance the evidence suggests that transplanting is likely to adversely a€ect the growth rates of transplanted corals at least in the short term (for 0.5±1 year after transplantation). 1 For example, a circular colony growing from 5 cm to 7 cm in diameter in a year would register growth of 196%, whereas a circular colony growing from 10 cm to 12 cm in diameter would register growth of 144%. Both, however, have a radial extension rate of 10 mm per year. Volume 37/Numbers 8±12/August±December 1998 4. Loss of transplanted colonies from the reef as a result of wave action (attachment failure) Where transplantation sites are exposed to wave action, a signi®cant proportion of transplanted colonies or coral fragments may be lost (presumed dead) during storms even when attached apparently securely with cement, epoxy resin, cable ties, nylon strings, plastic coated wires or large staples. When this occurs, the collateral damage done to relatively unimpacted reef areas to obtain the transplants has been to no avail, with large numbers of colonies or fragments being removed to no advantage. Thus the risk of attachment failure needs to be carefully assessed for exposed sites when evaluating whether transplantation is a sensible management option. Clark and Edwards (1995) reported the loss of 25% of transplanted colonies due to wave action during the ®rst seven months of their study on a shallow (0.8±1.5 m below LAT) reef ¯at. Birkeland et al. (1979) lost 79% of 643 colonies transplanted at an open coast site and Plucer-Rosario and Randall (1987) mention the problem of high losses, particularly from their more exposed transplant site. Auberson (1982) reported 20± 50% survival over a year at relatively exposed shallow sites but 70% survival at deeper lower-energy sites. Similarly, Alcala et al. (1982) recorded only 40% survival of transplants in 1.2±1.5 m depth over one year at a relatively exposed site. Gil-Navia et al. (1999) partly attributed the low mortalities of their transplants to selection of environmentally compatible low-energy sites that reduced losses from wave action. In general, it appears that wastage of colonies is likely to be signi®cant at high-energy sites despite best e€orts to attach colonies securely. However, once the bases of transplanted colonies have naturally accreted to the underlying substrate then losses of colonies from wave action appear to be low; for example, after natural accretion occurred, only 5% loss over approximately 2 years was recorded by Clark and Edwards (1995). Recently Bowden-Kerby (1997) and Lindahl (1998) have advocated Ôlow-costÕ (although actual costs are not stated) transplantation techniques for reef management. High costs derive from two primary sources which relate to loss reduction. Firstly, if the degraded reef area to be transplanted is largely mobile rubble and sand, such as areas that have been mined or parts of ship-grounding sites, then stabilization may be necessary prior to transplantation to provide sites for secure attachment. This may involve repair of the reef framework using concrete (Miller et al., 1993; M/V Alec Owen Maitland grounding site in Florida Keys National Marine Sanctuary (FKNMS): NOAA, 1999a; Miller and Barimo, 1999), use of arti®cial structures (Clark and Edwards 1995) or removal of crushed reef limestone rubble and other debris if patches are small (M/V Elpis grounding site in FKNMS: NOAA, 1999b; M/V Horizon: Goldberg and Caballero, 1999; M/V Maasdam: Jaap, 1999). All options are costly and budgets for restoration of the Alec Owen Maitland and Elpis grounding sites were each in excess of US$ 1 million. Secondly, if the site is at all exposed to wave action or currents then transplants need to be attached to prevent high losses (see above). Attaching fragments or colonies to the substrate is a time consuming and labour intensive process and thus is likely to be expensive if done on a large scale (remembering that even volunteers have opportunity costs associated with them). We found that to detach, transport (200±450 m) and cement 500 colonies on a Maldives reef ¯at required about 250 person-hours on site (i.e., excluding travel time to the site and preparation). Associated consumable costs were about US$400 and 140 h of boat time were needed. To achieve low-costs a site should require neither physical remediation nor that coral be attached. Bowden-Kerby (1997) and Lindahl (1998) have concentrated on transplantation on relatively sheltered shallow back reef and reef ¯at rubble areas without attachment or with minimal attachment (i.e. tying together fragments with polythene strings anchored to 5 kg stones at either end; Lindahl 1998) to reduce costs. Lindahl (1998) transplanted Acropora spp. fragments (having best success with A. formosa which naturally forms extensive thickets on sand) and achieved 51% increase in transplanted coral cover over about 2 years. Bowden-Kerby (1997) had 79±96% survival of Acropora fragments on rubble after 3 months. However, even on relatively sheltered sites Bowden-Kerby (1997) and Lindahl (1998) found attachment to signi®cantly improve fragment survival. Further, Harriott and Fisk (1988b) found that although in the short term (10 months) unattached transplants in sheltered backreef areas survived well, after the ®rst storm only three among hundreds of unattached colonies could be found. 5. Reduced fecundity of transplanted colonies due to stress Another sub-lethal e€ect which the stress of transplantation may cause is reduced fecundity (Rinkevich and Loya, 1989). There appears to have been little research on this aspect, but given that transplantation can increase mortality and decrease growth rates, it is likely that the fecundity of whole transplanted coral colonies may be a€ected for at least several months after transplantation. However, preliminary data from our Maldives site suggested that Acropora colonies surviving 6 months after transplantation had similar gametogenic development to undisturbed colonies. Szmant-Froelich (1985) showed that for Montastraea annularis fecundity was size related, with small colonies and parts of larger colonies that had su€ered partial mortality not being fully reproductive below a threshold size. Smith and Hughes (1999) reported that experimental fragments of three Acropora species had substantially reduced fecundities relative to intact control colonies. Thus, for some coral species both donor colonies and fragments taken from them might have signi®cantly reduced sexual reproductive capacity. This is an issue that merits further research. 479 Marine Pollution Bulletin 6. Raised public expectations followed by disillusionment when transplants do not survive well Transplantation is very labour intensive but can be carried out by competent recreational SCUBA divers or snorkelers who have received training in the techniques, and thus lends itself to volunteer support. Examples of this are the Singapore transplantation work of Newman and Chuan (1994) and part of the work carried out on the M/V Maasdam grounding site on Grand Cayman (Jaap, 1999). Involving the public increases awareness of the threats to coral reefs and may do much to allow cost-e€ective restoration of localised reef damage. However, if a signi®cant number of transplants are seen to die and donor areas are not seen to recover well, there is a risk of disillusionment. A risk averse strategy with careful planning, site selection and consideration of the various bene®ts and drawbacks should avoid unsustainable transplantation projects, although nature, in the form of toxic algal blooms, warm water anomalies and cyclones, may defeat the best laid plans. Potential bene®ts The potential bene®ts of transplantation are now examined with reference to research ®ndings where available. 1. Immediate increase in coral cover and diversity Transplantation of corals will clearly provide an immediate increase in coral cover and diversity at an impacted site, hopefully creating a community that resembles what was previously present. From a political point of view, an immediate and visible demonstration of environmental concern and executive action is achieved. Something has been done, and can be seen to have been done. Furthermore, if volunteers from local communities have been involved in the work, there are bene®ts of environmental awareness building and community involvement. However, if a site is suitable for coral growth, has a good supply of larvae and does not su€er excessive post-settlement mortality, it should, in due course, recover naturally. In such cases, the bene®ts of an immediate rather than longer-term increase in coral cover and diversity need to be assessed carefully. Are there better alternative uses of funds? A few studies have shown a marked increase in coral cover following coral transplantation (e.g. Lindahl (1998) indicated a 51% increase in Acropora cover over 2 years; Guzman (1991) reported a doubling of coral cover over 3 years at his Platanillo site). However, Clark and Edwards (1995) found that, because of increased mortality and reduced growth rates of transplanted colonies, the initial percentage cover achieved immediately after transplantation declined over 7 months and was not reached again until almost 2 years after transplanting (Fig. 3). Compared to many studies their overall mortality rates were quite low. Furthermore, within about 3.5 years, sites that had not had corals 480 Fig. 3 Mean percentage live coral cover (squares) and mean numbers of live colonies per m2 (circles) on three 18 m2 transplanted sites on a reef ¯at in the Maldives over 28 months. Error bars are‹ SE. transplanted to them had communities of branching corals of similar size and diversity to those on the transplanted areas. In less than 8 years, naturally recruited Acropora colonies of up to 135 cm greatest diameter were established. However, although branching species recruited well and grew fast, common reef-¯at massive species, such as Porites spp., were under-represented at sites that had not been subject to transplantation. Our experience in the Maldives indicates that although transplantation can speed up the initial recovery (<4±5 years) of a degraded site (at some cost to donor sites), over a 10-year timescale it may have a negligible e€ect on the branching coral (Acropora, Pocillopora) cover and diversity achieved. Therefore, where recruitment is satisfactory, there may be little biological justi®cation for transplanting branching corals although there may be special cases where such an approach is justi®able. By contrast, transplantation of certain massive species may have a much more lasting impact on recovery because these species grow more slowly and may recruit to degraded areas in fewer numbers (<10% of over 3000 recruits recorded on arti®cial reefs in Maldives over 3.5 years). In addition, massive species, such as Porites and Pavona, appear to survive transplantation well and so wastage is likely to be relatively low. The downside is that removing a slow-growing massive colony from a donor reef represents a greater negative impact than removing a similar-sized branching colony. 2. Increased recruitment of coral larvae as a result of presence of transplants It has been suggested that transplanting corals to a site might enhance local recruitment (e.g. Harriott and Fisk, 1988a) and thus bene®t rehabilitation. Two possible mechanisms would be a local increase in larval supply (particularly in planulating species) and asexual reproduction via fragmentation. There is also the possibility of established corals stimulating settlement in Volume 37/Numbers 8±12/August±December 1998 some way. In the Maldives study, we thus compared natural recruitment of corals on Armor¯ex concrete mats with transplants, to that on mats without transplants to see if any enhancement was evident. Coral recruits on both transplanted and bare Armor¯ex areas were dominated by branching species in the genera Acropora and Pocillopora. Survival was high and growth rates fast, with some Acropora cytherea colonies attaining a colony diameter approaching 20 cm within 12 months of ®rst being recorded and of up to 135 cm within 7.5 years. Twenty-eight months after deployment, coral recruitment on the concrete mats with coral transplants was compared to that observed on those without transplants. The numbers of visible recruits per unit area becoming established on the Armor¯ex mats, and on the vertical edges of ¯ooring slabs used to anchor them, did not di€er signi®cantly between areas with and areas without transplanted corals (Fig. 4). Furthermore, there was no signi®cant di€erence between the relative numbers of Pocillopora (planulating) and Acropora (broadcast spawning) recruits on areas with and without transplants. The conclusion for this site, where larval supply was not limited, is that the presence of transplants had no bene®cial e€ect on recruitment. However, at a site with poor water circulation at the time of spawning or planulation, a di€erent result might have been obtained. At the site studied, the nearest sources of larvae were only hundreds of metres away and thus one might not expect transplants to enhance recruitment signi®cantly. For more isolated sites, where larval supply may be a problem, the potential for enhanced recruitment remains. However, Willis and Oliver (1988) working on the Great Barrier Reef found that coral planulae were transported from one reef to another 26 km down current within two days of spawning and Williams et al. (1984) indicated that planulae of broadcast spawners could be transported hundreds of kilometres. Thus transplantation is only rarely likely to be useful as a means of enhancing larval supply to a damaged area. At sites where it is, transplanting gravid hermaphroditic Fig. 4 Comparison of mean recruitment (number of visible recruits per m2 ) to Armor¯ex concrete mats and the paving slabs anchoring them at sites with (n ˆ 3) and without (n ˆ 3) corals transplanted on them. Sites were surveyed 28 months after emplacement. Error bars are ‹ SE. brooders capable of self-fertilization (Gleason and Brazeau, 1999) may be a cost-e€ective option. In the speci®c case of ship groundings where damaged areas are only tens to hundreds of metres in extent, it is unlikely to be of bene®t in this context (see also Maragos, 1974). 3. Survival of locally rare and threatened coral species when primary habitat is destroyed Where the primary habitat for locally rare coral species is being unavoidably destroyed, there seems to be a clear case for transplantation to a safer area. In Florida, repair of an outfall damaged by Hurricane Andrew threatened a few hundred colonies; these were collected pre-constructrion and then transplanted back postconstruction (Dodge et al., 1999). In Guam and in Singapore transplantation has been used to save species threatened by pollution or loss of habitat due to reclamation (Plucer-Rosario and Randall 1987; Newman and Chuan 1994, respectively). In the Singapore case, reclamation threatened the whole reef and other reef invertebrates were transplanted as well as corals. The key issues arising from these studies are the importance of selecting an appropriate receiving area (as similar in environment as possible to the donor site) where the transplants will survive well and can be securely anchored to the substrate (where required). Plucer-Rosario and Randall (1987) reported high losses of transplants, particularly from their more exposed receiving area. Newman and Chuan (1994) do not appear to have rigorously monitored their transplants and it is unclear what mortalities were sustained. On the political side, there is the danger that if transplantation of corals (and associated sessile invertebrates) is seen as a generally acceptable and easy option, then it can used to legitimise coral reef habitat loss and avoid debate of other potentially better management options. 4. Reintroduction of corals to areas which are larval supply limited or have very high post-settlement mortality Perhaps the strongest case for transplantation is in areas that have poor larval supply or very high postsettlement mortality, leading to little recruitment of juveniles. Such areas are unlikely to recover well without assistance. Isolated inlets and bays where coral reefs have been severely degraded due to past pollution and into which there is poor current ¯ow from undamaged reefs would fall into this category, as would areas where changes in algal biomass or grazing have led to lack of recruitment (Ebersole, 1999; Gleason, 1999; Miller and Barimo, 1999). In such cases natural recruitment is unlikely to generate signi®cant recovery on even a decadal timescale and transplantation to help re-establish a viable population appears the only option. However, if post-settlement mortality is the key limiting factor then there remains a serious question as to whether a transplanted community can be sustainable. 481 Marine Pollution Bulletin A similar strong case may exist for transplantation where a dominant reef building species relies primarily on fragmentation to reproduce as for Pocillopora in Costa Rica (Guzman, 1991), which was decimated by the 1982±1983 El Ni~ no and a subsequent dino¯agellate bloom. Without transplantation, there seemed no chance of natural recovery on the reefs restored by Guzman (1991, 1993) and indeed there appeared to be a risk of local extinction of P. eydouxi. However, after 14 years of monitoring it became apparent that coral sexual reproduction in the region and hence reef recovery could naturally take 10±12 years and that the restoration efforts had perhaps been premature (Guzm an, 1999). Again the bene®ts from transplantation have to be evaluated on a site-speci®c basis and even if the best available scienti®c advice may later turn out to have been ¯awed, if restoration has been carried out with due care, then there should be no net damage and hopefully some net bene®ts. 5. Instant increase in rugosity and shelter for herbivores in bare areas Severely degraded reefs that have been mined (Clark and Edwards, 1994), su€ered ship-grounding damage (Gittings et al., 1988; Hudson and Diaz, 1988; Jaap, 1999; Miller and Barimo, 1999) or had 8 ton suctiondredge heads dragged over them (Miller et al., 1993) may have very little topographic relief left. The lack of shelter may severely reduce number of herbivores and thus grazing, potentially leading to excessive algal growth and a lack of surfaces onto which coral planulae will settle. The relatively ¯at surface may also reduce the chance of coral larval settlement. These problems may be exacerbated by the presence of unconsolidated rubble and ®ner sediment (Brown and Dunne, 1988). Transplantation of corals and deployment of arti®cial reefs, where the substrate is unstable, help to increase rugosity and provide shelter for herbivorous ®sh (e.g. ÔThe Sunny Isles Reef Restoration ProjectÕ reported in Miller et al., 1993). Miller and Barimo (1999) found that coral recruitment to structures used to stabilise the M/V Alec Owen Maitland grounding site was positively associated with roughness elements of the structures. Within a few months of deployment of arti®cial reefs on a severely degraded reef ¯at in the Maldives, ®sh populations returned to levels comparable to those on undegraded reefs (Edwards and Clark, 1992). However, Ebersole (1999) noted that remediation e€orts seemed ine€ective at restoring the diversity of ®sh assemblages at ship-grounding sites in Florida over two years. At present it is unclear whether rugosity on the scale provided by transplanted corals will signi®cantly a€ect recruitment, or how large a severely degraded area has to be before roving herbivore grazing is signi®cantly reduced. Further research is needed to determine in what circumstances transplantation and/or use of arti®cial reefs is likely to generate bene®ts in terms of directly or indirectly enhancing recruitment. 482 6. Improved aesthetics of areas frequented by tourists Transplantation has been suggested as a means of enhancing areas frequented by tourists. Bouchon et al. (1981) transplanted large coral heads in the Gulf of Aqaba to an area devoid of reefs but used by tourists. In this issue van Treeck and Schuhmacher (1999) suggest the use of coral nubbins transplanted onto steel mesh structures electrolytically coated with calcium carbonate (van Treeck and Schuhmacher, 1997) to attract SCUBA divers away from sensitive reef areas at sites where diving pressure is causing concern. Such activities are not really rehabilitation but may form part of an overall reef management strategy. In terms of our discussion here, the issue is whether provision of arti®cial substrates to which corals can recruit is enough, or whether transplantation is necessary. Unless the areas where the structures are deployed are failing to recruit, transplantation is unlikely to be necessary. Interestingly, in Bouchon and co-workersÕ study, within one year, of 42 colonies transplanted, 15 (36%) were dead or `decaying', whilst in the same period 16 newly settled colonies had recruited to their arti®cial reef (Bouchon et al., 1981). Conclusions It is clear from the studies above that reef restoration (returning to pre-disturbance condition: Pratt, 1994) and reef rehabilitation (re-establishment of selected ecological attributes: Pratt, 1994) based around coral transplantation are still largely at an experimental stage. This is in contrast to mangrove (Field, 1996) and saltmarsh (Zedler, 1984) rehabilitation, which are orders of magnitude cheaper to accomplish per unit area (Spurgeon, 1999), and where large-scale rehabilitation has been successfully carried out. Seagrass restoration also appears more advanced in terms of application (Thorhaug, 1987; Fonseca, 1994). However, useful guidelines have been established (Harriott and Fisk, 1988b; Miller et al., 1993; Rinkevich, 1995) and the use of coral transplantation and related techniques to aid reef rehabilitation is developing fast (NCRI, 1999). Evidence suggests that transplantation can in certain instances be a useful management tool and a few successful applications are reported. On the other hand, without careful consideration of the environment and ecology of sites targeted for rehabilitation and the biology of the coral species being transplanted, there is a real risk of misguided meddling (or `techno-arrogance'; see Grimes (1998) discussing marine stock enhancement) given the various dis-bene®ts discussed above. Spalding and Grenfell (1997) estimated the global area of coral reefs at around 2.5 ´ 107 ha. One of the largest scale reef rehabilitation projects to date was that of Guzman (1991, 1993, 1999) in which 7.1 ha were restored using almost 85 000 coral fragments. Given this di€erence in scale and the costs of transplantation (even of the `low-cost' variety), it is clear that coral transplantation is only likely to be e€ective on a small scale Volume 37/Numbers 8±12/August±December 1998 (in the order of 101 ha or less). It does not seem appropriate to suggest that coral transplantation is a potential solution for: `increasing coral growth/reef accretion to help reefs keep up with projected sea-level rise; accelerating coral reef responses to changing environmental conditions by introduction of heat, UV, sediment, or pollutant tolerant clones to replant reefs negatively impacted by such factors; or increasing the carrying capacity, recruitment, or survival of ®sh on reef ¯ats and in back reef areas' (Bowden-Kerby, 1997). Accretion of reef ¯ats in the face of rising sea-level is a global issue and coral mortality following sea temperature anomalies is a regional/global issue (Edwards, 1995; Wilkinson et al., 1999). Transplantation is not going to tackle problems on these scales (>104 ±105 ha). In areas where high sedimentation is normal, sediment tolerant corals are already present (for example, Ko Phuket, Brown et al., 1990). If the sedimentation is of anthropogenic origin, e.g., dredging or land-use changes, then improved management of these activities might be a better use of funds. Creation of coral-dominated habitats for ®sh in backreef areas, where they have not so far developed naturally, again seems to be a technology looking for a problem to solve. By contrast, a project to establish sustainable village-based coral aquaculture in Melanesia to produce Acropora in lagoons to supply the betel lime and ornamental trades (Bowden-Kerby, 1999) and reduce the estimated 2000 tonnes harvested annually from the wild, addresses a real problem. There are plenty of localised anthropogenic impacts on reefs at sites all around the world, where carefully focused transplantation may have a role to play in rehabilitation, without looking for applications. Challenger (1999) discussed scale from an opposite viewpoint, noting that reef damage due to ship groundings is orders of magnitude less than that sustained from severe storms. He raised the question of whether such damage is thus signi®cant enough to be worth restoring, given the high costs involved. Whilst recognizing that physical restoration e€orts such as repair of damage to the reef framework are sometimes warranted, he suggested that it was reasonable to consider alternative more farreaching approaches to use of recovered damages than just restoration. Such points require consideration and will perhaps stimulate more discussion of the scienti®c goals and socio-economic costs and bene®ts of restoration. Two issues stand out from our review of coral transplantation. Firstly, is the need for clear criteria for assessing reef rehabilitation/restoration `success' at a site. Such criteria should re¯ect the timescale of natural recovery, which will itself be related to the incidence of natural impacts such as storms, sea temperature anomalies, etc., the risks to transplants of which also need to be taken into account. This suggests that transplantation success needs to be assessed on 5±10 year timescales as a minimum and provision made for monitoring to allow this. Secondly, is the lack of information on the costs of transplantation. Many studies state that it is a `cost-e€ective' solution but almost none state the costs or how the bene®ts were quanti®ed. Such information is essential if scienti®c studies are to be translated into management technologies. Given the various potential dis-bene®ts of transplantation, we suggest that one should, where possible, let natural recruitment drive recovery (`passive rehabilitation'; Woodley and Clark, 1989) and secondly, one should focus less on fast-growing species to `jump-start' recovery and more on slow-growing species that are slow to recruit. If water quality is satisfactory (i.e. previous pollution has been remedied, or no decrease in water quality is expected) and the substratum is stable (Fig. 5), then natural recruitment processes may allow a degraded site to recover unaided over 5, 10, 30 or more years depending on the scale of the impact (Alcala and Gomez, 1979; Connell, 1997; Curtis, 1985; Maragos, 1974; Shinn, 1976). Because of the timescale, it seems that intervention (e.g., restoration by transplantation) has been assumed to be an appropriate response. We argue that unless a site is expected not to recover naturally, perhaps it should be left alone. If intervention is planned then the case for it needs to be evaluated with respect to the recoverability (Done, 1995) of the damaged site. Surveys of the coral community surrounding unimpacted areas will provide information on the type of coral assemblage that may be expected following recovery and on local sources of propagules (broadcasters and brooders). This can inform a judgement, based on reviews of coral life-histories (e.g., Richmond and Hunter, 1990) and local conditions (e.g., currents, isolation, herbivore biomass, sediment scouring, algal growth), as to whether natural recruitment is for some reason likely to be inadequate for recovery. Unfortunately, given the patchy nature of recruitment in both space and time (Hughes et al., 1999), surveys of recruits in neighbouring areas may not provide much guidance on the prospects for future recruitment. To test whether larval supply or post-settlement mortality are likely to be limiting would need a pilot study at the degraded site itself using settlement plates and monitoring of experimental patches on the degraded reef. One might fairly quickly (<1 year) ascertain whether brooders were likely to colonise but could wait a few years to decide for broadcasters. Unfortunately, the costs of such a study may be similar to those of transplantation. One may thus have to rely on inferences from the adult population and the experience from other studies to decide whether the costs of transplantation are likely to be justi®ed at a particular site and for which species (Fig. 5). A few general patterns regarding recruitment emerge from the literature. In the Indo-Paci®c, pocilloporids and acroporids have generally been found to be wellrepresented among recruits (Atrigenio and Ali~ no, 1994; Clark and Edwards, 1995; Maida et al., 1994; Sudara et al., 1994; Wallace, 1985; Yeemin et al., 1992; Yeemin 483 Marine Pollution Bulletin Fig. 5 A simpli®ed decision ¯ow-chart for examining whether coral transplantation may be a useful option. and Sudara, 1992), with poritids often third. Although most studies found recruits of planulating pocilloporids in abundance (33±66% of recruits in many cases), recruitment success of the primarily broadcast spawning acroporids may be sporadic (<1% to 87% of recruits). Some studies found rather few acroporid recruits (Ali~ no et al., 1985; Gleason, 1996; Harriott and Banks, 1995; 484 Thongtham and Chansang, 1999) and poritids as the second or most abundant recruits (contributing 23±80% of recruits). In summary, the fast-growing Indo-Paci®c branching species that have been favoured in many transplantation studies are often those that are likely to recruit well naturally and for which it is thus more dif®cult to justify transplantation. Their relatively high Volume 37/Numbers 8±12/August±December 1998 mortality rates following transplantation compared to certain massive species compound this (Fig. 2). In the Atlantic, brooding agariciids and Porites astreoides appear to have high recruitment rates with broadcast spawning Acropora spp. and massive species in the genera Montastraea and Diploria having uniformly low recruitment rates (Bak and Engel, 1979; Rogers et al., 1984; Smith, 1992). If there is a case for transplantation, then more emphasis on species that are expected to recruit poorly, su€er negligible mortality following transplantation, and are readily available in the unimpacted adult community, seems warranted. Transplantation of gravid massive brooders, where such exist, may be particularly e€ective (Richmond and Hunter, 1990). Even with low recruitment, if sucient time is allowed, populations may recover naturally (Guzm an, 1999). So, perhaps the primary guideline is `only transplant if there is a strong case for transplantation'. The research in Maldives was funded by the UK Department for International DevelopmentÕs Renewable Natural Resources Research Strategy and Engineering Research Programmes. We would like to thank Mr Hassan Maniku Maizan, Director of the Marine Research Centre and all his sta€ who assisted with that project. We also thank GCRMN South Asia for an opportunity to survey our study site in August 1998, various colleagues for helpful discussions and two reviewers for their helpful comments on an earlier draft of the manuscript. Alcala, A. C. and Gomez, E. D. (1979) Recolonization and growth of hermatypic corals in dynamite-blasted coral reefs in the Central Visayas, Philippines. Proceedings of the International Symposium on Marine Biogeography and Evolution S. Hemisphere 2, 645±661. Alcala, A. C., Gomez, E. D. and Alcala, L. C. (1982) Survival and growth of coral transplants in Central Philippines. Kalikasan, Philippines Journal of Biology 11(1), 136±147. Ali~ no, P. M., Banson, P. V., Yap, H. T., Gomez, E. D., Morales, J. T. and Bayoneto, R. P. (1985) Recovery and recolonization on a damaged backreef area at Cangaluyan Is. (Northern Philippines). Proceedings of the the International Coral Reef Congress 4, 279±284. Atrigenio, M. P. and Ali~ no, P. M. (1994) Seasonal recruitment of scleractinian corals at Puerto Galera, Oriental Mindoro, Philippines. In Proceedings, Third ASEAN-Australia Symposium on Living Coastal Resources, Vol. 2: Research Papers, eds. S. Sudara, C.R. Wilkinson, L.M. Chou, pp. 157±160. Chulalongkorn University, Bangkok, Thailand. Auberson, B. (1982) Coral transplantation: an approach to the reestablishment of damaged reefs. Kalikasan, Philippines Journal of Biology 11(1), 158±172. Bak, R. P. M. and Engel, M. S. (1979) Distribution, abundance and survival of juvenile hermatypic corals (Scleractinia) and the importance of life history strategies in the parent coral community. Marine Biology 54, 341±352. Birkeland, C., Randall, R. H. and Grim, G. (1979) Three methods of coral transplantation for the purpose of reestablishing a coral community in the thermal e‚uent area of the Tanguisson Power Plant. University of Guam Marine Laboratory Technical Report No. 60, p. 24. Bouchon, C., Jaubert, J. and Bouchon-Navaro, Y. (1981) Evolution of a semi-arti®cial reef built by transplanting coral heads. Tethys 10 (2), 173±176. Bowden-Kerby, A. (1997) Coral transplantation in sheltered habitats using unattached fragments and cultured colonies. Proceedings of the Eighth International Coral Reef Symposium 2, 2063±2068. Bowden-Kerby, A. (1999) The community coral reef initiative: coral reef restoration in rural Paci®c island settings. In Program and Abstracts. International Conference on Scienti®c Aspects of Coral Reef Assessment, Monitoring, and Restoration, p. 60. National Coral Reef Institute, Nova Southeastern University, Florida, USA. Brown, B. E. and Dunne, R. P. (1988) The environmental impact of coral mining on coral reefs in the Maldives. Environmental Conservation 15 (2), 159±166. Brown, B. E. and Howard, L. S. (1985) Assessing the e€ects of stress on reef corals. Advances in Marine Biology 22, 1±63. Brown, B. E., LeTissier, M. D. A., Scon, T. P. and Tudhope, A. W. (1990) Evaluation of the environmental impact of dredging on intertidal coral reefs at Ko Phuket, Thailand using ecological and physiological parameters. Marine Ecology Progress Series 65(3), 273±281. Bruno, J. F. (1998) Fragmentation in Madracis mirabilis (Duchassaing and Michelotti): how common is size-speci®c fragment survivorship in corals? Journal of Experimental Marine Biology and Ecology 230, 169±181. Challenger, G. E. (1999) Questions regarding the biological signi®cance of vessel groundings and appropriateness of restoration e€ort. In Program and Abstracts. International Conference on Scienti®c Aspects of Coral Reef Assessment, Monitoring, and Restoration, pp. 66±67. National Coral Reef Institute, Nova Southeastern University, Florida, USA. Clark, S. and Edwards, A. J. (1994) The use of arti®cial reef structures to rehabilitate reef ¯ats degraded by coral mining in the Maldives. Bulletin of Marine Science 55(2), 726±746. Clark, S. and Edwards, A. J. (1995) Coral transplantation as an aid to reef rehabilitation: evaluation of a case study in the Maldives Islands. Coral Reefs 14, 201±213. Clark, S. and Edwards, A. J. (1999) An evaluation of arti®cial reef structures as tools for marine habitat rehabilitation in the Maldives. Aquatic Conservation: Marine and Freshwater Ecosystems 9, 5±21. Connell, J. H. (1997) Disturbance and recovery of coral assemblages. Coral Reefs 16 (Suppl.), S101±S113. Curtis, C. (1985) Investigating reef recovery following a freighter grounding in the Key Largo National Marine Science Sanctuary, (Florida Keys, USA). Proceedings of the Fifth International Coral Reef Congress 6, 471±476. Dodge, R. E., Anderegg, D., Fergen, R. and Cooke, P. (1999) Coral transplantation following repair of outfall. In Program and Abstracts. International Conference on Scienti®c Aspects of Coral Reef Assessment, Monitoring, and Restoration, p. 80. National Coral Reef Institute, Nova Southeastern University, Florida, USA. Done, T. J. (1995) Ecological criteria for evaluating coral reefs and their implications for managers and researchers. Coral Reefs 14, 183±192. Ebersole, J. P. (1999) Recovery of ®sh assemblages from ship groundings on coral reefs in the Florida Keys National Marine Sanctuary. In Program and Abstracts. International Conference on Scienti®c Aspects of Coral Reef Assessment, Monitoring, and Restoration, p. 82. National Coral Reef Institute, Nova Southeastern University, Florida, USA. Edwards, A. J. (1995) Impact of climatic change on coral reefs, mangroves and tropical seagrass ecosystems. In Climate Change: Impact on Coastal Habitation, ed. D. Eisma, pp. 209±234. Lewis Publishers, Boca Raton. Edwards, A. J. and Clark, S. (1992) Rehabilitation of coral reef ¯ats using pre-cast concrete. Concrete 26(1), 16±19. Edwards, A. J. and Clark, S. (1993) Re-establishment of reef ®sh populations on a reef ¯at degraded by coral quarrying in the Maldives. Proceedings of the Seventh International Coral Reef Symposium 1, 593±600. Field, C. D. (Ed.) (1996) Restoration of Mangrove Ecosystems. International Society for Mangrove Ecosystems, Okinawa, Japan. 250 pp. Fonseca, M. S. (1994) A Guide to Planting Seagrasses in the Gulf of Mexico. Texas A and M University Sea Grant College Program TAMU-SG-94-601. Fox, H. E., Caldwell, R. L. and Pet, R. S. (1999) Enhancing coral recovery after destructive ®shing practices in Indonesia. In Program and Abstracts. International Conference on Scienti®c Aspects of Coral Reef Assessment, Monitoring, and Restoration, p. 88. National Coral Reef Institute, Nova Southeastern University, Florida, USA. Franklin, H., Muhando, C. A. and Lindahl, U. (1998) Coral culturing and temporal recruitment patterns in Zanzibar, Tanzania. Ambio 27(8), 651±655. Gil-Navia, M. F., S anchez, J. A. and Alvarado, E. (1999) Transplantation of reef-building corals on the Rosario Archipelago, Colombian Caribbean. In Program and Abstracts. International Conference on Scienti®c Aspects of Coral Reef Assessment, Monitoring, and 485 Marine Pollution Bulletin Restoration, pp. 91±92. National Coral Reef Institute, Nova Southeastern University, Florida, USA. Gittings, S. R., Bright, T. J., Choi, A. and Barnett, R. R. (1988) The recovery process in a mechanically damaged coral reef community: recruitment and growth. Proceedings of the Sixth International Coral Reef Symposium 2, 225±230. Gleason, D. F. and Brazeau, D. A. (1999) Can sel®ng coral species be used to enhance restoration of damaged reefs. In Program and Abstracts. International Conference on Scienti®c Aspects of Coral Reef Assessment, Monitoring, and Restoration, pp. 92±93. National Coral Reef Institute, Nova Southeastern University, Florida, USA. Gleason, M. G. (1996) Coral recruitment in Moorea, French Polynesia: the importance of patch type and temporal variation. Journal of Experimental Marine Biology and Ecology 207, 79±101. Gleason, M. G. (1999) The importance of algal-grazer interactions in early growth and survivorship of sexual recruits and transplanted juvenile corals. In Program and Abstracts. International Conference on Scienti®c Aspects of Coral Reef Assessment, Monitoring, and Restoration, p. 93. National Coral Reef Institute, Nova Southeastern University, Florida, USA. Glynn, P. W. (1984) Widespread coral mortality and the 1982/83 El Ni~ no warming event. Environmental Conservation 11, 133±146. Glynn, P. W. (1993) Coral reef bleaching: ecological perspectives. Coral Reefs 12, 1±17. Glynn, P. W. and DÕCroz, L. (1990) Experimental evidence for high temperature stress as the cause of El Ni~ no-coincident coral mortality. Coral Reefs 8, 181±191. Goldberg, W. M. and Caballero, A. (1999) Reef damage by large vessel impact and its mitigation by site cleanup: methods and results after one year. In Program and Abstracts. International Conference on Scienti®c Aspects of Coral Reef Assessment, Monitoring, and Restoration, p. 94. National Coral Reef Institute, Nova Southeastern University, Florida, USA. Grigg, R. W. and Maragos, J. E. (1974) Recolonization of hermatypic corals on submerged lava ¯ows in Hawaii. Ecology 55, 387±395. Grimes, C. B. (1998) Marine stock enhancement: sound management or techno-arrogance? Fisheries 23(9), 18±23. Guzman, H. M. (1991). Restoration of coral reefs in Paci®c Costa Rica. Conservation Biology 5, 189±195. Guzman, H. M. (1993) Transplanting coral to restore reefs in the eastern Paci®c. In Spirit of Enterprise. The 1993 Rolex Awards, ed. D.W. Reed, pp. 409±411. Buri, Bern, Switzerland. an, H. M. (1999) Large-scale restoration of Eastern Paci®c reefs: Guzm the need for understanding regional biological processes. In Program and Abstracts. International Conference on Scienti®c Aspects of Coral Reef Assessment, Monitoring, and Restoration, p. 97. National Coral Reef Institute, Nova Southeastern University, Florida, USA. Harriott, V. J. and Banks, S. A. (1995) Recruitment of scleractinian corals in the Solitary Islands Marine Reserve, a high latitude coraldominated community in Eastern Australia. Marine Ecology Progress Series 123, 155±161. Harriott, V. J. and Fisk, D. A. (1988a) Coral transplantation as a reef management option. Proceedings of the Sixth International Coral Reef Symposium 2, 375±379. Harriott, V. J. and Fisk, D. A. (1988b) Accelerated regeneration of hard corals: a manual for coral reef users and managers. GBRMPA Technical Memorandum 16. Hatcher, B. G., Johannes, R. E. and Robertson, A. I. (1989) Review of research relevant to the conservation of shallow tropical marine ecosystems. Oceanography and Marine Biology Annual Review 27, 337±414. Highsmith, R. C. (1982) Reproduction by fragmentation in corals. Marine Ecology Progress Series 7, 207±226. Highsmith, R. C., Riggs, A. C. and DÕAntonio, C. M. (1980) Survival of hurricane-generated coral fragments and a disturbance model of reef calci®cation/growth rates. Oecologia 46, 322±329. Hudson, J. H. and Diaz, R. (1988) Damage survey and restoration of M/V Wellwood grounding site, Molasses Reef, Key Largo National Marine Sanctuary. Proceedings of the Sixth International Coral Reef Symposium 2, 231±236. Hughes, T. P., Baird, A. H., Dinsdale, E. A., Moltschaniwskyj, N. A., Pratchett, M. S., Tanner, J. E. and Willis, B. L. (1999) Patterns of recruitment and abundance or corals along the Great Barrier Reef. Nature 397, 59±63. Jaap, W. C. (1999) Reef restoration and monitoring: SotoÕs Reef, Georgetown, Grand Cayman Island, British West Indies. In 486 Program and Abstracts. International Conference on Scienti®c Aspects of Coral Reef Assessment, Monitoring, and Restoration, p. 112. National Coral Reef Institute, Nova Southeastern University, Florida, USA. Lindahl, U. (1998) Low-tech rehabilitation of degraded coral reefs through transplantation of staghorn corals. Ambio 27 (8), 645±650. Maida, M., Coll, J. C. and Sammarco, P. W. (1994) Shedding new light on scleractinian coral recruitment. Journal of Experimental Marine Biology and Ecology 180, 189±202. Maragos, J. E. (1974) Coral transplantation: A method to create, preserve and manage coral reefs. University of Hawaii Sea Grant Prog. AR-74-03. p. 30. Maragos, J. E., Evans, C. and Holthus, P. (1985) Reef corals in Kaneohe Bay six years before and after termination of sewage discharges. Proceedings of the Fifth International Coral Reef Congress 4, 189±194. Miller, M. W. and Barimo, J. (1999) An assessment of juvenile coral populations at two coral reef restoration sites in the Florida Keys National Marine Sanctuary: indicators of success? In Program and Abstracts. International Conference on Scienti®c Aspects of Coral Reef Assessment, Monitoring, and Restoration, p. 139. National Coral Reef Institute, Nova Southeastern University, Florida, USA. Miller, S. L., McFall, G. B. and Hulbert, A. W. (1993) Guidelines and recommendations for coral reef restoration in the Florida Keys National Marine Sanctuary. National Undersea Research Center, University of North Carolina at Wilmington, p. 38. Mu~ noz-Chagin, R. F. (1997) Coral transplantation program in the Paraiso coral reef, Cozumel Island, Mexico. Proceedings of the Eighth International Coral Reef Symposium 2, 2075±2078. NCRI (1999) Program and Abstracts. International Conference on Scienti®c Aspects of Coral Reef Assessment, Monitoring, and Restoration. National Coral Reef Institute, Nova Southeastern University, Florida, USA, 218 pp. Newman, H. and Chuan, C. S. (1994) Transplanting a coral reef: a Singapore community project. Coastal Management in Tropical Asia 3, 11±14. NOAA (1999a) Maitland coral reef restoration. http://www-orca.nos.noaa.gov/darp/seregion/ maitland.htm. NOAA (1999b) Elpis coral reef restoration. http://www-orca.nos.noaa.gov/darp/seregion/elpis.htm. Oren, U. and Benayahu, Y. (1997) Transplantation of juvenile corals: a new approach for enhancing colonization of arti®cial reefs. Marine Biology 127, 499±505. Pearson, R. G. (1981) Recovery and recolonization of coral reefs. Marine Ecology Progress Series 4, 105±122. Plucer-Rosario, G. P. and Randall, R. H. (1987) Preservation of rare coral species by transplantation: an examination of their recruitment and growth. Bulletin of Marine Science 41(2), 585±593. Pratt, J. R. (1994) Arti®cial habitats and ecoystem restoration: managing for the future. Bulletin of Marine Science 55(2±3), 268± 275. Raymundo, L. J. H., Maypa, A. P. and Luchavez, M. M. (1999) Coral seeding as a technology for recovering degraded coral reefs in the Philippines. Phuket Marine Biology Center Special Publication 19, (in press). Richmond, R. H. and Hunter, C. L. (1990) Reproduction and recruitment of corals: comparisons among the Caribbean, the Tropical Paci®c, and the Red Sea. Marine Ecology Progress Series 60, 185±203. Rinkevich, B. (1995) Restoration strategies for coral reefs damaged by recreational activities: the use of sexual and asexual recruits. Restoration Ecology 3 (4), 241±251. Rinkevich, B. and Loya, Y. (1989) Reproduction in regenerating colonies of the coral Stylophora pistillata. In Environmental Quality and Ecosystem Stability. Vol. IVB. Environmental Quality, eds. E. Spanier, Y. Steinberger, M. Luria, pp. 259±265. Israel Society for Environmental Quality Sciences Publication, Jerusalem, Israel. Rogers, C. S., Fitz III, H. C., Gilnack, M., Beets, J. and Hardin, J. (1984) Scleractinian coral recruitment patterns at Salt River submarine canyon, St. Croix, U.S. Virgin Islands. Coral Reefs 3, 69±76. Salvat, B. (Ed.) (1987) Human impacts on coral reefs: facts and  recommendations. Antenne Museum, Ecole Pratique des Hautes  Etudes, French Polynesia, p. 253. Sammarco, P. W., Brazeau, D. A. and Lee, T. N. (1999) Enhancement of reef regeneration processes: supplementing coral recruitment processes through larval seeding. In Program and Abstracts. Volume 37/Numbers 8±12/August±December 1998 International Conference on Scienti®c Aspects of Coral Reef Assessment, Monitoring, and Restoration, p. 169. National Coral Reef Institute, Nova Southeastern University, Florida, USA. Shinn, E. A. (1976) Coral reef recovery in Florida and the Persian Gulf. Environmental Geology 1, 241±254. Smith, L. D. and Hughes, T. P. (1999) An experimental assessment of survival, re-attachment and fecundity of coral fragments. Journal of Experimental Marine Biology and Ecology 235, 147±164. Smith, S. R. (1992) Patterns of coral recruitment and post-settlement mortality on BermudaÕs reefs: comparisons to Caribbean and Paci®c reefs. American Zoology 32, 663±673. Spalding, M. D. and Grenfell, A. M. (1997) New estimates of global and regional reef areas. Coral Reefs 16, 225±230. Spurgeon, J. (1998) The socio-economic costs and bene®ts of coastal habitat rehabilitation and creation. Marine Pollution Bulletin 37, 373±382. Sudara, S., Yeemin, T. and Amornsakchai, S. (1994) Recruitment of scleractinian corals at Pha-Ngan Island, Gulf of Thailand: an experimental approach. In Proceedings, Third ASEAN-Australia Symposium on Living Coastal Resources, Vol. 2: Research Papers, eds. S. Sudara, C. R. Wilkinson, and L.M. Chou, pp. 157±160. Chulalongkorn University, Bangkok, Thailand. Szmant, A. M. (1999) Coral restoration and water quality monitoring with cultured larvae of Montastraea `annularis' and Acropora palmata. In Program and Abstracts. International Conference on Scienti®c Aspects of Coral Reef Assessment, Monitoring, and Restoration, pp. 188±189. National Coral Reef Institute, Nova Southeastern University, Florida, USA. Szmant-Froelich, A. (1985) The e€ect of colony size on the reproductive ability of the Caribbean coral Montastrea annularis (Ellis and Solander). Proceedings of the Fifth International Coral Reef Congress 4, 295±300. Thongtham, N. and Chansang, H. (1999) In¯uence of surface complexity on coral recruitment at Maiton Island, Phuket, Thailand. Phuket Marine Biology Center Special Publication 19, (in press). Thorhaug, A. (1987) Large-scale seagrass restoration in a damaged estuary. Marine Pollution Bulletin 18(8), 442±446. Van Treeck, P. and Schuhmacher, H. (1997) Initial survival of coral nubbins transplanted by a new coral transplantation technology ± options for reef rehabilitation. Marine Ecology Progress Series 150, 287±292. Van Treeck, P. and Schuhmacher, H. (1998) Mass diving tourism ± a new dimension calls for new management approaches. Marine Pollution Bulletin 37, 499±504. Wallace, C. C. (1985) Seasonal peaks and annual ¯uctuations in recruitment of juvenile scleractinian corals. Marine Ecology Progress Series 21, 289±298. Williams, D. M., Wolanski, E. and Andrews, J. C. (1984) Transport mechanisms and the potential movement of planktonic larvae in the central region of the Great Barrier Reef. Coral Reefs 3, 229±236. Willis, B. L. and Oliver, J. K. (1988) Inter-reef dispersal of coral larvae following the annual mass spawning on the Great Barrier Reef. Proceedings of the Sixth International Coral Reef Symposium 2, 853±859. Wilkinson, C., Linden, O., Cesar, H., Hodgson, G., Rubens, J. and Strong, A. E. (1999) Ecological and socioeconomic impacts of 1998 coral mortality in the Indian Ocean: an ENSO impact and a warning of future change? Ambio 28(2), 188±196. Woodley, J. D. and Clark, J. R. (1989) Rehabilitation of degraded coral reefs. Coastal Zone '89 Proceedings of the Sixth Symposium Coastal and Ocean Management 4, 3059±3075. Yap, H. T. and Gomez, E. D. (1984) Growth of Acropora pulchra II. Responses of natural and transplanted colonies to temperature and day length. Marine Biology 81, 209±215. Yap, H. T. and Gomez, E. D. (1985) Growth of Acropora pulchra III. Preliminary observations on the e€ects of transplantation and sediment on the growth and survival of transplants. Marine Biology 87, 203±209. Yap, H. T., Licuanan, W. Y. and Gomez, E. D. (1990) Studies on coral reef recovery and coral transplantation in the northern Philippines: aspects relevant to management and conservation. In Proceedings of the First ASEAMS Symposium on Southeast Asian Marine Science and Environmental Protection, ed. H. T. Yap, pp. 117±127. UNEP Regional Seas Reports and Studies 116, United Nations Environment Programme, Nairobi. Yap, H. T., Ali~ no, P. M. and Gomez, E. D. (1992) Trends in growth and mortality of three coral species (Anthozoa: Scleractinia), including e€ects of transplantation. Marine Ecology Progress Series 83, 91±101. Yeemin, T. and Sudara, S. (1992) The role of coral recruitment in problems of coral reef development: a perspective view. In Third ASEAN Science and Technology Week Conference Proceedings, Vol. 6, Marine Science: Living Coastal Resources, 21±23 Sept 1992, Singapore, eds. L. M. Chou, C.R. Wilkinson, pp. 15±25. Dept. of Zoology, National University of Singapore and National Science and Technology Board, Singapore. Yeemin, T., Sudara, S. and Amornsakchai, S. (1992) Distribution and abundance of juvenile corals at Pha-Nga Island, Tao Island and Nang Yuan Island, Gulf of Thailand. In Third ASEAN Science and Technology Week Conference Proceedings, Vol. 6, Marine Science: Living Coastal Resources, 21±23 September 1992, Singapore, eds. L. M. Chou, C. R. Wilkinson, pp. 63±66. Department of Zoology, National University of Singapore and National Science and Technology Board, Singapore. Zedler, J. B. (1984) Salt Marsh Restoration: a Guidebook for Southern California. California Sea Grant Report No. 7-CSGCP-009. 46 pp. 487