Population Structure as a Response of Coral Communities to Global Change 1

Transcription

1 AMER. ZOOL., 39:56-65 (1999) Population Structure as a Response of Coral Communities to Global Change 1 ROLF P. M. BAK AND ERIK H. MEESTERS Netherlands Institute for Sea Research (N10Z), P.O. Box 59, 1790 AB den Burg, Texel, The Netherlands SYNOPSIS. Coral population size structure is generally highly skewed, with a preponderance of the smallest colony size class in populations. To begin to assess possible effects of global change on coral populations, which will be largely controlled in the next decade(s) by "non-climate variables" such as sedimentation, turbidity and nutrient load, we compared degraded reef environments with less degraded reefs. To consider population dynamics within and between parts of coral species metapopulations we use data over two spatial scales (10 and 2,000 km). Colony size distributions appear to be affected in degraded/marginal reefs. This implies changes in mortality patterns (or recruitment) that result in relatively fewer small and more large colonies in populations. We predict that the short-term effects of global change, deterioration of local conditions, will not affect the occurrence of large coral colonies (in terms of absolute size, and possibly mean size) but will limit the abundance of small corals. Long-term global change will increasingly include a component of climate change and the effect on coral populations may become more diverse, although effects such as a decreasing calcium carbonate saturation state will also first affect the abundance of coral recruits. We hypothesize that over the next decade(s) coral populations will become increasingly skewed toward larger colonies. INTRODUCTION Populations are commonly defined as dynamic groups of individuals of the same species {e.g., Hastings, 1997). There are two important concepts in population biology concerning scale and interaction between individuals. At the largest spatial scale there is the metapopulation, which consists of the total grouping of all spatially separated, local populations or subpopulations (Hastings and Harrison, 1994). Within the metapopulation, interaction and exchange between subpopulations are generally low. The spatial units which form the local or subpopulations are the groups of interacting individuals within a species, and they are distributed over a range of separate locations. Subpopulations have largely independent dynamics and density regulation (Taylor, 1991). Subpopulations are engaged in con- ' From the Symposium Coral Reefs and Environmental Changes Adaptation, Acclimation, or Extinction presented at the annual Meeting of the Society for Comparative and Integrative Biology, January 3-7, 1998, at Boston, Massachusetts. tinuous processes of colonization of empty habitats, maintaining subpopulation densities in some, while going extinct in others (Hastings, 1997). Such local extinctions are probably more strongly influenced by demographic or environmental stochasticity (Birkeland, 1988) than by population genetics (Schaffer, 1987; Lande, 1988; Lacy, 1988; in Johnston et al, 1995). Global change, whether as change in climate, cloud cover, sea-level change, terrestrial run-off or otherwise, will change local environmental conditions. Because global environmental change first acts at the scale of the location {e.g., Pandolfi, 1999; Pittock, 1999), the subpopulation is the first level affected by global change (Fig. 1). In the geographic realm of the metapopulation, a mosaic of responding subpopulations will be the result of the heterogeneous environment. In tropical coastal seas corals will recruit, grow, maintain colony integrity, suffer partial mortality or die and disappear completely, depending on local conditions. Such ecological processes will shape the characteristics of the coral sub- 56

2 CORAL POPULATION STRUCTURE AND GLOBAL CHANGE 57 GLOBAL CHANGE SUBPOPULATIONS Environmental Change SPECIES METAPOPULATION and Stochasticity ' line* Colonization Extinction FIG. 1. Overview of species metapopulation, subpopulations and impact level of global change. Arrows indicate interaction frequency within and between subpopulations. population. Over long time scales global changes in magnesium calcite and aragonite saturation, intensity of tropical storms, coastal turbidity, etc., will influence rates and threshold values of these processes. On the short time scale of decade(s) non-climate changes such as increased sedimentation, turbidity and nutrient load of the water column will be most prominent. Such environmental changes will consequently change the composition of the subpopulation. The relation between environmental variables and population characteristics is poorly known. We do not understand how population and community structure/function differ in degraded and un-degraded reefs (e.g., Done, 1992), but we need such information to be able to predict changes in populations with global change. Here we will review the relevant literature and present new data. The question is: how do subpopulations of coral species vary with the environment? Also, can such differences then be used to predict responses in subpopulations under specific environmental conditions such as are expected to occur with global change? A first step is to understand variation among groups of individuals of a coral species living at different reef localities. In the literature such groups of corals, representing individuals spread over distances of 10' to 10 2 km, are generally referred to as populations. Such spatial scales are fractions of the total habitat available to a species and such coral populations, covering only parts of the total metapopulation distribution, are actually synonymous with subpopulations. CORAL POPULATION STUDIES Coral population biology is an amazingly neglected field in reef studies. Descriptive studies are rare, and there is even less known about the dynamics and mechanisms of change in coral populations. Much of the data available has emerged as part of larger community studies in terms of coral cover per species, number of colonies of species, and sometimes, (mean) sizes of colonies (e.g., Edmunds et al, 1990). This information can lead to intriguing questions, e.g., why do certain species at all sites always occur in very low densities compared to "normal" species which occur in varying densities, depending on the environment? However, in terms of understanding population dynamics and the stage of development of a coral species population, the information that has been traditionally gathered in reef coral community surveys is very limited. To understand dynamics, separate censuses over various time scales are thought to be necessary (Hughes and Jackson, 1985; Babcock, 1991; Bak and Nieuwland, 1995; Aronson and Precht, 1997). Also we need the intraspecific in addition to the interspecific point of view. To be able to discuss global change and predict change in coral populations we must understand the response of coral populations to environmental variation. To try to understand and predict the possible effects of global change on coral populations, we will first discuss characteristics of coral populations and their relationship to the environment. We use some of our own data sets in an attempt to link population structure to population dynamics. We will consequently formulate a hypothesis linking change in population structure to general deterioration of the environment. Finally, we discuss some of the specific environmental changes predicted for global change and their likely specific effects on coral populations. How do we see if coral subpopulations and their demographic histories differ, and how populations respond to the environment? An obvious variable to study in coral populations is size of colonies. Size of corals is related to processes of settlement,

3 58 R. P. M. BAK AND E. H. MEESTERS growth, survival, reproduction and mortality (Connell, 1978; Loya, 1976; Szmant, 1991; Soong, 1993) though the relationship between size and age is distorted in some species (Hughes, 1984; Hughes and Jackson, 1985). Interest in the relationship of life history characteristics to colony size has stimulated studies on the size frequency distributions of coral species (Babcock, 1991; Soong, 1993; Endean, 1997; Bak and Meesters, 1998) and these allow two important conclusions. First, all size distributions are extremely skewed to the right, having most colonies (the mode) in the smallest size class. Secondly, the maximum size of colonies in populations differs enormously among species. There are some exceptions to this general pattern. Tomascik (1996) found a population size distribution skewed to the right but with fewer colonies in the smallest size class. This distribution represents the grouped data for several species of tabulate Acropora colonizing newly available substratum (lava flows). These populations result from a major recruitment event, with subsequent rapid growth of colonies to high coral cover (>60%). High coral cover and the tabulate growth form may have inhibited survival of subsequent settlers, explaining the atypical shape of the size frequency distribution. Such deviant size distributions are probably more common in reefs where space is a crucial limiting factor (Done and Potts, 1992). The preponderance of the smallest size class in populations highlights the important role of small corals in population dynamics. However, analysis of processes acting on small-sized colonies is limited because in all data sets the smallest size category has been defined rather widely. Coral recruits at settlement, and during the crucial period following, have a size of millimeters to a few centimeters. In the available data sets, all small corals are included with much larger colonies, e.g., a size class from 0-50 cm 2. Such data lack the necessary resolution and do not allow analysis of the dynamics of these vulnerable, smallest size classes. TABLE 1. Total number of colonies measured in 9 massive/submassive coral species from Curacao and Florida. Species Agaricia agaricites Colpophyllia natans Dichocoenia stokesii Diploria labyrinthiformis Diploria strigosa Montaslraea annularis Montastraea cavernosa Porites astreoides Siderastrea siderea Curasao Florida VARIATION IN POPULATION SIZE DISTRIBUTIONS AT DIFFERENT SPATIAL SCALES A relevant question, in view of the potential effects of global change, is how do environmental variables shape the colony size distribution of coral subpopulations? If colony size reflects processes of growth and of partial and total colony mortality the environmental variables must be important. We need data on the same species from different localities to see if and how differences in characteristics of size distribution of populations are influenced by the environment. There are few such data available (Babcock, 1991; Lewis, 1997) and none offers the detail needed in the smallest size class. We have data to compare populations at two different spatial scales: at the scale of km for populations along the fringing reefs of Curacao (Bak and Meesters, 1998) and at the scale of 1,000s km for a comparison between Curasao and the northern part of the Florida reef tract (Ginsburg et al., unpublished data). We measured colony size of 9 massive/submassive coral species in Cura ao and Florida (Table 1) at horizontal reef bottom between 4 and 9 m. The communities were similar in composition of scleractinian and gorgonian corals, fishes, etc. We wanted to use narrow size categories to present small-sized colony frequencies. In Florida colony diameter was measured in ten centimeter intervals. In Curacao additional measurements were made to allow calculation of total living surface area and partial mortality. At the smallest scale, in Cura ao, there must be genetic ex-

4 CORAL POPULATION STRUCTURE AND GLOBAL CHANGE 59 Diameter (cm) FIG. 2. Coral colony size frequencies for nine coral species at Curacao and Florida. White bars: Curacao population, black bars: Florida population. For abbreviations, species names, and sample size see Table 1. change between sample sites (for scales of sites and currents see van Duyl, 1985) and we are dealing with the same subpopulation. At the larger scale, between Curacao and Florida, gene flow is unlikely (Roberts, 1997) and these are separate subpopulations within the metapopulation of each species. At all locations colony size distributions tend to have the usual, right-skewed distribution (Fig. 2). Comparing Curacao with Florida shows that there were significant differences in size distributions between subpopulations of the same species (all species except Montastraea cavernosa, Kolmogorov-Smirnov, P < 0.001). Colonies in the smallest category are more common in Curacao for all species (except Dichocoenia stokesii; Fig. 2). In Florida larger colonies are much more dominant. In this environment colonies apparently grow to much larger sizes. Crucial in the size regulation of coral colonies is partial mortality (e.g., Hughes, 1984; Meesters et al, 1996, 1997). Partial mortality, the death of part(s) of the living tissue, will cause large colonies to move back into smaller size class categories. In the smallest size classes partial mortality is of minor importance because these colonies are so small that most damage will result in total mortality (Soong, 1993; Bak and Meesters, 1998). The relatively large size most coral species reach in Florida suggests a relatively low impact of partial mortality in this environment. Low partial mortality would translate to low total mortality in small corals. Consequently the low abundance of juveniles in Florida would suggest a low input of recruits in Florida. This hypothesis is confirmed by studies reporting low juvenile densities for the Florida reef tract, 1.2 to 3.7 m~ 2 (Chiappone and Sullivan, 1996), compared to 15 m~ 2 in Cura ao (Bak and Engel, 1979). Rates of recolonization are important to maintain populations and minimize the chance of local extinction in suitable habitats (Lasker, 1999). Large colony size would correlate to a high number of reproductive modules (Johnston et al, 1995) but the Florida example shows that there is no obligate relationship between reproductive yield and recruitment.

5 60 R. P. M. BAK AND E. H. MEESTERS Demographic statistics in Honda differ from Curacao and the difference indicates low local recruitment in Florida that is not remedied by influx of emigrants from the wider Caribbean (Roberts, 1997). Florida reefs are presumably more marginal and environmentally stressed than reefs in Curasao, being subjected to a variety of disturbances, e.g., temperature, salinity/turbidity variations and nutrient enrichment (Hudson, 1981; Lapointe and Matzie, 1996). The differences in demographic characteristics of subpopulations at such a large spatial scale may have a genetic as well as an environmental basis. Both possibilities are relevant in the frame of global change events because there has been either adaptation of different subpopulations (a genetic phenomenon) or the data show how environmental variables can structure different parts of the same subpopulation, which is sharing a common, interchanging gene pool. To examine variation in a common interchanging gene pool we look at a much smaller spatial scale, viz., at size frequencies of subpopulations along the fringing reefs of Curacao. We assume that in these reefs over a distance of a few km (van Duyl, 1985), exchange of genetic material between these populations must be a frequent phenomenon. We have data from four sites (Avila, Marie Pompoen, Seaquarium and Cornelis Bay) with distances between sites being 2, 2.5 and 0.5 km, respectively. We can compare the same 9 species used in the Florida-Curacao comparison. Testing for differences in the size frequency distributions, for each species between sites, it appears that there are significant differences (G tests, P < 0.001) for all species (except Porites astreoides). Such differences between neighboring populations clearly suggest that the environment, and not genetic variation between subpopulations, is responsible for the differences in colony size between these parts of the metapopulations. That the impact of the environment is forcing the structure of the populations is confirmed when we look into the environmental differences between the four Curasao reef sites. Two sites are degraded, being subjected to enhanced turbidity and eutro- n c O 0,0 Q. O Control A. agaricites Polluted A. agaricites JklL 1218 ioe«ib3 oae-o UUJ C. natans C. natans 'M 403*3 22E*G6 12E-0T 006*08 D.strigosa D. strigosa \MJ «03*3 32E-O5 12E-OTME-O8 Size (cm ) FIG. 3. The size frequency distributions for three coral species Agancia agaricites, Colpophyllia natans, Diplona labyrinthiformes at control and degraded reefs in Curacao (colony size log scale). phication. The other two are upcurrent of these and can serve as controls for relatively unspoiled reefs. We can compare size distributions between degraded and control reefs. Figure 3 shows the size distributions for three species, Agaricia agaricites, Colpophyllia natans, Diploria labyrinthiformes. We see the same sort of variation between populations as encountered between Florida and Curasao: a decrease in the number of colonies in the smallest size class and in the dominance of large-sized colonies in the degraded areas. AN ALTERNATIVE VIEW OF SIZE DISTRIBUTIONS Transformation of colony size to natural logarithms will often remove the asymmetry and result in normal distributions. Such transformation will provide size distributions with a number of useful mathematical properties, e.g., the statistics of the frequency distributions can be compared (Underwood, 1997). Such data are available only

6 CORAL POPULATION STRUCTURE AND GLOBAL CHANGE 61 for Curasao. Here all living parts of colonies were measured precisely as geometric shapes and the surface area for each colony was calculated. The resulting size frequencies of species can be modeled by log normal distributions and analyzed in terms of descriptive statistics (Bak and Meesters, 1998). Figure 4 shows the size frequencies of coral species (Avila, Curacao) at linear and at log scales and highlights the enormous difference in perspective. Log transformation clearly has two advantages. First, it allows a much closer view of the partition over size categories of the smaller colonies. At the linear scale the highest numbers of colonies are generally in the smallest size class. At log scale a species known as a frequent recruit, e.g., Agaricia agaricites, is clearly relatively numerous in the smaller size classes. Transformation facilitates the comparison of the geometric mean. This is a speciesspecific variable but it varies greatly with environment in some species {e.g., Montastraea annularis, Meandrina meandrites, Colpophyllia natans) while such variation is small in others {Montastraea cavernosa, M. faveolata) (Fig. 5). This is important in view of the scope for adaptation and possible effects of global change. Throughout the Neogene, species with large colony size have lower extinction rates than species with smaller colonies (Johnston et al., 1995). The mechanisms involved here are unclear. That higher numbers of reproductive polyps per population result in higher recolonization is refuted by the large mean Florida colony size which was accompanied by low influx of small colonies. The descriptive statistic coefficient of variation (V) allows comparison of variation of colony size in coral populations of different mean colony size. V is highest in species with colonies having a small mean size. Within species higher V is also associated with the smallest size classes (Bak and Meesters, 1998). This indicates a relationship with fluctuations in recruitment and juvenile mortality. These processes are probably strongly related to fluctuations in environmental variables. Small mean colony size may be related to higher species origination rates (Johnston et al., 1995). Also, selection cannot act without variability (Hastings, 1997). The magnitude of V, in relation to size-frequency characteristics such as skewness (see below), could be of importance in evolutionary processes. This suggests that a decreased V in larger-sized coral populations, in relation to global change environmental stress, would tend to counter phenotypic plasticity in coral populations, to limit genetic variance, and to decrease speciation processes. The skewness of the size frequency distributions reflects the proportion of small versus larger colonies, representing juvenile input and longevity, respectively. In additional data sets from Curacao (14 species at 4 sites, n = 56) 51 data sets were skewed to the left (at log scale), meaning that generally large-sized colonies dominate in populations, although the proportion of large to small colonies differs between species. In species with large mean colony size, small colonies are underrepresented. This has consequences for the genetic variation of the reproductive output of species. Potts et al. (1985) propose that the rate of evolutionary processes in large-sized populations could be very different from populations with small colonies. Reproduction rates in corals are related to size with larger colonies having much higher fertility rates (Soong and Lang, 1992). Also, because the living surface area of colonies relates to potential reproductive output, the presence of relatively large colonies means that fewer genotypes are reproductively active (gametic dominance; Potts et al, 1985). This change in skewness could be a means by which populations can adapt: shifting towards smaller size colonies, meaning decreased gametic dominance by large colonies, higher turnover and greater genetic diversity. Skewness varies significantly between the Curacao reef environments and at degraded reefs there were fewer small colonies (Fig. 3; Bak and Meesters, 1998). At these sites distributions are more skewed to the left. There is a similarity here with the relatively low presence of small colonies in the stressed or marginal environments of the Florida reefs.

7 62 R. P. M. BAK AND E. H. MEESTERS 500 1C0 ' A. humilis 00 aso - a labyrinthifonvs re.q!u Q. MO O40 1r TOOQOO 11E-O5 14C-0S OOO OOO D. stokesi D. strigosa 1» I OOO I I 040 I " l 100 CWO I -80 O 20 - I O2O- I aio- ^ aio- 1EKJ6 22E-O5 QOO 2S0 M cavemosa P. astreoides A<. meandrites»» OBO- 200 O7O - I I OSO- I I ''"MO -l 030 I 040- I I too I I OSO y aoo OOO EO5 27E*O BOOOO ODD EKX5 000 Size (cm 2 ) 020 ato "f- 1 r T E*08 1BE*06 ;«06 fi2e*o5 000 M. faveotata LJO o 3 A.agaricites C. natans D. labyrinthiforms M o» fi.41 40U77 1S *O S E»OB 0OO2 «E«O6 O *0 1O0SS3 22E>06 (0 UJ D. stokesi D. strigosa M. annuiaris. faveolata ^ Q2O Q. o«h 010- c O aio 20 OO8- E 004 O O2 Q. O 9002 OE*OS 2-7S 1B0BXM 1ZEKJ E-06 M. cavemosa M. JJ meandrites ttalbk-j QS1 &*BO E.06 OI4 1Z10 10B&S3 O0E-C6 ON BE*O9 014 Size (cm ) P. astreoides :*O6 S. siderea FIG. 4. The size frequencies of coral species measured at one site (Avila) in Curacao at linear scale (A) and at log normal scale (B). J52E*O0 o

8 CORAL POPULATION STRUCTURE AND GLOBAL CHANGE 63 CD O 1 o tr FIG. 5. Geometric mean of colony size of coral populations at Curacao (4 sites grouped). Bars indicate variability between sites (95% CL). Note log scale. RESPONSES OF CORAL POPULATIONS TO CHANGE Global change will force changes in local environmental conditions and changes in the local conditions will influence the reefs (Pandolfi, 1996). To be reflected in coral population structure, such changes must be sufficient to alter coral recruitment, growth or (partial) survival. We predict changes in coral populations in terms of size-frequency characteristics. The initial effect of the deteriorating environment will not be great mortality in large colonies, but will have an impact on recruits and the smallest size classes. In deteriorating environments recruitment will decrease and mortality of smaller-sized colonies will increase. The distributions will become more skewed to the left and the size of mode and geometric mean colony size will increase (Fig. 6). The coefficient of variation will decrease. Obviously, catastrophic environmental change belongs to events in another category, e.g., FIG. 6. Global Change ^ the impact of tropical storms can annihilate colonies irrespective of size but depending on microhabitat, colony shape and attachment, etc. (Woodley et al, 1981; Bythell et al, 1993; Massell and Done, 1993). The changes we find in populations are intrapopulation changes, not to be confused with differences in properties that are inherent to intrinsically smaller and larger species. Smaller-sized species are in general subjected, possibly adapted, to strong variation in the processes of recruitment and related mortality. A higher rate of speciation would correlate with such characteristics (Johnston et al, 1995). The persistence of large colonies, with large numbers of polyps, theoretically results in a high output of gametes. However, our Florida example suggests that in more marginal environments such output may have no effect. Recruitment rates may suffer under increased total mortality of recruits and juvenile colonies. We have treated global change as unspecified deterioration of the environment and our predictions are for reefs that in the next decade(s) will be exposed to increased sedimentation, turbidity and nutrient loads. These represent the primary, local threat to reefs. Forcing factors that will act in global change in the long run, which are not covered by our discussion, are temperature increase and decrease in calcium carbonate saturation state of the ocean. There are no indications that a temperature increase, while changing the geographic distribution of species, will have a specific impact on population structure in most coral species. On the other hand the change in ocean cal- I,.1..IF h... I.. III ll Log (size class) Model of effect of global change (environmental deterioration) on coral size frequency distribution. ^ Log (size class) 1

9 64 R. P. M. BAK AND E. H. MEESTERS cium carbonate saturation will reduce calcification rates. There are two possible effects. Calcification in corals may decrease 9 to 30% (Gattuso et al, 1999). Calcification rates are not related to colony size or age (Buddemeier and Kinzie, 1976), nevertheless such a decrease will change the size distributions of coral populations. Recruits and corals in the smallest size class are most vulnerable and need to "escape in size and height" (Jackson, 1977; Meesters et al., 1996). Any disadvantage at this juvenile stage will greatly increase their mortality rates relative to those of large colonies, and this will affect their relative abundance in the populations. Size distributions will become increasingly skewed, with large colonies more dominant in populations. Coral recruits may also be affected through a decrease in preferred or obligatory substratum. Magnesium calcite, a component of the skeletons of crustose coralline algae, will experience relative decreases in saturation state greater than aragonite. Because many corals, in major families, are largely depending on specific crustose corallines for settlement (Morse, 1994; Morse et al., 1996) a decrease in availability of this substratum would also tend to limit the presence of recruits and small colonies in future coral populations. These hypotheses/predictions cannot be checked easily in records of reef global changes of the past. Size frequencies in the geological record do not readily offer the precision that is needed, e.g., in recording small colonies. Our predictions should be tested through construction of high precision size-frequency distributions in degrading and marginal environments. Subsequent analyses of descriptive statistics using logscaled normal distributions will confirm our hypotheses or, at least, offer new insight into coral population dynamics. ACKNOWLEDGMENTS We thank staff of Carmabi, M. Hilterman, E. Kardinaal, M. Keetman, M. de Vries and other students for their help in surveying the Curacao reefs. Dr. R. N. Ginsburg supported R.P.M.B. in Florida with funds from the Hachette Filipacci Foundation. R. N. Ginsburg, W. E. Kiene, V. Kosmynin joined R.P.M.B. in collecting the Florida data. We thank Dr. R. W. Buddemeier for his comments on the ms. Support of SCOR in attending meetings of the Working Group 104 is gratefully acknowledged. Interaction with Working Group members and participants in the Boston workshop was a stimulating experience. REFERENCES Aronson, R. B. and W. B. Precht Stasis, biological disturbance, and community structure of a holocene coral reef. Paleobiol. 23: Babcock, R. C Comparative demography of three species of scleractinian corals using age- and size-dependent classifications. Ecol. Monogr. 61: Bak, R. P. M. and M. S. Engel Distribution, abundance and survival of juvenile hermatypic corals (Scleractinia) and the importance of life history strategies in the parent coral community. Marine Biology 54: Bak, R. P. M. and E. H. Meesters Coral population structure: The hidden information of colony size-frequency distributions. Mar. Ecol. Prog. Ser. 162: Bak, R. P. M. and G. Nieuwland Long-term change in coral communities along depth gradients over leeward reefs in the Netherlands Antilles. Bull. Mar. Sc. 56: Birkeland, C Geographic comparisons of coralreef community processes. Proc. Sixth Int. Coral Reef Symp, Townsville, Australia, 1: Buddemeier, R. W. and R. A. Kinzie Coral growth. Oceanogr. Mar. Biol. Ann. Rev. 14: Bythell, J. C, M. Bythell, and E. H. Gladfelter Initial results of a long-term coral reef monitoring program: Impact of hurricane Hugo at Buck Island Reef National Monument, St. Croix, U.S. Virgin Islands. J. Exp. Mar. Biol. Ecol. 172: Chiappone, M. and K. M. Sullivan Distribution, abundance and species composition of juvenile scleractinian corals in the Florida Reef Tract. Bull. Mar. Sci. 58: Connell, J. H Population Ecology of reef-building corals. In O. A. Jones and R. Endean (eds.), Biology and geology of coral reefs, pp Academic Press, London. Done, T. J Phase shifts in coral reef communities and their ecological significance. Hydrobiologia 247: Done, T. J. and D. C. Potts Influences of habitat and natural disturbances on contributions of massive Porites corals to reef communities. Marine Biology 114: Duyl, F. C. van Atlas of the living reefs of Curacao and Bonaire (Netherlands Antilles). Found. Scient. Res., Surinam, Netherlands Antilles. Edmunds, P. J., D. A. Roberts, and R. Singer

10 CORAL POPULATION STRUCTURE AND GLOBAL CHANGE 65 Reefs of the northeastern Caribbean I. Scleractinian populations. Bull. Mar. Sci. 46: Endean, R., A. M. Cameron, H. E. Fox, R. Tilbury, and L. Gunthorpe Massive corals are regularly spaced: Pattern in a complex assemblage of corals. Mar. Ecol. Prog. Ser. 152: Gattuso, J. P., D. Allemand, and M. Frankignoulle Photosynthesis and calcification at cellular organismal and community levels in coral reefs: A review on interactions and control by carbonate chemistry. Amer. Zool. 39: Hastings, A Population biology. Springer Verlag, New York. Hastings, A. and S. Harrison Metapopulation dynamics and genetics. Ann. Rev. Ecol. Syst. 25: Hudson, J. H Growth rates in Montastrea annularis: A record of environmental change in Key Largo coral reef marine sanctuary, Florida. Bull. Mar. Sci. 31: Hughes, T. P Population dynamics based on individual size rather than age: A general model with a reef coral example. Am. Nat. 123: Hughes, T. P. and J. B. C. Jackson Population dynamics and life histories of foliaceous corals. Ecol. Monogr. 55: Jackson, J. B. C Competition on marine hard substrata: The adaptive significance of solitary and colonial strategies. Am. Nat. 111: Johnston, K. G., A. F. Budd, and T. A. Stemann Extinction selectivity of neogene Caribbean reef corals. Paleobiol. 21: Lacy, R. C The effects of inbreeding on isolated populations: Are minimum viable population sizes predictable? In P. L. Fiedler and S. K. Jain (eds.), Conservation biology, pp Chapman and Hall, London. Lande, R Genetics and demography in biological conservation. Sci. 241: Lapointe, B. E. and W. R. Matzie Effects of stormwater nutrient discharges on eutrophication processes in nearshore waters of the Florida Keys. Estuaries 19: Lasker, H. R. and M. A. Coffroth Responses of coral reef taxa to environmental change. Amer. Zool. 39: Lewis, J. B Abundance, distribution and partial mortality of the massive coral Siderastrea siderea on degrading coral reefs at Barbados, West Indies. Mar. Pol. Bull. 34: Loya, Y Skeletal regeneration in a Red Sea scleractinian coral population. Nature 261: Massel, S. R. and T. J. Done Effects of cyclone waves on massive coral assemblages on the Great Barrier Reef: Meteorology, hygrodynamics and demography. Coral Reefs 12: Meesters, E. H., I. Wesseling, and R. P. M. Bak Partial mortality in three species of reef-building corals and the relation with colony morphology. Bull. Mar. Sci. 58: Meesters, E. H., I. Wesseling, and R. P. M. Bak Coral colony tissue damage in six species of reefbuilding corals: Partial mortality in relation with depth and surface area. J. Sea Res. 37: Morse, A., K. Iwao, M. Baba, K. Shimoike, T. Hayashibara, and M. Omori An ancient chemosensory mechanism brings new life to coral reefs. Biol. Bull. 191: Morse, D. E., et al Morphogen-based chemical flypaper for Agaricia humilis coral larvae. Biol. Bull. 186: Pandolfi, J. M Limited membership in Pleistocene reef coral assemblages from the Huon Peninsula, Papua New Guinea: Constancy during global change. Paleobiol. 22: Pandolfi, J. M Responses of Pleistocene coral reefs to environmental change over long temporal scales. Amer. Zool. 39: Pittock, A. B Coral reefs and environmental change: Adaptation to what? Amer. Zool. 39: Potts, D. C, T. J. Done, P. J. Isdale, and D. A. Fisk Dominance of a coral community by the genus Porites (Scleractinia). Mar. Ecol. Prog. Ser. 23: Roberts, C. M Connectivity and management of Caribbean coral reefs. Sci. 278: Schaffer, M. L Minimum viable populations: Coping with uncertainty. In M. E. Soul6 (ed.), Viable populations for conservation, pp , Cambridge University Press. Soong, K Colony size as a species character in massive reef corals. Coral Reefs 12: Soong, K. and J. C. Lang Reproductive integration in reef corals. Biol. Bull. 183: Szmant, A. M Sexual reproduction by the Caribbean reef corals Montastrea annularis and M. cavernosa. Mar. Ecol. Prog. Ser. 74: Taylor, A. D Studying metapopulation effects in predator prey systems. Biol. J. Linnean Soc. 42: Tomascik, T Rapid coral colonization of a recent lava flow following a volcanic eruption, Banda Islands, Indonesia. Coral Reefs 15: Underwood, A. J Experiments in ecology. Cambridge University Press, Cambridge. Woodley, J. D., E. A. Chornesky, P. A. Clifford, J. B. C. Jackson, L. S. Kaufman, N. Knowlton, J. C. Lang, M. P. Pearson, J. W. Porter, M. C. Rooney, K. W. Rylaarsdam, V. J. Tunnicliffe, C. M. Wahle, J. L. Wulff, A. S. G. Curtis, M. D. Dallmeyer, B. P. Jupp, M. A. R. Koehl, J. Neigel, and E. M. Sides Hurricane Allen's impact on Jamaican coral reefs. Science 214: Corresponding Editor: Kirk Miller