SUPPLY-SIDE ECOLOGY WORKS BOTH WAYS: THE LINK BETWEEN BENTHIC ADULTS, FECUNDITY, AND LARVAL RECRUITS.
N.A. MOLTSCHANIWSKYJ 
J. E. TANNER 
Key words: coral reefs; dispersal; fecundity; gene flow; Great Barrier Reef, Australia; mass spawning; population dynamics; recruitment limitation; spatial scale; stock size; supply-side ecology.
Abstract. "Supply-side" ecology recognizes the potential role that recruitment plays in the local population dynamics of open systems. Apart from the applied fisheries literature, the converse link between adults and the production of cohorts of recruits has received much less attention. We used a hierarchical sampling design to investigate the relationships between adult abundance, fecundity, and rates of larval recruitment by acroporid corals on 33 reefs in five sectors (250-400 km apart) stretching from north to south along the length of the Great Barrier Reef, Australia. Our goal was to quantify patterns of recruitment at multiple scales, and to explore the underlying mechanisms. Specifically, we predicted that large-scale patterns of recruitment could be driven by changes in the abundance of adults and/or their fecundity, i.e., that corals exhibit a stock-recruitment relationship. The amount of recruitment by acroporids in each of two breeding seasons varied by more than 35-fold among the five sectors. Adult density varied only twofold among sectors and was not correlated with recruitment at the sector or reef scale. In contrast, fecundity levels (the proportion of colonies on each reef that contained ripe eggs) varied from 15% to 100%, depending on sector, year, and species. Spatial and temporal variation in the fecundity of each of three common Acropora species explained most of the variation (72%) in recruitment by acroporids, indicating that the production of larvae is a major determinant of levels of recruitment at large scales. Once fecundity was accounted for, none of the other variables we examined (sector, reef area, abundance of adults, or year) contributed significantly to variation in recruitment. The relationship between fecundity and recruitment was nonlinear, i.e., rates of recruitment increased disproportionately when and where the proportion of gravid colonies approached 100%. This pattern is consistent with the hypothesis that enhanced fertilization success and/or predator satiation occurs during mass-spawning events. Furthermore, it implies that small, sublethal changes in fecundity of corals could result in major reductions in recruitment.
Dispersal plays a crucial role in the ecology and evolution of many organisms, particularly in marine systems where many animals and plants exhibit alternate benthic and planktonic life-history stages (e.g., Thorson 1950, Reed et al. 1988, Strathmann 1993). At the end of the dispersal phase, the abundance of larvae at settlement is often highly variable, both spatially and temporally (e.g., Barnes 1956, Milicich 1994). Recent attention has focussed on the causes of variation in settlement, particularly on larval mortality (e.g., Houde 1987, Underwood and Fairweather 1989), transport mechanisms (e.g., Gaines et al. 1985, Black et al. 1991, Milicich 1994) and larval behavior before and during settlement (e.g., Boicourt 1982, Grosberg 1982). "Supply-side ecology" recognizes the role that variable larval input plays in determining the size of local adult populations (e.g., Underwood and Denley 1984, Hughes 1984, 1990, Gaines and Roughgarden 1985, Roughgarden et al. 1985, Caley et al. 1996). However, the converse linkage between adult stocks and the production of larvae is much less clear (Grosberg and Levitan 1992, Eckman 1996). At small scales (less than a few meters), fertilization rate in sessile or sedentary broadcast spawners is often crucially dependent on adult density, i.e., on the distance traveled by sperm before they encounter an unfertilized egg (in echinoids, Pennington 1985, Levitan 1991; starfish, Babcock et al. 1992; and cnidarians, Yund 1990, Coma and Lasker 1997). At larger scales, the relationship between adult abundance (stock size), fecundity and recruitment in noncommercial species remains virtually unknown.
The issue of spatial scale is crucial for understanding stock-recruitment relationships. If larvae are widely dispersed, the local production of propagules by sessile or sedentary adults will not be correlated with local recruitment; locally derived larvae go elsewhere, and recruits come from afar. Nonetheless, at larger spatial scales there may be some detectable relationship between the size of the spawning stock (i.e., the amount of larval production) and the amount of recruitment. Certainly, temporal correlations between spawning by adults and settlement are well established, with settlement peaks corresponding to earlier cycles of breeding (e.g., corals, Wallace 1985a, b; barnacles, Barnes 1956; crabs, Christy 1982; fishes, Doherty and Williams 1988). The fisheries literature highlights the importance of large-scale stock-recruitment relationships for understanding population dynamics and for management of marine resources, although in most cases "recruit" refers to harvestable adults (e.g., Lipcius and Van Engel 1990, Hilborn and Walters 1992, Peterson and Summerson 1992). However, for most marine organisms, the spatial scale at which adult stocks and recruitment are coupled is unknown. This lack of data most likely stems from the difficulties of measuring recruitment, adult abundances and fecundities at large spatial scales which approach or exceed the extent of larval dispersal (Hughes et al. 1999).
It is often tacitly assumed that stochastic variation in fertilization success, planktonic duration, or mortality rates will destroy any relationship between stock size, the production of larvae, and the number of propagules still alive at settlement (e.g., Houde 1987, Underwood and Fairweather 1989). However, this argument is not valid at larger scales, where global recruitment must diminish if stock sizes are greatly suppressed (e.g., due to disease, Karlson and Levitan 1990, Peterson and Summerson 1992; climatic variation, Cushing 1986; or overfishing, Hilborn and Walters 1992). Moreover, temporal variation in recruitment (for a given stock size) or density-dependence do not preclude a stock-recruitment relationship. Rather, the former will simply increase the variance about the relationship, while the latter will alter its shape. Thus, at larger scales stock-recruitment relationships must exist, although they may be statistically messy and difficult to detect (e.g., Ricker 1954, Lipcius and Van Engel 199 0).
We demonstrate here that large-scale variation in the density of coral recruits on the Great Barrier Reef, Australia, is strongly associated with spatial and temporal changes in the fecundity of adults. This study investigates adult-juvenile relationships at a very large spatial scale ([sim]1800 km), and provides a unique insight into the coupling of benthic and planktonic processes in sessile marine organisms.
We examined the relationship between the abundance of adults, fecundity, and recruitment of corals along the length of the Great Barrier Reef (GBR), in 1995/1996 and 1996/1997 (year 1 and 2). Corals are either brooders, which release internally fertilized planulae, or broadcast spawners, which release eggs and sperm (Harrison and Wallace 1990). We focus here on spawning species in the most abundant scleractinian family, the Acroporidae, hereafter termed "acroporid". We designed our study to take advantage of the predictable, annual mass spawning of corals on the Great Barrier Reef, where more than 130 scleractinian species release their eggs and sperm over a period of a few days in November/December (see Harrison et al. 1984, Harrison and Wallace 1990).
To examine spatial patterns at a hierarchy of scales, we partitioned the GBR into five adjacent sectors, each one 250-400 km long from north to south (Fig. 1, see also Hughes et al. 1999). Eighteen reefs were sampled in year 1 (three reefs per sector, except for sector 4 which had six), and 15 other reefs in year 2; i.e., reefs were independent of year. We deliberately chose a wide range of reef sizes in each sector because bigger reefs should have larger stocks of breeding adults (although settlement can also occur over a larger area). Four sites were established on each reef on the reef crest (1 m depth at low tide), [sim]1-4 km apart. The abundance of corals (counts of colonies [greater than]1 cm in diameter, and percent cover) was measured using ten 10-in line intercept transects a few meters apart at each of the 132 sites (33 reefs X 4 sites). On each of the 1320 transects, each colony lying underneath the tape was identified, and the intercept was measured to the nearest centimeter. More than 30 000 co lonies were censused.
To assess fecundity, tissue samples were collected synchronously each year 10 d ([+ or -] 1 d) before the predicted annual mass spawning of corals. Three species that are abundant throughout the GBR were selected for fecundity analysis: Acropora hyacinthus, A. cytherea, and A. millepora. Two or three branches containing several hundred polyps were taken each year from 20 large colonies ([greater than]30 cm diameter) of each species on each reef, a total of 1980 colonies. Samples were decalcified and dissected and the proportion of colonies with mature (pigmented) eggs was scored for each species and reef (see Hall and Hughes 1996). Additional samples collected after the predicted spawning were empty of eggs. We assume here that the fecundity patterns (in space and time) of these three species are broadly representative of acroporids as a whole, and that they may be able to predict patterns of acroporid recruitment. It is not possible to compare fecundity and recruitment at the individual species level becaus e of the limited taxonomic resolution of newly-settled recruits.
To assess recruitment by spawning acroporids, ten replicate recruitment panels (11 X 11 cm) were deployed each year on the reef crest at each of the four sites per reef (a total of 1320 panels), also 10 d ([+ or -] 1 d) before the predicted annual mass spawning of corals (e.g., Harrison et al. 1984, Harrison and Wallace 1990). The panels were unglazed clay tiles attached individually by a bolt drilled into the substratum. Panels were retrieved after eight weeks (86% were recovered undamaged), and the coral recruits were counted and identified to family or genus. The synchronous deployment and retrieval of panels on large numbers of midshelf reefs (typically 30-80 km offshore) along nearly 2000 km of coastline was essential for ensuring that spatial patterns in recruitment were not confounded with (1) the timing of deployment and (2) the duration of exposure of the panels. Based on pilot studies, most of the recruits at the time of collection would have settled in the previous 5-6 wk; i.e., beginning [sim]2 wk after deployment of the panels and 3-5 d after the mass spawning. This relatively short duration was long enough to allow modest taxonomic resolution, while minimizing losses through early mortality. Acroprid colonies do not develop species-level morphological features until a minimum age of 2-3 yr.
Spatial variation in adult abundance, fecundity and recruitment were examined using hierarchical (nested) analysis of variance, with appropriate transformations where necessary. We used nonlinear regression models (Generalized Additive Models, Hastie and Tibshirani 1990) to examine the relationship between recruitment by acroporids on each reef versus the abundance of established acroporids, fecundity of each of the three targeted Acropora species, and several other potentially important variables (reef size, sector, and year). The size of each reef (projected area in [km.sup.2]) varied by 1-2 orders of magnitude in each sector, and was measured from GIS data. Sector (n = 5) and year (n = 2) were entered into the model sequentially as categorical variables to examine spatial and temporal variation that was independent of fecundity. All other variables were continuous, and preliminary analyses revealed, not surprisingly, that they were not linearly related to recruitment. Consequently, they were entered into the model as smoothed, nonlinear variables using spline functions with four degrees of freedom. (One degree of freedom would fit a straight line whereas n degrees would join all points. Four produces "modest" smoothing, see Hastie and Tibshirani 1990).
RESULTS AND DISCUSSION
The mean abundance of adult acroporids on all reefs was 12.3 [+ or -] 1.3 (mean [+ or -] 1 SE) colonies per 10-m transect (or 20.56% [+ or -] 3.7% cover, n = 33 reefs). Adult density varied only twofold among sectors (Fig. 2), with most of the variation occurring at much smaller spatial scales (Table 1). A hierarchical analysis of variance indicated that 19% and 5% of the variation was attributable to sector and reef, respectively. The remaining 75% of the variation in adult density occurred within reefs (i.e., among sites on the same reef, and among adjacent transects). Similarly, 82% of the variation in the percent cover of spawning acroporids also occurred at very local scales, i.e., within reefs.
In contrast, fecundity showed much more substantial large-scale variation among sectors, with different patterns occurring among years (yielding a significant sector X year interaction, P = 0.0 12; Table 2, Fig. 3). All three species exhibited a similar spatial and temporal pattern (Table 2), with markedly lower fecundities at both ends of the Great Barrier Reef in year 1, and in the northernmost sector in year 2. For the three species combined, the proportion of colonies containing eggs varied spatially by six-fold among sectors in year 1 and by twofold in year 2. The largest difference between years was in the far north and southern-most reefs (sectors 1 and 5), where the average proportion of colonies containing eggs was two to three times higher in year 2 (Fig. 3).
Recruitment by spawning acroporids also varied greatly in space and time (Fig. 4, Table 3). A total of 58471 coral recruits were recorded on 1135 panels, for both years combined. Spawning corals comprised 83% of the recruits and brooders made up the remainder, with 96% of the spawners being juvenile acroporids. The mean number of acroporid recruits per panel varied by more than 100-fold among sectors in year 1, and by 35-fold in year 2 (Fig. 4). In year 1, fewer recruits were found at the northernmost and southern sectors (1, 4, and 5). In year 2, recruitment increased in all regions of the Barrier Reef (except for sector 4, which remained unchanged). The northernmost and southernmost sectors (1 and 5) exhibited the most marked increase, by a factor of about 10 and 100, respectively (yielding the highly significant sector X year interaction, P [less than] 0.001; Table 3). For all sectors combined, the total amount of recruitment by acroporid corals along the Great Barrier Reef in year 2 was three times highe r than in year 1.
The large-scale spatial and temporal variation in recruitment was clearly linked to patterns of variation in fecundity rather than adult abundance, i.e., when and where fecundity increased, so too did the density of recruits (Figs. 3 and 4). Among the variables we examined, the fecundities of Acropora hyacinthus, A. millepora, and A. cytherea on each reef were the three best predictors of recruitment by spawning acroporids, accounting for 49%, 45%, and 36% of the variation in recruitment among reefs, respectively (Table 4, Fig. 5). When the fecundities of each of the three species were entered sequentially into regression models, they collectively accounted for 72% of the variation in total recruitment by spawning acroporids (all species of recruits were combined, because they cannot be identified). This very high explanatory power implies that the fecundities of these three species are broadly representative of the reproductive output of acroporids in general, and that fecundity drives recruitment. A furthe r 16% of the variation was attributable to the remaining variables (sector, 9.4%; reef area, 5.1%; adult abundance, 1.2%; and year, 0.2%), although none of these was statistically significant (Table 4). Only 12% of the variation in recruitment remained unexplained by our model.
The very high correlation between fecundity and recruitment of spawning corals at the scale of individual reefs (Table 4, Fig. 5), would seem to indicate that many of the recruits are of local origin. However, this conclusion is unlikely to be correct, since most of the variation in recruitment and in fecundity occurred at the largest spatial scale, i.e., among sectors rather than reefs. Depending on the year, 55-57% of the variation in recruitment was attributable to sector (see Fig. 4), while only 0-6% occurred among adjacent reefs (nested within sector). The remaining 39-43% of the variation occurred at smaller scales among sites and panels. Similarly, 91% of the variation in fecundity (for A. hyacinthus, A. cytherea; and A. millepora combined) occurred at the sector level in year 1, and 82% in year 2. Furthermore, regressions of fecundity in each sector versus the mean recruitment per sector (identical to Fig. 5, but with adjacent reefs combined) were also highly significant. We conclude therefore that t he high correlation between fecundity and recruitment (Table 4) occurs at the scale of sectors rather than adjacent reefs (which show no significant differences in fecundity, Table 3).
A sector-level correlation between fecundity and recruitment implies that larvae are distributed relatively uniformly among adjacent reefs, but do not undergo larger-scale movements among sectors of the Great Barrier Reef (e.g., transported by the East Australian Current, see Wolanski 1994). Spawning acroporids are generally capable of settling [sim]3-7 d after spawning, although some may remain competent for much longer (Babcock and Heyward 1986, Wilson and Harrison 1998). After mass spawning, coral larvae are often aggregated into buoyant surface slicks that are usually blown or washed off a reef within a day or two, after which they dissipate (e.g., Willis and Oliver 1990, Oliver et al. 1992). Three-dimensional hydrodynamic models predict that a portion of larvae could be retained on natal reefs (e.g., 5% after 10 d; Black et al. 1991), but most are likely to be dispersed among neighboring reefs. Allozyme variation within and among reefs on the Great Barrier Reef points to modest amounts of gene-flow in s pawning corals (Ayre and Hughes 2000) that approach the levels for fishes and other organisms with longer larval durations (e.g., Benzie 1994, Doherty et al. 1995). Brooding corals, in contrast, have a shorter precompetency periods (usually 1-2 d, see review by Harrison and Wallace 1990), and lower rates of gene flow (e.g., Ayre and Dufty 1994, Benzie et al. 1995, Ayre et al. 1997). Consequently, connectivity among reefs is likely to vary greatly among species, depending in part on their respective reproductive strategies and larval durations (e.g., Black et al. 1991, Milicich 1994, Ayre and Hughes 2000).
Our results suggest that the causes of large-scale spatial and temporal variation in fecundity of corals are crucial to understanding subsequent patterns of recruitment. This conclusion mirrors smaller-scale experimental studies that have demonstrated the effects of gamete dilution and aggregative spawning on rates of fertilization (Pennington 1985, Yund 1990, Levitan 1991, Babcock et al. 1992, Coma and Lasker 1997). Of necessity, our result is based on correlating large-scale spatial and temporal patterns, and the possibility remains that sectors and years which had higher fecundities also had higher rates of recruitment, but that there is no causal relationship between the two. However, we cannot formulate an alternative hypothesis that could explain large-scale variation in recruitment, independently of variation in fecundity. Conceivably, spatiotemporal variation in temperature could simultaneously enhance both egg production and reduce the length of larval life (resulting in more recruits). However, the temperature gradient along the Great Barrier Reef is not likely to explain lower fecundities or recruitment in both northern and southern regions (Fig. 3). Similarly, year to year variation in temperature is unlikely to have influenced the substantial temporal patterns of fecundity and recruitment documented in this study, since mean monthly sea surface temperatures were close to normal ([less than]0.5[degrees]C anomalies) in both 1995/1996 and 1996/1997 (IGOSS 1998), throughout the 8-9 mo period which encompassed the gametogenic cycle and brief planktonic phase of spawning acroporids. If large-scale patterns of recruitment by spawners were primarily due to meteorological, climatic, or hydrodynamic variation rather than fecundities, we might expect that sectors or reefs with large numbers of spawning recruits should also have had greater than average recruitment of juvenile brooders. However, this was not the case, at any spatial scale: sectors, reefs, sites, and recruitment panels that had high numbers of s pawners did not necessarily have high numbers of brooders (Hughes et al. 1999).
The most likely cause of spatial and temporal variation in fecundity in this study is "split-spawning" (sensu Willis et al. 1985), i.e., some corals may have lacked eggs when we collected them because they had released their gametes before the major mass spawning in early December. To test this hypothesis, we collected monthly samples of coral tissues in 1996/1997 on a subset of northern and southern reefs (three reefs each in sectors 2 and 5). These data indicate that a portion of adult corals did not release eggs at all in year 2. For example, up to 35% of large A. cytherea in sector 5 did not spawn that summer, nor did they contain immature eggs. Consequently, corals that belong to species which participate in multispecies mass spawning on the GBR may nonetheless miss one or more years, or release gametes in other months (see also Wallace 1985b, Willis et al. 1985). Based on two years of fecundity data on 33 reefs, we tentatively conclude that mass spawning is more synchronized within and among species in the central portions of the Great Barrier Reef than at the extremities (Fig. 3).
Our results indicate that corals which do not participate in mass spawning may be at a selective disadvantage. Our field evidence shows that as the proportion of gravid colonies approaches 100%, there is a disproportionate increase in recruitment (shown by the upward sloping curve in Fig. 5). We postulate that this pattern could arise if higher densities of eggs and sperm lead to greater levels of fertilization (and hence more larvae) after spawning (e.g., Pennington 1985), or if high numbers of larvae lead to satiation of predators (e.g., Westneat and Resing 1988). Both of these phenomena may be instrumental in the evolution of multispecies mass spawning (Harrison and Wallace 1990, Pearse 1990). Indeed, mass spawning of corals on the Great Barrier Reef occurs each year during a period of neap tides when dispersion of gametes prior to fertilization should be minimized, and at night when predation on eggs and larvae should be reduced (Harrison et al. 1984, Babcock et al. 1986).
In conclusion, this study demonstrates that reproductive processes occurring in the benthic phase of marine organisms (i.e., the production of eggs) may have a fundamental impact on the distribution and abundance of recruits. The change in fecundities among years, which affected most of the Great Barrier Reef, indicates the potential impact of large-scale phenomena such as climate change on rates of recruitment and replenishment of coral reefs. Natural and human perturbations are normally measured in terms of their effects on adult abundances and rates of mortality of adults, while less obvious impacts on reproductive biology and regenerative processes are usually ignored (e.g., Richmond 1993, Hughes and Connell 1999). The linkage between benthic and larval stages, as demonstrated here, means that apparently localized changes that affect reproduction at one location may also have important effects on downstream populations. Moreover, small changes in fecundity could result in disproportionately larger change s in recruitment. Currently, it is often assumed that reductions in the size of open populations due to natural events or human impacts are readily reversible because of a virtually inexhaustible supply of recruits (but see, e.g., Karlson and Levitan 1990, Peterson and Summerson 1992, Hughes and Tanner 1999). Our results indicate that large-scale degradation of adult breeding stocks could also impinge on their ability to recover, potentially resulting in recruitment failure in areas that are in most need of replenishment. Understanding the dynamics of coral reefs and other open marine systems requires a better awareness of the two-way links between plank-tonic and benthic life history stages.
We are grateful to 41 graduate student volunteers from James Cook University who provided critical assistance with the large-scale fieldwork in 1995-1997. GIS data on reef area were generously provided by M. L. Puotinen, I. Poiner, the Great Barrier Reef Marine Park Authority, and the Commonwealth Scientific and Industrial Research Organization. We thank D. Ayre, J. Caley, H. Choat, P. Doherty, J.E. Duffy, and two anonymous reviewers for comments on the manuscript. Research was funded by a grant to T. Hughes from the Australian Research was funded by a grant to T. Hughes from the Australian Research Council. This is contribution number 164 of the Coral Group at James Cook University.
(1.) E-mail: email@example.com
(2.) Present address: School of Aquaculture, University of Tasmania, Launceston, P.O. Box 1214, TAS 7250, Australia.
(3.) Present address: SARDI Aquatic Sciences, P.O. Box 120, Henley Beach, SA 5022, Australia.
Ayre, D. J., and S. L. Dufty. 1994. Restricted gene flow in the viviparous coral Seriatopora hystrix on Australia's Great Barrier Reef. Evolution 48:1183-1201.
Ayre, D. J., and T. P. Hughes. 2000. Genotypic diversity and gene flow in brooding and spawning corals along the Great Barrier Reef, Australia. Evolution, in press.
Ayre, D. J., T. P. Hughes, and R. J. Standish. 1997. Genetic differentiation, reproductive mode, and gene flow in the brooding coral Pocillopora damicornis along the Great Barrier Reef, Australia. Marine Ecology Progress Series 159: 175-187.
Babcock, R. C., G. D. Bull, P. L. Harrison, A. J. Heyward, J. K. Oliver, C. C. Wallace, and B. L. Willis. 1986. Synchronous spawnings of 105 scleractinian coral species on the Great Barrier Reef. Marine Biology 90:379-394.
Babcock, R. C., and A. J. Heyward. 1986. Larval development of certain gamete-spawning scleractinian corals. Coral Reefs 5:111-116.
Babcock, R. C., C. N. Mundy, J. Keesing, and J. Oliver. 1992. Predictable and unpredictable spawning events: in situ behavioural data from free-spawning coral reef invertebrates. Invertebrate Reproduction and Development 22:213-228.
Barnes, H. 1956. Balanus balanoides (L.) in the Firth of Clyde: the development and annual variation in the larval population and the causative factors. Journal of Animal Ecology 25:72-84.
Benzie, J. A. H. 1994. Patterns of gene flow in the Great Barrier Reef and Coral Sea. Pages 67-79 in A. R. Beaumont, editor. Genetics and evolution of aquatic organisms. Chapman and Hall, London, UK.
Benzie, J. A. H., A. Haskell, and H. Lehman. 1995. Variation in the genetic composition of coral (Pocillopora damicornis and Acropora palifera) populations from different reef habitats. Marine Biology 121:731-739.
Black, K. P., P. J. Moran, and L. S. Hammond. 1991. Numerical models show coral reefs can be self-seeding. Marine Ecology Progress Series 74:1-11.
Boicourt, W. C. 1982. Estuarine larval retention mechanisms on two scales. Pages 45-56 in V. S. Kennedy, editor. Estuarine comparisons. Proceedings 6th Biennial International Estuarine Research Conference. Academic Press, New York, New York, USA.
Caley, M. J., M. H. Carr, M. A. Hixon, T. P. Hughes, G. P. Jones, and B. A. Menge. 1996. Recruitment and the local dynamics of open marine populations. Annual Reviews of Ecology and Systematics 27:477-498.
Christy, J. H. 1982. Adaptive significance of semilunar cycles of larval release in fiddler crabs (Genus Uca): test of an hypothesis. Biological Bulletin 163:251-263.
Coma, R., and H. R. Lasker. 1997. Effects of spatial distribution and reproductive biology on in situ fertilization rates of a broadcast-spawning invertebrate. Biological Bulletin 193:20-29.
Cushing, D. H. 1986. Climate and fisheries. Academic Press, London, UK.
Doherty, P. J., and D. McB. Williams. 1988. The replenishment of coral reef fish populations. Oceanography and Marine Biology Annual Reviews 26:487-551.
Doherty, P. J., S. Planes, and P. Mather. 1995. Gene flow and larval duration in seven species of fish from the Great Barrier Reef. Ecology 76:2373-2391.
Eckman, J. E. 1996. Closing the larval loop: linking larval biology to the population dynamics of marine benthic invertebrates. Journal of Experimental Marine Biology and Ecology 200:207-237.
Gaines, S., S. Brown, and J. Roughgarden. 1985. Spatial variation in larval concentrations as a cause of spatial variation in settlement for the barnacle, Balonus glandula. Oecologia 67:267-272.
Gaines, S. D., and J. Roughgarden. 1985. Larval settlement rates: a leading determinant of structure in an ecological community of the marine intertidal zone. Proceedings of the National Academy of Science, USA 82:3707-3711.
Grosberg, R. K. 1982. Intertidal zonation of barnacles: the influence of planktonic zonation of larvae on vertical distribution of adults. Ecology 63:894-899.
Grosberg, R. K., and D. R. Levitan. 1992. For adults only? Supply-side ecology and the history of larval biology. Trends in Ecology and Evolution 7:130-133.
Hall, V., and T. P. Hughes. 1996. Reproductive strategies of modular organisms: comparative studies of reef building corals. Ecology 77:950-963.
Harrison, P. L., and C. C. Wallace. 1990. Reproduction, dispersal and recruitment of scleractinian corals. Pages 133-207 in Z. Dubinsky, editor. Ecosystems of the world. Volume 25. Elsevier, New York, New York, USA.
Harrison, P L., R. C. Babcock, G. D. Bull, J. K. Oliver, C. C. Wallace, and B. L. Willis. 1984. Mass-spawning in tropical reef corals. Science 223:1186-1189.
Hastie, T. J., and R. Tibshirani. 1990. Generalized additive models. Chapman & Hall, London, UK.
Hilborn, R., and C. J. Walters. 1992. Quantitative fisheries stock assessment. Chapman & Hall, New York, New York, USA.
Houde, E. D. 1987. Fish early life dynamics and recruitment variability. American Fisheries Society Symposium 2:17-29.
Hughes, T. P. 1984. Population dynamics based on size rather than age: a general model with reef coral example. American Naturalist 123:778-795.
Hughes, T. P. 1990. Recruitment limitation, mortality, and population regulation in open systems: a case study. Ecology 71:12-20.
Hughes, T. P., and J. H. Connell. 1999. Multiple stressors on coral reefs: a long-term perspective. Limnology and Oceanography 44:932-940.
Hughes, T. P., A. H. Baird, E. A. Dinsdale, N. A. Molchaniwskyj, M. S. Pratchett, J. E. Tanner, and B. L. Willis. 1999. Patterns of recruitment and abundance of corals along the Great Barrier Reef. Nature 397:59-63.
IGOSS 1998. The Climate Data Library of the International Research Institute for Climate Predictions/Lamont Doherty Earth Observatory. [less than]http://ingrid.Idgo.columbia.edn/SOURCES/.IGOSS/[greater than].
Karlson, R. H., and D. R. Levitan. 1990. Recruitment limitation in open populations of Diadema antillarum: an evaluation. Oecologia 82:40-44.
Levitan, D. R. 1991. Influence of body size and population density on fertilization success and reproductive output in a free-spawning invertebrate. Biological Bulletin 181:261-268.
Lipcius, R. N., and W. A. Van Engel. 1990. Blue crab population dynamics in Chesapeake Bay: variation in abundance (York River, 1972-1988) and stock-recruit functions. Bulletin of Marine Sciences 46:180-194.
Milicich, M. J. 1994. Dynamic coupling of reef fish replenishment and oceanographic processes. Marine Ecology Progress Series 110:135-144.
Oliver, J. K., B. A. King, B. L. Willis, R. C. Babcock, and E. Wolanski. 1992. Dispersal of coral larvae from a Iagoonal reef. II. Comparisons between model predictions and observed concentrations. Continental Shelf Research 12:873-889.
Pearse, J. P. 1990. Lunar reproductive rhythms in marine invertebrates: maximizing fertilization? Advances in Invertebrate Reproduction 5:311-316.
Pennington, J. T. 1985. The ecology of fertilization of echinoid eggs: the consequences of sperm dilution, adult aggregation, and synchronous spawning. Biological Bulletin 169:417-430.
Peterson, C. H., and H. C. Summerson. 1992. Basin-scale coherence of population dynamics of an exploited marine invertebrate, the bay scallop: implications of recruitment limitation. Marine Ecology Progress Series. 90:257-272.
Reed, D. C., D. L. Laur, and A. W. Ebling. 1988. Variation on algal dispersal and recruitment: the importance of episodic events. Ecological Monographs 58:321-335.
Richmond, R. H. 1993. Coral reefs: present problems and future concerns resulting from anthropogenic disturbance. American Zoologist 33:524-536.
Ricker, W. E. 1954. Stock and recruitment. Journal of the Fisheries Board of Canada 11:559-623.
Roughgarden, J., Y. Iwasa, and C. Baxter. 1985. Demographic theory for an open marine population with space-limited recruitment. Ecology 66:54-67.
Strathmann, R. R. 1993. Hypotheses on the origins of marine larvae. Annual Reviews of Ecology and Systematics 24:89-117.
Thorson, G. 1950. Reproductive and larval ecology of marine bottom invertebrates. Biological Reviews 25:1-45.
Underwood, A. J., and E. J. Denley. 1984. Paradigms, explanations, and generalizations in models for the structure of intertidal communities on rocky shores. Pages 151-180 in D. R. Strong, D. Simberloff, L. G. Abele, and A. B. Thistle, editors. Ecological communities: conceptual issues and the evidence. Princeton University Press, Princeton, New Jersey, USA.
Underwood, A. J., and P. G. Fairweather. 1989. Supply-side ecology and benthic marine ecology. Trends in Ecology and Evolution 4:16-20.
Wallace, C. C. 1985a. Seasonal peaks and annual fluctuations in recruitment of juvenile scleractinian corals. Marine Ecology Progress Series 21:289-298.
Wallace, C. C. 1985b. Reproduction, recruitment and fragmentation in nine sympatric species of the coral genus Acropora. Marine Biology 88:217-233.
Westneat, M. W., and J. M. Resing. 1988. Predation on coral spawn by planktivorous fish. Coral Reefs 7:89-92.
Willis, B. L., R. C. Babcock, P. L. Harrison, J. K. Oliver, and C. C. Wallace. 1985. Patterns in the mass spawning of corals on the Great Barrier Reef from 1981 to 1984. Proceedings of the 5th International Coral Reef Congress, Tahiti, 4:343-348.
Willis, B. L., and J. K. Oliver. 1990. Direct tracking of coral larvae: implications for dispersal studies of planktonic larvae in topographically complex environments. Ophelia 32:145-162.
Wilson, J. R., and P. L. Harrison. 1998. Settlement-competency periods of larvae of three species of scleractinian corals. Marine Biology 131:339-345.
Wolanski, E. 1994. Physical oceanographic processes of the Great Barrier Reef. CRC Press, Boca Raton, Florida, USA.
Yund, P. O. 1990. An in situ measurement of sperm dispersal in a colonial marine hydroid. Journal of Experimental Zoology 253:102-106.
Nested analysis of variance of adult acroporid densities (number of colonies per 10-m transect). Percent- age of Source of variation df MS F ratio P variation Sector 4 6.76 8.377 0.003 19 Reef (sector) 10 0.81 3.185 0.004 5 Site (sector X reef) 45 0.25 2.794 0.000 6 Residual 1126 0.09 69 Notes: Reefs are nested within sectors, and sites within reefs (see Fig. 1 for locations of sectors and reefs). Data were log(x + 1) transformed. Note that most of the variation occurred at the smallest scale, among replicate transects (see also Fig. 2). Three-way analysis of variance of acroporid fecundities, with reefs as replicats and percentage of individuals with eggs present as the response. Source of Variation df MS F ratio P Sector 4 9376.3 4.545 0.056 Species 2 1419.7 2.272 0.165 Year 1 12952.3 7.714 0.050 Secor X species 8 628.9 2.427 0.115 Species X year 2 834.1 3.229 0.093 Sector X year 4 1693.2 6.534 0.012 Secotr X species X year 8 259.2 1.415 0.206 Residual 69 183.2 Notes: Data were not transformed becuase the ANOVA assumptions were met. See also Fig. 3. Two-way nested analysis of variance on density of acroporid recruits. Source of variation df MS F ratio P Year 1 69.273 5.525 0.780 Sector 4 48.542 3.623 0.120 Sector x year 4 13.364 10.321 0.000 Reef(sector x year) 20 1.433 1.801 0.033 Site(sector x year x reef) 85 0.826 8.958 0.000 Residual 1018 0.092 Notes: Site is nested within reef and reef within sector (see Fig. 1). Data were log (x + 1) transformed. Year represents 1995/ 1996 and 1996/1997. See also Fig. 4. (A) Analysis of the relationship between recruitment of spawning acroporids (all species combined) and fecundity of Acropora hyacinthus, A. millepora, and A. Cytherea (see Fig. 3). Fecundity of each of the three species was first examined separately, and then entered sequentially (A. hy. + A. m. + A. c.) into multiple regression models. Once fecundity of each species was independently accounted for, we used a hierarchy of regression models (B) which successively incorporate additional explantory variables (sector, area of each reef, cover of spawning acroporids, and year). A. Percentage variation Model Residual deviance df explained Null 78 477 32 A. hyacinthus fecundity 39 706 28 49 A. millepora fecundity 43 129 28 45 A. cytherea fecundity 50 168 28 36 Fecundity (A. h. + A. m. + A. c.) 21 976 20 72 A. Model P Null A. hyacinthus fecundity [less than]0.001 A. millepora fecundity [less than]0.01 A. cytherea fecundity [less than]0.01 Fecundity (A. h. + A. m. + A. c.) [less than]0.001 B. Change in percentage variation Model Residual deviance df explained F P Fecundity 21 976 20 Fecundity + sector 14 590 4 9.4 1.52 0.24 + reef area 10 590 4 5.1 1.03 0.43 + adult abundance 9 681 4 1.2 0.17 0.95 + year 9 480 1 0.2 0.13 0.73 Note: The null model assumes that none of the variation in recruitment is explained by any of the variables were examined.
|Printer friendly Cite/link Email Feedback|
|Author:||HUGHES, T. P.; BAIRD, A. H.; DINSDALE, E. A.; TANNER, J. E.; MOLTSCHANIWSKYJ, N. A.; PRATCHETT, M. S|
|Article Type:||Statistical Data Included|
|Date:||Aug 1, 2000|
|Previous Article:||ECOLOGICAL PROCESSES UNDERLYING ONTOGENETIC HABITAT SHIFTS IN A CORAL REEF FISH.|
|Next Article:||RECRUITMENT FAILURE, LIFE HISTORIES, AND LONG-TERM DECLINE OF CARIBBEAN CORALS.|