Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
00001.tif
(USC Thesis Other)
00001.tif
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
INFORMATION TO USERS This manuscript has been reproduced from the microfilm master. UMI films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter free, while others may be from any type of computer printer. The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely affect reproduction. In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion. Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand comer and continuing from left to right in equal sections with small overlaps. Each original is also photographed in one exposure and is included in reduced form at the back of the book. Photographs included in the original manuscript have been reproduced xerographically in this copy. Higher quality 6” x 9” black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order. UMI A Bell & Howell Information Company 300 North Zeeb Road, Ann Arbor MI 48106-1346 USA 313/761-4700 800/521-0600 ANALYSIS AND SIMULATION OF SPATIAL DISTRIBUTIONS OF CORALS ON UNCONSOLIDATED REEF SUBSTRATES by Wilfredo Roehl Ybanez Licuanan A Dissertation Presented to the FACULTY OF THE GRADUATE SCHOOL UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements for the Degree DOCTOR OF PHILOSOPHY (Biology) May 1995 UMI Number: 9621622 UMI Microform 9621622 Copyright 1996, by UMI Company. All rights reserved. This microform edition is protected against unauthorized copying under Title 17, United States Code. UMI 300 North Zeeb Road Ann Arbor, MI 48103 UNIVERSITY OF SOUTHERN CALIFORNIA THE GRADUATE SCHOOL UNIVERSITY PARK LOS ANGELES, CALIFORNIA 90007 This dissertation, written by W ilfredo Roehl Ybanez Licuanan under the direction of h..i5....... Dissertation Committee, and approved by all its members, has been presented to and accepted b y The Graduate School, in partial fulfillment of re quirements for the degree of DOCTOR OF PHILOSOPHY Dean of Graduate Studies D a te .. A j)ri.0.81 ..m:>. DISSERTATION COMMITTEE !S..*.... •— v ^ 1 1 .......... 7 Chairperson This work is dedicated to the late Dean Francisco Nemenzo, D.Sc., University of the Philippines, a most patient and inspiring mentor. I wish we had more time to learn from you. Acknowledgments As the culmination of years of research on marine organisms, this work has benefited from interactions with countless mentors (both within and outside the academic community) who have provided support, as well as challenges and adversity. I thank Drs. Perry Alino, David Green, Ang Put Ong, John McManus, and Saul Saila for their ideas, suggestions and support. Prof. Edgardo Gomez, Drs. Helen Yap, and Mike Fortes have made additional facilities and equipment available. Benny Campos, Menchie Ablan, Nancy Bermas, Marl Villanoy, Hildie Nacorda, Jong Cuaresma, and Mike Atrigenio made various aspects of the field and laboratory work lighter and have been very generous with their time and views. To you all, heartfelt thanks. Special thanks to my adviser, Prof. Jerry Bakus, for his insights, patience, and support. This work could not have been completed without the support and inspiration provided my wife, Suzanne, and my parents, Dr. Lino (deceased) and Minda Licuanan. I pray that I may be able to repay them. Finally, I thank my daughter Ardea, whose photos have served as a constant reminder of the many reasons I should finish this work and go home. ii TABLE OF CONTENTS Page Introduction .................................................................................................... 1 Review of Literature .................................................................................... 8 Coral competition, diversity and distributions ......................................... 8 Reproductive processes and spatial distributions..................................... 14 Disturbances and spatial distributions .................................................... 16 General comments .................................................................................. 18 Demographic and spatial models of corals .............................................. 18 Population projection-matrix models ................................................ 18 Cellular automaton models ............................................................... 21 The study site and study organisms ........................................................ 23 The Field Experiments .................................................................................. 27 Materials and Methods ............................................................................... 27 Cross-transplantation studies ............................................................... 31 Coral settlement studies ....................................................................... 34 Physical factor measurements ................................................................ 37 Analyses ............................................................................................... 40 Results ........................................................................................................ 40 Cross-transplantation studies: Branch growth rates and related observations ....................................................................................... 40 Cross-transplantation studies: Colony size and survival ........................ 44 Coral settlement studies: Recruit distributions ..................................... 49 Coral settlement studies: Recruit transplantation .................................. 52 Physical Measurements ........................................................................ 55 Discussion .................................................................................................. 59 The Computer Simulations .......................................................................... 75 Material and Methods ................................................................................. 76 Field collection of parameter estimates ................................................. 79 Construction of transition matrices for Acropora subglabra and Anacroporapuertogalerae .............................................................. 81 Evaluating contributions of asexual and sexual reproduction ................ 85 iii Adjustments of size class intervals to a common definition for Acropora subglabra and Anacropora puertogalerae....................... 86 The cellular automaton model................................................................. 86 Results ........................................................................................................ 92 Population projection matrices: Growth, sensitivity, and elasticity ...... 92 Population projection matrices: Evaluating modes of reproduction with size-structure...................................................... 100 Cellular automata: Competition and coral spatial distributions ............. 102 Cellular automata: Recruitment patterns and coral spatial distributions .................................................................................... 102 Cellular automata: Disturbances and coral spatial distributions 1 IS Discussion .................................................................................................. 124 General Conclusions and Future W o rk ....................................................... 131 Literature Cited ............................................................................................. 134 Appendices ..................................................................................................... 148 Appendix 1: Comparisons of branch diameters and spacing in Anacropora puertogalerae ................................................................... 148 Appendix 2: Competitive pairing experiments ........................................... 152 Appendix 3: Density manipulation experiments......................................... 155 Appendix 4: The alizarin experiments ....................................................... 159 Appendix 5: Distribution of Acropora subglabra and Anacropora puertogalerae rubble ............................................................................ 162 Appendix 6: Predation of recruits on settlement tiles .................................. 164 Appendix 7: Comparison of projection matrix and cellular automaton projections ............................................................................................ 167 Appendix 8: Source code of the cellular automaton model ...................... 168 Appendix 9: Plates .................................................................................... 194 LIST OF TABLES Table Description Number 1 Summary of the major components of the field studies. 2 Descriptions of the experimental blocks (=sites) used in the coral transplantation and recruitment experiments. 3 Distribution of taxa among treatment patches in the First and Second Plateaus. 4 Summary of the rules used in the various cellular automaton simulations. 5 Upper limits of size classes used in the projection matrix and cellular automata projections. 6 Transition matrices used in the projection matrix models. 7 Transition matrices used in the projection matrix models and their population growth rates (X). 7-1 Correlation coefficients between abundances projected by matrix models and by cellular automata (Appendix 7). 7-2 Size-class distributions for Acropora subglabra and Anacropora puertogalerae from both models (Appendix 7). Page 28 30 53 78 82 84 93 171 171 v LIST OF FIGURES Figure Description Page Number 1 Diagrammatic representation of the elements of a projection 20 matrix 2 Map of the main study sites in Puerto Galera Bay, Oriental 24 Mindoro. 3 Map of the experimental blocks used in the First and Second 29 Plateaus. 4 Diagram of an experimental block and the treatments used in the 32 cross-transplantation experiments. 5 Diagram of a settlement tile rack and the layout of the 35 recruitment monitoring for each experimental block. 6 Layout of treatments of the recruit transplantation experiment 38 for each experimental block. 7 Comparison of growth rates of branches of Anacropora 41 puertogalerae transplants and controls. 8 Branch growth rates of Anacropora puertogalerae averaged 42 over the duration of the study. 9 Comparison of growth rates of branches of Acropora subglabra 45 transplants and controls. 10 Branch growth rates of Acropora subglabra averaged over the 46 duration of the study. 11 Comparison of colony sizes of Acropora subglabra transplants 48 and controls. 12 Settlement rates of scleractinian corals averaged over the five 50 samplings in one year for each treatment or site. vi 13 Settlement rates of scleractinian corals in the various treatments and sites over the five samplings in one year. 14 Counts of live scleractinian coral recruits among treatments in the recruit transplantation experiment. 15 Mean sediment load in the three treatment patches in Block A, First Plateau. 16 Average clod card diffusion factors (i.e., weight loss in the field relative to weight loss in enclosed containers) in the three treatment areas. 17 Water movement rates (as measured by rhodamine dye patches) among the different treatment patches in Block A. 18 Secchi depths between the First and Second Plateaus. 19 Average water temperatures in the study area. 20 Grain-size analyses of sediments collected from the three treatment patches in Block D (10 m depth) and Block E (7 m depth), both in the Second Plateau. 21 Sensitivity of Acropora subglabra transition matrix (ignoring reproduction) to small changes in transition probabilities. 22 Sensitivity of Anacropora puertogalerae transition matrix (ignoring reproduction) to small changes in transition probabilities. 23 Sensitivity of Acropora subglabra transition matrix (including fragmentation or asexual reproduction) to small changes in transition probabilities. 24 Elasticity, or proportional changes in population growth rate of Anacropora puertogalerae in response to a 10% increase in transition matrix probabilities. 51 54 56 57 58 60 61 69 94 95 97 98 vii 25 26 27 28 29 30 31 32 33 34 35 36 Elasticity, or proportional changes in population growth rate of 99 Acropora subglabra in response to a 10% increase in transition matrix probabilities (excluding observed fragmentation or asexual reproduction). Observed and projected size-structures of Acropora subglabra 101 and Anacropora puertogalerae if they reproduced mainly by asexual means (fragmentation of SC3 to produce SC2) or sexual means (SC3 producing planulae that start as SCI colonies). Initial reef configuration with random distribution and high 103 density of corals (500 individuals per size class). Initial reef configuration with random distribution and a low 104 density of corals (100 individuals per size class). Initial reef configuration with corals distributed in several bands 105 or zones. Initial reef configuration with corals distributed in a few bands 106 or zones. Segregation indices for various competition simulations with 107 random recruitment to SC2. Map of model reef after 100 years of the competition simulation 108 with random settlement to SC2. Map of model reef after 5 years of the competition simulation 109 with random settlement to SC2. Segregation indices for various recruitment simulations with 110 random settlement or aggregated settlement to either SCI (sexual reproduction) or SC2 (asexual reproduction) and with no competition. Map of model reef after 100 years of the recruitment simulation 111 with random settlement to SCI and no competition. Map of model reef after 100 years of the recruitment simulation 112 with aggregated settlement to SCI and no competition. viii 37 Map of model reef after 100 years of the recruitment simulation 113 with random settlement to SC2 and no competition. 3 8 Map of model reef after 100 years of the recruitment simulation 114 with aggregated settlement to SC2 and no competition. 39 Segregation indices for two recruitment simulations (without 116 competition) wherein only Anacropora puertogalerae exhibited aggregated settlement to SC2, while Acropora subglabra settled randomly (upper graph), or vice versa (lower graph). 40 Map of model reef after 100 years of the recruitment simulation 117 with aggregated settlement to SC2 (for Anacropora puertogalerae only) and no competition. 41 Map of model reef after 100 years of the recruitment simulation 118 with aggregated settlement to SC2 (for Acropora subglabra only) and no competition. 42 Segregation indices for two disturbance simulations (without 119 competition nor aggregated settlement). 43 Map of model reef after 100 years of the disturbance simulation 120 with random settlement to SC2, no competition and 10 random 1-cell disturbances per year. 44 Map of model reef after 50 years of the disturbance simulation 121 with random settlement to SC2, no competition and 200 random 1-cell disturbances per year. 45 Map of model reef after 10 years of the disturbance simulation 122 with aggregated settlement to SC2, no competition and 500 random 1-cell disturbances per year in columns 20-40. 46 Map of model reef after 100 years of the simulation wherein 123 corals that are surrounded by 7 or 8 empty cells die due to exposure. ix 1-1 Diagram showing the positions of the measurements made in 150 Anacropora puertogalerae colonies to compare branch diameters and distances (Appendix 1). 1 -2 Comparison of branch diameters of Anacropora puertogalerae 151 release colonies and control colonies at the edge of the patch and 0.5 m interior to the edge (Appendix 1). 3 -1 Diagram of two of three treatments in the Acropora subglabra 156 density manipulation experiment (Appendix 3). 3-2 Comparison of branch growth of Acropora subglabra in 158 quadrats of different colony densities (Appendix 3). 4-1 Branch growth rates of Anacropora puertogalerae treatment 161 colonies measured using Alizarin Red S marks incorporated in the skeleton (Appendix 4). 5-1 Distribution of live Anacropora puertogalerae and Acropora 163 subglabra patches, and their respective rubble. 6-1 Comparison of coral settlement rates on outer, exposed tiles and 166 on inner, sheltered tiles of racks in Anacropora puertogalerae patches, Acropora subglabra patches, and areas devoid of corals (release area). 7-1 Comparison of projected abundances of Acropora subglabra by 169 cellular automata and projection matrix models with low and high initial densities (Appendix 7). 7-2 Comparison of projected abundances of Anacropora 170 puertogalerae by cellular automata and projection matrix models with low and high initial densities (Appendix 7). x Abstract A series of field experiments and computer simulations was initiated to investigate the role of competition, recruitment patterns, and disturbances in the formation and maintenance of monospecific patches (3 to 30 m in diameter) of Anacropora puertogalerae and Acropora subglabra growing on unconsolidated sediments in Puerto Galera Bay, Philippines. Recruitment rates of scleractinian corals (mainly pocilloporids) were shown to be lower within A. puertogalerae and, to a lesser extent, A. subglabra patches compared to adjacent areas devoid of corals. Reduced growth of branches o f A. puertogalerae transplants to barren areas outside the patch, and to patches of A. subglabra, was also found. Transplants of A. subglabra showed no differences in branch growth, but the transplants to A. puertogalerae patches did exhibit significantly higher mortality rates. This occurred despite the dominance of A. subglabra over A puertogalerae in digestive interactions as revealed by pairing experiments. These results may be related to the significantly lower water flow and sedimentation rates existing within A. puertogalerae patches, which in turn may be caused by hydrodynamic changes created by the coral aggregations themselves. Evidences for this include a decrease in branch diameters of A. puertogalerae colonies transplanted from the patch interior to areas outside the patch and the lack of any differences in recruitment patterns if settlement tiles were conditioned outside A puertogalerae patches and subsequently transplanted to within the patch. Conditioning of the settlement tiles appears to have been delayed by the lower water flow within the coral patches. Computer simulations using probabilistic cellular automata show competition is unlikely to lead to the formation of the monospecific patches of the two species, although interspecific competition is required if these pure stands are to be maintained after the patches are formed. Evidence from projection matrix simulations and genetic studies (published elsewhere) indicate that A. puertogalerae and A. subglabra reproduce mainly by asexual means (fragmentation). Using the cellular automaton model, it was shown that the limited mobility and higher survival rates of these asexually produced propagules lead to the formation of the species-patches described. Simulations of the effects of physical disturbances on large fragments and colonies of these corals show that disturbances of various magnitudes are unlikely to lead to any significant patch formation unless these disturbances are strongly selective and localized. Introduction A criticism of contemporary ecology is its limited predictive ability. This, according to Peters (1991), can no longer be justified by the relative youth of the field, nor by the complexity of its subject matter. Predictive ability in ecology is made more important by the even bigger need to contend with increasingly difficult environmental problems (often human-induced) confronting most ecosystems. Ecologists must be able to make more definitive statements about ecosystem responses even in the face of insufficient information about the system components. This requires that ecologists focus on the analyses of empirically identified patterns, with the aim of maximizing predictive power and practicality rather than the creation or validation of global models (Peters 1991). However, predictive ability will be limited by the degree to which the system is understood, the same way that predictive ability determines the level of control one will have over a system (Bradbury and others 1983). So, while there is the need for phenomenological approaches that allow for some progress in ecology, reductionist mechanistic studies still need be done. Ecologists must strike a balance between treating ecological entities as black boxes and being concerned with pulling these boxes apart to reveal their workings. In addition, novel approaches must be developed to supplement traditional methods. New paths need to be cut while old routes of thinking are reevaluated (see Price and others 1984). The realization that most ecological concepts (especially model 1 formulations based on the Lotka-Volterra equations) have relied on the assumption that individual organisms are homogeneously distributed across space has led to the development of landscape ecology with its explicit treatment of heterogeneity in organismal distributions and abundances, and the effects of scale of observation (see Forman and Godron 1986, Hatcher and others 1987, Turner 1987, Turner and Gardner 1991, Kolasa and Pickett 1991). Related to this is the application of hierarchy theory (Allen and Starr 1982, O'Neill and others 1986, 1991, Allen and Hoekstra 1992) and fractals (Burrough 1981, Sugihara and May 1990) to ecology. Novelty and sophistication must not be confused with scientific rigor, however, or else the same lessons will be relearned, but more expensively and with yet another layer of unnecessary terminology. As chaos theory cautions (see May 1974, also Gleick 1987), the extent to which we can understand and predict nature has real limits, hence the need for stable empirical footing as ecologists try to find these limits. Finally, ecologists must also recognize that, like most ecological conceptions, pure or basic ecology and applied ecology do not form a perfect dichotomy as in the branches of a tree, but define nodes on a continuum like the veins of a dicot leaf. Both have much to contribute to, and gain from each other. For example, resource management efforts can be designed to serve both as experiments for some ecologists, and as adaptive management schemes to others. One preeminent tenet of community ecology is that interactions between and within component species of a community play a major role in the distribution and relative abundances of organisms. These interactions include intra- and interspecific competition, herbivory, and predation; all of which modify, or are modified by environmental factors. The effects of competition on ecological communities have received considerable attention from western science (see Keddy 1989 for a general review). However, there has been surprisingly little progress in the development of competition theory (Keddy 1989). One difficulty in competition research is the limited generality of its basic principles. This stems from the limited number of species studied per taxonomic group, and the tendency to focus on certain localities and taxa. From random surveys of articles in the journal Ecology, Keddy (1989) found that about 80 percent of studies published on competition were done in the mainland U.S. Most of the research surveyed was on vertebrates, which not only comprise a small percentage of the world's species, but possess several unique characteristics that make generalizations to other taxa untenable. One could only speculate on the form modem ecology would have taken if most major research universities were located in the high-diversity tropics. Perhaps an even more important issue is the tendency to emphasize competition over other forms of inter- and intra-specific interactions. Keddy (1989) attributes this to the ffee-market orientation of the West, where modem science is most developed. Many also regard competition theory as a natural extension of the theory of evolution. On the other hand, Karlson and Hurd (1993) believes this bias is merely the result of the tendency to study assemblages of organisms of the same trophic level. Whatever the 3 motivation, the emphasis on competition has been at the expense o f investigations of other processes such as mutualistic interactions and recruitment, which may be equally important in determining the distribution and relative abundances o f organisms. Fortunately this trend is beginning to be reversed. For instance, see Sale (1988) for some contributions of coral reef science to ecology. Coral reef ecology is only about 40 years old, and even within this field the insights presented are incomplete and still skewed. Since the distance of these systems was so far from most important research universities, most early conceptions were based on short-term (relative to most reef organisms' life-spans) expeditions. Even today, most major reef research centers and field stations are located in areas peripheral to the major Indo-Pacific center of diversity (Potts 1983). The early studies on eastern Pacific atolls have led to the conception of ideas such as leeward and windward differentiation in offshore reefs, and coralline algae-paved reef crests that are rarely seen in the monsoon dominated reefs of the western Pacific (Licuanan and Gomez 1988) and the Indian Ocean (Stoddart 1969). Moreover, most reefs are in continental shelf areas, and therefore are subject to terrestrial factors like runoff (see Potts 1983) unlike eastern Pacific oceanic reefs. Today, most reef research is conducted by scientists who have greater access to reef organisms. It is still important, however, to consider the peculiarities of the reefs on which scientists base their experiences when interpreting their findings. For example, reefs in the Red Sea are obviously affected by the conditions in the nearby deserts, 4 whereas reefs in Hawaii are less diverse but can be as large and structurally complicated as western Pacific reefs. Biogeographic differences in the Caribbean are also emphasized by differences in relative importance of taxa compared to the Indo-Pacific, e.g., the gorgonians and sponges of the former are more abundant, and grazers are mainly echinoids and large mollusks rather than fishes (although there is evidence that this may be partly due to overfishing, Hughes 1994). Australian, and Australia-based researchers have become a dominant force in reef research, but most of their work has focused on the Great Barrier Reef, which is mainly composed of offshore reefs and islands. Although these reefs share strong affinities with reefs in Southeast Asia, important distinctions remain (Licuanan and Gomez 1988). The most important difference is the fact that SE Asian reefs are mainly fringing reefs and non-reefal coral communities (McManus 1988) that are very much subject to terrestrial and human impacts. The studies presented here focus on fringing inshore reefs in the northwestern Philippines. Philippine reefs are among the richest and most diverse in the world. Nearly S O O species and varieties of scleractinian corals alone have been reported (Nemenzo 1981), with several remaining undescribed (Veron and Hodgson 1989). Indo-Pacific floral and fauna! range distributions often have the Philippines at their centers, and the Austro-Asian region is recognized for its unusually high species richness (McManus 1985), supporting about half of the world's reefs (Potts 1985). Coral reefs are essential for the survival of most Pacific island ecosystems. An estimated 11-29% of Philippine 5 total fisheries production is contributed by reefs (Carpenter 1977, Murdy and Ferraris 1980, Alcala 1981), with the worldwide figure at 12% (Munro and Williams 1985). Most reefs in the region are in a rapidly deteriorating state, principally because of increased human use (Yap and Gomez 1985a). About 32-38% of Philippine reefs are in poor to fair condition (0-24.9% to 25-49.9% live coral cover; Yap and Gomez 1985a, 1985b, Alcala and others 1987). The validity of these numbers is already in question because of methodological constraints and the age of the data set (Gomez and others 1994). Because of this imminent ecological crisis, major reef research centers in the region have turned their attention to vital topics such as ecosystem rehabilitation and replenishment, and reserve management; which are applied aspects of community ecology. Often these projects are in the form of regional cooperative programs by groups such as the Association of South East Asian Nations (see Alcala and others 1991). Critical to these efforts, however, is the ability to evaluate the relative importance and interactions o f abiotic and biotic factors on community composition, and to project successional trajectories that a result of increased human use (Yap and Gomez 1985a). About 32-38% of Philippine reefs are in poor to fair condition (0-24.9% to 25-49.9% live coral cover; Yap and Gomez 1985a, 1985b, Alcala and others 1987). The validity of these numbers is already in question because of methodological constraints and the age of the data set (Gomez and others 1994). Because of this imminent ecological crisis, major reef research centers in the region have turned their attention to vital topics such as ecosystem rehabilitation and replenishment, and reserve management; 6 which are applied aspects of community ecology. Often these projects are in the form of regional cooperative programs by groups such as the Association of South East Asian Nations (see Alcala and others 1991). Critical to these efforts, however, is the ability to evaluate the relative importance and interactions of abiotic and biotic factors on community composition, and to project successional trajectories that a given coral community can take in response to perturbations. This introduction and background serve as context for this present research -- why certain questions were asked, why certain species and habitats were chosen, and why certain approaches were taken. These pages define the outline of the work and why it is significant, relevant, and timely ~ the hallmarks of a good dissertation research (Friss 1994), while remaining engaging to the researcher. My research has focused on spatial arrangement of organisms, and the reasons for such distributions, as well as their consequences. This work has struck a balance between basic and applied ecology, and has included mechanistic and phenomenological approaches. It has a firm empirical footing but also an exploration of theory. It has used a familiar and important system that was studied using new, unfamiliar methods and approaches. The immediate goals set were achievable in the given period, while the research raised many additional questions for future work. This research focuses on the role of interspecific interactions, recruitment and reproductive processes, on the patchy arrangement and relative abundances of two scleractinian corals, Anacropora puertogalerae (Nemenzo 1964; referred to from now 7 on as AP) and Acropora subglabra [(Brook 1891); see Veron and Wallace 1984 for descriptions of both species; referred to here as AS] in a small embayment in the Philippines. The research has the following objectives: 1 . To examine the role of interspecific competition, recruit supply and settlement, and disturbances in determining the small-scale (within- zone) distributions of AP and AS colonies, specifically their tendency to form large monospecific thickets; 2. To develop demographic and spatial models of the coral species studied to examine the effects of interactions of the above factors on the spatial distributions of the coral colonies, permitting a series of "what-if1 thought experiments. Review of Literature Hutchinson (1953) categorized spatial pattern of organisms as vectorial (caused by external environmental forces), reproductive (determined by genetic continuity), coactive (from competition among individuals of different and of the same species), social (determined by signaling of some kind) and stochastic pattern (i.e., caused by random events such as disturbances). Much of the present work revolves around the evaluation of the relative importance of these factors in determining the distributions of AP and AS. * Coral competition, diversity and distributions The discovery of consistency in the victors in direct competitive interactions (sensu Lang and Chomesky 1990) among scleractinian corals (Lang 1973) has led to the realization that competition for space in a reef is not just a scramble for space, but 8 involves direct interference between coral species. Several such studies, aimed at describing digestive hierarchies in different areas (using both direct observations and experiments), have since followed (Sheppard 1979, Cope 1981, Bak and others 1982, Logan 1984, Dai 1990, Tanner 1992; see Lang and Chomesky 1990 for a comprehensive review of coral competition). These hierarchies have been postulated to contribute to the maintenance of diversity in reef habitats, especially in cases wherein intransitive networks exist (e.g., species A being competitively superior to species B, B being superior to C, but C being superior to A) since no single species, or small groups thereof, are able to monopolize the available resources (defined for most benthic habitats as space; Jackson and Buss 1975, Connell 1978, Buss and Jackson 1979, Tanner 1992; see also the simulations of Karlson and Jackson 1981, and Karlson and Buss 1984). However, subsequent work has shown that outcomes of such encounters depend on several factors such as the relative positions of the corals (Bak and others 1982), the presence of intervening organisms (Chomesky 1983, Rinkevich and others 1992), nutrient and predation levels (Alino and others 1992 with soft corals), and even seasonal weather conditions (Lang and Chomesky 1990), and not just the identity of interacting species. Species can also adjust their competitive responses leading to reversals (Wellington 1980, Chomesky 1989). A victor in one encounter may yield space to the same competitor in a subsequent encounter. Even if the outcomes of competitive encounters are predictable, diversity patterns depend greatly on the arrangement of interacting organisms, as Silvertown and others 9 (1992) show for the terrestrial environment. It is thus difficult to consider competition in isolation of the factors and components in the organisms' immediate "neighborhood," e.g., local substrate conditions, neighboring taxa, predators and grazers, and disturbances (see Karlson and Hurd 1993). Hence the need to consider distributions at the level of individual colonies and not just assemblages, and at the appropriate temporal and spatial scales (see additional comments in the disturbance effects review, page 16). Failure to do so results in much confusion and little progress, as efforts to demonstrate a link between competition and coral distributions illustrate. Observations showing redirection of growth (Romano 1990) suggest that competitive interactions could also affect distributions of corals, since subordinate species will grow toward space not controlled by competitive superiors. However, the effect of dominance hierarchies and networks (Lang 1973, Cope 1981, Dai 1990) on spatial patterns has been difficult to demonstrate, especially in the species-rich reefs of the western Pacific. Temporal, and possibly, spatial variability in the competitive hierarchies (Chomesky 1989, Lang and Chomesky 1990) as well as some taxonomic uncertainties within Scleractinia, complicate the situation. Sheppard (1979) showed that competitively aggressive species dominate certain zones in the Chagos Atolls, and Chadwick (1991) demonstrated this among three species of ahermatypic (non reef-building) corals and a corallimorph off California. Bradbury and Young (1983), on the other hand, found that only 5 of 545 records of coral-coral pairs sampled (at 10 cm intervals) around Heron Island, Australia were not randomly mingled with each other and that the contribution of competition to community structure is weak at best. A subsequent report (Reichelt and Bradbury 1984) showed that there is little propagation of patterns, ascribed to competition, to higher spatial scales. On the other hand, Green and others (1987) showed that pattern does emerge at larger spatial scales despite the "anarchy" at smaller scales. Using another method and at a finer spatial scale, Licuanan and Bakus (1992) found consistent patterning (in terms of repeating sequences along transects) of corals, soft corals and algae within parts of a coral community in the Philippines. They speculated that the anisotropic nature of the patterns is due to physical factors influencing the outcome of these interactions in certain directions only. The apparent lack of consensus on the role of interactions in determining coral distributions is due to three factors. One factor is methodological /statistical difficulties. Green and others (1987) indicated that community-level studies often use multivariate pattern extraction techniques that suppress variability in the data, whereas the ANOVA (analysis of variance)- oriented approaches used in population-level studies emphasize this unpredictability. It may also be that standard statistical approaches are not applicable when analyzing what may be events that are not independent, such as the settlement of corals at certain positions and in certain sequences. Reichelt and Bradbury (1984), for instance, pointed out the issue of double counting of occurrences leading to inflated error terms. Licuanan and Bakus (1992) believed their randomization approach overcomes this, but had to use a test statistic that does not allow for the confirmation of whether a 1 1 given taxon or group is causing the departure from randomness. On the practical side, isolation of biological from environmental (abiotic) effects on the outcome of competitive interactions is very difficult (see Bak and others 1982) and limits the kinds of conclusions one can make. New approaches are clearly required to deal with this limitation. The second factor leading to disagreement is the uneven appreciation of the scale specificity of certain phenomena. For example, if competition is important, competitive subordinates can be "squeezed out" from areas where dominants are superior. Thus, depending on the degree to which this has occurred unperturbed over a given time period, distributions may vary from random to completely patterned, and these distributions may occur at a scale of a few meters (e.g., patches in a reef zone), to whole zones (Sheppard 1979), or to whole reefs. Though broad scale zonations of reef benthos and the factors controlling these are relatively well understood (Sheppard 1982, Done 1983), distributions within zones remain enigmatic. Digestive interactions at the level of two co-occurring species may reflect on compositions and distributions at the level of reef zones but not necessarily at the level of the larger reef community. The degree to which these may be detected will vary. For example, the effect of a disturbance will decrease over time from the event that reduced abundances and made space non-limiting to recruitment (Connell 1978, Green and others 1987). Thus, the findings of studies must always be considered appropriate only for the spatial scales and temporal scales examined. 12 Regarding the temporal scale, note that the studies described were limited to "snapshots" at a given time, with no assurance that areas compared had competitive processes operating at equivalent lengths of time. In addition, it is unclear whether time should be considered in an absolute scale, or relative to generation times of participating species (as in Connell and Sousa 1983). Unfortunately, the temporal (e.g., relative to the organisms' life-spans) and spatial components (partly a function of both distance and of level of organization examined, i.e., whether population or community) are not orthogonal (= independent) and preclude generalizations across the continuum that these define, even for systems of comparable species diversity. Definitive answers to questions of relative importance of processes cannot be given without an understanding of the interactions between the spatial and temporal scales. Thirdly, interactions studied to date mainly considered interactions between adjacent colonies in isolation from all other neighboring benthos. This is unrealistic since the composition of the community as a whole, and coral-associated organisms in particular, can affect the outcome of any interaction (Lang and Chomesky 1990). Since faunal compositions will differ from zone to zone, and reef to reef, it will be more difficult to find consistent patterns at different scales. Unfortunately studies of diffuse competition in reef corals are lacking (Lang and Chomesky 1990). This is mainly due to the lack of more rigorous sampling designs and theoretical constructs needed to comprehend such processes in high diversity habitats (but see Keddy 1989). Consideration must thus be given to the number and kinds of neighboring species as 13 another scaling criterion, apart from spatial and temporal considerations. Ecologists need to see things from an individual coral's point of view. And this is a big challenge. Even if these three factors are discounted, it must be remembered that all the studies reviewed are just interpretations of patterns and do not directly test the link of competitive interactions and distributions. ♦ Reproductive processes and spatial distributions Aggregated coral distributions have been noted in both the Pacific (Grassle 1973) and the Caribbean (Dana 1976, Lewis 1970) and among the explanations advanced is that these distributions may be the result of reproductive events leading to aggregated juvenile, and eventually, adult distributions. Carlon and Olson (1993) suggested three ways by which larval or juvenile behavior can lead to such aggregated patterns: a) gregarious settlement - where larvae preferentially settle near conspecific adults; b) selection of microhabitats - where a limited number of suitable substrates attract larvae; c) limited dispersal - juveniles are incapable of dispersing far and thus settle near parent colonies. Differences among these three may not always be clear cut however, since adults will be common on suitable substrates whereas it may be difficult to recognize suitable substrates without the adults. In addition, post-settlement mortality may enhance or obviate the expression of these events, or may even produce similar patterns without any aggregating reproductive processes. For instance, disturbances may produce gaps between patches of newly settled larvae. Carefully planned experiments and detailed observations are clearly required to make these distinctions. 14 Corals exhibit a variety of mechanisms that can lead to aggregated distribution patterns. The massive coral Goniastrea produces negatively buoyant gamete clusters unlike most other broadcast spawning corals (Kojis and Quinn 1981). Fertilization and settlement thus occur mainly near adults. Chemical cues from coralline algae induce different species of the Caribbean coral Agaricia to metamorphose in an algal-species and site-specific manner, thereby contributing to niche diversification (Morse and others 1988). Various chemical cues emanating from adult corals are known to attract a number of predators, symbionts, or parasites like nudibranchs (Pires and Hadfield 1993), barnacles (Lewis 1992), and boring clams (Mokady and others 1992), often in a host species-specific manner. It is possible that these cues may also be detected by conspecific coral larvae. Other corals show remarkable site selectivity and are even capable of reversing metamorphosis (Sammarco 1982, Richmond 1985) when conditions become unsuitable in the site of settlement. Larval dispersal and settlement clearly cannot be assumed as passive events. Asexual reproduction, more commonly through fragmentation (Highsmith 1982), but also through production of asexually produced planulae (Stoddart 1983), may also be more common than previously believed. Fragmentation is especially advantageous in marginal areas since the large size of the fragments often means better survival rates and greater stability on unstable substrates (Highsmith 1982). Skeletal fusion between adjacent clones also enhance the latter factors (Hildemann and others 1974, Highsmith 15 1982). Proximity to conspecifics also increases the chance that local conditions will be more suitable for growth. ♦ Disturbances and spatial distributions General reviews on the effects of, and recovery from, disturbances on corals and coral reefs include Pearson (1981), Brown and Howard (1985), and Grigg and Dollar (1990). Karlson and Hurd (1993) examined the role of disturbances in relation to community ecology paradigms. Although there is little known about the effects of disturbances on the spatial arrangement of reef organisms, disturbances do mediate the effects of competition and can modify reproductive patterns, and thus merit quantification. One prevalent view of coral communities is that they are composed of several patches at different successional states, formed by a variety of disturbances leading to a temporal mosaic in space (Grassle 1973). A related concept is the intermediate-disturbance hypothesis (Connell 1978) which holds that competitively superior species are kept from dominating communities by intermediate-level disturbances. Validation of these two theories requires delineation of patches, and an objective unambiguous scale for defining the degree of perturbation. Grigg and Dollar (1990) recommended that estimates of stress on reefs caused by disturbances be partitioned into natural and human-induced mortality, and that the size of the area disturbed be scaled relative to total habitat area. However, progress in this regard also depends on the incorporation of spatial information, i.e., the positions of organisms 16 relative to each other. This is especially important in marine benthic organisms since their patchy distribution and strong dispersal capabilities act to stabilize ecosystem responses (Steele 1985). Studies on disturbance effects also need to focus on the level of individual organisms and not just on population averages (e.g., abundances and percentage cover) or community parameters (such as species diversity and evenness). The level of the individual is where the effects of the biological interactions and those of the environmental factors can be easily differentiated (Colwell 1984, Huston and others 1988). The increased awareness of the importance of spatial heterogeneity in ecological systems (see Kolasa and Pickett 1991) makes this individual-level approach even more necessary. In addition, findings at the level of the individual can be more easily generalized to the population or community levels by, for example, the application of hierarchy theory to ecology (Allen and Starr 1982, O'Neill and others 1986, Kolasa 1989, Allen and Hoekstra 1992). An implication of this theory is that multidimensional aspects of the niche (Hutchinson 1957) can be mapped to two or three-dimensional habitat space as "... a way around the complexity and continuous character of ecological variables" (Kolasa 1989). This allows for quantitative and qualitative projections on the possible effects of disturbances on species relative abundances and the development of null models of community organization akin to the use of the Hardy-Weinberg principle in population genetics (Kolasa 1989). In all cases, temporal and spatial scales involved must be explicitly stated. 17 ♦ General comments The short reviews presented illustrate some of the variety of processes that may influence fine-scale, within-zone distribution patterns of benthic organisms such as corals, and the intricacies involved in understanding these processes. The literature also shows that one way the effects of various interacting factors may be distinguished is to focus at the level of individual corals, rather than at the level of populations and communities. Also, experiments are called for, both in the field (to define the realm of the probable) and in computer simulations (to synthesize the results of the field studies, examine their interactions, and define the realm of possibilities). Spatially-explicit demographic models are thus used in this work and are described below. ♦ Demographic and spatial models of corals Two types of models used in this work are reviewed here as they are applied to reef settings: matrix population models (see van Groenendael and others 1988 for an overview), and cellular automata (see Czaran and Bartha 1992, Villa 1992 for general reviews of spatial models): • Population projection-matrix models Matrix models are powerful extensions of classical life table analyses that allow for understanding of the demographic effects of changes in vital rates at various stages of the life cycle (Caswell 1989). As applied to corals, these models involve the construction of a transition matrix describing a) the probabilities for growth or shrinkage of each coral colony in a given set of size classes, b) the mortality rates of each coral colony in each size class, and c) the fecundity or recruitment rate expected from members of the various 18 size-classes of the species being modeled. Multiplication of this matrix with a vector describing the initial size structure of the population permits the estimation of the future size structure and abundance of that population, assuming the transition matrix holds for each time step (see Figure 1). This allows for the simulation of disturbances like typhoons or harvesting (by varying the elements of either the transition matrix or the population vector), as well as changes in the demographic parameters. This type of model has been used previously in estimating recovery rates of coral populations after crown-of-thoms starfish infestations (Done 1987, 1988) and hurricane damage (Hughes 1984). Caswell (1989) provided a useful guidebook on matrix models. Besides projecting the size structure and abundances of organisms, analysis of the matrix model allows for the estimation of the relative importance of the various size classes in determining the growth rate of the populations as measured by A , the largest eigenvalue of the projection matrix [lambda (A ) is equal to the natural logarithm of the more familiar r in Lotka's equation; Caswell 1989], This importance is usually expressed as a sensitivity index for each matrix parameter (growth, mortality, and recruitment rates per size class), which estimates the degree to which the population growth rate is modified by a small change in that parameter (Caswell 1978). Hence, a sensitivity measure could be computed for each element of the transition matrix (see Figure 1). Since matrix elements are often expressed in different or incomparable scales (for example, transition probabilities cannot exceed unity whereas recruitment rates can), proportional effects on population growth rates of proportional changes in matrix 19 SCO ✓ \ ✓ \ / V f0 + L f ^ S f2+ s f3 +s Xo Xo 9 L, s s X x, — X, 9 9 1 -2 s X2 X2 9 9 9 L a / x3 < / X3 < / TRANSITION POPULATION POPULATION MATRIX VECTOR AT VECTOR AT TIME 0 TIME 1 Figure 1. Diagrammatic representation of the elements of a projection matrix, f represents fecundity or settlement rates, g is the probability that an individual grows to a larger size class, s is the probability of shrinking to a smaller size class, and L the probability of remaining in the same size class. Survivorship is the sum of all L, g and s for a given size class (the columns of the transition matrix). Figure is modified from Hughes (1984). 20 parameters are better reported as the elasticity values (Caswell and others 1984, de Kroon and others 1986, van Groenendael and others 1988). Both indices are used here to give insights on the potential responses of corals to disturbances and competition. • Cellular automaton models Cellular automata (CA) are discrete, spatially-explicit models that represent entities as identical cells whose fate on a subsequent time period (t+J) depend on their state on the previous time (f) and those of their immediate neighbors. This model is better known among recreational mathematics circles in the form of John Conway's game "Life" (see Gardner 1970, 1971; Cartwright 1993 describes a spreadsheet implementation). In his two-dimensional "universe" (a grid of square cells like that of a chessboard), cells can only have two states ~ live or dead (represented as either black or light-colored squares). Births occur in an empty cell surrounded by three live neighbors, and life persists as long as there are two or three live neighbors. Other situations ("overexposure" or "overcrowding") result in death. Despite this simple formulation, this universe exhibits an extremely rich and interesting behavior (Cartwright 1993). Wolfram (1984) has explored the properties of one-dimensional (i.e., a line of cells) cellular automaton models and has shown that they are capable of very complex behavior, including stable limit cycles, fractal growth, and chaotic patterns. Molofsky (1994) used such a one-dimensional model in simulating intraspecific competition in plants. There are several previous implementations of cellular automata in benthic settings, although frequently the automaton approach was used only in incorporating 21 arrangement of entities in space, their local interactions and processes, and the consequences of these. Few studies focused on how the distributions or arrangements are generated in the first place. Reichelt and others (1985, 1988, 1990) and Green (1990) used automata to simulate the spread of the corallivore starfish Acanthaster planci in reefs, and the effects of cyclones. These models are extensions of earlier work on fire-effects on zonation of Australian vegetation (Green and others 1985, Green 1989). In both cases, the model included an essentially stationary phase that may or may not grow (i.e., the corals or the reefs they form, the vegetation) and a mobile phase (starfish, coral planulae, fire) moving among the stationary phases. Karlson and Jackson (1981) used cellular automata to explore the consequences of intransitive competitive hierarchies and disturbance on the diversity of clonal organisms. This allowed for the examination of the large-scale (global) effects of interactions at the local level and confirmed that intransitive hierarchies do allow for more species to coexist. But this depends on the arrangement of individuals, as Silvertown and others (1992) has shown with a cellular automaton model of a grass community. In a subsequent work, Karlson (1985) also examined the impact of competitive reversals and phenotypic variation on diversity. 22 Maguire and Porter (1977) also focused on competition among corals although their model was more at the colony level, i.e., with the cells representing polyps. Again, cover of interacting taxa, rather than distributions, was the focus of the study. More recently, cellular automata were used to examine the interactions between coral life-history patterns (broadcast spawning versus brooding), degree of connectivity between reefs, and the feeding behavior of the Acanthaster in determining the recovery rates of coral communities (Johnson and Preece 1992, Preece and Johnson 1993). As was stated earlier, most ecological applications of cellular automaton models reviewed here used these models mainly to represent an essentially static spatial arrangement of entities, rather than allow the model processes to generate the spatial arrangement. This work emphasizes the projection of the arrangement of points in space over time. * The study site and study organisms Puerto Galera Bay is a small 4 km2 inlet on the northeast coast of Mindoro Is. (Figure 2). Despite its small size, it contains a wide variety of coastal marine habitats including mangroves, seagrass beds, soft-bottom and coral communities. The area is presently a reserve of the United Nations Man and Biosphere Programme. The increased tourism associated with the reserve status has led ironically to several problems and conflicts, and thus the Bay presents an opportunity for developing management strategies appropriate for such small embayments. 23 T h i r d p l a t e a u MEDIO IS P A N IQ U IA N IS. F i r s t p l a t e a u S e c o n d p l a t e a u E S C A R C E O P O IN T PHILIPPINES / t NAUTICAL MILE Figure 2. Map of the main study sites in Puerto Galera Bay, Oriental Mindoro. Each of the First and Second Plateau sites contain at least two experimental blocks of Anacropora puertogalerae, Acropora subglabra, and release treatments. Coral settlement was also monitored in the Third Plateau and near Escarceo Point. See Figure 3 for a detailed map of experimental blocks. Two sandbar- like projections called "plateaus" extend from the south end of the Bay and these are inhabited by a diverse set of coral communities living on unconsolidated substrates of sand and silt (described below). On the west side of Medio Is. is a wider platform (the "Third Plateau") containing a different assemblage of corals in addition to a small seagrass bed and mangrove patch. Currents are mainly tidal driven, forming a gyre in the middle of the Bay (Villanoy and others, in press). The Bay is in the typhoon belt of the Philippines, with about 21-30% of the annual average 19 tropical storms in the Philippines passing through this area (Lim 1988, Philippine Atmospheric, Geophysical, and Astronomic Services Administration, unpubl. data). It is a good anchorage site, however. The name of the Bay and the town is derived from the Spanish galleons involved in commerce between the Philippines and Mexico. Puerto Galera Bay was the focus of studies by many early marine biologists working in the Philippines, mainly because of the availability of a small marine station established in 1930 by the University of the Philippines. Most of these earlier studies were on the taxonomy of hard corals (Nemenzo 1955a, 1955b, 1959, 1960a, 1960b, 1964, 1967, 1971, 1976), soft corals (Roxas 1932, 1933a, 1933b), and other organisms (e.g., sponges: De Laubenfels 1935; asteroids: Domantay and Roxas 1938). This has led to the area being the type locality of several new species. One example is Amcropora puertogalerae (Nemenzo 1964), an exquisite branching coral that is rare throughout the Indo-Pacific (Veron 1986,1992) but common in one section of the bay. This species is the focus of the present work. Recent studies concerning reef ecology in the area include 25 Gomez and others (1988), Hilomen and Gomez (1988), Licuanan and Gomez (1988), Licuanan (1988, 1991a, 1991b), Bermas and others (1992), Atrigenio and Alino (in press), and Licuanan and Alifio (in press). A fragile and biologically unique ecological community found in Puerto Galera Bay is the series of silt-based coral patches found on First and Second Plateaus. These communities are dominated mainly by Acropora subglabra (AS, a bottlebrush coral) patches interspersed with those of Anacroporapuertogalerae (AP), Seriatopora hystrix (both finely branching forms), Montipora cf. foliosa (a foliose/plate-like form) with Acroporapulchra, A. carduus and A.formosa thickets, and Porites nigrescens colonies in the shallower areas. These species are unique in that most are free-living (unattached to the substrate) on unconsolidated substrate, mainly sand and silt. They are mostly found in the site as monospecific patches or thickets, unlike in typical reef slope communities in which colonies (and not patches composed of several hundred colonies) seem randomly interspersed among each other. Only growth and aspects o f reproduction of A. pulchra have been intensively studied of the Philippine species mentioned (see Yap and Gomez 1981, 1984, 1985b, 1985c, Yap and others 1990). The lower diversity of the community, and the importance and ease of manipulation of the free-living corals that inhabit it, make these ideal for field study and simulation for purposes of pioneering approaches that may then be applied to more complex situations. As listed on page 8, the aim of this work is to examine the various processes that are responsible for AS and AP forming monospecific stands, using both field experiments 26 and computer simulations. Experiments were conducted to verify the presence of interspecific competition (by cross-transplantation of adults) and detect differential recruitment patterns and survival (using settlement tiles). Monitoring studies, meant to evaluate the impact of disturbances and other environmental factors, were also initiated. In addition, data were collected for use in constructing demographic and spatial models to further examine the interaction of these different processes in determining the distribution patterns of these corals. The different components of this work are summarized in Table 1 . The Field Experiments Materials and Methods All field experiments were conducted in experimental blocks demarcated on the First and Second Plateaus, although not all blocks were used for a particular experiment. Each block consists of a haphazardly chosen Anacropora puertogalerae (AP) patch, an Acropora sublglabra (AS) patch, and an area devoid of coral (the "release" area). All patches for a given block are within 10-15 m of each other and differ in depth by no more than 2 m. Six experimental blocks were established, three on February 1993 (Blocks A, C and E in Figure 3), two on May 1993 (Blocks B and D), and Block F on December 1993. Depths and locations of the patches are summarized in Table 2. 27 Table 1. Summary of the major components of the field studies. PROBLEM: CAUSE(S) OF PATCHY DISTRIBUTION OF SINGLE -SPECIES STANDS?: HYPOTHESES: I. COMPETITION? 2. REPRODUCTIVE EVENTS? 3. DISTURBANCES? EXPERIM ENTS/ OBSERVATIONS: A. INTER-SPECIFIC COMPETITION B. INTRA-SPECIFIC COMPETITION Al. Cross transplantation page 31 (methods) page 40 (results) A2. Pairing experiments see Appendix 2 B. Density manipulation see Appendix 3 A. PRE-SETTLEMENT EVENTS? B. POST-SETTLEMENT EVENTS? A. Settlement distribution page 34 (methods) page 49 (results) B. Recruit transplantation page 37 (methods) page 52 (results) A. Mapping of rubble see Appendix 5 B. Fish predation see Appendix 6 to 00 PU ER TO GALERA BAY S E C O N D PLATEAU FIRST PLATEAU M UELLE Figure 3. Map of the experimental blocks used in the First and Second Plateaus. Information regarding each block is summarized in Table 2. Table 2. Descriptions of the experimental blocks (=sites) used in the coral transplantation and recruitment experiments. Each block includes an Acropora subglabra (AS) patch, an Anacropora puertogalerae (AP) patch, and an area devoid of corals (release area, REL) — all within 15 m of each other. Site Location Depth range (m) A First Plateau 13-15 B First Plateau 13-15 C Second Plateau 13-15 D Second Plateau 10-12 E* Second Plateau 7-9 F* First Plateau 7-9 * used in recruit transplantation experiment or certain minor experiments only 30 ♦ Cross-transplantation studies For the coral cross-transplantation studies, Blocks A to D were used. For each block, haphazardly chosen colonies 14-53 cm in diameter for AS, and approximately 25 cm for AP were used in four treatments as shown in Figure 4. For AP, since colonies (i.e., corals with contiguous skeletons) were difficult to distinguish within the AP patches, fragments were obtained with the minimum breakage from a clump and loosely tied together. Cross transplants are those colonies moved from the original positions and patches to another species patch (usually in the space made available by colonies used in the other treatments) within the same block. This treatment was meant to detect possible interspecific interactions. Release colonies were relocated to areas devoid of other corals to examine the effects of the reduction in density of both conspecifics and potential competitor species. The transplant control treatment was composed of colonies subjected to the same handling and transport regime as the cross transplants, except that they were immediately returned to their original positions and orientations. This treatment was meant to detect any effects of handling. Controls were colonies not subjected to any handling. In each of four blocks, two to five colonies for each of four treatments and two species combinations were used. All colonies were interspersed with each other in the patch interior, each colony being separated by a distance of at least 0.5 m, with intervening colonies of the host patch (except in the release area). Growth rates of the branches of treatment colonies (of both species) were monitored at five-week intervals 31 Release treatment Transplant c o n tro l™ :; Figure 4. Diagram of an experimental block and the treatments used in the cross- transplantation experiments. Cross transplants are colonies moved to another species- patch while transplant controls are also relocated but immediately returned to their original positions and orientations. Release colonies are those moved to a nearby area devoid of coral. True controls are unmanipulated colonies. Colonies of both Acropora subglabra (AS) and Anacropora puertogalerae (AP) were used in this experiment. 32 using the following methods: One or two of the longest branches in each colony center were marked with a zip tie 1 to 5 cm from the tip. Then in each monitoring visit, a flexible measuring tape was run from the zip tie mark to the branch tip. Drawings were made to serve as guides and ensure consistency in the remeasurements. Missing ties/branches and branches that have grown too long for accurate measurements were replaced by retagging the same colony at another branch. AS colonies have radial corallites that are longer toward the base of the colony. Because branch length in this species is effectively the length of the axial corallite, measurements were more difficult since the measuring tape cannot be put closer to the branch main axis, especially nearer the base. The measurement method described thus had to be applied more carefully to minimize parallax errors. Error in measurements, as estimated from repeat measurements, was no more than 0.5 cm for AS and negligible for AP Besides branch extension rates, colony sizes (mainly for AS) were also estimated using ruler measurements o f the largest colony diameter and of the largest diameter perpendicular to the latter. Projected circular area was computed using the formula (from Yap and others 1992): ( JhngthxwiW) \ 2 area = — j ------j * 71 33 ♦ Coral settlement studies The coral settlement study was meant to examine whether the patchy adult distributions are due to juvenile settlement patterns and the modification of these by preexisting adults. There were two component studies, one aimed at describing settlement patterns (the recruit distribution study), and the second designed to distinguish whether the patterns shown by the latter were due to pre- or post- settlement factors (the recruit transplantation study). For both studies, terra-cotta tiles 10 x 6.4 x 0.6 cm in dimensions were used to provide settlement surfaces for hard corals. Each tile has a smooth surface and a ridged one. A hole was bored near one end of each tile, through which a steel rod was inserted to allow the tile to be hung perpendicular to the substrate. Six tiles were then suspended in each rod to form a rack, with each tile separated from the next by 2 cm segments of 1 cm diameter plastic hoses. Three tiles of each rack were set with their flat sides facing the nearest end of the support rod. Solid copper wires (AWG #14) and additional spacer hoses were used to keep the tiles and hoses pressed against each other and prevent adjacent tiles from colliding (see Figure 5). The racks were then set in the middle of the treatment areas such that the tiles were in a vertical position with their tops at about 20 cm from the substrate - approximating the height of adjacent coral colonies (if any). Additional steel rods were used to support the rack and keep the tiles off the substrate. In all cases, the racks were positioned such that they were perpendicular to the slope. 34 DIAGRAM OF A RACK OF SETTLEMENT T ILE S Coral-free areas ("Release" treatment) I * ij & S g g : m Figure 5. Diagram of a settlement tile rack and the layout of the recruitment monitoring for each experimental block. For the recruit distribution study, two experimental and a control treatment were used in each block (as defined in the adult transplantation experiments). For the experimental treatments, racks were set at the middle of AP and of AS patches to quantify coral settlement in these monospecific stands (Figure 5). Racks were also set in nearby areas, 10-15 m away from either of the two experimental treatments, which were lacking corals to serve as the release treatment. These treatment areas corresponded to the release patches used in the cross-transplantation studies. A total of five complete experimental blocks were set in both the First and Second Plateaus. Racks were also set in each of two coral communities (the Third Plateau and in Escarceo Pt.) where AP and AS do not normally occur (see Figure 2). The Third Plateau is closest to the study areas but has a very different reef community dominated by faviids, mussids, and acroporids. Poritid and faviid hard corals and xeniid soft corals dominate Escarceo Point, which is 2.5 km away from one of two channels into Puerto Galera Bay. Racks were immersed for approximately 91 days, beginning on the first week of June 1993, and replaced after such intervals. Collected tiles were air dried. In the laboratory, both surfaces of each tile were examined under a dissecting microscope and recruits were marked, identified, and counted. Since genus- and species-level identifications were possible only for older recruits, most identifications were done at the family level following the guide prepared by Babcock (in English and others 1994). Live (when collected) and dead recruits were also distinguished when counting. Coralla of the dead recruits were often fouled with algae. 36 To distinguish between pre-settlement and post-settlement effects on the recmitment of corals, three additional racks of settlement tiles were set in the release area of each block on March 11-13, 1994 simultaneously with the last of the recruitment monitoring set just described. This was at about the peak spawning period of corals in the area (Bermas and others 1992). After three months (when the monitoring set was collected), one rack from this release set was transplanted to the middle of the AP patch of the blocks (the recruit transplant set; see Figure 6). The second additional rack was similarly moved to the AP patch, but was immediately relocated to its original position in the release area, as a handling control analogous to the transplant controls in the adult transplantation experiment. The third rack was left in the release area as an unmanipulated control. The three racks for each of the six blocks (including Block F; see Table 2 and Figure 3) were then left immersed for an additional month. The same procedures described earlier were followed for processing and analyses of tiles. ♦ Physical factor measurements Water temperature was monitored during each monitoring visit using maximum - minimum thermometers set in the middle of the AP, AS patches, and release area for 24 hours. Measurements were made in at least two study blocks on succeeding days of a monitoring visit. Between 11:30 a.m. and 1:30 p.m. of each monitoring day, water transparency was measured with a 20-cm white secchi disk (with two quarters painted black) at a point midway between the First and Second Plateaus. Lack of strong currents over the 37 Coral-free ("Release”) area Transplant control tiles | Control Transplant tiles Am Recruits counted as sample of numbers before transplantation tiles h i Figure 6. Layout of treatments of the recruit transplantation experiment for each experimental block. sites made determination of the angle of the secchi disk line unnecessary. Cloud cover (in octas or eighths of the sky) was estimated at the same time. Sedimentation rates were measured on July 1994 at the onset of the rainy season. During each measurement, glass test tubes 13.5 cm long and with a 1.25 cm mouth were placed among colonies in the AS and AP patches, and around the release area, with about 10 cm of the tube embedded in the substrate. Diving activities were planned to minimize diver-derived resuspension of sediments in a given study site. After 48 hours, tubes were plugged and collected. In the laboratory, sediments were decanted onto pre-weighed filter paper, air dried, and weighed. Water movement was measured using two methods. Current speeds through the coral patches and the release area were estimated using rhodamine B dye solution squirted downwards from a wash bottle. The time required for the front of the cloud to travel 1 m (marked with transect tapes) was then recorded for each run. In addition, clod cards, following the formulation of Doty (1971), were set in the treatment areas for 24 hours. At least 10 cards were distributed around the patch interior following a complete blocks design as in the transplantation and recruitment studies. Control cards were maintained in sea water containers on land, taking the precautions described by Jokiel and Morrissey (1993). In the laboratory, cards were dipped in freshwater to remove settled sediments and other loose material, and air dried in an air-conditioned room to constant weight. 39 ♦ Analyses The coral transplantation and the recruit transplantation experiments were designed as randomized complete blocks experiments and were analyzed using the appropriate fixed-effects analysis of variance (ANOVA) model (Gomez and Gomez 19S4, Winer and others 1991). The same holds for the clod card measurements. In some cases noted in the text, the appropriate form of t-tests (Pagano and Gauvreau 1993) were used for comparisons between two variables. Survival rates of the transplantation treatments were compared using the log-rank test (Pagano and Gauvreau 1993). The coral settlement monitoring was analyzed as a completely randomized design with the racks as replicates and tiles as subsamples (Gomez and Gomez 1984, Winer and others 1991). Recruit counts were square root (x+0.5) transformed to approximate a continuous distribution (Winer and others 1991). In all other cases, if analyses of transformed and untransformed data yielded the same results, the latter are presented to simplify interpretation (see Maxell and Delaney 1990). Tukey's test (Winer and others 1991) was used for multiple comparisons of all significant ANOVA results. Results ♦ Cross-transplantation studies: Branch growth rates and related observations Figure 7 summarizes the growth rates of AP colonies subjected to different treatments. Data averaged for the entire 18 month study period showed highly significant differences between treatments (p=0.007; Figure 8 and accompanying table), with linear extension rates of AP colonies transplanted to AS patches and those moved 40 25 -10 ■1.5 2 5 2.0 1 5 1.0 0 5 00 Ja -0.5 - 1.0 -15 AP CONTROLS • 4 • • • • 1 a 1 • l • • • • 9 • • i • • 1 • • 93 ' Ma) -93 Au| •93 * N o* •93 Mai 94 Juf 94 Sef • AP TRANSPLANTS a a a t a a t t i ) i 1 a a 1 a k i 93 M ai a -93 * A uf ■ 93 * Not -93 * Ma 94 J im a 9 4 Sef a a a date 2.5 2.0 1.5 1 0 0 5 0.0 Jai -0 5 - 1,0 -1.5 AP TRANSPLANT CONTROLS Ma) 93 o Au( 1 -93 o 8 •93 •94 Jur -94 -94 a 9 A t i a a k t * 1 i i 4 * a a a 93 Mq 9 3 Auj 9 3 N o> -93 Ma 94 Jut •94 Sif k «i a 94 date Figure 7. Comparison of growth rates of branches ofAnacropora puertogalerae (AP) transplants and controls. Averages are presented in Figure 8. Note the generally slower growth in the AP transplants and release colonies, i.e., those outside the source patch. 0.80 AP CONTROL AP RaEASE AP AP TRANSPLANT TRANSPLANT CONTROL treatment Figure 8. Branch growth rates oiAmcropora puertogalerae (AP) averaged over the duration of the study. Randomized complete blocks ANOVA show significant differences among all treatments. Bars represent standard errors. ANOVA Source o f Variation SS d f M S F P-value Replication 0.0816 3 0.0272 1.8836 0.2108 Treatment 0.3808 3 0.1269 8.7934 0.0065 Error 0.1155 8 0.0144 Total 0.5779 14 42 to the release areas being about half that of both the unmanipulated and transplant controls. Multiple comparison tests show highly significant differences between all possible pairs of treatments (Tukey's test, alpha=0.01) indicating some handling but large treatment effects. In all four treatments, there was no apparent seasonality in linear extension rates that could be associated with the wet (June-November) and dry seasons (Figure 7). This is in spite of, or may be due to, the record number of typhoons and tropical storms (32) entering the Philippine area of responsibility in 1993, leading to an extended rainy season (see water transparency results on page 59). Some initial stunting of transplants was observed. The AP release colonies grew thinner branches (p<0.05) compared to colonies in the patch interior, but remained comparable to colonies near the patch edge (see Appendix 1). Only three cases of tissue damage to AP cross transplants were noted, two at the start of the experiment and the third when an AS fragment fell on an AP. This low number was unexpected since a preliminary experiment involving pairing of equally-sized fragments of both species, AS either damaged or killed AP in most pairings in less than three months (see Appendix 2). Branches of AP were also observed to curve away from AS (including transplants), but as with the tissue damage, these were not as common as expected. There were also two cases of AS branches curving away from AP. One of these was an AS cross transplant, but the other was an AS control that was in a patch partially taken over by AP cross transplants that had coalesced. Growth of the latter 43 transplants was sufficiently high that spaces between AS colonies within their own patch were being filled in by AP (see colony size analyses below). Results for AS showed no significant differences in linear extension rates (p=0.34; see Figure 9, ANOVA table in Figure 10) although 18-month averages (Figure 10) resemble the AP results with unmanipulated controls growing fastest and release colonies being slowest. No apparent changes in morphology of the cross and release transplants were observed. ♦ Cross-transplantation studies: Colony size and survival Both AP cross transplants and AP release colonies experienced partial mortality and fragmentation during the experiments. Fragmentation of AP treatments (except true controls) was common, largely because of the collection method. In addition, blennies in two o f the four release areas often carried off small (less than 8 cm long) AP fragments to line the outer rims of their burrow openings. Most of these fragments were already dead or dying when observed. Since "colonies" of AP studied here were just loose tangles of branches of several fragments tied together, scattering of AP release colonies and cross transplants sometimes lead to the formation of small but apparently thriving clumps of up to 0.37 m2 in area from a single release transplant, and 0.29 m2 from a cross transplant. In the AS patches, AP cross transplants in two blocks were coalescing and filling in the spaces between the AS colonies by the end of the study. Disturbances by fishes were not as 44 branch grow th (cm /m o) branch grow th (cm/mo) AS CONTROLS AS TRANSPLANT CONTROLS 2 5 2.5 0 5 0 5 Nov *93 Jur 94 Nov -93 Jur -94 •93 ■1.5 -15 -2.5 -2 5 25 05 -1 5 -2 5 A A A i t i * A * b A i J b . - j - . - j 8 I i 93 Mat - 8 3 A uo -93 Nov-93 1 Ma A -9 4 a Jun k 94 Sec i date 2.5 05 -15 -25 1 A ‘ I : 1 A A A A 4 A ! I 93 Mat i -93 Am -93 * Nov -S3 * Ma . •94 Jun [ 9* Sac A A b date •94 Figure 9. Comparison of growth rates of branches ofAcropora subglabra (AS) transplants and controls. Averages are summarized in Figure 10. Growth did not differ significantly between these treatments. 0.60 S 0.20 0.10 0.00 AS CONTROL AS RELEASE AS TRANSPLANT AS TRANSPLANT CONTROL treatment Figure 10. Branch growth rates of Acropora subglabra (AS) averaged over the duration of the study. Randomized complete blocks ANOVA reveal no significant differences among all treatments. Bars represent standard errors. ANOVA Source o f Variation SS d f M S F P-value Replication 0.1549 3 0.0516 1.4580 0.2900 Treatment 0.1361 3 0.0454 1.2810 0.3387 Error 0.3187 9 0.0354 Total 0.6096 15 46 apparent with these AP cross transplants as they were in the two release areas. One colony in Block C did appear to have been cropped by an unknown organism. Since size changes per colony, mortality of fragments, and asexual reproduction (to the periphery of the marked colonies) were confounded in the AP treatments, no detailed analyses of these parameters were conducted. In the case of AS treatments, there were no apparent trends in colony sizes over time (Figure 11). Survival rates were more revealing however. Correcting for different transplantation dates, comparison of AS survival rates show a significant difference (log-rank test, p=0.03) between AS transplants and AS transplant controls. Eight of the 17 AS cross transplants were dead by the end of the study. AS cross transplants that survived often had large, bleached sections. Rapid reduction in tissue area often preceded colony death. Usually the necrosis began at the colony base and extended to the branch tips. Since colony sizes were based on distances between living branch tips, this pattern resulted in the colony size measurements not reflecting these changes until whole branches were dying off. Survival rates of AS release and AS control colonies were not significantly different (log-rank test, p=0.54). In January 1994, widespread partial mortality (bleaching followed by algal overgrowth) of AS colonies was observed in most blocks. In a monitoring quadrat at 7 m depth, partial mortality ranging from S to 50% of a colony's tissue area was observed in some AS colonies. AP cross transplants remained unaffected however, even if the 47 r- AS CONTROLS 1200 1000 m 800 ► » 600 400 200 Nov-94 Od-92 Mar-93 Aug-93 Jan-94 Jun-94 r AS TRANSPLANTS p g 800 J Aug-93 Jan-94 date AS TRANSPLANT CONTROLS — Aug-93 o i a D o a 1 □ □ ° o i i B a a a j 1 & B o □ ? 8 F _______i 1 i * — Aug-93 Jan-94 date Figure 11. Comparison of colony sizes of Acropora subglabra (AS) transplants and controls. No strong trends are evident despite significantly higher mortalities in the AS transplants. -u 0 0 host AS patch was decimated (which occurred in Block C). In all cases, AS colonies were physically undamaged but were found bleached or already coated with algae. These deaths probably did not affect the results of the AS transplantation experiment, since about half of the deaths in that study occurred before this time. Also, 80% of the AS colonies that died during the study were cross transplant colonies (i.e., those relocated to AP patches). No unusual mortality of AP was observed. ♦ Coral settlement studies: Recruit distributions The number of live scleractinian recruits (averaged per treatment over the whole year studied) showed significant differences among the patches (one-way ANOVA, p=0.011; see Figure 12). This was due to large differences between settlement rates on AP tiles compared to the release and Escarceo set, and among the Third Plateau set and all other treatments (Tukey's test, alpha=0.01). If individual sampling results are examined (Figure 13), AP consistently had the lowest number of recruits settling of the five treatments in all five sampling periods (extending from May 1993 to June 1994). The Third Plateau control racks, on the other hand, always had the highest settlement levels observed, leading to the time-averaged results described in the previous paragraph. The lowered settlement rates in AP patches were most pronounced during the fifth sampling (June 1994), at about the period during which most corals in the area are believed to spawn (Bermas and others 1992). In this period, AP tiles received only 7 of the 100 live recruits counted in the AS, AP and release patches. This was significantly 2.00 1.80 1.60 1.40 » 1.20 | 1.00 | 0.80 0.60 0.40 0.20 0.00 Figure 12. Settlement rates of scleractinian corals (square-root-transformed counts per tile) averaged over the five samplings in one year for each treatment or site. Per-sampling breakdowns are shown in Figure 13. Bars represent standard errors. ANOVA Source o f Variation SS d f M S F P-value Between Groups 2.2760 4 0.5690 4.3034 0.0113 Within Groups 2.6444 20 0.1322 Total 4.9205 24 3RD AP AS RELEASE ESCARCEO PLATEAU PT. treatment 50 3.00 □ 3RD PLATEAU May-93 Sep-93 Dec-93 Mar-94 Jun-94 Figure 13. Settlement rates of scleractinian corals (square-root-transformed counts per tile) in the various treatments and sites over the five samplings in one year. Note the consistently lower settlement rates in iheAnacropora puertogalerae (AP) treatment. Rates per tile were averaged per block. Averages per treatment are shown in Figure 12. Bars represent standard errors. 51 (Tukey's test, p<0.05) different from the level in the release areas, which got 67% of the total recruits. Also, AP tiles were often the least fowled by tubeworms, bryozoans, and even algae, with the 3-month old tiles often appearing like the 1-month old tiles of both the AS and release tiles (see also Licuanan and Alifio, in press, for results of early analyses). Settlement rates in AS patches (which got 26% of the total recruits in the same peak recruitment season) was intermediate between that of AP and release treatments, and did not differ significantly from these two treatments (p>0.05). Part of the differences in recruitment rates among treatments may have been due to the way recruits were allocated among the study patches. Seven of the 13 acroporids found in the peak season were in the AS patches, and none were in the AP patches. This difference is not significant (p>0.05) although cell frequencies of the AP treatments were often too low to put much confidence in the Chi-square test (see Table 3). However, 61 of the 100 recruits were pocilloporids, and their relative abundances among the patches did not differ from expected either (p>0.05). * Coral settlement studies: Recruit transplantation Results of the recruit transplantation experiments suggest that the differences in settlement rates could be due to differences in the suitability of the settlement surfaces (i.e., pre-settlement effects), and not post-settlement mortality. This could be inferred from the comparable numbers of surviving recruits on the transplanted tiles, compared with the two controls (handling and unmanipulated) per block (p=0.22, Figure 14). In 52 Table 3. Distribution of taxa among treatment patches in the First and Second Plateaus. Tiles were collected on June 13-15, 1994 after about 90 days of immersion. Note that cell frequencies may be too small to ensure an accurate test. AP AS RELEASE TOTAL Pocilloporidae expected Acroporidae expected Others expected TOTAL 40.87 4.27 15.86 0.91 3.38 8.71 17.42 1.82 6.76 100 Computed Chi-square 7.2190 p>0.05 53 CONTROLS TRANSPLANTS TRANSPLANT CONTROLS treatment Figure 14. Counts of live scleractinian coral recruits (square-root transformed) among treatments in the recruit transplantation experiment. No significant differences were found. Bars represent standard errors. ANOVA Source o f Variation SS d f MS F P-value Between Groups 0.6597 2 0.3298 1.6917 0.2175 Within Groups 2.9247 15 0.1950 Total 3.5844 17 54 fact, numbers of live recruits on the tiles transplanted to the AP patches were larger, though not significantly so. Most of these recruits were pocilloporids. Also, unlike in the recruit distribution study, tiles that were transplanted to AP patches had comparable levels of fouling by tubeworms and other taxa compared to the control and transplant control tiles. Conditions within the AP patches thus do not seem to be adversely affecting biota already on the settlement tiles. ♦ Physical Measurements Results of the two sets of sedimentation studies are shown in Figure 15. On both dates, there was a significantly lower sedimentation rate within the AP patches (p<0.001) even when the traps were set in areas where cover of the ubiquitous zoanthid Palythoa sp. was comparable (these are the measurements made on July 17, 1994). This result could be related to the sediment resuspension allowed by the resident corals in a given patch, that is, the higher density of AP colonies could be minimizing the effect of water movement on the underlying sediments. Clod card measurements reveal a significantly lower diffusion rate inside the AP patch (p<0.01) suggesting mass flux was lowest near the bases of AP corals (Figure 16). Remember that AP also has the most densely packed colonies. Measurements using rhodamine dyes do not show any apparent trend though (Figure 17). The clouds of the dye often got trapped in between coral branches, leading to variable results. 55 1.60 □ Jul-14-94 □ Jlll-17-94 1.40 1.20 N 1.00 0.80 0.60 0.40 0.20 0.00 RELEASE AS AP patch Figure 15. Mean sediment load in the three treatment patches in Block A, First Plateau. Sediment traps set on July 17 were positioned in areas of comparable zoanthid cover. Differences among patches are very highly significant (p<0.001). Bars represent standard errors. 56 8.00 AP PATCH AS PATCH RB.EASE PATCH Figure 16. Average clod card diffusion factors (i.e., weight loss in the field relative to weight loss in enclosed containers) in the three treatment areas. Data are from four experimental blocks with 12 clod cards per patch. Differences are highly significant (p<0.01). Bars represent standard errors. 57 7.0 6.0 5.0 2.0 1.0 0.0 - AP AS RELEASE p atch Figure 17. Water movement rates (as measured by rhodamine dye patches) among the different treatment patches in Block A. Error bars represent ranges. 58 Water clarity was very variable for most of the study (Figure 18). This was largely due to off-season storms and typhoons that increased runoff-borne sediments flowing into the bay. Water temperature did not differ between the treatment patches (and were just averaged), but showed the usual seasonal trend of being higher in summer (Figure 19). Discussion The results of the field experiments show that single-factor explanations for the patchy distribution of single-species stands studied are too simplistic and that the observed patterns are due to several processes that may or may not be interrelated. The transplantation (pages 40-49, Figures 8 and 10) and pairing experiments (Appendix 2) indicate that the outcomes of one-to-one coral interactions and those of many-to-one interactions may not be the same, as other processes beside digestive interactions come into play. These two experiments do show that competition, either direct or indirect, can help maintain pure stands of one species, especially in the case of AP. The density manipulation experiments (Appendix 3) also show that, at least with AS, the formation of single-species patches or stands was not detrimental to the growth of these corals by increasing intra-specific competition. In the case of AP, the species may even benefit from such high densities since this allows them to overcome the digestively superior AS by "peripheral encircling" (Lang and Chomesky 1990), along 59 meters 14 12 10 8 6 4 2 0 — Nov-93 Jun-94 Aug-94 Mar-94 Jan-94 Figure 18. Secchi depths between the First and Second Plateaus. Water transparency was very variable due to run-off borne sediments. 6 0 temperature (Celsius) 30 29 28 27 26 25 24 23 22 21 20 Mar-94 Apr-93 Jul-93 Oct-93 Dec-94 Jan-93 years Figure 19. Average water temperatures in the study area Bars represent ranges. 61 with other possible advantages speculated on below. It is unclear at the moment whether AP cross transplants can displace AS in their own patches however. The mortality rates seen in these competition experiments also show that the outcomes (judged on the basis of branch growth rates) are unlikely to be reversed since death of one of the competing species is often rapid. AS dominance over equally sized AP fragments was already apparent after three months (Appendix 2), whereas two AS cross transplants to AP patches died even before growth measurements were made a month after transplantation (the others that were still alive after 18 months were often partially bleached). Wellington (1980) and Chomesky (1989) found that competitive dominance judged on observations of tissue damage may be reversed as interacting corals adjust competitive responses (e.g., develop sweeper tentacles), hence the difficulty in relying on "instantaneous diagnosis" when studying competition (Lang and Chomesky 1990). The few digestive interactions seen in the cross-transplantation experiments, coupled with the clear outcomes, suggest interactions are occurring through less obvious, indirect means (see below). The depressed settlement rates in the AS patches, and especially in the AP ones, demonstrate that adults of both species can also affect the establishment of planulae of other coral species within these patches. It is uncertain though how this is achieved. Differences in sedimentation rates among the patches could not explain these results. If sediments hinder settlement of corals (see Babcock and Davies 1991, Te 1992), then the AS and release tiles should have lower, not higher settlement rates. Results of the recruit transplantation experiment (pages S2-SS and Figure 14) reveal that tiles conditioned outside the AP patches can attract and maintain comparable numbers of coral recruits regardless of whether they are moved to the AP patch interior or remain outside in the release area. This shows that planulae can still reach tiles inside the AP patches thereby discounting larval predation by the host corals (or else the additional month of recruitment could lead to higher numbers on tiles outside the patch) and that mortality rates are comparable within and outside the AP patches (otherwise recruit numbers outside the patch would again be higher). The possible role of bioactive compounds, already found in some hard corals (Gunthorpe and Cameron 1990), is also eliminated by these results. The findings thus indicate that depressed settlement is a key factor in the maintenance of single-species stands (see also the simulations presented below). Differences in conditioning of the settlement surfaces may explain the observed results and these differences may be due to water movement effects. Although the experiments show that certain competition and recruitment processes may be responsible for the maintenance of single-species stands or patches, these may not be sufficient to explain the formation of the patches in the first place (see also the computer simulations below). Unfortunately, the reasons behind the formation of patches are harder to elucidate since this has a strong historical component, i.e., isolated events in the past could be sufficient to generate present day patterns. Processes maintaining the patches are ongoing and are thus more easily detected. 63 Disturbances and predation events are likely explanations for patch formation, especially if their effects favor one species, or groups thereof, over others. The observed die-off and partial-mortality of AS colonies on January 1994 (see page 47-49) show pure stands of one species could be destroyed or formed without affecting other coral species in them. The cause for this event is unknown, although there is a possibility that this is due to human activities. Fishermen using cyanide to collect aquarium fish have been observed to work around the Third Plateau (V. Hilomen, pers. comm.) although effects of this have never been seen. A warehouse-barge (containing beer or brewery materials) used the Second Plateau as shelter from typhoons late in December of 1993, just before the mortalities were observed. The outcome (or lack thereof) of the alizarin-staining experiment (Appendix 4) and growth patterns revealed by the transplantation experiments (Figures 7-10) suggest that there are differences in sensitivities of AP and AS to changes in water conditions, so it is conceivable that certain perturbations could selectively affect AS and AP differently (see also computer simulations below). If the present day patches were produced by disturbances that selectively killed species, distributions of rubble should reflect the earlier distribution patterns. However, patches of rubble were found to be often associated with live patches of the same genus, and there was little intermingling with other genera (Appendix S). Analyses of sediments underlying the patches suggest that the patches are fairly old, judging from the thick layer of conspecific rubble underneath AS and AP patches. Thus if patches were generated by such selective disturbances, this probably happened so long ago that evidence may 64 already have been buried or transported. More detailed investigations are clearly required to properly evaluate this possibility. Effects of fishes also need to be evaluated. The transport of coral fragments by gobies was mentioned earlier (pages 44). This could also modify distribution patterns of adults corals, especially if the fishes form aggregations themselves and prefer one type of coral over others. These gobies are not widespread in the area though. Also, since most of the coral fragments they used were already dead or dying, it is unlikely that these fishes have a strong impact on the coral distributions. Fishes will have a greater impact if they can modify distributions of juvenile corals. Pomacentrids are known to tend algal farms, thereby indirectly affecting coral survival (Sammarco 1981) while several other groups are known predators. Again, if certain species of coral are selectively eaten, these fishes will have consequences on distributions of adults corals. Comparison of coral settlement rates on the outer face of tiles on either end of settlement racks (Figure 5) with those immediately interior to these show no departures from the general recruitment trends described earlier (see Appendix 6). If these inner tiles (which are spaced about 2 cm apart) are assumed to be inaccessible to fishes, this result is evidence that there are only minor effects of fish predation on the distributions of coral planulae in the blocks. Other predators like crabs or mollusks may be important if they prefer certain species of coral over others. An aggregation of druppelid corallivores was observed 65 attacking a patch of Pachyseris on the Second Plateau, while ignoring other nearby species of corals. Druppela species are known to prefer acroporid corals (see Fujioka and Yamazato 1983). Outbreaks of these gastropods have been reported in Japan and the Philippines (Moyer and others 1982) and may be partially responsible for the decline of Acropora pulchra on the reef flats of Bolinao, northern Philippines (Licuanan and Hilomen, unpubl. data). Acropora pulchra is common on shallow parts of the First Plateau and druppelid densities are still very low in Puerto Galera Bay. This does not preclude past effects though. Results of the field studies thus indicate that competition and recruitment processes operate in maintaining dominance of AS and AP in their respective patches (although AP show stronger effects). Patch formation scenarios presented earlier may merit more investigation. Alternatively, there is a simpler explanation for how these patches were formed — asexual reproduction through fragmentation. The lack of sexually derived juveniles of AS and AP, and observations of fusion among small fragments of AP lead to studies of the distribution of clones in these corals (Ablan and others, in press; Ablan and Licuanan in prep.) These reveal, among others, that the AP patch in Block E is dominated by at least two clones that are 6 m apart, i.e., almost at opposite ends of the patch. This suggests that the species reproduces asexually, with establishment of propagules near parent colonies leading to the growth of patches. Since clones can fuse as readily as they can break apart, the individual fragments may also enhance each other's survival by supporting smaller fragments off the substrate (Hildemann and others 1974). If both AS and AP reproduce mainly through such asexual means, then the role of competition and disturbances may be less important in determining the distributions of these species. Interactions of this mode of reproduction with competition and disturbance effects are examined in the computer simulations of the next section. The results of the transplantation and recruitment studies share a common thread and need not be seen as outcomes of independent processes. Although speculative, the following discussion will try to show that most findings may be explained by the corals' modification of water movement and sedimentation in their immediate vicinity. The lowered growth rates of AP when relocated outside the confines of its patches (both to the release area and to the AS patch; Figure 8) indicate that environmental differences, rather than direct competitive interactions, are important in determining the distribution of this species. Although the AP transplant controls also showed a significant decrease in growth relative to the AP controls, this was not to the same degree as the two other treatments. Thus, the handling and transplantation process account for only part of the observed differences in growth rates of AP transplants outside the patch. Based on the few environmental parameters measured, it appears that AP may be sensitive to increases in the sedimentation and water flow rates (Figure 15 and 16) ~ two parameters that may be expected to vary in the scale of a few meters, and thus define 67 microhabitats in the study sites. Water movement is important to corals since it replenishes planktonic food and nutrient supply, removes waste and sediments, and moderates temperature, salinity and dissolved oxygen levels (Wells 1957). The effects of water flow on prey capture (Patterson 1984, McFadden 1986, Sebens and Johnson 1991) and on metabolic rates (Dennison and Barnes 1988, Patterson and others 1991) have been documented for soft and hard corals, and are believed to be determinants of the zones these corals can occupy on the reef (Sebens and Johnson 1991). Note that both water movement and sedimentation can be modified by the presence of the corals themselves in a given position in the reef. Morphologies of corals modify boundary conditions such that feeding is maximized (Sebens and Johnson 1991) and this may explain the low diffusion factors measured within the AS and AP patches (Figure 16). The finer sediments found at the bases of both AS and AP patches (Figure 20) may be due to wakes created by the corals in these patches, similar to what Eckman (1983) found with blades of artificial marsh-grass. Once particles settle, the rubble produced by the corals probably helps in retaining the finer particles and prevent their resuspension. The significantly lower yields (p<0.001) of the sediment traps in AP patches could be due to the effect of AP colonies themselves in causing particles to settle out of the water as they enter the patch and keeping them from being resuspended. All these results suggest that the reduced growth of AP release and cross transplants may just be due to the higher water flow and sedimentation rates outside the immediate vicinity of their source patches. The initial stunting of AP transplants may 68 Figure 20. Grain-size analyses of sediments collected from the three treatment patches in Block D (10 m depth; upper graph) and Block E (7 m depth; lower graph), both in the Second Plateau. The standard Wentworth size classes (with higher phi representing finer sediments) were used. 69 also be related to this modified flow regime. The release transplants may have had to achieve sufficient packing of branches to reestablish reduced flow and resume feeding, especially on laterally transported particles trapped in eddies behind branches (Patterson 1984, McFadden 1986). The higher numbers of branches (but not number of branches per colony; see Appendix 1) in the native AP patches thus may have provided some advantage. McFadden (1986) found that at high velocities, particle capture rate in soft corals is enhanced by the presence of neighboring colonies. This advantage may be lost as the patches grow larger though. Large AP patches usually have barren portions in their centers that are covered only with nibble and a few free-living fungiid corals. An analogous situation may be found in kelp forests where it appears that an upper limit to patch size is set by the nutrient requirements of innermost individuals (Dayton and Tegner 1984). It is unclear why AS linear extension rates do not differ among the treatments (Figure 10). One may speculate that the bottlebrush form of the species allows it to change flow patterns such that it does not require close proximity to other AS colonies so that velocities for optimal feeding are obtained. Thus, flow conditions within and outside the AS patches may not be different enough for the coral's growth to be affected (see results of density manipulation experiments, Appendix 3). The lack of any strong differences in AS growth is inconsistent though with its high mortality rates when transplanted to within AP patches. These mortalities are probably not due to significant changes in water flow conditions, shading, or the 70 availability o f food (Buss 1979), since increased density of AS colonies within their own patches did not result in growth and survival differences (see Appendix 3). The observation that death of AS transplants begins from the base and proceeds upwards means the effects of this unknown factor are less, nearer the branch tips. This may explain why growth of the tips seemed unaffected by the slow death of the colonies but this can only make sense if the level of integration within the AS colony is low. This is unlikely to be the case in acroporids because of specialization of axial and radial corallites (Soong and Lang 1992). However, the same authors also pointed out that the bases of branching corals (which are often composed of senile, infertile polyps which have limited access to light and water, and are subject to encroachment by fouling organisms) may be more energetically expensive to maintain, especially for a coral involved in a competitive interaction. Redirection of growth away from competitive dominants has been described previously (Romano 1990). If growing upward was the only available option, then the increased allocation to branch growth at the expense of maintenance of the base may be required. Note that AS colonies are significantly taller than AP colonies (average 24 cm vs. 10 cm, respectively; p<0.001). Note also that for asexually reproducing corals, bases may need to be maintained only long enough for the branches to be of sufficient size to be viable propagules. At any rate, this phenomenon needs to be further investigated. The slower water flow in the AP patches may affect successional processes in the patch such that conditioning o f the substrates for other coral recruits will be delayed, if not prevented. This conditioning may just be chemical, e.g., the leaching of noxious 71 substances out of the settlement tiles used. Pocilloporid planulae are available year round and yet the tiles (as indicated by preliminary sampling) remained unoccupied for at least a month. Conditioning may also involve species such as coralline algae that may modify the texture o f the substrate and/or provide settlement cues to corals (Morse 1992)- This scenario assumes that succession on such surfaces follows the facilitation model (Connell and Slayter 1977). The differences in conditioning may thus be the reason three-month old settlement tiles positioned inside AP patches are fouled only at the same level that is achieved in a month outside these patches, whereas tiles conditioned outside the patch remained comparable to one another even after some are transplanted to the AP patch interior (Figures 6 and 14). Whatever the mechanism, it is apparent that the presence o f large aggregations of AS and AP have important implications on the local environmental conditions in their "neighborhood." This property of these corals deserves more investigation for implications on two related concepts, that of ecological fields and o f characteristic size, in corals: Ecological field theory, originally described in terrestrial ecology (Wu and others 1985, Walker and others 1989) seeks to predict the outcomes of plant-plant interactions on the basis of the plant's influence on its local environment. A premise is that in the course of normal functioning, the plant acquires some measure of control (both direct and indirect) over the resources it needs to exploit. Such a concept will be useful as 72 topics concerned with diffuse competition (i.e., simultaneous interactions of individuals with several other individuals and species) are pursued. Characteristic size may be defined as the size or age at which an organism is able to acquire the full measure of capabilities needed to maintain its ecological field. This may also be defined as the size at which the correspondence between the potential and realized niche is greatest. Most digestive hierarchies are created based on one-to-one pairings of colonies of approximately the same size, both in controlled conditions and in the field (Logan 1984, Dai 1990, Tanner 1992; see also Lang and Chomesky 1990). However, digestive hierarchies give only a snapshot of the potential outcomes of competitive interactions. As this, and other studies have shown, an organism has a plethora of tools it can use during these interactions, the availability of which depends on the size the organism can achieve. Competitive pairings should thus include experiments in which the common size range of the species is used in the pairings just to determine how the outcomes of one-to-one interactions affect distributions at different spatial scales. The extent to which competition will be important in a community will be a function of the degree to which the component species can utilize its full range of competitive responses. In a diffuse competitive environment of many-to-many interactions, this may ultimately mean that individual species will be limited to certain size-ranges that are the result of the balance between their competitive ability (=size) and 73 that of its neighbors. Transect studies, using the now common "life-form" method (Bradbury and others 1986, Licuanan and Gomez 1988), have shown strong correspondence in multivariate patterns between cover and counts data of reef benthos. This shows that these groups do occur at certain limited size classes across a variety of reef habitats, leading to the correlation between the number of colonies or patches and their cover. Nelson (1992) suggested deviations from this pattern with cover and number of occurrences may be due to disturbances such as cyclones (typhoons), but as seen in her Figure 1, there is still a correlation between cover and number of patches even in the cyclone-damaged reefs that she studied. If anything, one lesson that this study provides is that if we are to properly appreciate the importance of one-to-one interactions in corals in a many-to-many world, we should look at these organisms both as a whole and as a collection of interacting modules. As was mentioned in the Introduction, a balance between mechanistic and phenomenological approaches need to be achieved in ecological studies. The field experiments just described illustrate how even a simple question about coral distributions can lead to a complex web of interactions and interdependencies that generate more questions than answers. Although some level of mechanistic understanding of natural systems is required, it is also essential to explore where the present understanding leads 74 to, i.e., to know what we can and we cannot expect from a given system with our limited knowledge. Such exploration is the topic of the next section. The Computer Simulations The issue of scale specificity of the generalizations made in the last section introduces the question of how factors such as competition, reproductive patterns, and disturbances interact to affect distributions and abundances of corals, and how these effects are propagated to higher spatial and temporal scales. Related issues are whether the statistically significant findings described in the field studies have any demographic significance to the populations studied (see Weinberg and others 1986) and whether they influence the distribution patterns in the community concerned. Since these are difficult to examine experimentally and would require more time and resources than are currently available, computer simulations are used here to synthesize aspects of the field studies and produce testable hypotheses, rather than definitive conclusions, about the community examined. The specific objectives of the simulations presented in this section are: 1 . To estimate the relative contributions of sexual and asexual reproduction in maintenance of AP and AS populations; 2. To assess the role of competition, reproductive patterns and disturbances on the spatial distributions of the colonies and patches of these species. 75 To achieve these objectives, two modelling approaches, projection matrices and cellular automata (see review of literature on pages 18-23 for background information), are combined to draw on the strengths of both and develop a spatially-explicit, individual-oriented, multi-species simulation of the reef community that retains properties of population-level models. Material and Methods The following is the summary of procedures that were followed to produce the simulation models of the reef community studied: 1 . Size-specific mortality, colony growth (i.e., expansion rates of the colonies and not just the branches), and fragmentation rates were estimated by monitoring tagged AS and AP colonies in the field and noting their fates over time. 2. The field-derived parameters were then used to construct projection matrix models (one for AS and another for AP) o f the coral populations studied using the size classes defined on page 81. Population growth rate (as measured by X , the largest eigenvalue of the projection matrix), sensitivity, and elasticity were computed analytically from the projection matrices (after Caswell 1989) for comparisons between the two species (results on pages 92-100). 3. The asexual reproduction rate (SC3 to SC2 transitions) or the sexual reproduction rate (SC3 to SC 1 transition) was then adjusted to produce zero population growth (X of 1.0). To infer the most likely mode of reproduction, stable size distributions produced by 100-year projections of these adjusted 76 matrix models were compared with size distributions observed in the field (results on pages 100-102). 4. New projection matrices, with modified size-class intervals, were constructed for both species using the field data so that their size classes were made equivalent; e.g., a SCI AP was redefined to be of the same size as a SCI AS. Sexual or asexual reproduction rates of these new matrices were also adjusted to produce zero population growth to avoid filling the model reef (described below). 5. The new, adjusted transition matrix probabilities were then used as the basic rules defining the fates of each coral in a cellular automaton model of the AS-AP community. For example, the transition matrix element that defined the proportion of the AS population that grew from SC2 to SC3 (row 3, column 2 in the matrix) was used as an estimate of the probability that a single SC2 colony will grow to SC3 in the next time step. The model of the community was thus made probabilistic and individual-oriented, compared to the deterministic, population-level matrix models used to construct it. Additional rules governing cell transitions or fates were also used for simulating recruitment (random or aggregated), competition (digestive or by peripheral encircling), and disturbances. These are summarized in Table 4 and the results presented in pages 102-124. Note that the transition probabilities used were derived from the monitoring of corals that are different from those used in the field experiments described earlier. Since the simulations were designed to investigate factors affecting distributions 77 Table 4. Summary of the rules used in the various cellular automaton simulations. The neighborhood of a given cell is defined as the eight cells that surround it. 1 . Competition: The coral occupying a given cell dies if: a. It is a SCI AP that has at least one AS SCI or SC2 neighbor b. It is an AS colony (regardless of size) that has at least five SC2 or SC3 AP colonies around it. 2. No competition: The appropriate transition probabilities defined in the matrix models are used (see Table 6). The matrix used depends on whether sexual (planular) or asexual (by SC2 fragments) reproduction is being simulated. 3. Random settlement: A random-number generator provides a row and column number, and if the corresponding cell is available (i.e., has no occupant), settlement occurs there. 4. Aggregated settlement: New recruits (SCI planulae or SC2 fragments), depending on the mode of reproduction being simulated, are allowed to occupy available space adjacent to at least one adult or at least two SC2 fragments of the species. The settlement next to fragments was necessary to allow settlement to occur when the eight cells surrounding an SC3 adult is filled. 5. Physical disturbance: Individual cells (regardless of whether it is empty or occupied) are selected by a random number generator. If that cell contains an SC2 or SC3 colony of either species, the corals are made to shrink to the next smaller size class. The program prompts the user for a number of disturbances to be introduced per year in the model. The range of cells chosen for the disturbance can also be specified. 6. Others. Death of corals due to "overexposure" occurs when a colony is surrounded by seven or eight empty (space) cells. 78 and patchiness of corals, an index of segregation (Pielou 1962) was used to help judge whether corals of the two species were randomly mingled. Details of all these procedures are described below, but may be skipped for later reading. ♦ Field collection of parameter estimates To estimate parameters (probabilities of growth to larger size classes or shrinkage to smaller ones, recruitment and mortality rates) of the transition matrices, coral colonies were tagged, mapped, and monitored at approximately regular intervals in the First and Second Plateaus. For AS, the corals were monitored in their patches in quadrats 0.5 m wide (to allow for measurements without disturbing adjacent colonies in the quadrat) and with length set parallel to depth contours. The quadrat was 3 m long in the First Plateau (16 m depth; between Block A and Block F in Figure 3) and initially contained 29 colonies while the Second Plateau quadrat was 5 m long (10 m depth; near Blocks D and E) and had 41 colonies initially. Both quadrat lengths were set based on the maximum number of measurements possible within the bottom time allowed by SCUBA no-decompression limits. Parameter estimates from the larger (Second Plateau) quadrat were used, with the data from other quadrat (First Plateau) used only for comparisons. Monitoring was conducted five times, from August 17, 1993 to July 13, 1994. During each monitoring visit, length and largest perpendicular width (parallel to the substrate) of each fragment or colony was measured in situ (i.e., without removing it 79 from the substrate) using a flexible ruler. Area was later derived from these measurements using the formula described earlier (on page 33, from Yap and others 1992). Since colonies / fragments were often in contact with each other (thus obviating the use of photographic techniques), they were distinguished by movements when different colonies were slightly jiggled. Colonies were thus defined on the basis of skeletal connection regardless of whether living issues were contiguous or not. Colonies, which have living tissues separated by fouling organisms at the surface, have been observed to contain coral tissues within skeletal cavities under the fouled surface, thus justifying this definition of colonies (especially since non-destructive sampling was necessary). Also, skeletal connection means physical disturbances will likely affect all points of the colony equally. Besides the measurements, parentage of new fragments encountered during each monitoring visit was inferred from their relative positions and the scars left by the fragmentation process. These new fragments were then tagged and the maps updated. The difficulty in distinguishing colonies of AP within their patches required a different approach for this coral. Fragments of three height classes (approximately S, 10 and 20 cm) were carefully extracted from the patch, tagged, and scattered haphazardly in the release area (devoid of corals) of the experimental block that includes the source AP patch (see design of the transplantation and recruitment studies described earlier). 80 Measurements and monitoring methods were the same as in AS. Note here that initial height becomes measured length since the AP fragments were most stable lying on their sides with some branches buried in the sediment. Naturally produced fragments at the periphery of AP patches have a similar orientation; therefore the method was not deemed unusually artificial except for the distance from the parent colony or patch. The movement to the release area was needed to avoid confusion with the many naturally-produced fragments encountered up to 3 m from the patch margins. Fragments were also moved to the rubble areas in the middle of the AP patches in a subsequent experiment and, in this situation, the distance from the source colony was greatly reduced. A total of 84 AP fragments was monitored in three blocks (Block A in the First Plateau, and Blocks C and D in the Second Plateau). Size (in terms of area) of each colony or fragment was computed using the same formula described above. * Construction of transition matrices for AS and AP The transition matrices for each species were estimated from tabulations of sizes of each tagged colony over time. Both species were arbitrarily divided into three size classes each (see Table 5), to represent small fragments and sexually produced propagules (SCI), medium-sized colonies and fragments produced asexually (SC2), and adults (SC3). The size interval of each class was set such that the three size-classes for each species had approximately equal members, and that distribution of counts per size class is approximately unimodal. Moloney (1986) describes a more sophisticated algorithm for deciding the number and sizes of categories, but this procedure was 81 Table 5. Upper limits of size classes (in square cm) used in the projection matrix simulations. AP AS Size class 1 (SCI): Small fragments & sexually-derived recruits 10 100 Size class 2 (SC2): Medium sized colonies and asexually produced propagules 70 300 Size class 3 (SC3): Large adult colonies over 70 over 300 82 deemed unnecessary for an exploratory analysis such as this. Note that at this stage, the size classes for each species were defined independently, and so SCI colonies of AP may not be of the same size as AS SC 1 colonies. Counts of transitions over all the sampling periods were then aggregated, and a maximum likelihood estimate of each transition probability was computed by dividing the number of transitions from a size class j at time t to another at time t+1 by the total number of colonies in j at time t (Caswell 1989, p. 81). For example, if there were 10 SC2 colonies at time t and 3 of these were found to have shrunk to SCI at time t+1, then the estimate for SC2 to SCI transitions for the species is 0.3. Two sets of transition matrices were produced from the field data. The first matrix (Table 6) summarizes the fates of tagged colonies found in the monitoring sites but does not consider recruitment into the quadrats, both sexually- and asexually produced. Thus, the SC3 to SCI transitions (row 1, column 3) in these matrices refer only to the proportion of SC3 colonies that shrank to SCI in the next time step, but do not include any reproduction of SC3 to give rise to SCI individuals (compare with Figure 1). These "no reproduction" matrices were constructed for both species. In AS however, fragmentation occurring in the quadrats was obvious and the "parentage" of individual fragments could be inferred from the positions of the breakage scar and the orientations of the fragments. An additional "asexual reproduction" matrix was thus constructed for AS for comparisons. Although fragmentation of AP was more 83 Table 6. Transition matrices used in the projection matrix models. a. Acropora subgalbra (AS) matrix without reproduction: SCI SC2 SC3 b. Anacropora puerlogalerae (AP) matrix without reproduction: SCI SC2 SC3 a. Acropora subgalbra (AS) matrix with reproduction: SCI SC2 SC3 SCI SC2 SC3 0.5625 0.5238 0.0588 0.0417 0.4762 0.4510 0.0000 0.0476 0.9216 SCI SC2 SC3 0.6316 0.3519 0.0476 0.0789 0.4815 0.2381 0.0000 0.0370 0.6429 SCI SC2 SC3 0.5625 0.3810 0.0196 0.0417 0.4762 0.2157 0.0000 0.0476 0.7255 84 common, it was difficult to trace parentage (for example, whether a given fragment came from an SC2 or a SC3 coral). Thus, no asexual reproduction matrix was constructed for this species. For all matrices, measurements at approximately regular intervals were used in the computations to ensure comparability in the rate of transitions. In effect though, seasonality is averaged out. Also, it was assumed that the eight month data for AP is sufficient to estimate annual transition rates (AS had a full year of monitoring). This assumption is needed since the two species are made to operate in the same time step in the community models. Sensitivity and elasticity values (see pages 19-21) of the transition matrices were computed analytically (following the formulae of Caswell 1989) to show the degree to which changes in the individual transition matrix parameters affect population growth rates. Note that the AS asexual reproduction matrix was used only for comparisons and not in the cellular automata. ♦ Evaluating contributions of asexual and sexual reproduction Adjustments were made in the matrix elements of the no-reproduction matrices to examine whether the two species reproduced sexually or asexually. The first set of scenarios involved adjusting the SC3 to SCI rate (row 1, column 3 in the transition matrices) to achieve zero population growth. This adjustment was meant to simulate a stable population of the species, with the adults (SC3) reproducing by means of sexually derived (i.e., pianular) recruits (SCI). The second set of scenarios involved "tweaking" 85 the fragmentation+shrinkage rate of SC3 to SC2 (row 2, column 3) to produce zero population growth. This was meant to simulate the species reproducing by fragmentation / asexual reproduction, with the largest (SC3) colonies producing SC2 fragments. The two matrices per species were then projected for 100 years until a stable size-distribution was achieved. These projected size structures were then compared with size structures observed in the field to see which of the former was closer with the latter (see pages 100-102 for results). ♦ Adjustments of size class intervals to a common definition for AS and AP To allow the integration of the transition data and other information into a general model of the reef community, size classes of both AS and AP transition matrices need to be defined similarly such that an individual AP SCI colony should be of the same size as an AS SCI individual. Since AS is larger than AP, this was achieved by using the SCI limit of AP (<10 cm2 ) and the SC3 boundary of AS (>300 cm2 ). Field data were then retabulated and recomputed following the procedures outlined in step # 2 (Construction of transition matrices) to produce two new, no-reproduction matrices. These matrices were further adjusted so that zero population growth was achieved, mainly by increasing asexual reproduction (row 2, column 3) in the transition matrices. Sexual reproduction rate was adjusted instead for some recruitment simulations described below. ♦ The cellular automaton model A cellular automaton model was constructed using the new, adjusted matrices described in the last section. The two-dimensional reef surface was represented as an 8 6 80-cell x 80-cell array of square cells, with each cell representing a unit of space that was either occupied by a coral (either AS or AP) or empty. Besides the occupancy of a given cell, the size of the occupying coral was represented by using the same common size classes (three for each species) in the adjusted transition matrices. Since it was assumed that only one coral, of either species, can occupy a cell or unit space, it was also assumed that the size of the coral (whether it is an SCI recruit or an SC3 adult) does not preclude it from interacting with the residents of the eight neighboring cells around it. For computational purposes, the model reef (i.e., the two-dimensional matrix) was wrapped into the shape of a torus or donut, i.e., the first and last rows of the model reef meet, likewise for the first and last columns. This means that all cells have eight neighbors since comers and edges have been eliminated. At each time step, the fate of each cell was determined from its previous state and the states of its eight neighbors. For example, a cell occupied by an SC2 AP colony could become occupied by an SC3 coral in the next time step (i.e., the coral "grows") with the probability defined by the adjusted transition matrix. In other words, the transition matrix was considered as containing estimates of the probability that each colony will suffer a certain fate, rather than as estimates of the proportion of a given population that experience that fate. Assume, for example, that the transition matrix element that defined the proportion of the AS population that grew from SC2 to SC3 (row 3, column 2 in the transition matrices) is 0.8, i.e., 80% of all AS SC2 corals grow to SC3 in a year in the projection matrix model. In a cellular automaton, the probability that a given AS SC2 individual colony will grow to SC3 in a year is set correspondingly at 80%. Recruitment in the automaton model was determined similarly. Fecundity in transition matrices (fin Figure 1) is defined as the number of juveniles produced by each colony in a given size class. The same definition was applied in the automata, except that instead of having every colony of the same size class producing the same number of juveniles each year, the numbers of juveniles produced by each coral were allowed to vary in the automaton model as a Poisson variate with a mean identical to the relevant number used in the matrix models. Thus, for example, if the projection matrix describes that each AP SC3 coral produces an average 4.1 SC2 fragments per year; in the cellular automaton model, this number is used to compute the probability that each colony produces no fragments, one fragment, two fragments, and so on, using the Poisson probability distribution (modified from Pagano and Gauvreau 1993): where jc is the number of asexual fragments produced, y the mean fragmentation rate, and P(X=x) is the probability X that x fragments are produced. The resulting frequency distribution is then used with a random number generator to assign a integer number of recruits produced to each reproducing coral. Thus different SC3 colonies produce different numbers of fragments but the average number for all AP SC3 colonies 88 will still be 4.1. The same procedure was used for determining the number of planulae each coral produced. For program simplicity, settlement was assumed to occur simultaneously for all colonies of a given species, but the sequence in which AS and AP colonize space was randomized. Thus, for example, all AP recruits can colonize available space at the same time for a given year before all AS recruits, but in a subsequent year AS may be first to colonize space. This randomization was introduced so that a species does not gain advantage by consistently preempting settlement space (except in one simulation described below). All planulae and fragments were assumed to recruit, that is these were not subject to mortality before recruitment. The automaton model of the community was thus made individual-oriented and probabilistic, as opposed to its deterministic, population-level precursors. Despite these modifications and assumptions, the basic cellular automaton model just described still exhibits properties of the two projection matrix models used to construct it (see Appendix 7). Various combinations of other rules were used to modify the basic transition probabilities defined by the AS and AP projection matrices. These rules, summarized in Table 4, allowed for the simulation of the effect of competitive hierarchies and networks, recruitment patterns, disturbance levels, and various combinations and intensities thereof, in the automaton model. Most of these rules involved changing transitions probabilities 89 of each cell depending on the neighborhood conditions, i.e., the number, species and size of corals in the neighboring cells. For instance, to simulate peripheral encircling, a cell occupied by an AS colony could be made to revert to space (i.e., the coral dies) whatever its size, if more than four of the eight neighboring cells are occupied by AP adults (SC3). Note that the transition probabilities in all simulations were derived from monitoring of corals separate from the field experiments described earlier. The rules used are just qualitative, not quantitative, approximations of the results of the field experiments and were designed to simulate those processes observed or deemed likely to operate in the study sites. The ability of competition, recruitment patterns, and disturbances to generate patchy distributions of single-species stands of AS and AP were initially evaluated individually. Intensity of competition was varied by changing the number and sizes of AP needed to kill AS colonies by peripheral encircling, and by the effect of AS on small AP SCI fragments. The effects of recruitment patterns were explored by simulating both sexual (to SCI) or asexual (to SC 2) reproduction, by using the appropriate transition matrices described earlier, and by making one or both species either settle adjacent to conspecifics or randomly. Various levels of physical disturbance (assumed here to be strong waves or boat anchors that cause shrinkage of SC2 and SC3 colonies to the next smaller size class) per year were also simulated. In addition to these rules, various initial configurations of the reef (i.e., various densities and arrangements of corals) were used to start the simulations. The most commonly used initial configurations are described on 90 page 102. Unless stated otherwise, the results described are for model runs in which the starting configuration is of randomly arranged corals, with no competition nor disturbance, and with settlement occurring at randomly selected empty cells. To quantify the effects of competition, reproductive patterns, and disturbances on the distributions of individual corals, a test for segregation of species was used (Pielou 1962). This segregation index was computed from 10 randomly selected transects (i.e., columns in the model reef) sampled every model year. This test involves comparing the observed length of runs (i.e., uninterrupted sequence of cells occupied by the same species) with what would be expected if the two species were randomly mingled (as generated from a geometric distribution). An index significantly less than unity indicates that the individuals of the two species were found in patches, forming longer runs than expected. An index significantly larger than one indicates individuals of two species "mingling" more with each other than with conspecifics. Graphs of this index per model year, along with its 95% confidence intervals, are presented to show how this index behaves over time for each simulation run. Note however that mingling as defined by this index is based on run lengths of the two species relative to each other, whereas the random initial configurations used here were randomly distributed relative to space. Hence, due to the difference in statistical distributions mentioned, the random initial reefs used here yield indices significantly higher than unity at first. Examination of the reef maps shows this distinction is unimportant in these analyses. 91 Results ♦ Population projection matrices: Growth, sensitivity, and elasticity Table 7 summarizes the transition matrices and population growth rates (X), both "no-reproduction" (Tables 7a and b) and asexual reproduction (Table 7c), of AS and AP. Population growth rates (X) for both no reproduction matrices are below 1.0, suggesting that population size would decline quicidy without reproduction. The similarity in X of both species (0.766 for AP and 0.773 for AS, or an intrinsic rate of increase of about -0.27) suggests that both species would decline in numbers at approximately the same rate if neither reproduced. No symmetrically-shaped small colonies with branches emanating from the colony base (suggesting growth from planulae) of AS or AP were seen during the study, but fragmentation was evident for both. If the asexual reproduction matrix is considered for AS, the monitored population showed little change in numbers, with a X of 0.971. Note that the data for this matrix came from the Second Plateau, where widespread partial and complete mortality was observed on January 1994 (see pages 47-49). Therefore, despite the death of a nearby patch, and partial mortality within the monitoring quadrat itself, the effect of these deaths on the monitored population size was small and localized. Sensitivities of the no-reproduction matrices of the two species are shown on Figures 21 and 22. Population growth rate for AS appears most sensitive to changes in transition probabilities of SCI and SC2 growing to SC3, and SC3 corals remaining in that size class. Transitions to SCI had the lowest sensitivities, being almost negligible 92 Table 7. Transition matrices used in the projection matrix models and their population growth rates (lambda). a. Acropora subgalbra (AS) matrix without reproduction: SCI SC2 SC3 SCI SC2 SC3 0.5625 0.3810 0.0196 0.0417 0.4762 0.2157 0.0000 0.0476 0.7255 lambda=0.773 b. Anacropora puertogalerae (AP) matrix without reproduction: SCI SC2 SC3 SCI SC2 SC3 0.6316 0.3519 0.0476 0.0789 0.4815 0.2381 0.0000 0.0370 0.6429 lambda=0.766 a. Acropora subgalbra (AS) matrix with reproduction: SCI SC2 SC3 SCI SC2 SC3 0.5625 0.5238 0.0588 0.0417 0.4762 0.4510 0.0000 0.0476 0.9216 lambda=0.971 SC1 SC2 SC3 SC1 0.062 0.033 0.033 SC2 0.315 0.165 0.166 SC3 0.767 0.767 0.772 Figure 21. Sensitivity of Acropora subglabra transition matrix (ignoring reproduction) to small changes in transition probabilities. The matrix is most sensitive to small changes in transitions to SC3 and in retention in that size class . 94 SC1 SC2 SC3 SC1 0.199 0.093 0.054 SC2 0.666 0.310 0.181 SC3 0.840 0.840 0.491 Figure 22. Sensitivity of Anacropora puertogalerae transition matrix (ignoring reproduction) to small changes in transition probabilities. The matrix is most sensitive to factors affecting growth into SC2 and SC3. 95 compared to transitions to SC3 (Figure 21). A similar trend of exists in the sensitivity table of the AS asexual reproduction matrix (Figure 23) except that transitions to SCI are much less important. With the AP no-reproduction matrix, sensitivity values show that change in its X would be largest if transition rates from SCI to SC2 and from SCI and SC2 to SC3 are altered (Figure 22). Note also the decreasing trend of sensitivities, with growth to the larger size classes being more sensitive than loop growth (retention in the same size class) and loop growth being more sensitive to shrinkage. This trend is seen in all size classes. The latter trend also exists for the AS projection matrix sensitivities (with or without reproduction). Both species therefore depend on fast growth to larger size classes although AS is more affected by factors affecting adult (SC3) survival. Note that sensitivity is measured based on absolute changes in growth rate (A ) for small changes in each transition matrix element. The elasticities of the matrices studied (Figures 24 and 25), i.e., the proportional changes (in this case, a 10% increase) of growth rate relative to proportional changes in each transition matrix element, showed a different pattern compared to the sensitivities. Elasticities for both species emphasize the importance of retention within a given size class. With AP however, A is most affected by a proportional increase in the survival of SCI fragments / colonies (Figure 24), whereas with AS it is the retention of SC3 that is important (Figure 25). This indicates that population growth in AP would be most affected by factors acting on the 96 SC1 SC2 SC3 SC1 0.014 0.010 0.010 SC2 0.141 0.099 0.096 SC3 0.918 0.918 0.887 Figure 23. Sensitivity of Acropora subglabra transition matrix (including fragmentation or asexual reproduction) to small changes in transition probabilities. Like the matrix that excludes observed reproduction (Figure 21), the matrix is most sensitive to small changes in transitions from SCI and SC2 to SC3 as well as survival of SC3 colonies. 97 SC1 SC2 SC3 SC1 4.68% 0.82% 0.04% SC2 0.86% 2.11% 0.29% SC3 0.00% 0.33% 2.48% FROM SCI Figure 24. Elasticity, or proportional changes in population growth rate of Anacropora puertogalerae in response to a 10% increase in transition matrix probabilities. A 10% change in the retention rate in SCI, not SC3 as in Acropora subglabra (Figure 25), will have the most impact on population growth for the species. No transitions from SCI to SC3 were observed, hence elasticity is set to zero for this element. 98 SC1 SC2 SC3 SC1 0.63% 0.17% 0.01% SC2 0.17% 1.16% 0.45% SC3 0.00% 0.47% 7.79% FROM SC1 Figure 25. Elasticity, or proportional changes in population growth rate of Acropora subglabra in response to a 10% increase in transition matrix probabilities (excluding observed fragmentation or asexual reproduction). A 10% change in the retention rate in SC3 will have the most impact on population growth for the species. No transitions from SCI to SC3 were observed, hence elasticity was set to zero for this element. 99 smallest fragments and sexually-derived juveniles, whereas in AS it is the survival of the largest colonies that will have the largest impact on population growth rate. ♦ Population projection matrices: Evaluating modes of reproduction with size-structure If sexual reproduction rates (i.e., by SC3 adults to SC 1) were adjusted in both projection matrices such that abundances were maintained (A=1.0), each AS adult (SC3) colony would need to produce an average of 27.2 SCI colonies, while each AP adult (SC3) colony needs to produce 18 SC colonies per year. If AS reproduced exclusively by sexual means, the stable size-distribution would be 91% SCI, 8% SC2, and 1% SC3. AP would have a size structure o f 85% SCI, 14% SC2, and 1% SC3 with this sexual reproduction assumption (see Figure 26). If both species reproduced only by asexual means (fragmentation), AS adults need to produce 2.6 SC2 fragments per year, whereas the rate for AP should be 4.03, in order for both to maintain a constant population size (A=l .0). This would mean stable size distributions with larger SC2 proportions (43% SCI, 49% SC2, and 8% SC3) for AS, and 47% SCI, 48% SC2, and 5% SC3 for AP (see Figure 26). The observed initial size structure for the First Plateau AS quadrat (data not used in estimating AS transition matrix used in these models) was 35% SCI, 48% (SC2), and 17% (SC3), which is closer to the asexual reproduction proportions (Figure 26). For AP, painstaking dissection of a small part of the patch (an operation that required four hours of bottom time) yielded a size structure of 10% (SCI), 41% (SC2), 1 0 0 1 0 0 % A. subglabra OBSERVED ASEXUAL SEXUAL 90% ----------------------------------------------------------------------------------------------------------------- A.puertogalerae iffl'itiiiii OBSERVED ASEXUAL SEXUAL Figure 26. Observed and projected size-structures of Acropora subglabra and Anacropora puertogalerae if they reproduced mainly by asexual means (fragmentation of SC3 to produce SC2) or sexual means (SC3 producing planulae that start as SCI colonies). 101 and 49% (SC3) if colonies with fresh breaks are excluded. This is even larger than that of the asexual reproduction simulation (Figure 26), suggesting transplantation outside the patch led to overestimation of mortalities of large colonies, or inaccurate size-structure data from the field sampling. ♦ Cellular automata: Competition and coral spatial distributions The initial reef configurations used in the competition simulations included high (Figure 27) and low densities of randomly arranged corals (Figure 28), and those with corals in bands or zones (Figures 29 and 30). Regardless of the initial configurations, the levels of competition used in the simulations were unable to produce any significant patch formation on the model reef if recruitment was random (see Figure 31). For example, the reef shown on Figure 27 looked like Figure 32 100 years later, also showing no patterning in the intervening period. Bands on the reef shown in Figure 29 disappeared by year S (Figure 33), showing how initial configuration had no long-term effect on distributions in the competition simulations. Changes in the number of AP needed to encircle and kill AS, modification of the effect of AS fragments and adults (SC2 and SC3 respectively) on AP SCI, and the order of settlement of the two species also did not lead to the formation of patches (not shown), although these did affect the rate at which one or the other species went extinct. ♦ Cellular automata: Recruitment patterns and coral spatial distributions Aggregated settlement, either to SCI (as planulae) or SC2 (as fragments) for both species quickly led to patchy distributions, even if initial reef configurations had randomly distributed corals (see Figures 34-38). Patches formed by sexual reproduction 1 0 2 Figure 27. Initial reef configuration with random distribution and high density of corals (500 individuals per size class). Only the first 50 rows and 70 columns of the 80x80 reef are shown. L e g e n d : s p a c e , A S : * : S C 1 , ;ij:S C 2 , | : S C 3 , A P : 1 : S C 1 ( 2 : S C 2 , 3 :S C 3 » R u n # 0 C o v e r : S 0 0 5 0 0 5 0 0 5 0 0 5 0 0 5 0 0 r e s p e c t i v e l y . Figure 28. Initial reef configuration with random distribution and a low density of corals (100 individuals per size class). Only the first 50 rows and 70 columns of the 80x80 reef are shown. L e g e n d : - : s p a c e , A S : * : S C 1 , ii;:S C 2 , | : S C 3 , A P : l : S C l , 2 : S C 2 , 3 :S C 3 » R u n # 0 C o v e r : 1 0 0 1 0 0 1 0 0 1 0 0 1 0 0 1 0 0 r e s p e c t i v e l y . -1-2- -1 2 *- . 3— * . -* 2- -2----- *— -III----- 1- -1 11- -1-*- - 3 - 3 — -1 *- - 3 -------- 1 - -2- 1 --------3 -------- 1 3 - -1 *--- !---- 2- 2 2- —3-*— -2 -* |— 3 --------* -------- - 1 - 1 2 - * - - 3 3 — 3 - -11- 104 Figure 29. Initial reef configuration with corals distributed in several bands or zones. Only the first 50 rows and 70 columns of the 80x80 reef are shown. L e g e n d : - : s p a c e , A S : * : S C 1 , i;j:SC 2, | : S C 3 , A P: 1 : S C 1 , 2 : S C 2 , 3 :S C 3 » R u n # 0 C o v e r : 3 2 0 3 2 0 1 6 0 3 2 0 3 2 0 1 6 0 r e s p e c t i v e l y . 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 2222222222222222222222222222222222222222222222222222222222222222222222 1111111111111111111111111111111111111111111111111111111111111111111111 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 2222222222222222222222222222222222222222222222222222222222222222222222 l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l l 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 2222222222222222222222222222222222222222222222222222222222222222222222 1111111111111111111111111111111111111111111111111111111111111111111111 105 Figure 30. Initial reef configuration with corals distributed in a few bands or zones. Only the first 50 rows and 70 columns of the 80x80 reef are shown. L e g e n d : - : s p a c e , A S : * : S C 1 , jjj:S C 2 , | : S C 3 , A P : 1 : S C 1 , 2 : S C 2 , 3 :S C 3 » R u n # 0 C o v e r : 8 0 8 0 4 0 8 0 8 0 4 0 r e s p e c t i v e l y . 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 3 - 2222222222222222222222222222222222222222222222222222222222222222222222 1111111111111111111111111111111111111111111111111111111111111111111111 106 200 1.60 140 1.20 5 100 060 0.60 HIGH DENSITY, RANDOM REEF o .« o 0.20 0.00 “ " e s a s s s s s s s s s s s s s a s g HIGH DENSITY, ZONED REEF o » ° » a s s s s s 8 8 8 ° R 8 S 8 8 8 Vv w LOW DENSITY. RANDOM REEF O K > O IO RSSS $ 3 8 8 8 S £ j e S S 8 8 8 200 1.80 1.60 1.40 5 1 120 I 1.00 | 0 80 060 040 0.20 0 0 0 QoOo LOW DENSITY, ZONED REEF o < o e M > S S 8 ! ! ! S S 8 S 8 S e K 8 S 8 S 8 Figure 31. Segregation indices for various competition simulations with random recruitment to SC2. No patches were formed, regardless of initial density or arrangement of corals, over the 100 year projection. Paired lines are 9S% confidence intervals. Indices significantly lower than unity indicate aggregation of conspecifics. Figure 32. Map of model reef after 100 years of the competition simulation with random settlement to SC2. See Figure 27 for initial reef configuration. Only the first SO rows and 70 columns of the 80x80 reef are shown. L e g e n d : s p a c e , A S : * : S C 1 , |||:S C 2 , | : S C 3 , A P : 1 : S C 1 , 2 : S C 2 , 3 :S C 3 » R u n # 1 0 0 C o v e r : 1 1 5 1 1 5 5 2 2 5 1 0 0 4 0 7 6 0 r e s p e c t i v e l y . — 2 -1H1SS -1S1I1? - I l l — iiiSSl - B l i l l i — 3 - - - - - i l l — l i l i l l - |2 — III- - - - - 2III— I - - - - - - - - III2 ■ - - 2 - 2 III | | | 2 | | | - ! | | - - i | i - - - - - - - - - IH —III—2 l l l ———- - - - l l l l l l - I I ! - I l l * - - - - - - - - - 2 - - - - 1 1 1 - 1 2 - 1 1 * 1 1 1 1 1 - - - - - 1!! I l l - - - I I I p - _ - - - - - H l 3 1 - - 2 - - l l l 2 | - - - - - - H I- - - - 1 2 - - - - 1 —2 —— III- - - - - - - - - - - - - - - - - llllll— 111312-111111 lilill— - 2 ----------------------2— I -------------- IIIIII2 - 3 - ------------ --III-IIIIII-2— 2 --j;l------- 2---------- H I------ H I ill - -H I -HI— 2 3 — I I I -llllll-------------1 -IH --2-------- 1 — jii--------- llll H I 1 1 1 * 1 1 1 ------ 2 2 — 2 2 ---------- 2I I I - I - - - 2 -III2 - * — 7--------I I I — llllll 2—| - * ----------— ---2III-III-----------|- - d i! 2 - 2 - - jlill!-------- — * ----------- --— H i— H I-----------3 * ----— — 2— — — ( 2— — H I—— H illi---------------------- I 7H IIII------ 2 1 1 1 - 1112-7-------- *1 1 1 ------1 — 1 1 1 !— - — ------*2--------- III — I I I - * — II I-I II — ill- 2 * * -----------1 — H llll - * — III - - I — III— * - I I I - r P ^ - i i i | | | - - - - - - - - - - - - l l i i — - i i i i i l — — III— i l l - i l l i i l — n i ? - ~ - 2 - - - ~ l i | T ^ * - - - - - l- lll 2liiBii— 1— -ill - * - 2 — 1 - 1 1 1------- 1 1 1 — ini-ill— huh— m 2-------------------- 1 ----------iiii-Hiiiii------- p m - | l l l l - 222 | ; 2 7 - - 2j l | - - — - - | - - - - * — 1 ||“ — 2 - 2 r j l j — — - - 2 - l j l - l l l l l l — — - p l l j j III —III—3 - - - - - - I I I - - - - - - — - - - I I I - — III - 2 - - - - l l l l l l ——2 —III * - - - - - ^ 3 - I I I - l l l l — — - I I I - - — - 2l l l ! i - 2 — j - --III 1----- 2------III------------------H I-1 1 1 -------|- lli— H I------- 11122---------------- 1112221111— 1 1 1 ---------- - — III— H I-*— III------H I— III2 2 -IH-------HI-2 - --------2-3111-ill-* 2 * - llllll---------------- 2 ------H i— 2 — -llllll-— lOlll----------- 3-----lll|2lll— III— 2 1 1 1 — 2 1 1 1 ------- 2---------------- 1 1 1 1 1 1 1 1 1 2 -------- 2------------ 2-ill - I I I - I I I - * - 2 — I I I - 2III- - - l l l l - l l l - — III- | - - - - - - - - - - - - - - - - - - III— 2IIIIII— III- - - - - - - III III2 - I — 2 — - * — ------- llllll 3 | - * ----- - i l l — III -111-------- llllll----------------------- III-— H i----------------------------------H I - 1 2 - 1 * 2 ------ H i - 2 1 1 1 - - - - - - - - - 2 * - - * - - 1 1 1 - - - - - - I I I * — H I * - - - - - - - - - - - III— 2 - 2 - - - - - I I I - E l l — HI— HI— U S — * 2 - - - - - - 2 - — 2 - - - - - - - | i i 2 - - - - - - III— - * — 111111-111-- - - - - - - - - - 2 - I I I — U — — - — - - I l l - I l l — -III 1 - 1— - 1 III— l l l - l — 2 — III— !!!!!!!!!— P - m m - m - ! ! ! — !!!— - — 1 1 1 1 1 1 1 -2 — 12 - — 2I I I - - - - - - - - - - - - - - 2 i i i - - - - - - - - - - - 2 — i i i i i i ii i - i i i | — ii i 2 - i i i 2 ---- : : : 2 - - - - - - - - - - - - : i i l —=:£ 1 - 2 i i i l - - - - - - - - ■ I l l - I l l - l l l l l l - - — III 2 - l l l | l — - l l l - l l l l l l l l l - — - - - - - - - - - - - 3 - - - 3 - I I I 3 - I 2 - — - 2 — 2=liil= f 3 2 -------- 2-1112------*2-111-*---------- III-III— ill—* ! ----H I 2--------------- * 2 — 1 1 1 1 1 1 -------- — 2 I I I ---------- * * -2 ------ nil—I ------llll-lll-------4 |ll^ lll---------- 1 2 --------2----------------- I I I -I I I 2_lll III-!!; - - - - - - - i l l | - i l l i l l -2 - *— -i-lll -2-^-111------ i p - 2 2 1 - 3 - - ---- llllll —llllll — I — H— 2 2 -— — llllll 3 — I I I - llllll * — — — - - - l l i p - 1 — - I l l - I l l 12III - - - - - - - - - - - - - - - - - - - - - - - - - 3 I H H 2 — i - i l ! III „ □ { [ I I I -----1 — - i — I I I ----- - ||||I I - -------2— I I I ---------------------- — — 2 - — I I I -- — Ill * - I I I - I I I - I I I — III - I I I 2 - I I I — I — i i ! - * - - - - - - J - l l l - p l l l l l l l l - - - - - - - - - - - - - - III22 III2 — — -H I- - - - - - — - I I I - - - - - - - - - - - - - III*-- - - - - - - * — 3 - H l — 2F - - - - - - - 2 - I I I - 2 2 — — 2 l l l - 1 — llllll — 1 2 2 - H I ----------------- j j j----- jjj— 2-----------H I —H I ---------- 1 —H I-H I—llll------H I 2 -3 -" -2 |-* -H !------- - 1 1 1 — ill— III 2 - ! ! ! ■ * ! ! ! 1 — 1 — - - - - - I I I - - - - - - - - - - - - - H i— - I H im - - - - - - 2III— - - I I I — I I I - - - - - - - l l i i — — - I I I - I I I - m - m i l l — - — h i -2 — * - i i i i i i 3 i i i - — III— - - - - - - - - - - - - - - -111-111-111— 1 -1 1 1 2***212111— 2----------------2 3 IIH H -— IlllilB— H H H IH 2 -III * ----- - - 1 ----- llllll -llllll -llllll- 2 - 2 — llllll - I I I— * l l l - | - l l l - l l l l l b - - - - - - 2 2 — — HI— IH — 111- - - - - - - J - 2 - j l j - - - HI——2 ——jii— * ———i |j —- - - - - - - - - * — - I I I — - — - I I I - - - P i l l - ! ! ! jjf — § 3 - f t - - 2 -ill - I I I - - - - - - - - - - 2 - 2 - - - - - III-i. — — — 2 — — | l 2 - - - - - l l l l l l - - - - - - - — 1 1 1 2 — ill-----------1-iii— 2 1 1 1 -Hi-HI*-------- 111-2-------------------- III— 2 - * - 2 ^ ! l--------2 — !H2— lillil-*2— 108 Figure 33. Map of model reef after S years of the competition simulation with random settlement to SC2. See Figure 29 for initial reef configuration. Only the first 50 rows and 70 columns of the 80x80 reef are shown. L e g e n d : - : s p a c e , A S : * :S C 1 , jji:S C 2 , l : S C 3 , A P : 1 : S C 1 , 2 : S C 2 , 3 :S C 3 R u n # 5 C o v e r : 5 8 5 5 8 1 3 0 2 2 4 6 7 9 1 1 9 r e s p e c t i v e l y . RANDOM SETTLEMENT TO SCI 2.00 V 0 0 1.60 140 ! 1.20 1 . 0 0 k 0 60 0.60 0.40 0 . 2 0 0 . 0 0 2 8 8 S 8 S 3 8 S 8 8 S f f S 3 8 8 8 RANDOM SETTLEMENT TO SC2 8 5 8 8 8 8 8 8 8 8 S 52 8 8 8 8 8 2.00 1.60 1.40 | 1 . 2 0 | 1 . 0 0 AGGREGATED SETTLEMENT TO SCI 040 0 . 2 0 0 0 0 m o o 2 . 0 0 1.60 AGGREGATED SETTLEMENT TO SC2 1.40 K i 1 . 0 0 060 0.40 0 . 2 0 0 . 0 0 8 e m e Figure 34. Segregation indices for various recruitment simulations with random settlement or aggregated settlement to either SCI (sexual reproduction) or SC2 (asexual reproduction) and with no competition. Patch formation was strongest for aggregated settlement to SC2. Patches formed by aggregated SCI recruits were weakly defined because of low persistence (high mortalities) of smaller corals. All simulations shown here had the same initial reef (see Figure 27-High d e n s ity random reef). Maps of these runs are shown onFigures 35-38. F igure 3 5 . M ap o f model reef after 100 years o f the recruitment simulation with random settlement to SCI and no competition. See Figure 28 for initial reef configuration and Figure 34 for the segregation indices. Only the first 50 rows and 70 columns o f the 80x 80 reef are shown. Legend: - : s p a c e , AS: *:SC1 , iii:SC2 , l:S C 3 , AP: 1 :SC1 , 2 :SC2 , 3 :SC3 Run # 100 C over: 520 264 62 1008 291 39 r e s p e c t iv e ly . - 1 ----- l - l - § — 2----------* -1 ----- 1-------------- 2*------------------ 1 1 1 2 -* ----- 1— * 2 - - — 1- — 1 * !!! 12— 2 — 1------ *-!!! 1 * 1 1-2!!!!;!-* 1-----------------------1!!!— 1- - 2 — 2) 1 !--------1—)))-l* ~ — 2------- 1--------- 2- 11- 11—§ - 3- 2- 2 1 1 - 1— 1 -2 ------ ill 1-1 1- 11- 12221- * — 1 -1 -----3- * -------------- **321- 2— 3— *!!!1-----11- 11- 1-211111------ *1 - * - 1 -2 -----2 -1 — 2 1 -* --------- 11----- *1------* -* ----------Hi-1— i 11*— 13------- 1------ 12*— 11--------1— 1- 1- 2 -2 ----- 2--------- 11------ 2--------------------12----- 2 -2 1 -----2 1 1 3 1 ------- -1 3 ---------------- 1*----- 1 * 1 -* —Ijj-i— 1— 2— 1— 1------- 2--------------- 2- —jjjlljjjl— 1 *-------- '*-21— * l - l |- l l l ! i * ----- 1-------**------l - * * * - * l------ *— 11—I s !----- 2 3 1 - 1— *— 1------- 1---- - 1 ------------ *----- 1— ));121-211--------2 —| - * * - l ----1-----------— 1---------11— * - l* * _ * -------- *_2— i — i n ------ 1 2 -* -l* * — *--------- !!!----------------- 1— * -1 ---------------!!!l— *— *§1- * * _ * _ ! | i | l - l l : ! j | H ill* -------1 -2 — I I I ------1-----1 -1 -1 —11121*— III-T— -2 1 -------- 1*-|-1;;!— 2 * -!* * — 2 -1 ----------1 -2 — - * — !!!— 2--------- 2 — 1-----*— |j!----- - — 1 H I — 1-------------- 1 -1 ------- *----------------*-----------13----j : j * — 1----*----- **--- 1----- *--- *2 * 1*1----- I I I *— *— *---- 1 !)-------------1-----* -3 ----- *--------- *1 1 ) 221-----1 - * ~ 1--------- 2-----1------11— 11 1 — 1-----1 * 1 -1 -!!!---------------- l — l ---------1 -1 1 1 ------------ 11------- l!!!-l— ** --------2*- -1-!!!-----1 -* 1 ----- *1*— *----- *-----1 |- * 2 - 1 - * 1 -2 -1 -1 1 1 -1 — * - 1 ------ !!!-l------11-*!!!----- i l l — 11 -!!!!!!*§- 2 2 -!!! -1 * *-------- 12— 2 * - - l- ! ! ! - l* 2 1 ~ 2 —!!!--------- !!!— *— 1 * -2 — *-!!!— ! ! ! 1— — )))— *— 11------- **— 1 -1 2 — *----- * *----------- 11— 1 - * — *------1 * 1 -1 — *---- 1 * -— -----2— 1-)))—§-*-))) 1--------------------- 1— 1-------1— *)))-----))) 1— * 2 - *1— 1-------------- 1------- 1----- 1 - * ------ **!!!-*-221*l----- *!!!!!!■*-------1 -1 1 -* 1 — 1-----*— !!!** *11— *------ !!!----- 1 * -* * -------*111!!!)))-------------1— 1— i — ;;;-l-------------- 1 - 1 - 1 ------!!!------ **-------2— 2- 1- 2— *----- * -* — *111-1--------------------- 1— 23---------- *----- 2 -1 1 ------121----- 2 *1— *— 2- * l — H i-* -* ------------ 11* - 1- * 1*------- 2---------------- 1- 1— *----- !!!---------- ------------ H I — * - - i i — i * - i * * — 2 - 1 - 1 -----------1----------*— 1— 1— 1— 1— *-----l)))-J----- 1 — *— I I ;------------* l - l - l --------------*!!!!!!|-----!!!— i — i - * i 1-------- 2-------l!!!2— * jjj * -------- 1— 1— *— ! ! ! *11-----1*— 1— l-!!!l----------- 1-------2 -* * -----------1 - 1 - 1 -------- 1---- *— !!!— !!)-------* - i — 2 * l* * * ------------------ 1- 1- 2- 1*— I I I ---- 3 -3 -------1*21- 12 — 2 — 2---- !!!l -12-!!! 1 - ---------- *------- 1--------------- 1*— !!!_i_*------- !!!-!!! l!!!-------------- 12-----------2!!!— 11— *_jjj---------- 1---------1-------1------- * - 11-11------- 2 — 1----- 1-----2 3 — * -1 ------2 - 3 1 ------ ------------ : -------*------- 11— 2-------- ***-----1- 1-!!! 2-2 * - 1*1 -1 -----------!!!ll---------------3 - 1 ----- 2 - 2 — 1!!)------------ 2 -2 ------- l)))l-l— l * - § ^ 2 - l— * 2 2 l!!!-|----- 11------------ 1— 2----- ))!------11 ---------- 21— !!!— 1-!!!------------*1—!!!— l-l«l!!-l- 3 -------- 31-----12— 2 1 — 1-------- 1— 1--------- !!!-!!! 1 -1 — l l l l - * 2 -----1— **----------- 1 -* -1 2 -* -* — *— * -1 1--------------- 1— 11--------------* 1---------11*----- 1 -1 -----i!!*l— *------- 2------ 31*1- * ----- 21---------------1 -1 ------12- § * 1— 2 -1 * 1 ------------ 1-------§ ------ ll-l-!!!2 * -----*------1 -1 -------------- 1 -2 ----- 1----------- :!!*— — 2— i ---------H 2 ----- 1-!!!*1---------! ! ! - l |- l ------------ 1---------11!!!—!!!------- 1---------*------1--------- - 2 -1 ---------- l- * - * - ! ! ! - l---------1- ------------ *3*1*2------- !!!2---------- * - i — *—| -----112-------- - 1 - * -----1------- 1---- * - * - i i — **3------------i - i i -----l-* 2 !!!* -l-2 ------- 2— 11----------1!!!12*2 2311— !)!-------*----- l l -!!!-12-----1— 1----- *-----!!!-l----- * § ------------------l l - 12- 3 * - l! ! ! l- l- * — 1!!!2*---------1-------11----- *— **!!!— *------- 1**— 21------ * - 12! ! ! l l - l — 31-------------- *-----! ! ! ) ) ) - l l l - l — 2-------- 2 -1 1 ------ 1---------- 1 -------*-----* 1 1 -* 1 2 -1 ---------- 1 1--------- - !!!§!!-!!!-*— * 1 * 1 -* — !!!--------- | l 2 — *------121**1*-!!!------------------12-!!!-2----- 2 — 1*1— * - 1— 1-----22------- *------- 3- 21- * ---------1— *2---------2 * 1*---------- * - 1- 11 -2-----* 1 2 ---------! ! ! — *— 1---------1----- *1— *1----l-i!!!— l ------------------------ 1--------------------* -* --------*2 — 1— 1— 1— *!!!-!-----1------1 |---------— 12---------------- *-------ll - ! ! ! - l-------!!!*-21------------*1- *— 1-----* -* 2 -------------11----------------1*1---------11-------------l l - l ) ) ) - l - l — 111------- 11------- 31— 2— !!!22--------------------------1— 1- * ------ 122----------------------1----- 12------- 1----------- * - 1 - 2 - 1 — l - * 2 - |l ! ! ! l - * l J ---------- 2 - 1 - 1 -----!!!-----------* - * - - 1 - 1 ----- 11— 11----- *------1*— 2 3 *!!!1— 2 1 -1 1 ---------- *1!!!21- 1— 3- 1-!!!!!!— —H i*— *— 1- 1— 1— ll!!i— 2 * *----- )))— 31*!!!-1 111 Figure 36. Map of model reef after 100 years of the recruitment simulation with aggregated settlement to SCI and no competition. See Figure 28 for initial reef configuration and Figure 34 for the segregation indices. Only the first 50 rows and 70 columns of the 80x80 reef are shown. L e g e n d : - : s p a c e , A S : * : S C 1 , :I;:S C 2 , l : S C 3 , A P: 1 : S C 1 , 2 : S C 2 , 3 :S C 3 R u n # 1 0 0 C o v e r : 6 2 2 2 5 3 6 4 1 1 1 3 2 7 6 5 1 r e s p e c t i v e l y . * _ * * -------------------------------------------------- * 1 2 - 1 1 2 1 -------------------2 — 1 — * — * --------------------- *!i!---------- * * ! l l |* - l — 1 1 1 -------------------------------------- * - * ------------------------ 1 2 ----------- 1 — 1 1 ---------------* — *-------- * * '" ------12---------------------------------------------------— * ------------------ H I — 1----------2 1 --------------* * — 1 1 1 1 *----------- 1 1 1 1 1 ------------------------------------ * ------Ills;;*------- 1 - 1 ----- 1 — 1 * 1 3 2 1 --------------- lli*-2 1— 2 ---------------* - 1 2 --------------------------------------------------H i--------- 1 2 1 -------1----------III— * 1 — 1 1 ---------* - * * 2 - 1 1 * * ----- * 1 * * 1 1 1 -------------------------------------* — --------- 1221------------------— ill— 1- 1 — 7 — ; ---- 1 1 1 2 - * |j------* | ---------------------------------------------------------------------- 1 - 1 ----------------* * * r _ i 2 i - j j j j ||* -j|j * * - 1 2 - 1 * ■ — * _ ----------- 21 — * * ---------------------------------------------- jjj*jjjfjj-------- ------- — j|!--j|| * 1 1 1 1 — * * *!!! 2 !!; 1 * * - - - - - * * — * * * - 2 2 ■ ----------------- * * * ------------2 1 1 1 * — III---------- 1 -------------1 2 1 — * |||------- i||---------------1 2 1 -------- 1 7 * 1 1 8?— * * _ i_ * _ * ;! i---------- 1 1 2 1 1 — * ------ * * ------------- 1 2 H 1 - * * * - * * * - * * — - 1 1 1 - * * - * ; ! ----- - 1 * ^ * - ;;! — 2 1 ----- * 1 1 1 1 — 1 1 1 ------------1 * 1 1 ; * ---------- *-*111-'*------ — I — :!;— 1 1 - 1 - ; ;; * ----------------------- 2 ---------------------------------------3 -------1 — 1 -1 1 1 1 3 1 1 -1 1 1 — * ;;! * - l * * '* * 1 3 1 1 1 - - — * ---------------------- 1 — 1 — 1 ------------1 -------1 2 ------------- 1 2 3 2 1 1 2 ------ *------ ill— * -* -* * jiiii:1 1 2 -------- *;;;--------------- 1 ---------- 1 ------------------------------------- 1 1 1 - 1 1 1 - 3 1 1 1 1 ------ * ■ - * - * _ * |- * 1 _ 1 3 _ 1 _ * * _ : * n 2 2 1 - 1 - 2 ----------------------------2 1 — 1 1 1 1 1 1 3 1 2 -------- i l l ------ - * * — 1 1 1 - 1 - * * ! ! ! * — 1 - 1 1 2 1 ------------- 1 1 - 1 1 1 1 2 1 1 - 1 - 2 1 2 2 1 1 1 1 1 — * — ----- — 1 — 1 2 ! ! ! - * - * - 1 1 1 1 2 1 2 1 1 1 1 1 1 1 - 1 1 2 - 1 - 1 ------------------------ 1 2 ------1 1 1 — * * '* ------------- 1------ 1 - 1 1 1 1 1 1 - 1 ------------------------1 1 3 3 1 2 1 1 2 ------------------------------- *— 1 1 1 1 1 - * - ! ! ! * - * * -1 1 — 2 2 1 1 1 2 1 1 1 — 1 - 1 — 1 -----------------------* ---------------- --------------------- l i p ---------------- 1 - 2 2 2 1 1 2 1 1 2 1 -----1 ------------------------------- 1---------------------*!jjj|j----------------- ---------- ----__*j|! * _ _ * _ * * * ----1 1 1 1 1 - 1 1 — 2 1 — _____ _ _ _ _ _ _ ^ _ _ _ _ j |; _ _ * _ _ - * _ - 1 - 1 - 1 ------* — *;;;■**;:;--------------1 1 1 2 ------1 - 2 -------------------------------------- *— *-------- *-111111-*— l l l l l — 1 2 1 ------* * - * * -- -* - * . ------------1 1 1 - 1 -------- 2 2 ---------------------------------------- | | j *11!---* 7 j |i * l * — 1 1 1 1 * *!!! *!;!!!! * ------ * *!!! - * — * * — H - 1 --- — * —■ * — — — — *!!! - * * B l - 1 1 - 2 — * | - - - -!!!*!|j*— !jj**!!!*— ------- ------— _ — ----------------------------&*— — * * * - * ' * * - * 2 * — — — 1 — 2 1 — * — *jjj * — — * — * | — *j||------ ■ -------------------------------------------- - - — 11*!;: *!!:!;; .*'★★★11!*-------------- 112 ***** ------------------------------- _-- --------- --------- ,------------------ 3 1 * 1 -------i l l - * - * — 1 - 1 -------------------1 1 ----------* --------------------------------------------------------------------- 1 ---------- l!!!2 1 — * 1 * * ---------1 2 1 - 1 1 --------------- 1 ---------- * ------------------------------------------------------------ 1 1 - 1 2 --------- * * * 1 — *:••* 1 2 1 1 1 1 2 1 1 1 * * *!!;* 1 1 1 2 1 1 ------ *;;;*-------- 111- 3- * -------11 * | * | ------* - ------ 1------------------------------------------------------- 11----------- _ * ---------- * --------------- 1 3 2 2 1 -----------!!!” * " ---------------1 - 1 - 1 1 1 ----------- 1 -------------------------- 1 1 --------------- * ------------------ * * - 1 1 2 1 - 1 * ! ! ! * !!=■-*------------- 1 2 2 1 2 1 2 2 - 1 - 1 ------------------------------1 2 --------------- * |- ! ! ! ------------* *!!! - 1 2 1 * — iiiiii— *!!!!!! *i!!--------------- 1 - 1 2 - 1 - 1 1 2 2 2 1 ------------------------- 1 1 - 1 ------------- — * ------------*!!!!!! * 1 1 — Bi; *■;! » - * * * ■ * * ---------------2 1 1 1 -------2 1 1 2 1 1 - * — * - * ----------------------------- *!|!------------ * S l -------* * — * !!! 11 llii — *--------------- 1------ 1113121— 1 1 **1 ;:*---------------- *---- * - i -------------- * 2 1 - 2 — 1 -------* i||.1 2 1 1 — * * ---------------------------------- 3 1 2 1 — -*ji!* * 7 ------- .*— * * * — * - 1 * - * * 1 2 1 - 1 2 1 !;;-!!!**!!!* 1 1 - 1 ------------ * * ! * * -!!!* ------ * ----- 21;;!- I 1 2 1 2 2 -1 * ;;!;!;* * — f * * ------------------------------------------------* ;;;i;l* ils - ( ! ! - ;s * l------- * 1 * * 1 * * ---------- 2 1 1 - 2 * l ! ! * l — l - * - * - J * l - l -------------------------- 1 1 1 1 1 1 2 * * - * ! i! -* - ! ! s 2 3 1 --------- * — !!;!!!” ------- 1 1 3 -j|| * 2 1 - 1 *!!! * * * * | - — 3 ------------------------------2 1 1 2 2 lj|! * **** liij * * ----- * — H I *|j|*--------------- 1 1 * **21!!!— * ! ! ! * * ! ! ! - * - lll -----------------------1 1 2 - 1 1 1 1 - 1 ;;* — *!!!*-**!;;*------- - * - - * ------* ----------- 1 2 1 2 2 * — * j ! l * - l l ! i ! - l l -----------------------1 1 1 1 ---------* 1 ! ! - ! ;;* - * ----------------- ------------------- !|jj||— * 1 1 1 — 1 — 1 1 1 1 ------- * — !iil— -2 3 2 --------------------1 1 3 1 2 1 1 1 - 1 - * ;;! - ! ;; * - * * - * - ! ;! * — 1 1 1 1 * * - * - 1 3 1 1 2 1 --------------------- * - 3 1 1 1 ------------------- 1 3 - 1 1 1 1 1 — * - * * ! ; ! * - * * * * |* — * 112 Figure 37. Map of model reef after 100 years of the recruitment simulation with random settlement to SC2 and no competition. See Figure 27 for initial reef configuration and Figure 34 for segregation indices. Only the first 50 rows and 70 columns of the 80x80 reef are shown. L e g e n d : - : s p a c e , R u n # 1 0 0 C o v e r : A S : * : S C 1 , 1 6 5 1 6 0 3 jj:S C 2 , | : S C 3 , A P : 1 : S C 1 , 2 : S C 2 , 3 :S C 3 3 7 1 1 0 3 4 2 0 3 8 3 1 4 r e s p e c t i v e l y . 3 1 22! 2 2 2 1 1 1 1 ! 2 - 12! li!i 3 2p - 1 - 21311121! 12Z21221p * 2i i i ! 22iii!iil3- 2i!f-2iiii! 2221! 2 2 2 1 1 1 : 222112! £ 22: 2 ll!l 21:111 | - 2i;i 2 2 | — 2 — !;; 2 2 -I! 32;1123L21— lll21 2 — f!Tl22— *i|l2 2[ 22III22— *!ii3H ll32— 2 1 — 1 — 1| 213:;::!! - 2 p 2 - p i i i liii 2 2 22 2III22— !!!=!!2221X2!!!2F - p i i i 2 2 1 2 2 2 - 1 * 2 1 ; -2 iii2 3 iii2 2 |3 iii2 2 2 2 [ -!!!2 !!!liii-l” 2 - 3 2 - 2 — !l;33!!!il!!3— 2 1 2 — III2 1 2 1 1 2 — 1 1 1 II! 21........ Ill2 2 -22:1! -ill22211111* liii 2 2 l i i i ^ 11: !!: 12i;i 1 2 32 2 2 | - 2 2 2 2iii 2 -2 iii|2 ;:: liiiiii: 2iii * 3 liii 2 - 2 - H i 2 2 1 - 1 1 1 - 2 - 2 1 1 liii 2 p 2 1 1 2 iii|liii liii 3 1 2 *iiiiiiiii 2 2iii 1 - 1 3 p ! 12iiiiii 11112 2 1 liii2 iii|- 2 — iii 3 2 — iii 3111! 1 2 2 2 *iii 2iii 1 1 2 - 2 1 2| 3*2 2 222 *! _..........— iii— — l::::::2 :i 332:::2:::*2 i:i212ii 1223112:; f l 3‘ l | !2r ' ! l « ! f t p 122 lliii* 2| — iii 2 2iii 2212jii 3 2 — 2 ! |- 2 1 1 H 2 2 * 2 1 liii 3! 2 — iiiili2112iij 1 2 2 1 2 2“ : : 122-21 « ::« -312P 2 2 1 2 | l ; i i : i : 2 2 l i i i — 2 1 1 1 | 1 1 2 l 2 2 i i i 2 1 2 1 - l i ; * 2 2 - 2 3 2 2iii - 2 -iiiiii 2 2 - 3iii 2 liii -2i! [-1212^!! 2- 2iiil-*iiiiiiiliii21J ii2iiiiiii-i! ... 12 2 - 2 3 2 3 2 2 — iii 2 llH'l 2: 2 1 3 p i i i *iii 1 -iiiiii2 2 2! * • “ - ... 1 ! 2122iiim -iii 3 liii 2 2 2iiiiii 2 2 - 1 1 2 3 1 2 - 1 2 2 2 1 iii2 1 2 3 2 2 1 2 1 2 2 liji 2iij -iiiijiiiip * 3 2 * 1 1 2 ;:! 2 2 22:::2-12 ^22 " f l s i i s i s t t M M j 2B 11 1 2 2 1 !!!!! 222- ! 1 2 1 1 :1 ; 3 l p 1 2 iiiiiipiiiiiiiii 3ii 2iii 2 1 2 p 2 1 * - 3 2 2iii -iii *:ii 2iiiiii - i« 3iii 2 2 — 2iiiiii - 2 -2iiiiii— 3 2 2 1 2 2 2 1 2 2 i;;|2 3 H - i ! ; 2 -B -iii 2iii liii 22 - 2iiiiiil*223|-j! liiiiii 2 l p 2-*■!!!!! *iii 2|iiiiiiii 2iii 2 -iii 2 2ii 2i!I-22=lii 2p iii3 2— 3iii2— 21*22 3! ! ! 2 2| - i i i - 22-iii 2 212 - | l 1112 22iii 122? 2222iii2223- 221* - 12222i ii! ii 2 2iii liiiiii -iiiiii 32 3iii 2 2iii 2 liii|- 2iii 3 21- 2iii-232213- 22- 2112” 1 1 2 - 2 1 1 2 1 2 ! 1 1 - p i i i *211112! !3 : iH : 3(1: 331: 2 2 i i i p 3 W i 2 - p i i 112212— 2 2iii 212-21:1222122:!! 2 2 - 1 3iii2 2 - 1 2 — 2 - l - l i : : l ~ 3 i 2 i i i l l 2 2 2 2 p 2 2 2 * - 2 2 12iiiiii 1 2 2 * 3iii liiiiii -iii2 2 1 2 2 - 12iii 3 - liii:;: - 2 - 2 1 — | 32iii— 2 2 2i;i 2 2iiiiiiiiipiii 2iii 2iiiiii liii2 - 2 2 2 - liiiiiiiii 2 2iii » ; ; 2 1:;::::::: 2::; 212;;; -2 ;;; 2 2;;;;;;;;; 1 3 - 2 ; ; ; p -2;;;;;; 2 2;;;;;;;;; 2;;; 2 1 H i -2;;i;;;;;l 2 2 liiiiii 2 - — 3 2 1 1 *iii2 2 2iii 1221113iii - 1 - * 2 — 12ii' 2iii 2 2 2 1 I B — iii - ■ liii- 2 2 2 1 2 2'if 21111— iii 2 1 2 2 “ iiiii 2 - 2 B 2 2 1 1 1 liii 2 1 2 - 1-1113122-211 :::: ± : i: 4 i l p : ii 12211 2| 2p 213iiiiiiiiiiiiliiiiii22 --2X22 1 1 1 2 2 1 1 1 2 -2 2 1 1 1 W 22i: 221! 1-: i f 22: 2111-371 lBiiiiiiii liii 22222 -liliiiiii 2lif liii liii -iii * 2 -iiiiiiiiiiii 2 2 1 1 1 -iiiiii 13 1 1 1 2 2iii 21— III 2 3;;il-i;i; 12 2 -2 *2 2 31 liii liii2 221- 2- lp iii 1 2 1 1 1 1 1 1 liii223221113 1311— 1122 liii 222 liii 2 1 1 1 1 1 1 2 1 1 1 1 liiiiiiiii -2 liii 112;; ; ; ii 12 liii- 2322111- 1- f 1 - 1 - 2 B |1 - 2 | 2 2 - 1 2 1 2 i i i 2 p l ! 1 1 1 2 2 3 2 - 2 P 1 * 2 lip ii 3 1 1 1 — 12— 121! 222 2 iliii — 2 1 1 1 lBil 2 1 21- 1 1 1 1 2 -1 1 1 1 1 1 2 iilB X 2 - *221212 liii 2— 1 2iiip iiip iii-2-iii23| iii2= 1 = 2 2 3= 1122 liii 2- 1; |2— iii 3221221111— 112221111 2111*1 1121-2 liii* 2 2 2 ftp iiiiiii! l '* |2 2 2 2 2 l : i ; ; : l 2 i!ll2 2 i i i p 22111 - : i | i - 2p 22-'2 1 l f l l l 21 212 l i i i l l - p 2211: ‘ ~ B l ! * * 2( 3 — 2;!! 111:11311! |222- 2 1111112111122— 23l l 2iiiiii*2 1 - |2 2 2112 3 1 1 1 2 -iiiiii 2 liii - 2 - ill! - *iii 2 liiiiii 112- 3” l 2iiiiiiii[2 2 1 1 1 -^ p iii 2 l i t : 3221112 2 1 — iii 1 1 2 2 1 ■ - 21231- l p 2122:;:*23i J ii 1— * Iii; liiiiii 213iii 2 -T 2! 2222 I [2 2 3 2 2 2 1 112— ii -3-21! * 12 - ; 2i*r -2;:::::1;:;2::;iB|122- - *!:: 2 2 p i i i 2 p 2 2 1 1 1 1 1 1 22:iiiii223*2H iiiii;2- 122*1113211 1221112-12321; 211112223— B|l2ii '2222:;!—M i! 3iii2pii=22ii * ...2 2 1 1 1 1 m 1 [23- 2 liii 2 1 1 1 2 7-pii — iii iii — iii - 2 1 ! -2 3::;;::p 222 2; ! ; 12 * * 1- _ * liii liiiiii 2131112 -iiiiiiiii 3 1 1 1 2llii=72!iii!i|*p:;; 22- *iiiiii -iiiiiiiiiiiiii - 22T22M-2l 3- 2i:;;;;l:ji2;;:lH l22- 2- 2 1 *ii;21:i;*yy:;2;;:1212;;:2 - 2- l - 22; 2 = 1 22 *213- * - 221112-iii 2 3 * 2 22222212 liiiiii -iiiiiiiiiiii 222 l i i i 22-iii 22p4;22j J 12 2 22l l 2 2 - 3| 2|i= 2 2 2 222-;!: 2 2l * 2 !!! - li in 213. 2. 2- 332jii iii i p 2iij 2 2 £ 212 2 H i! 2 - * H i 2 ^ 1 1 — 1 liiiiii 2— 11— 1 21 1 1 1 — 2 liii * 1 1 1 2 1 1 1 liiiiii 1 2 1 1 1 *2211111123211- 12! [ 3221: 2p 2 1 1 1 1 1 1 2 2 3 * 2 3 liiiiii 2 2 1!!!2 12!!! 12 1 1 1 1 1 1 1 2 2 1 1 1 3 | : 12 —III2III22 III2H I1 2 121111 — 2! - 22232211* - 11-12222223 2 1 1 1 1 — |ii— 2 1 1 1 1 ii! '221- i * - 32! 22; ! ! ! ! ! - 11131.2 -2 l|j ! 2]j| -2 l|||l 2 l!j! 2 !|! 2: ! s 1 1 1 1 2 1 1 .... 2 l ! ! : i : ; ; 2 i i 2222111111132312211 Iliii-2 liiiiii -12221 * ;;: -1 2111-11112-22-22—2 ***1211112*1- 1 -1 1 1 -2 2- 22iii22p 2- 222iiiiii [ 2121112-2 liii2221112= 111222- *2 2 322 * -iii 2 2 *iii -2 - 3 3iii -1 2 2iii 212p l - 3iiiiii liii 3i!i-3i!i!iii?2iii|* 3 12iii - 222 lT 2iii 2 2132: ! ! 1 -2 2::!'iiiil^ 2= = ; - * - p 122 22 - 222 * - 112213 312 2;:;212 ;;;;;;2 -2 ■ ! ! 212121 — !!! 2!!!!!! 3!!! 22 - - ! ! ! - ! ! ! i ! ! ! p i 2 2 - ! ! ! * 2 -2 2!!! 312 !!! 2 1 222123 iii * 2 2 l i i i i i i2 l i ii i i i i i i2 2 1 1 2ii i2| l 3 - i i i i i i i i i 212 " - p i 212122211112 i i i i ! i | i i i i i - i i i — 22 2 l j i i 2 2 2 i i i l : : : 2 p ; : : - * : : : 2 2 ; j | 2 - ; : ; 3 * 1 2 2 1 2 : ; : 1 2 2 - | 2 - 1 $ 3 ; ; : - 1 l : ; : 2 - 2 j ; ; 2 2 jjj l i n j j j - 2 l l ; ;; 2iiili;22*2— l ! 2 2 3 *2iiil2!iip l - 32p - l ! 223222” 2p l i i i 2ii 2 13ii! 32 2 l l - i l liiiiii 2 2 - 2p 3iii— 212 2 2 !!!” ! ; ^ p -11 1 2 1 1 1 -ii! 2 2 —1 1 ! ... 322-iii liii— 2223211 iii * p 1121 liiiiii 321112- 13- 11- 111111- 1 1 1 1 3 * 1 1 2311121121! 2|i i 212 iiiiiiiiiiii * 13 212 iiiiii -T - 3^1 ! iff* 211322:;:pi!i 2iiiiii- 22132ii" 13 1 - 3212- 122122-1 ! ! ! i ! ! ! ! ! ! | ! ! 2! 222221- 22jjjpiji 1 2 2iii 2 liiiiiiiii 22 -2 liiiiii 1- *2 2 -2 3p!!! 2 2 3! ii 21 lBiiii: * 212 2iiiiii 1 l i i i - 22112 *H ii -22 2 1 1 2 2 *iii *!! 2 2 2 2iii 1 2 1 1 1 23iiiiiiliiili -1 2 liii - 2lii 2 *iii 2iiiiiiiii lZlii 2 -iii— 2ii!iiiiii 3 - liii iii 1 - 3iii222 3iii 2iiiiii 2 - 2iii - * 2 liii 212 liii - 2-2 22pi!!iii!!j 1 *” 2 1J2- 2 1223| — p liii 2 * -iii 2 - 2 3 -i!!!!! 12 21212 - 21— 1 --!!! -1 1 2 2li! 21 113 Figure 38. Map of model reef after 100 years of the recruitment simulation with aggregated settlement to SC2 and no competition. See Figure 27 for initial reef configuration and Figure 34 for segregation indices. Only the first SO rows and 70 columns of the 80x80 reef are shown. L e g e n d : s p a c e , A S : * : S C 1 , :i-:S C 2 , R u n # 1 0 0 C o v e r : 1 7 0 1 5 5 S 3 5 9 1 1 ( :S C 3 , A P : 1 : S C 1 , 2 : S C 2 , 3 :S C 3 6 2 1 5 4 3 5 2 r e s p e c t i v e l y . * :H : i i2 ! iiii _ 2222—H i l l ! 3111!!l2 2 2 1 ~ !!!-!!!!!!* -l 1 2 2 2 1 2 2 2 — ill* 2 3 3 2 3 1 2 1 2 1 2 122 2 2 2 2 0 1 ■ liiiiii:Jllillill2 2 1 * 1 2 1 2 2 2 2 2 1 1 - ! « — III:!:- : 2 2 1 2 2 2 2 2 2 2 2 3 2 — i:H::i* — ::: ™ s« — 1 2 2:::|iy::::iyi;::::: 2 2 2 2 1 2 1 1 1 " -12322222322 Iiiiiiiiiiii! * | iilll-2213*- II * jjjjiiijjllll2— 1222 2 ! ! ! 2 2liii|llilllll|rjilill *21322 *L ■ II12 2 liii liii *lii * |iiilliiiiii 1 2 2 1 2 2 1 1 1 l i- i i ig i i l i l l l l l * — l i i l l l i i l g l l l l |5 ||l i i - * 2 1 3 2 2 1 2 2 1 - 2 - l i l l l l S l * 2 1 2 2 2 2 1 3 2 2 1 2 2 2 2 1 ■iiiiiiiiillg||iiiiii-12322223211-i— 222322-12211122212 iiiiilllllllFilllllilillllli 11221221211211222223-22132312212 liii 1 2 2 2 2 2 2 2 1 2 2 _______________ - 2 1 3 2 1 i i l l i - 2 2 3 2 2 1 2 2 1 2 2 2 2 1 2 f ||lii|i|iiiiiil2 2 1 2 2 2 2 2 2 2 2 2 2 1 - 2 2 2 2 2 3 1 2 1 1 - 2 2 2 1 I -....*....... . iiii*iiiiiiiiiiliiiiH iiiii g n-|||!j|j||3122j!!l!li 1 -2 1 2 2 2 22T*iii|iiiii *iiiiiiiiiiiii!Iiii -iiiiii 1323131 3 1 2 1 -2 2 212111- * i l i i i *|iiiiiiii|ii|ii|ii 121212 2213212 212iiiiiiiiiiiliiiiiiiii — iiig^iiiii 122 222 212-2112112!!;!!! — ill -ii! - *0!®! *!i!0! -B 223 llll[ 2212212313222111-llllllillp iiliiiill — T l 32211212221132!i!llllll"li| 22 312 22 2122 2 3iiiiiiiiiiiiiii * f iiiiiiiiiiiiiiiiiiiiiiiii-*iiiil;*iiiiiiiiiiiiiii 1 1 3 - 3 2 1 2 2 - 3 2 2 1 2 2 2 2 1 1 3 2 2 1 1 2 1 3 1 2 2 1 Iiiiiiiiiiii 1 2 1 2 3 2 3 2 2 2 2 1 2 1 1 3 1 2 2 1 3 1 2 1 3 3 2 2 1 3 ■ ii- * iii2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 3 2 2 1 2 2 2 1 iiT — flT — iii— 1 1 1 2 2 2 2 2 2 2 1 2 1 2 2 2 3 2 2 2 2 — 2 3 2 1 3 1 2 1 2 2 iiiiiiiii|B li2 2 3 1 1 3 1 2 2 2 1 2 1 2 2 2 1 3 1 2 2 1 2 1 1 1 1 - i::i ii l 2 2 2 2 3 2 2 2 - 2 2 3 1 2 2 2 2 - 3 1 1 2 1 2 3 1 - ii i - ! l[ * 2 1 2 2 1 2 1 2 2 1 2 2 2 1 2 2 2 2 3 2 2 2 2 2 1 2 111!-Ill :Liii-2:j|l21 1 2 1 2 2 2 2 2 1 2 2 2 1 2 2 - 1 - 2 1 2 1 2 2 2 l H ii2 1 2 iii2 2 1 2 1 2 2 3 1 1 2 2 3 1 1 3 2 1 2 2 2 2 2 2 2 iii3 2 ! » [ 2 2 2 2 I 2 2 1 1 2 2 3 1 3 2 1 1 1 2 2 2 - 1 2 2 1 2 1 2 - 1 2 2 P 3 1 2 2 1 - 2 2 1 1 2 1 3 1 1 2 2 1 3 3 2 2 1 2 1 2 3 3 1 3 2 2 2 " ! ! 3 2 2 2 1 1 2 1 2 2 13222iiiiiiiii 2 1 2 2 2 3 1 3 1 1 1 2 2 2 2 3 1 1 -ii H it 1 1 5 ^ ■ ll:;:8 iM - 2 3 2 2 2 1 3 3 2 2 2 1 2 ::ll:l2 2 2 3 2 2 2 2 2 2 2 1 2 2 2 1 1 2 2 2 liiiiii J i i ii i ii i li i ii i p liiiii2 2 1 2 2 2 - 2 2 2 1 2 3 2 3 2 2 1 2 2 3 2 1 3 2 2 2 2 2 1 2 1 2 1 2 iiii|| 1 2 1 1 2 2 1 2 1 2 1 2 3 2 2 2 1 2 0 ? — B - i i ! illi! iH - i2 2 1 2 2 1 2 3 1 2 3 3 1 2 1 2 3 3 1 1 - 1 1 3 1 1 2 3 3 2 2 2 2 2 l l l M 2 2 2 2 2 1 2 2 2 2 2 2 2 2 2 1 2 1 - i ii i ii! ! lK iiiii* K iiiiif 2 2 3 1 3 2 2 1 2 1 2 2 1 1 2 2 1 2 1 1 3 2 2 2 1 2 2 2 2 2 2 2 1 3 2 il! f 2 1 2 1 - 2 2 2 2 2 1 2 2 1 1 2 2 1 ------!!!!!i“ iiBii==::!i-iilBii2 2 2 2 3 1 2 1 2 2 2 3 1 1 2 2 1 2 2 2 2 2 2 2 1 3 2 2 2 2 1 1 1 2i!!iii!!i 2 2 3 1 1 2 3 2 2 1 1 1 — 1 2 2 1 2 * — liiiiiiiii — — iiiiiiiilBii— 2 2 1 1 2 2 2 2 2 1 2 1 2 2 2 1 2 2 2 1 1 2 1 2 3 1 2 1 2 113222iiiiiiiii 1 3 2 2 - 2 2 1 1 2 1 2 - 2 1 1 2 1 2 - l- illllllll:llillB il* lll3 1 -2 1 2 1 2 - 3 3 2 2 2 2 2 2 2 1 2 1 1 2 2 2 2 2 1 1 2 1 3 1 l l l i r 1 1 2 2 2 | — 1 3 2 2 3 1 1 2 l l i l i i i i i H i i i : : : 1 1 1 2 2 1 2 2 3 1 1 2 2 2 ^ - - 1 1 1 1 2 2 2 2 1 2 2 1 2 2 2 1 2 1 3 2 2 2 1 3 - 2 - 3 - 2 2 2 2 2 2 2 1 1 1 1 1 2 2 1 2 2 1 2 2 1 2 2 3 1 2 3 2 12iiiiiilL >132232222222122322*1111111" 1 2 2 2 2 2 2 32lll?ITllllll 1 1 1 3 2 2 2 2 2 2 1 * 1 1 1 1 1 1 — B **!!!— ill 2 1 2 2 1 1 1 2 — 3 2 2 2 2 2 2 — 222221: 3 1 2 3 2 1 1 1 *iiiiiiiiiiiiiii2 2 1 2 3 1 2 3 liiiiiiiii . - i i i .........i i i i i -12222----:::--:»:»121311 212211— iiiiiiiiiiii *Bll-2- 2-!!!-*l!!lll::::ll-ill!!!lll!!nilllllll-2 12- 2112232113111** - 12122— lllllllli— T*l-lll*l!l-i!l*gllllll!i!!!-iiiiii III * 2 - 2 132122231112221!!!| — 22221-iiiijjliiiiiiii32- 112iiiiiiBiiiiil!!iiiiiiiii*iiil322232122222iii22222212-liiii 112 llll-!!!p!;!!!!!-32ll! 1212— !ii!!!-!!!!!!i-!!!-!!;222233122!!!H^123221122322 2 2 2 2 2 1 2 2 1 2 2 2 1 1 2 - 2 1 3 3 1 2 2 1 3 - 2 3 2 2 2 2 1 1 2 2 2 1 2 1 2 1 2 1 1 2 2 2 1 2 1 1 1 2 2 1 2 1 2 3 1 2 111122222- 1 -2 2 _ 2 2 2 3 1 2 2 2 2 1 1 2 2 - 1 2 2 2 2 2 * ii! ! * - 2 3 ! ! ! ii! 3 2 2 ! ! ! ! i! i|--------2 2 2 -2 2 2 2 2 -i!!!!i!!i!!i* * 1 1 2 2 1 2 2 2 2 3 1 2 1 1 2 3 2 2 2 2 1 2 3 1 - I llllll2 3 1 1 2 2 llllll2 2 2 * 2 2 2 - llill* if p - - 1 1 2 3 2 3 3 1 2 1 2 — |l!!l!lllllllll 1 1 2 2 2 1 2 - 2 2 2 2 - 2 2 2 2 3 2 2 2 2 1 S 22 i illillilil— 2 2 — 2 2 2 2 — 3 1 l ll l l2 3 l l — :i;2 2 1 1 2 1 2 2 2 - 2 lil2 2 i!!!!!| ■ iiiii2 2 2 2 1 1 3 1 2 2 2 2 2 1 1 2 ilil I! l:! l|l2 2 1 2 - 1 1 2 2 2 1 3 2 3 * 1 — !: — :::- 1 2 2 1 1 2 2 2 - 1 1 1 3 2 3 1 - il iiiiiiiiiiii 2 1222212 2 1 1 2 2 2 2 -Iliii iiig - 1 3 2 1 2 2 2 1 3 2 2 2 2 2 1 2 —-il lllii?2 2 2 -2 1 1 1 2 3 2 1 1 2 2 ll!iP “" 1 1 1 2 1 3 1 1 2 2 2 1 1 2 1 2 2 2 2 2 — Iiiiii . H i 3 3 3 1 1 2 2 3 3 2 2 3 1 2 1 3 2 liiilliiiil 2 1 1 2 2 1 2 2 2 3 — 1 2 2 2 3 3 -liiiiiii - 1 2 1 1 2 2 2 2 2 1 2 2 2 2 2 2 2 2 liii2i H y iii ill ;■ *-iiiiii 1 2 2 1 2 2 2 3 2 2 2 3 1 2 2 2 2 2 1 1 - 2 - llBH-lll 1 2 2 2 1 1 3 1 2 2 1 2 2 2 2 1 2 3 - l B - O H H ii * 2 3 2 2 2 2 2 2 2 3 2 3 2 1 3 1 1 2 — iii !!!!!~ - !!! 21122222222223lllllllllllllil ...................1 2 - 1 2 2 1 2 2 2 2 1 1 2 — i i i - | 121-2222211122222 *; : : : ; ; ■ ■ ilililill 1 1 2 2 2 2 1 2 2 2 2 — ii 112222221222121*11 1 * 1 3 2 2 1 2 2 1 1 1 2 2 — il 1 2 3 1 2 2 2 2 2 2 2 1 2 liiiiii iiiiii * 2 2 2 1 3 2 3 2 2 — *ii Iii-ill * 2 2 2 3 2 2 1 2 1 -1 1 1 1 1 *■11222312231111 i— Iiiiii::::lll2112221 2 — li | r J i 2 1 2 1 1 2 2 3 2 ---------- * - - 2 'l 2 1 — 2 1 3 1 1 - 1 2 1 2 2 ::::|!iM -il:32111221------------------- 2 1 2 - 1 - 1 1 2 2 1 2 2 2 2 2 liiiiii Iiiiii - d i i I2122iiiiii— 2 2 1 ------2*111111212222— 1 1 1 2 1 2 2 2 2 1 llllf li i::::::n:::iiiigiiiii|l3111------ l|!!!!i 3 2 2 3 2 - 2 2 1 2 2 3 2 1 1 2 2 2 1 ^ “' ■iiiiiiiiiiiiiiiii 1 1 2 2 2 1 2 2 - 2 2 1 2 2 2 2iilliiiii * | l i - ------- 2 2 2 1 2 2 2 1 2 1 2 2 1 2 1 - ill * ! ! ! | | 121 1 2 2 1 2 2 2 2 2 1 ? 2 - 2 2 2 1 2 2 2 2 1 3 2 liii-iiiilT— I*jiiiii2iiiiiiiiil-i!liii 1 2 1 2 2 1 1 1 2 - 1 2 2 2 2 - 2 2 1 0 _ _ _ _ _ _ _ _ I ll H H » N ! •••••{•••M il 1 2 2 2 2 1 1 2 2 2 1 1 1 2 1 2 2 1 1 1 3 2 2 -3 1 1 -* * !!!* — iiiii I I -gli 12 32 322 -2 3121-ijiiiiiiiil P ll2 114 (to SCI) however were not as well defined and contained colonies of the other species, even if initial coral densities were low (compare Figure 36 to Figure 38). A similar outcome was found if only AS or AP, but not both, exhibit aggregated settlement (Figure 39). Again, the patches formed were not pure stands (Figures 40 and 41). Distribution of patches in all runs showed little relation to initial configuration since patches, by chance, move across the reef by losing more colonies on one edge compared to the other. Sizes of these patches were more dependent on initial densities however, since the model parameters prescribed stable population sizes without competition. The large patches seen on Figure 38 were formed by the fusion of several smaller patches. ♦ Cellular automata: Disturbances and coral spatial distributions The various levels of non-selective, randomly-distributed disturbances used also did not lead to any patch formation, even on a crowded reef (see Figure 42, with sample maps on Figure 43 and 44 with the initial reef shown on Figure 27). If the disturbances were limited to a certain zone however, the numerically abundant species did have a slight advantage even if the disturbance affected both species equally. This can be seen in year 10 of a simulation (Figure 45) in which random disturbances (500 1-cell disturbances per year) were limited to a central band on the model reef (columns 20 to 60). The greater numbers of AP colonies effectively spread the risk of damage over a wide area. As can be seen in the projection matrix analyses of 115 2.00 1 8 0 1.60 1.40 | 1.20 1.00k 0.80 0.60 0.40 0.20 0.00 0 ,n S l £ 8 « S i R ? ! ? 5 ? f f i 8 8 S I 2 8 8 8 » 8 200 1.80 1.60 1.40 | 1.20 1.00 0.80 0.60 0.40 0.20 0.00 ou, ° S 8 « 8 8 ? S 8 ! 8 8 8 £ £ 8 8 8 8 8 years Figure 39. Segregation indices for two recruitment simulations (without competition) wherein only Anacropora puertogalerae exhibited aggregated settlement to SC2, while Acropora subglabra settled randomly (upper graph), or vice versa (lower graph). The random settlement of one species led to higher mingling between species, and thus smaller patches. The second graph was cut short by the model reef filling up. See Figures 40 and 41. years Aggregated A. subglabra settlement Random A. puertogalerae settlement fHiiiin m m i w iiih ihhiiiihhiii m m m in im u m 116 Figure 40. Map of model reef after 100 years of the recruitment simulation with aggregated settlement to SC2 (for Anacropora puertogalerae only) and no competition. See Figure 27 for initial reef configuration and Figure 39 for segregation indices. Only the first S O rows and 70 columns of the 80x80 reef are shown. L e g e n d : - : s p a c e , A S : * :S C 1 , iii:S C 2 , l : S C 3 , A P : 1 : S C 1 , 2 : S C 2 , 3 :S C 3 R u n # 1 0 0 C o v e r : 1 5 2 1 4 2 3 " 3 2 2 9 4 7 1 8 4 4 3 2 8 r e s p e c t i v e l y . 1 1 ! 2iilB!l222— 2122322222222 lill— lilll— 22— — 1 1 1 ------ 1 1 1 2 2 222!!!!!! —! ! ! ! ! ! -Ill * -Iii 2 * 11212BII---- '— II! - * * 12 2 3 1 ! ! 2!il 2 2 l | l 2 2 1 3 2iiiB— Bil— 1 — I — iii-iii22132iiiiii— iii-* P p iiii-2 iii2 3 1 3 T -2 1 - — - - * - 1 2 2 3 3 3 1 1 2!!!l2!!!!i!iiiiii2 2 2 -HT* 1 1 1 — l-lis-iiiiii - 2 1 liii 2 2 iii|-iiil— lilil! — ill 1 2 1 2 2 — 2 |iil -if ! iiK i- - iii23iiil332li!i2322iii332*iii-fl ii! 122 2 2 1-iii —iii!!;-iiiiiiiii * 2 31 |2 IN! 2Bii—! ; ! ! = ! 1 -iilH T -iii -iiiiii 2222222132iii 22222!iiiii2------------------- 2222322-llhli----------- 22iii-*123iii------212- ™ _ ^ 4 ! I J I ji j i2 i i 2 li j ii i 2 2 2 2 2 2 * - - - - # < i i! ! i ! ! ! ii;-iiilliii2 2 2 iiiiii* -iiiiii* -1 2 1 2 iiil-2 2 iiiiiiiiil2 3 2 iii iii * -iii -iii-iiiiii-2 22122!!!-222 iiiiii! — ill — iii-^B -iii -iii niiiii -mil!— * 1 1 1 1 2 2 22 — mil! -1 ini!!! 2 1 1 1 2 -iii---iii 2 |F liiil2 12 l - l — 1 2 2 iii--l-~ — iiiiii— ;;;-----Iiiiii -I!! *—! ! ! ! ! ! -!!! - 13222iii 2— 2 = 1 1 2 -----------2 -iiiiiiiiillT liiil2 2 2 2 liliiiill2 1 -4 ii-^ii--iiilli-iii--iiiiii2 2 i|-iiiiii-* ir 2 2 3 1 2 2 2 2 2 1 2 2 2 --|----- ^ i i - - 221 l i l 22iii22-*iiiiii21- i i i - ^ i i |^ - i i i - ^ i i - n 32^ ^ i - - i i i ! l ! 32221*122212l!!j!jfc-- - i i i - - - |l 2iii-2222- 2iii--2iii*-!ii4 i-"ii2 j!-i% !jjH!i221^i!!i!i-i!!” 21p 22l!!3!!!2222322-!!!--ii!-- H I ~ *1 tT22 2 2 1 1 1 ! 2 2ll! 2 212 2!!!iii —i;;l2;;;--^-iiiiii * -12 3iiil5;ii 2 * 2— * *22! ! ! - 12! ! ! 322112; — Plli-ill 11311122 3*1 ! ! 221— 22^BiiX22H— |ii||Hliii 12 J l 212 2 3 * — ill2221----- 1122l l 31iii— sssf ~ -iiiiiiiiiiii 1 2 2 liii2 1 2 2 1 2 2 2 2 2iii 3 p 2 2 3 3 iiiiiliH 2 ------- 2 |2 iii 22iii -iii2 2 2 - i i i l - 2 3 1 1B 2 2 i i i | l — - i i i | ii!— — W* iii! 1 1 2 2 2 2 2 2 2ii! 2 1 2 3 2 2 1 1 2 2 3 2 1 2 2 3 2 1 1 1 — 2 12iii 1 i 1-iii 2 - I i i - 1 2 2 1 li!T322!!; 2 2 liiiiii - 1 m * 2 1 1 3 2 2 2 2ii! 2 2 1 2 1 2 1 2 1 2 — 2 — 3 llli l H 3 1 2 2 liii 1 2 2 2 3 2 1 liii |2 * 3 i i i - i l 2 ? 2 1 2 1 i i i — iii ^ - i i i * 2 3 1 2 Iiiiiii2 iii2 2 2 2 l2 3 iii2 2 2 3 2 ii:2 1 -iii2 :ii2 1 3 2 2 iii2 * 3 3 2 1 2 2 2 1 1 - 2 2 1 2 2 -2 2 2 iii *322-;;- iii- *iii “ B riii2 2 3 2 2 iii3 2 2 liiiiii- 2 2 * 1 1 2 2 li!! 2iii - ( 2 2 2 2 2 1 l i2 2 1 2 2 2 3 2 2 2 2 i i i l 2 2 liii 2*22 *iiiiii 2 2iii------ iiiiii -iiiiiiiiiiii211322|-'i-iii— iii - liii 1-iii 2 2iii!ii| 2 12 2iiilii -iii liii 2i:i 1 2 3 |2 2 1 *iii 2222211----- 2p i---- iii -iii --iii222122 iiiiiiM Bii - 1 2 -1 2 1 -2 32 2 l l 2 2iiiliiiiiili liii2-222 3212 22iii222221------------ iii- — — — — iii! 2 3 1 | 2 2i!li!l!=nii 2 2 l i i i l * l l 2 2 2 2iiiii!2 2 2 3 2 1 I I * 12 2 2 2 2iii 1 1 1 2 1 3 2 2 2iii -iii 1 1 2iii -iiiiiiiii — iii - - | i | l 2 2 2 2 2 1 2 2 2 1 -iii 1 2 2 1 - — ill-iii 2 2 * i!i|ii 1 2 3 2I 2I 2 1 2iii2 2 2 2iii2 2 1 1 2 2 1 2iii— 2iii— Iiiiiiiiiiii * — I i i - F 2 2 i i i 2 2 3 i i i |i i 2 1 1 3 3 3 T 2 - i i i i i i - 'H |- * - - £ |2 2 3 2 l |l l 2 3 2 2 2 2 2 1 2 1 2 2 2 2 1 - i i i - i i i H - i i i i i i i i i " * - i i i - 2 1 1 2 l i i i 2 1 2 i l i l l l 2 1 2 3 2 i i i i i i — i ii - * - - - J F 2 2 3 2 ii i 2 2 2 3 ii i 3 1 - ii i 2 1 2 1 p ii i l2 — - 2 3 2 - iii- iii---------- iil21212-iii-12iil21iii22322iiiiii-iiliiiiii-*iii!ir^iiiii222322iiiili221121- 22iii222- | - 21iii---Hi*------ lB 2 iii* lliii* 1 3 1 2 1 2 2 ll2 2 -iii-iii;iiiii|ii;ii— iiiiii2 2 1 — 3 2 2 liii2 2 2 lH lliii2i::iii 1 * l:i= * * -------- l " 2 3iii 2 3 liii 1 ill 1 1 3 2 3 liiiiii- - - - - - - l i P - l l i 1 2 3iiiiii 2iiiiiiiii * 2 1 2 2 22iii3 2 2 2 1 2iii- - - - - iiiiiiiiiiii -iii -iii — iii 1 3iiiiii 22113'iiii iii! - 212iii 1 *— -iii -iiiiiiiii -iii-2 21222 2 --2 2 1 * l | l 2iiiiiiiii 2 2 1 1-iii —iiiP iii - f c 2 5 5 5 2 5 5 5 — *221311 - H 2 312.......: — 223122-122555121555 122 liii 12iii2-iii * iii *---2 2 = 5 5 2 2 2 i ii l ii i l3 i i i2 2 1 2 " 3 2 1 - ii i - ii i - - - i i n ii i i ii i i2 i i il l 2 2 1 2 1 - 2 2 1 3 ii i 2 3 2 2 p 2 3 i2 i i ii i i iiiiiiiii 1 2 2 2 2 - 1 1 2 3 1 2 2iii 2 2 liii liii -J liH 2iii —iiiiiiiii------ 2iii2 32 2 iii(3 2 liii 112i!i31i22122iiiiiiiii *ili— *iii 2 1 2 iii2 2 2 1 iii2 2 2 2 iiiiii- - - iii2 1 2 3 iiiiiiiii4 f liil- liii2 1 1 2 2 iiiT 2 2 li2 1 2 2 iil2 3 iii2 2 1 1 § - - liiip p 2 2 1 2 - l - 2 2 2 i i i 2 2 1 2 i i i i i i i l i - - 1 2 1 1 2 3 i - * |p i i - 2 i i i 2 3 2 2 3 2 1 - |l 2 - l - i i i 3 * 2 1 2 i l 2 1 - " i i i i i i i i - ! i i - ^ i i 2 3 2 2 1 2 iii- lliiiiiiiii2 iii- — ^ii!— xSli— 2 liliill * iiiK * --1 3 -2 2 1 2 1 2 iii2 2 1 iii— iii2 2 2 2 2 2 2 iii-iii-iii— iiiiii-iii-22iii 2 3 2 - - - - - - 2 3 2 3 i i i —iii — iii — iii— liii- - - - - iiiliili;2 2 3iii 2iii2 3 2 2iiiiii 22iiiiii 2 1 1 2 3iii2 2 2 2iii- - - - - - - iii- - - - - iii— iiil iiil2iiiiii— 2 2 2 3 2 iii *1iiiii— ii!B^— B^iiiiii— iiii!T 2223 2 iii2 iiiX 3 2 3 2 2 2 1 2 2 2 3 3 2 2 2 iii2 1 1 1 2 — I ; — • — iii— iii- 1 3 ii— 1 2 2 2 2 1 3 2 5 5 :------* — iii iiiiii-2 3 2 3 2 2 l5 ii2 iii2 3 2 1 2 2 ii5 2 2 l2 1 2 2 -2 3 2 2 i5 il2 ii5 2 -* -iii-iii— 3 12 32 2112iiiiii— * -iiiiii - *iii — 1 1 1 2ii! 12 liii - I i i 222 2iiilii 2 — I l f 121H2 liii 3iiiiii 12iiiiiiiii — iin il2 2 ii -1=112 2 2 2 p i 2iii -iii — iii -iii— iii — iii - — iiiiii l i i i - | l i 2 | 3 2iii 2 -iii l i i i 2 "iiiii 2 3 3 1 32iii2 2 2 -iii-M 2 1 ! ! " i i - liiiiii;;; 2 2 iii iiiiiiiiiiii — iii *iii — 2 p 3 1 2 l | 3 | * 1 3 2 2 1 1 2 2 2 * 2 2 1 - 1 3 2 1 1 2 1 2 3 2 2 ;= - = = = l- iiiK I illiiiiii- i- l- i! ! i! ! - * - iiii- iiil- - * iii|l3 2 2 2 lii2 1 * iii2 1 iii- - 1 2 iiiliii2 iii2 1 3 lii3 2 - 1 2 3 2 iiiii— ;ii * 2 2 iii-2 * 1 3 iii2 -iii-iii-iiil-2 2 p -iii-1 1 2 2 1 Z 2 2 iii-iiiiiiiii-iii iii 1 2 — iii— liii 1 3 2 1 2 1 2 2 2 2 1 ii lH — iii - 1 3 3 | 2 liii 1 1 2iii 1 -iii— *iii — iii-iii 1 2iii 2 2iiiiii 1 2 2 2iii2 2 1 2 1 2 2 1 2 2iiiiiiiii — iii2 2 2 1 2 1 2 2ili 3 - l ii i ii i 2 1 1 1 2 1 2 2 2 2 2iii 2 iii -iiiiii|l2 iii2 2 2 * 3 1 1 2 3 2 2iii 2 2iii 3 2 - 2 - 1 -iii- 2 2 5 ,1 3iii 1 3ii! 2 2 22| 22-;;;T3221221321222 l | - - J - 2;;il*3 2 123221- 2 1112iiil21211- * —iiiiii-ii! iii! 122112!!! 1 2 3 2 2 |2 1222 332221222 321—!iilliiii!ii!i2 2 2 3 *132!!il22 liii222221— iiiiii|l22— 22-1221222 ■ aaa aaaaa**aa aaaaaa aaa aaa aa aH ■ 1 3 3 lii= 2 3 2 2 iiil2 1 1 3 2 1 2 iiiiii2 * 2 1 1 2 2 2 2 1 1 2 2 1 2 2 2 iii-1 2 1 2 2 3 -1 2 2 2 iii-iii— 1 2 2 ------ 3 2 1 - 1 2 2 2 1 1131121112223321*2123-222321212 liii-iii22ill— 122iii23iiil3*-iiiliiiiiili3 |^ - 2 2 2 1 1 1 11)122 221113321132l i i 2322iiill2iiill222321— -iii22| i i l - 12liill2232iii-"il2222i i i - l l - - - i i i l i i i 223| 22iii2222iiii!5 i 2113222-iii223iii2i i l i i i l p l |l 22iiii2iii223122lii2iii-2222iii-iil--"iiiiiiiiiillii-2 117 Figure 41. Map of model reef after 100 years of the recruitment simulation with aggregated settlement to SC2 (for Acropora subglabra only) and no competition. See Figure 27 for initial reef configuration and Figure 39 for segregation indices. Only the first SO rows and 70 columns of the 80x80 reef are shown. L e g e n d : - : s p a c e , A S : * : S C 1 , jii:S C 2 , l : S C 3 , A P : 1 : S C 1 , 2: SC 2 , 3 :S C 3 » R u n # 6 3 C o v e r : 1 6 4 1 3 4 7 3 1 3 1 2 3 6 2 5 4 4 3 8 6 r e s p e c t i v e l y . 3 2 1 3 3 iii2 |ll2 2 1 1 3 2 [ 2 2 liii *2iii 5!ii -2 2 2 2 2 1 - 2 2H * 2 1 2 2 -2 2 1 2 2 3 2iiilin iiii 1 1 — *ii:222— 2 1 2 1 2 2 — 3!il!i!2*!l |ii2 2 2 2 2 2 2 12222111*22!! !!!2 * !!!2 !!!|!!!!!iii3 2 1 1 2 2 2 2 3 2 3 1 -2 -!ii|2 1 3 1 3 1 1 * -1 2 -1 2 i!i|iiiii K * V iiiiiiiii'2 3 2 1 - 2 2 2 2 2 2 2 2 2 1 2 l! ! ;2 2 2 1 1 1 2 1 1 3 2 2 2 2 iii! ! r i3 “ ! ! l i l 2 i ! ! 2 * - 1 1 2 1 2 1 2 2 2 2 2 1 2 i l 2 - l l - 1 1 3 2 1 2 2 r 2|22iiiiii|ii|2j»2*iii: 122!!!|!!"l!3p l | i ____________________________________________ _ 232iiii!iiii 1113*21 3 » !!iilll2 iii2 * -1 1 2 1 2 1 2 2 2 2 2 1 2 1 1 2 -ll-1 1 3 2 1 2 2 B i iii3 2 2 3 1 2 1 -2 3 iii3 p iii-1 2 2 2 2 2 2 1 2 2 3 1 1 -2 2 2 2 2 1 1 2 2 2 Ip 3 2 iii2 ! p 2 1 2 2 1 — 2 1 2 iii" ii2 iiil2 2 1 2 1 2 2 -2 2 -2 3 -2 -1 2 2 1 1 2 1 2 iiip l!!!3 222211122221 2 2 -p * 3 2 2-!ii3112112232iiiiii22-iiil21 2 2 3 1 1 2 2 3 1 3 2 2 1 -2 2 12331iii * 2 2 - - *— 223 ll-2 1 1 2 l2 2 2 i!!2 11211222-22-22113-!!:;;! 1223— 2222121-232123321-2IIIIII2211 ::;11112222312:l:Illlll22221 3 1 1 2 2 1 -2 1 2 -2 2 -2 * l p 2 2 2 1 2 1 2 2 1 3 2 2 -3 2 2 11222-11IIIIII2112 2222211-122*P !!!!!!p |2 iii3 iiiiii2 2 2 2 2 2 -2 3 -2 1 iiiiiip 2 2 3 1 2 -2 1 1 1 3 1 -1 1 1 2 3 3 2 2 2 iii2 * 2 2 2 2 2 2 3 2 1 2 -2 2 -2 liiiiiiiiiiiiiii!!?2piiiiiiiii21 -3 2 2 2 1 2 -2 1 2 2 llliiiii— 1 2 1 2 2 2 2 2 1 2 2 1 2 -3 -2 3 1 1 2 1 2 2 2 2 2 2 2 -2 1 2 3 -1 2 1 2 iii2iiiiii2iiiiiii3222iii22122222132iiiiii22312iii|212-2 * 2 1 1 3 2 2 2 2 1 2 2 2 1 1 2 2 2 2 122iii21 2 -1 2 2 2 l-2 iii? * 2 2 2 2 Ippiii-212222iiiiiiiiiiiiiii2iii32iiil-3iiiiii2ll2221 2 12iiiiilii! 121212 2 1— 232222::::::111— 111:::::— :::12*:::2:::::::::::il2— 223::::::2:::*llH221::::::|ilB::7222223— :::»:::::::222222 1 -l-2 1 1 2 2 p 2 2 2 1 2 iiiH iii!!ll!!!iii-liiB !!il2 -2 2 2 iiiil2 2 2 !iiiiil2 !ii* l2 1 2 2 -2 -2 2 !iiliiiii2 1 2 2 -2 2 211!223123=112111-1 lllii*!!! 12 liii 2 liiiSIliiiiiii22 2 p H 2 i i i 2iii 2 liiiiii *2 liii 2 liii 2 2 2 2 ^ iilii 12 2121 ii! 32 -122!!!!!: 122122 liiiyjp 3 !:!:!! 212 ^ iiiiiiiiiiii 2 3!i!iiii!!i -112!:! 2!!:-2=1=3 — iii 2 3 2 2iii 2ffl22 2 2 2 -1 2iiil21212!!!-2122222l|j|ii-H 2221l|2!!!i!ii!!!311l!!!*!!!l22212-!!!22!!!212*22122 *— 2!!!2 * !!! 32 2222iiiiii2ii!iii321iiiiii312iiiiii2illl-liii*22iii2iiiiiill2-iii2-22212222iiiiiil222-2222iiiiii2liiiii232 iii 2 1 2iiiiii liii 12 2;:!llllll 22 222 2 1 2 3 1 — 11121222!!!2112— 2211 2 2 1 2 2 2iii - * 2iii 1222 l K j i i i l - 1 -2 2 — 2 iliiiiii32iii * 3 2 2 -2 2 2 1 2 1 2 1 2 2 2 2 1 2 -1 2 2 2 2 1 2 1 2 1 2 1 1 2 2 2 13l!ii 12i!iliiiiii 12222Bii311iii22 3 2 1 2 1III::::::III221 1 2 2 1 2 2 2 -1 2 -2 1 2 1 1 1 2 2 1 2 2 2 1 2 2 2 lllll2 2 1 2 2 2 2 2 2 lH * -lllI-* l2 * ll2 * ;!:lll2 -!!:2 2 2 2 32iii 2 |i! 12 22 2 3 liii 132 21212 -2iii * 2122221212iii 222 2 2 -iii *iii 1 2 2 2 1-lil 2 = = = = iiiii 2 l= f!ii 2iii 12 liiiiii 2 2!!!!!!!!l!!!-!!!2222-22112-232222!!!!!i22211212112222222-l-2212-2232l!!!!!!!!!l;!23!!!!!!223 m i f : ii222-222223B ll 1 2 -2 2 2 2 2 2 2 1 1 2 1 2 3 1 3 1 2 -2 2 2 2 -2 3 1 1 2 2 2 2 2 llli2lKlllillillllllllli 13 Ii22lii*2— 1 3 1 1 2 1 2 3 1 l|ft-iii2 2 1 -1 2 2 iii* 2 3 1 1 -1 2 1 2 2 2 1 3 1 2 2 3 2 -l-3 2 1 2 !!!-* ll!!l!!!!!!!!2 2 _ |f l 2 1 1 2 2 2 2 2 2 2 1 2 2 2 122pj!!!!! 1-I212i!!i!ii!i2222222222221 2 3 2 2 2 2 1 1 2 1222;!!-lT222!!!!!!p ii!!iil32122222221311222i!i-2221122-iii21iii2||l32222-222-l-122222212!!!2!!!!!!2l!!!!!!!!:2 iii 1 1 2 1 -2 2 2 2 2 1 2 iiil-2 2 2 -2 2 iii-2 2 -iii* -liii2 ^ IiiH * 11221221112222312213^22!!!!!!!!!-1-2!!! 211121222222iii!!!Tl21233222123:::iiil22iiiiii2l!!!!=22HP12213-2112112*122iii2223!!!211 “ 222*1222-22iii2iii222-22222iii22iii2iiilll*2*iii*2iii|!!|ii232122222iiiiii211i 12!!!!!! -iii 2 1 2 2 3 2 1 2 2 1 2 1 - 12iii 2ii!l2iiiiiil - - 2 2 p 2 2 1 2 1 2 3 2 -||! I l l l|j!!!! * M jj||j|IS 2 !!! * !!!■ * !!! 122 2-2 2 322 2212-i!!™ !! 2 l 2 2 liii2:ii 3 2 1 2 2 2 2 2 2 !ii2 2 2 1 2 2 2 2 2 3 — Iiil 2 iiip * 2 2 2 2 2 2 1 2 2 - 2 2 2 2 3 1 2 3 2 - ! ! ! 2 2 : l i ” 2 2 - 2 2 1 2 1 2 1 - * - 1 3 2 y ! ! ! ! ! ! 2 2 p - ! i i | 2 - * 2 2 3 2 2 3 2 2 2 2 2 2 l | 3 Iiiiii!*!!! 2 3:!:-*!!! 3i iiiiii2 * 2 2 1 2 2 2 - 2 2 liiiajfeiii 2i!!|!i 1 3ii!i!i 3iiiiii[ 2121111 2 3 1 1 1 1 1 — ii!iiiK i 1 2 |ii!ii liii liiiiii2 p p l 2122221323!!! *iii li I 2I 2 12 2iii 2121.2 -Il:: 22 l!!!iii2222312 Iiiiii! 2 2 2!!!!!!! iiiiii 13 * !!!2 32 221 * i! ! 2 2 -ii!I piiiiiiiii *21222 - liii * p i i i i iiiiii! 2!!! * 12 2 -3i!!liT=! ::;2 2 2 1 2 2 2 2 [_ 7 !!222223112l!i iiiiiiiii 2iii 121223 liii 1 * 2ii -liii*|3iiiiiii l l l l l l l 12=ii222— iiii i ;!!32;li221-2122| 221111*12222111 2i 2 » ? 2 -l-iljl* tlsHBBBCtt. ■Ishii _ _ iiiiwii H jiip ip : 32 2^isi It:: * :t:2l:t * |: : 1 2 |12 11 iii— 2 2 T 1!!iiiH !i!iiiiiii! 2 * 1 ' - -•"K W « :8 ,»85- 2-32332312222iiiiii3iini[2lB i!Pp*2322iii!!ri22-12piii!22332222t!!22112-133!!!!!f*!!!2i!!2 2 2 121321212 Ilill::illl::2 3 2 2 i2 2 !l!2 2 1 1 1 2 ill[2 2 2 1 ll2 2 3 1 3 2 122221%:: 1 2 1 1 1 1 1 -3 2 2 1 2 — : ! ! : ; : 21112222-22-iii»«i!iiil!i!2122— 2 !lip lll2 2 2 |l3 1 2 p ii2 3 2 2 2 2 2 3 2 2 2 2 -2 2 2 1 2 2 2 1 3 1 2 !!! *! !!! !! 1222221-22iii*iiiii!piiiiiil|22-313!!!!iili!l*3iii32122iii-3212-22221222222122222!!!!!!!!!2!!!!!! 2 3 1 * iii2 2 i!!lp i!lj3 ljiiiii3 -l!ji-2 2 jij2 |ji!i]2 --3 1 2 2 1 1 2 1 2 1 2 --2 2 3 J 3 2 2 2 1 2 1 -2 2 2 2 M !!2 I!!!!!:2:;;212!!:|l2!::2!!!l::22!!!l-!!!222!!!:::!!!2p3212122112222-----2222!!!1222=!!2112123==f2==T-2 3iii^ii-liiiiiiT 2iii2iii*:::= iil2i32121-*iii2iiip22221-2-2212111212B !!p 22!!!23122!!!!!!lp l2 118 segregation index 0.60 0.40 0.20 0.00 10 1-cell disturbances per year 2.00 ° w ° s s « n ? « n 8 years S £ S 8 8 8 8 1.80 ■ 200 1-cell disturbances per year 0.00 Im n-H i i h i h i i h i h ii ii iii iii ii ii ii ii ii iii m inim ii ii ii 1 1 iiiniiiiim ii i ii ii i iii i m in i ii h i 0 ‘n ° “’ 8 R S 8 ? 8 8 f f i 8 8 ° S 8 8 8 $ 8 y e a n Figure 42. Segregation indices for two disturbance simulations (without competition nor aggregated settlement). No patches were formed at levels of disturbances ranging from 10 (upper graph) to 400 cells perturbed per year. See Figures 43 and 44. Figure 43. Map of model reef after 100 years of the disturbance simulation with random settlement to SC2, no competition and 10 random 1-cell disturbances per year. See Figure 27 for initial reef configuration and Figure 42 for segregation indices. Only the first SO rows and 70 columns of the 80x80 reef are shown. L e g e n d : s p a c e , A S : * : S C 1 , jii:S C 2 , l : S C 3 , A P : 1 :S C 1 , 2 : S C 2 , 3 :S C 3 » R u n # 1 0 0 C o v e r : 1 2 0 1 2 1 5 2 8 9 6 2 5 1 2 5 4 2 1 2 r e s p e c t i v e l y . Figure 44. Map of model reef after 50 years of the disturbance simulation with random settlement to SC2, no competition and 200 random 1-cell disturbances per year. See Figure 27 for initial reef configuration and Figure 42 for segregation indices. Only the first 50 rows and 70 columns of the 80x80 reef are shown. L e g e n d : s p a c e , A S : * :S C 1 , iji:S C 2 , | : S C 3 , A P : 1 : S C 1 , 2 : S C 2 , 3 :S C 3 » R u n # 5 0 C o v e r : 59 4 4 4 8 8 3 4 1 5 4 0 8 6 r e s p e c t i v e l y . Figure 45. Map of model reef after 10 years of the disturbance simulation with aggregated settlement to SC2, no competition and 500 random 1-cell disturbances per year in columns 20-40. See Figure 27 for initial reef configuration. Only the first 50 rows and 70 columns of the 80x80 reef are shown. L e g e n d : s p a c e , A S : * :S C 1 , |j : S C 2 , ( : S C 3 , A P : 1 : S C 1 , 2 : S C 2 , 3 :S C 3 R un # 1 0 C o v e r : 1 3 3 6 6 3 1 4 6 7 2 2 9 6 1 1 4 6 r e s p e c t i v e l y . 1*12-2212-1212---------- 1— lU i 2-----------1---------21------------ II I---------- 1-22-1222!!!-- P ' ” i 3 2 ~ 112212-2------------**!!!------- 1---------dP-------122----------------------------2?.? 3----- ijjB ij 2 1 — 2 1 — 12222322---------- * -— 11121----I ii— 2 -2 1 1 2 — 1 2 -2 ------------------iiiiii 2 2 —= :!■ !! :::— !!! 1— 2 1 — 1211132123— 21 ---------------------------------- *— 1---------------21 ---------1 -2 -----------V - 2 — 2 -W - — --* i-2 1 3 2 2 2 1 2 li;2 -2 1 -------------- 11 -1 1 !------------- Iiiiii---------- 1— 1-----------2 2 2 2 2 -2 ------- ---------- ■ 1 3 2 2 2 — *— 1------------ 1------- 1— 2---------- — | — 1 1 ! ------------------------ 1322221121111- -2 2-----11*21-2 2-iii------- 1-------- 1 iii 1-----------" -Iii------------------ 1 -1 -2 2 1 -2 1 1 3 2 - 232 liiiiiiiii 1 * liii-------------------------- iii------------------------iii-IP-ill-------- 1— 12132lliiii 2iiilii 2212---- — *iii“2221----------------------------------------------------iiiiiiiii--------------1— 12 3liiiii 1— 222 | -iii-2 2 2--------11------- *— -iii-iii---------------1 1 1 -2 2 -1 2iii2 *1111222------- 1--------1- B i l l - * * -------------2 1 - 1 2 1 2 2 1 1 - 1 -1 2 2 ............................ - 1 - 1 1 -------ill— ill— 1— **!l!----------------- 2— 21— 2 2 ----------21 212-Iiiiii -Bjllli 1 1 1 1 ! -iiiiiiiii - 1 1 1 1 -------1 - 1 --------- lllBiiiill------- ii!-------------------lll-* !!!-2 1 l!l-------- 12 1211-B!!-Il2-!!!2!!!l!!l-------------1 -1 ----------- lilii-2-----11— *--------------- 12-*1*— 4 ------3 - 1 1 1 2 l l l 2 2 1 2 - l t ------------------------ 1--------------III----------1 2 -* — 1------------- 1— illilllll-BIIIII1212 2 - 2lllli! * !!! 2 3 - 1 1 1 ! lilli -ii!----------- ill— 1----------- 1-----------1 -2 2 -1 ------------------ illl!!-l!!--=l!l-------- | 3 ill! * - il l 121-132 * (£ * ---------- * III--------------------------------------------12—I I I - — II!— ----------- -l--!!!* -2 2 -!!!-:!!!!i-« * ------------------ 1-------------2----------1-----------!!!!!!!!!-Iiil!! *ii!- 1 — *212222 2i!!!i!-iliRli-------------- ! ! ! * * * -------------------------------------1--------------- 1 1!!-!!!!!!!!!-*!!-* 2 M- — 2 3 21112211121-ill *lll|---------1 - - — *------------------ Iii------------------------------ il;— ill!!!!!!!!!!!!— 22 iiiiiiiii— 2 1 3 1 -2 2 — --li!***!!!— 1 -1 1---- !!!*— 1------- 1----------------^ 2— -212 2I 2 2 -2 2M ll 2 2l l l — ill * * * 1 1 1 1 1 1 — 11 -1 - * |jj *-------------- i||------2 -1 ---------------- I I I — 1- 122- 2 " 2 2 3 3 2 lT i2 2 2 2 -2 1 -* !!!p---- 11— * -* 1 1 -1 --------1 — 222—!!!-------------1 ! ! ! ! ! ! ! ! - 2 -III 222 12 2 -1 1 -2 3 2 1 2 2 (2 1 3 2 2 * -1 ----------- Iiiiiiiii*-------------- 1 — 22132*---------------— 2— - — 1*-2 1312122311112 Iiii2( l 2 2 2 1 1 1 - 111— *-*-*111111------------- 1------1 - * — 12------------------ III 2 - 2 - * 2 3 1 i- l- 2 ! ii 2 l( 3 * 2 i il - 1 2 2 1 1 2 ------- 1 * — l i i i f i 1 1 -------------------------- 2 1 ------------III 22lll! 3 1 2 — I iiiii 2 3 1 — = = ; 2 3 3ll! 1222122 - ---- Iliii! **12:H!!(!I- -- -- - -- -- -- - -- -- - -- -- * - - - - - -1 -1 ------- ! ! ! ( - — 21222 2 2 — III!!!!!! 1 Iiiiiiiii! 2 1 2IIIIII3 2 - 2( l l — 1 * !!! * * ------ *I?T -1----------------------------1— 2----l!!lj!|l - 11 3 2 2 liii-!!! Ill— 2-ll|Bil------- 1 2 - 1- T ------------------------ * !ll 1------1 - 1 ------ - ! ! ! ( -2 1 1 1 — 2 -* -!i !------ l — :ii 1 -1 2 1 ------------------------------------------ 1- iii- ----------------1 - 1 1 -------------- -n?1 2 2 2 -1 2 1 2 1 2 2 2 -2------- ill— 22---- 1-------- ( ------------------- 2l!!-!i!-------------- 1-------------**122--------- 11 2 1 2 -2 2 2 2 2 1 iiiiiiiiiiii------------------------------------------2 - ( l l ------------------------ 1 ------*!!!2 2 l ! ! l ------ 22 22 2 2 3 1 2 -3 1 ( i i *---- 2 ------------------------i l l --------------------- 1-------1 3 - 1 - 2 -!!!-ll — IIIB 1 2 2 2 2 --------------1 1 - 1 ------- 2 2 2 - 1 ----------------------------------------------------------- l l l - 2 2 2 - * 3 - 1— — 2 1 1 -2 ------ 111-122112-12-312---------- iiil------------------------------------------------- 312121 1-111232213221-1111221312— 231-1 2 ---------------1 -1 ---- 1 1 1 — 21----------------------------- 2 2 1 -2 2 - (-2 -2 -3 -1 1 1 1 1 1 1 — 2 1 2 2 2 3 2 — ( 2 -------- 1 -------------- 1 2 --------- 1------------- III2III l l - ! ! ! 2 - l 2— { I ! 2 2 2 2 3 - d p -* 22- 22( - -III----------- 11---- 1122!!!-------------------------- iii— 2! ! ! ! ! ! ! ! ! ! ! ! 2 1 1 1 :!!2 — 322 2llii:i -III— 1222*1!— -II!*-----111--------1132-------------------------- !!!---B!!-2-!!!B!!!l!!!l2 — Ill--------------1 1----- 1122!!!---------------------------------liii— 2jiL; * ii— in*— i n ------- 11 3 2 ------------------------------ i!!---B !!-2 -!!!B !!!l! ! ! ! 2 1 1 - 3 1 1 2 - 2 2 2 2 1 ----------------- 1 ---- 1 - 1 3 2 4 1 ---------------------------------- 1 -lulfl 2 2 1 — 2 2 : « l - l - :ii— 1 2 2 2 2 1 2 1 2 1 1 * * * -----------------------2 2 2 - ------ 1 ------------------ * 3 2 — -1 2 2 M II* H I(2 2 2 — 2!!lill!ii—1 2 2 2 1 2 1 1 2 — 11112*111 * ------------------------------2 ------------- !!!-!!!-------------- 2 B * 2 2 iii2 p i 122!!!!!l(j|2232-2-31---------2 1 2 ------1--------------------2---------- f -------------- ! ! ! ------- j!«ii212-!!!!!i — 32*i!!i— ( 2 2 2 — — 3 — 2 2 ----------------------1 ----------------- * — 1 3 2 -----------!!!(---------------------------------------------- -- - - 1 - 2 — 2 2 1 2 -— 2 1 2 -2 — 232-ill--------------------------------------2§2--------- !!!"----------------Hi!— 1---------- 2121 2 2 - 1 2 - 2 3 2 1 1 2 2 2 2 2 2 2 4 1 1 ------------------------------1-------- * -----------— ----------------- ill-------------- 2 2 ljll2 3 111312 1 1 1 2 1 1 2 2 2 1 1 1 2 2 2 1 ------------------ 1 2 -------------1 1 ------ * ------------------------------ !!!!!!----------- 2 2 2 ! ! ! ( |- *— 2 l 2 1 2 - 2 - 2 2 ( i ! 3 1 — * -------------------iliii! 1 ---- 1 1 - - ( * - 1 1 1 - 1 — III------------------ — --------------2 2 - — - lij— 2 2 1 2 2 — 2 2 1 1 3 2 i — 2 -------------2 ---------- - ( - 1 - 1 — 1 -:•• -* -iii * -------------------------------1 - 3 2 2 ^ —2 — 1 3 - 2 2 2 2 2 2 2 - 1 2 2 1 — ! |— 1 --------1 1 1 - * --------- ! ! ! ! ! !! j j------ --— 1-----------1 -------------------- 1132jjjj!!!!!- 122 Figure 46. Map of model reef after 100 years of the simulation wherein corals that are surrounded by 7 or 8 empty cells die due to exposure. Settlement to SC2 was random, and there was no competition and no disturbances. See Figure 27 for initial reef configuration. Only the first 50 rows and 70 columns of the 80x80 reef are shown. L e g e n d : s p a c e , A S : * : S C 1 , iii:S C 2 , l : S C 3 , A P : 1 : S C 1 , 2 : S C 2 , 3 :S C 3 B u n # 1 0 0 C o v e r : 4 9 4 8 7 " 1 0 0 4 0 4 6 7 9 1 0 6 1 1 2 — III— — 1 — 1 !!— — 2 ------1 — 2 - 1 3 -----------III— 1 2 -------- 1 - 1 - 2 2 - 1 - 2 -------- 1 1 1 ------1 — 1 1 1 -------- H i 2 ------ 2 - l - — lll|- lll2 1 - 1 - 2 3 -------------- 13111— 2 — 1 -------2 --------------- 2 1 -------- 2111-2-------- 2 1 2 — 1 --------- B - 2 — 2 - — ’ ---------Iiil 2 ----------------3 2 ------- 2 - 2 - 2 — ill-1 -------- 2 ------------------ 2 1 III------ 2 — i |] ----------- H i 1 - 2 — 2 - 2 1 ------------| | l - * - l ~ 2 ----------------------------- - 2 --------- - 2 1 -------------2211;--------- 1112 --------- lJ |iili_ 2 - lj i- 3 — * — 1 2 ---------- 3 - 2 --------— — 3=Ii------ I II I II — 2 - - ■ ------ --— 1 - 2 - 3 - 2 — 2 — ------ II I— 1 2 |j |- - : - - - - - - * — I I I - — j]j2- - - - 2 - - - - - - - 1 3 - - - - - - - - - 1 - - - - - - - 3 - jjl-22111-12- - - - - - 1 “ H I I - - - - - - — i: ---------------2 -------- - 2 H i 2 --------------3 ---------------l i i i - * — H i— i -------------- --1------------- 1 — 1 — 2 — 1 --------------------------Iii— — 11112------------ 2 ------------------------- 1 — B ii— 211112— 1 — i|l------------III-IIIIII-III — 2 ----------- * l | — ill-3 1 -ill— 1 ------ 1 - 1 ----------lillll-2— III------------- 2 2 - — Iii---------- l:i - * 2 — 1 -------------2 - 1 - 2 1 1 1 — 3 — I II---------------* - 3 - 2 — - I l l - I l l 1 2 - 2 --------------------3 -------1 --------------- 2 ----------- -------------2 -------------------1 — I I I — I I I 2 ------3 -2111---------I I I 2 - 2 -III 2 - 1 ------- I I I 2 ---------------III----------------------2 — 2 I l l - l ---------- 2 — III------1 2 - J - 2 1 2 ------I l l l 2 - 2 2 i 2 - l l l - 2 |2 1 1 — 2 — *21-1111------2 — J — 1 — * - — 2 - - 1 2 — 1 -III 3 — I I I 3 3 - ■ 1----- 2 ----------- 1 2 — -------- 2 -Hi-----------------B - ^ l l 2 2 2 — ii!«?----------- 2 - 2- - - - 2 2 - - - - - - - - - 2 — | - - - - - - - l i i i - 2 2 - - - - - - - - Hi |H 1 - - - - - - - - - - - 1 — il?- - - - - - - - - - - - - — 3 l l i - - - - - - - - ■ l - 2 -----------------1 -------- 2 ----- 2 - 2 - 2 — 1----------------------------- 1 ---------- 2 ----------------2 ---------2 — 2 — III— ■ --------------------- ;;i-------- 3 — 1 — 2 ------ 1112----------2 ----------------------2 ------ 1 ----------------- 1 * — 2 1 ------- 7 ------------------------ 2 - 2 2 — 1 — Ill-Ill----------------- 2 - l l l l — III---- 2 2 llil-2111--------------- * --------------- 1112-------- — 2 2 1 -------------- iii 12111----------- 2 - - i ||X ---------2 | - l l l - 2 2 -------------|||1 2 --------------------- III-III---------------- 1 — ------2 ------------------- 1 - ----------------- 2 - 1 -------- 1 - | ------ -------- 2jji * - ------ 1 ------ * 2 ---------2 -------1 — 2 1 - — ---------- 2 ---------- 1 -iii-------- i n --------- iil2 — 2 -ill------- -------------J — - 2 -iii-iii------------- 2 — — 1 -------------------1 — III— III— - 2 — 2 - — III— 2 - ---------- 2111-2------------- IIIH2 ----- ■ — 2 i i i - f c - 3 — 2 - - - - - - - - - - - - - - - - - - - - - 2 - - - - - - - - 22-11123- - - - - - 3 1 - - - - 1 — - - - - - 1 j l l l - - ” l i i 1 - 1 - 1 - 1 1 “ 1 — | — iii— ( - 3 — * ----------21--------------1-jjjl— i — 2 — 2 - i i — iii------------ 12 ----------- 1 ---------- 1 llillll-2 — 2 ---------------2 3 2 ------ 2 - 2 - 2 - ------ 3 - 2 --------------- 2 — 3 - l l l 2 - 2 | ----------------- 2 1 * - 2 — 1 - 1 — 1 — -III----------- I I I 1 --------------2 - 1 --------- 1 — III--------III -III 1 - 1 - p i i i - III- 3 * — 2 — 2 ------ 2 ---------| i | ^ - 2 ----------ljl— 1— 2 - — ------ 2 — 2 — 1 — ;ii-2 ------ ljjj------------- l l l r l - - 2 — 3 - 2 ------------------------------------ 2 ------------------- 1 - ----- 2 2 2 — 2 2 ill-2 -lll— 1 -------2 — 2 3 liil--------- !!!- 1 --------2111-2---------1 --------------- 2 --------------H i— III----- * — lli- 2 - iil- — 2 l i ! -------1 — iiiiii— - l l l l - l — 2 212IH iii3------------- 1 - | — n - - | — iii— n ii- - iii2 ----------i-iiiiii— 2 2 — - — 2 -iii-iii— — .— 2 --------------- 1 ------ III------I II— " ---------- 2 2 -----------I II---------- l - l l l l -------- 1 ------------------------ l l l l — 1 2 1 ill-ill 1 ------1 -----------2 1 2 -------------l l l l l — 1---------III------ 2 2 - 1 — 2 -IH— * 3 - 1 - 2 ------- 1 -------2 - " - --------- 2 H im III 2 - 1 -------------- -2 1 1 1 1 1 -2 2 --------- 22 -------- 2-|— * 2 2 * -----------2 - 1 ------------- - 2 ----------------- 1112— 2 -IIIIII-------------------- 2 — III-------2 l l l l - -----------liii 1 - 2 - 2 --------1 — 3 — 1 - 2 - 2 — 1 ---------------— 11112----------------- 2 III-— 1 3 ---------- ■ 12 — 2 — " - 2 ------- 2 - 2 ------------1 2 - 1 — 2 -------- 1 ------ ■ - ------ 2 ----------- ill- 2 — '--------- 2 1 ------2 ---------* - ■ III12III-2 2 III- — - * 1 1 ------1-22=1121-------------- llll-III-------- 2 -------1 — I I I------ l l l - f *-111-111— * ----------1-!!!— 2 1 1 - 2| ~ 2 2 * 1 — 2 — 21112------ 1 2 -------------------- I II I II — 2 - 2 - ------- l - l l l III-— 2 — 3 ----- I I I I I I 2 — IIIIII- 1 2 -----------2 -------- ( — l | i l - 21— 2 --------- ! H ( - 2 3 - l ---------------I I I I I I — III------231-111--------3 1 — llll-------— — -------------- 2 2 3 - 3 1 ------— — l-2!!lljj— 1-il!-----------III 1 2 ----------1 — 1 — * 3 3 - 2 ---------- - — 2 — 2 1 f - 2 ------- ||l 2 — 2 - 2 ------ 1 — I I I I I I — - as— 2 -------------1 - 1 — 2 3 ------------ I I I 2 ---------------- 3 III 1 - — I ------- 2 ------------1 -------- 1 2 2 2 -----------l - 2 - l l l ----------------- 2 -------------2 --------------------I I I llll— * ------ - 2 2 1 1 1 - 2 ---------- 22-2111 III 1 - 2 III— 2 III---------------- llll— III- 2 - 1 ------- 2 - 2 --------- I ll- - - - - - - - - - - - - - - - 1 1 1 2 3 3 2 2 - 2 - - - - - - 21 - - - - - - - - - - - - - - 1- - - - - 2 2 - 1 1 1 2I I I - H I2 2 - - - - - - - - - - - - - - - - - - - - - 2 123 AP (page 100), the species generally produces a large number of fragments, albeit with high mortality. In a simulation meant to examine the effects of death of recruits in some parts of the reef and force formation of patches, new recruits (SCI) were made to die when surrounded by seven or eight empty cells. This simulated death due to overexposure did not lead to patch formation either (Figure 46). Discussion Comparisons of the size-structure of AS and AP in the field with those resulting from projections of either sexual or asexual reproduction suggest that both species reproduce mainly by asexual means, i.e., by fragments breaking off from the mother colony. This supports the findings of the genetic studies (page 66) and observations made in the field. During the monitoring studies, no sexually derived colonies of either AS or AP were encountered. As can be seen in the projections of the sexual reproduction matrices, if the corals studied reproduced using planulae, they will need to produce more than a dozen new recruits per colony per year just to maintain their numbers. And this is likely to be an underestimate since SCI coral transition rates were based on monitoring of corals that were larger than the usual sexual recruits, leading to lower mortality estimates for this size class. It is unlikely that such high numbers of new recruits could have been overlooked during the monitoring dives. However, this does not mean that such sexually produced colonies are absent, since these would be symmetrical in form and with branches emanating from the base of 124 the corallum (the main diagnostic features of sexually produced colonies; see Wallace 1985) only when newly recruited, and thus easily confused with colonies from other sources when larger. And, as the automaton simulations demonstrate, they should recruit in aggregations next to adult colonies, if the patchy distribution is to be produced. Most of the substrates in the "plateaus" are unconsolidated and therefore unstable for planular settlement, which means that larvae that do recruit will have to do so near parent colonies, or on rock outcrops. This may be facilitated by the modified water flow created by the parent colonies. Note that higher (but not significant) densities of unidentified acroporids described earlier (page 52) were found settling on tiles in the AS patch - suggesting such aggregated settlement may occur. Although no AP and AS were found attached to the rocky outcrops in the sites (which are often dominated by other species of corals), this absence may just suggest low recruitment success in the years prior to the study. With AP, even if evidence shows a small number of genotypes comprise a patch (Ablan and others, in press), other studies show the population within Puerto Galera Bay are in Hardy-Weinberg equilibrium, indicating sexual reproduction is also occurring at that larger scale (Ablan and Licuanan, in prep.). But it is likely that sexual reproduction occurs rarely, given the size-structure observed, and the simulation finding that patches produced by aggregated settlement of planulae will be more irregular and remain interspersed with that of other species (page 102). The latter are true because the higher mortalities of these SCI recruits rapidly produced gaps in the forming patches (see Figures 40 and 41) and quickly changed patch boundaries. Thus, patch 125 formation will be weak with aggregated settlement of SCI recruits if competition does not operate, but asexual reproduction (production of SC2 fragments) can produce well-defined patches even without the competition. The latter statement applies only if AS and AP are considered and when both reproduce asexually, however. The finding of the preponderance of asexual reproduction was expected, since fragmentation is the most viable mode of reproduction for an environment such as this. Asexual reproduction leads to the production of propagules at near-adult sizes that can be expected to be more stable, and consequently have lower mortalities on the unconsolidated substrate (Harriott and Fisk 1988). This also maximizes the retention of such propagules in the parent's "proven" habitat (Highsmith 1982). Given the limited mobility of these asexually produced propagules and the higher survival rates associated with larger fragments (Highsmith 1982), the formation of patches can also be expected. Rough computations based on average growth of AS-SC3 individuals that showed increases in size during the study suggest that the species can sustain growth sufficient to produce 3.2 SCI colonies per year. Comparing this with the 2.6 recruits per year needed to stabilize AS population size shows that fragmentation could account for all of the reproduction this species needs to maintain its numbers. Comparison of the adjusted (asexual reproduction) matrices of AS and AP suggests differences in the life-history strategies that the two species follow even if both reproduce asexually. AS is more Porites -like in strategy, with lower recruitment levels 126 but higher survival of large corals (Potts and others 1985). The population studied is sensitive to changes in size and numbers of adult colonies, which makes sense if it is to survive on a substrate where large size means stability on an unstable substrate, and where there is a need to produce large fragments / propagules. AP, on the other hand, is more sensitive to changes in growth rates (transition rates to larger size classes) and survival of small colonies. It is tempting to speculate that this is due to a need for the species to quickly achieve a large size with several branches — possibly to make up for lack of branching complexity needed to modify water flow and facilitate feeding. In the case of AS, the bottlebrush form of the species, in which the radial corallites are elongate, encapsulates this three-dimensional complexity within a single branch. Three-dimensionality is also an important feature that determines survival of fragments on loose, unconsolidated branches since this increases the likelihood that parts of the fragment can remain unburied and grow, since those that are buried lend stability to the rest of the colony (Highsmith 1982, Harriott and Fisk 1988). AP must depend mainly on high asexual reproductive rates to maintain its population densities. These differences may be the reason the two species responded differently to disturbances (Figure 45, and see below). As was mentioned earlier, the automaton simulations show that reproductive events are sufficient to explain the formation of patches by both AS and AP. The formation of patches occurs regardless of the original arrangement of parent colonies (Figure 38). The sizes of the patches will depend on the population growth rate and the 127 initial densities of corals however, since large patches form by fusion of several smaller ones. This may be the reason individual patches are composed of a few genetic individuals or clones, but two patches never share the same clone (Ablan and others, in press, Ablan and Licuanan, in prep.) Competitive interactions will still need to be invoked, however, especially in the case of aggregated settlement of planulae (SCI) since these do not always form pure stands (as can be seen in Figures 40 and 41). Competitive interactions can help maintain the dominance of the host species in these patches but the patches will still have irregular boundaries because of higher mortalities of the SCI recruits. Patch formation is not just a matter of producing aggregations, but requires persistence of the colonies in the patches. The model presented only includes the two dominant species in the area. If the other coral species in the community are considered, competitive abilities will also remain essential in maintaining dominance in patches by dealing with planular recruits. The many pocilloporid corals on rocky outcrops and settling on the tiles show that there are other species that could be found in the AS and AP patches if the interactions found in the field studies where not operating, both at the level of adults and planular recruits. And, as the small patches of the pocilloporid Seriatopora hystrix prove in the First Plateau, the absence of these corals in the AS and AP patches is not merely because they are unable to settle and grow on the unconsolidated substrate. 128 As seen in the field experiments, competitive responses of the coral species studied (e.g., digestive dominance of AS but peripheral encircling in AP) can lead to different outcomes. Thus, one species cannot be expected to become dominant in the area solely through competitive means. If only AS and AP are considered, as has been done in this report, the rate at which these patches grow by asexual means may be too low to lead to widespread contact between the patches, when interspecific interactions become important. Measurements of patch extension rates showed no appreciable changes in the size of AS patches, although AP grew outwards at 1.3 cm/mo. The latter species does, however, tend to develop gaps in the patch centers, due to reasons speculated on earlier (see page 70). This, and the generally low (<30%) coral cover of the species studied, means lower frequencies of contact between the two species. Block C (at around 20 m in the Second Plateau) is the only area in the Bay known where the two large species-patches are in contact. Even if cover of the two species is higher, the patchy distribution due to asexual reproduction could still lead to coexistence between the two species even without the competitive intransitivity or lack of dominance of one species over the other. Metapopulation (i.e., "a population of populations"; see Hanski and Gilpin 1991 for a history of the concept) models show that the limited dispersal capabilities (that results from asexual reproduction here) will keep the patches of the subordinate species from being overrun by the dominant (McLaughlin and Roughgarden 1993). Also, if the patches are permanent (that is, if certain locations are more conducive to the growth of 129 one species over the other) these can act as refuges that could continually supply suboptimal areas with recruits (McLaughlin and Roughgarden 1993). The rubble distributions around and within the patches (see Appendix 5) suggest that the present patches have been in there for some time. This "foundation," plus the modification of environmental conditions within the patch (e.g., water flow and sedimentation rates), and the reduced growth of AP outside the source patches all indicate the present AP patches (at least) may be permanent refuges. Also, the simulations show that a few seed colonies are not sufficient to produce the large patches of several hundred colonies of AS and AP found in Puerto Galera. Parts of the reef had to be more conducive to growth of individual species (either by action of environmental gradients or the influence of pioneer conspecifics) such that several smaller patches survived to fuse into the present large patches. As the simulations reveal, disturbances (mainly storm damage or human impacts) can cause patchy distributions only if the effects are localized, and when one species is affected more than the others. Dana (1976) points to storm induced surges along irregularities on the reef slope as potential sources of such localized disturbances. Species that can quickly colonize disturbed patches, or those that are abundant enough to leave proportionately more survivors (while rarer forms go locally extinct) then can lead to disturbance-created patches. The life-history differences between AS and AP mentioned in this discussion (pages 126) may be one reason AP had a slight advantage in this regard in one simulation (Figure 4S). This may also explain the mortalities of AS 130 that were observed in small patches at the deeper areas where AP cross transplants seemed unaffected (pages 47-49). The source of this is mortality is unknown but does suggest that the role of disturbances in patch formation may be greater. The eigenvalue (A) of the AS matrix (considering reproduction) suggests that the Second Plateau stocks, where most of the partial and complete mortalities were observed, were approximately stable. This was unexpected since the year the monitoring studies were started (1993) is the same year a record number of typhoons and storms (32) entered the Philippines. Comparisons of changes in community structure in the area relative to more exposed areas suggest communities on the plateaus are more sensitive to storm effects (Licuanan 1988, 1991a). The localized death of a part of an apparently stable AS population, plus the simulation results presented here, call for more research on the role of disturbances in coral distributions. General Conclusions and Future Work The results of this research show that several factors and processes interact to produce the single species patches of AS and AP. Patch formation is more attributable to reproductive processes and patterns, and possibly disturbances. Dominance of a single species in a patch is maintained by competition, but is also affected by settlement rates and dispersal of recruits. Field studies, along with computer simulations paint a picture of patches that are conducive to the survival of the dominant species, especially AP. Growth of colonies is greater within the patch with conspecifics. Planular settlement of potential competitors is 131 low, while their fragments and colonies grow slower and suffer higher mortality rates when surrounded by the "resident" species. Asexual reproduction keeps the newer generations within "proven" ground. Water flow and sedimentation is lower. The home patch is the refuge. What remains to be demonstrated shown is how much of these attributes are due to the presence of the corals themselves, and how much are explained by the peculiarities of the particular locations of these patches. Speculation on the role of branch morphologies and colony densities in determining competitive dominance, through their effects on feeding and nutrition, require new experiments. The experiments need to combine measurements of energy budgets of individual corals with manipulations of water flow (by using transparent baffles that may or may not allow particulate food through), sedimentation (by using inert materials as bottom cover to hinder sediment resuspension), and densities of competitor species. Behavior of planulae also need to be directly observed in the presence of AP and AS colonies and patches. At the same time, finer mapping of recruit distributions (both in the water column and on settlement surfaces) is required to see if the effects of the resident corals vary at the edges of the patches and in the interior. Although the present cellular automaton simulations do not require that corals be homogeneously distributed across the reef, it does assume that all parts of the reef are 132 equally habitable to both species studied. New simulations should include permanent refuges that favor one species over the other. Also, presence of conspecifics (and not just competing species) should result in alteration of the transition probabilities. The field research was formulated around a multiple choice question — that of the factors responsible for the formation of single species patches of AS and AP. The answers found involved aspects of all factors that were considered. Single-factor mechanistic explanations for natural phenomena are rare in community ecology. Questions lead to more questions than answers. On the other hand, the computer simulations were based on a simple relationship — that of coral size and survival rates. Given the proper constraints and rules (which often do not require knowledge of mechanisms), the simulations were able to show patterns similar to those found in the field. But one will not find things unless one knows what to look for. Both the field studies and the simulation portions of this work greatly benefited from the insights allowed by the other. Beyond the details of the ecology of AS and AP patches, the research illustrates an important lesson - at least to a beginning ecologist. Start with a single hypothesis and the answer will be incomplete. Start with too many, and the answer will be vague. Start with none and get nowhere. The right balance can be achieved by mixing empirical data with theory, mechanistic explanations and phenomenological projections. 133 Literature Cited Ablan C, Licuanan WY, Alifio PM, Campos R. In press. Genetic variation within and between patches of the coral Anacropora puertogalerae in Puerto Galera, Mindoro: Preliminary results. In: Proceedings of the Third National Symposium in Marine Science; 1994 May 23-24; University of the Philippines in the Visayas, Miag-ao, Iloilo. Alcala AC, 1981. Fish yields of coral reefs off Sumilon Is., Central Philippines. National Research Council of the Philippines Research Bulletin 36(1): 1-7. Alcala AC, Gomez ED, Yap HT. 1987. Philippine coral reefs: Status and human responses to changes. Resource Management and Optimization 4(3,4):297-340. Alcala AC, Chou LM, Miclat R, Kastoro W, Fortes M, Gong W, Sasekumar A, Bifia R, Tridech S, editors. 1991. Proceedings of the Regional Symposium on Living Resources in Coastal Areas. Quezon City: University of the Philippines. S97 p. Alifio PM, Sammarco PW, Coll JC. 1992. Competitive strategies in soft corals (Coelenterata, Octocorallia). IV. Environmentally induced reversals in competitive superiority. Marine Ecology Progress Series 81:129-145. Allen TFH, Starr TB. 1982. Hierarchy: Perspectives on ecological complexity. Chicago: University of Chicago Press. 310 p. Allen TFH, Hoekstra TW. 1992. Towards a unified ecology. New York: Columbia University Press. 384 p. Atrigenio MP, Alifio PM. In press. Seasonal recruitment of scleractinian corals at Puerto Galera, Oriental Mindoro, Philippines. Proceedings of the Third Regional Symposium on Living Resources in Coastal Areas. Babcock R, Davies P. 1991. Effects of sedimentation on settlement of Acropora millepora. Coral Reefs 9:205-208. Bak RPM, Termaat RM, Dekker R. 1982. Complexity of coral interactions. Influence of time, location and epifauna. Marine Biology 69:215-222. Bermas NA, Alifio PM, Atrigenio MP, Uychiaoco A. 1992. Observations on the reproduction of scleractinian and soft corals in the Philippines. Proceedings of the Seventh International Coral Reef Symposium 1:443-447. Bermas NA, Alifio PM, Atrigenio MP. In press. Growth and morphological variations in Anacropora puertogalerae (Scleractinia) transplants. Proceedings of the Third Regional Symposium on Living Resources in Coastal Areas. 134 Bradbury RH, Reichelt RE, Green DG. 1983. Explanation, prediction and control in coral reef ecosystems 3. Models for control. Proceedings of the Inaugural Great Barrier Reef Conference, p 165-169. Bradbury RH, Loya Y, Reichelt RE, Williams WT. 1986. Patterns in the structural typology of benthic communities on two coral reefs of the central Great Barrier Reef. Coral Reefs 4:161-167. Bradbury RH, Young PC. 1983. Coral interactions and community structure. Marine Ecology Progress Series 11:265-271. Brown BE, Howard LS. 1985. Assessing the effects of "stress" on reef corals. Advances in Marine Biology 22:1-63. Burrough PA. 1981. Fractal dimensions of landscapes and other environmental data. Nature 294:240-242. Buss LW. 1979. Bryozoan overgrowth interactions— the interdependence of competition for space and food. Nature 281:475-477 Buss LW, Jackson JBC. 1979. Competitive networks: nontransitive competitive relationships in cryptic coral reef communities. American Naturalist 113(2):223-234. Carlon DB, Olson RR. 1993. Larval dispersal as an explanation for adult spatial pattern in two Caribbean reef corals. Journal of Experimental Marine Biology and Ecology 173:247-263. Carpenter K. 1977. Philippine coral reef resources. The Philippine Journal of Fisheries 15:95-126. Cartwright TJ. 1993. Modelling the world in a spreadsheet: environmental simulation on a microcomputer. Baltimore (MD): Johns Hopkins University Press. 423 p. Caswell H. 1978. A general formula for the sensitivity of population growth rate to changes in life-history parameters. Theoretical Population Biology 14:215-230. Caswell H. 1989. Matrix population models. Sunderland (MA): Sinauer Associates. 328 p. Caswell H, Naiman RJ, Morin R. 1984. Evaluating the consequences of reproduction in complex salmonid life-cycles. Aquaculture 43:1231-34. Chadwick NE. 1991. Spatial distribution and the effects of competition on some temperate Scleractinia and Corallimorpharia. Marine Ecology Progress Series 70:39-48. 135 Chomesky EA. 1983. Induced development of sweeper tentacles on the reef coral Agaricia agaricites: a response to direct competition. Biological Bulletin 165:569-581. Chomesky EA. 1989. Repeated reversals during spatial competition between corals. Ecology 70(4):843-857. Colwell RK. 1984. What's new? Community ecology discovers biology. In Price PW, Slobodchikoff CN, Gaud WS, editors. A new ecology: Novel approaches to interactive systems. New York: Wiley, p 387-396. Connell JH. 1978. Diversity in tropical rain forests and coral reefs. Science 199:1302-1310. Connell JH, Slayter RO. 1977. Mechanisms of succession in natural communities and their role in community stability and organization. American Naturalist 111:1119-1144. Connell JH, Sousa WP. 1983. On the evidence needed to judge ecological stability or persistence. American Naturalist 121:789-824. Cope M. 1981. Interspecific coral interactions in Hong Kong. Proceedings of the 4th International Coral Reef Symposium 2:357-362. Czaran T, Bartha S. 1992. Spatio-temporal dynamic models of plant populations and communities. Trends in Ecology and Evolution 7(2):38-42. Dai C. 1990. Interspecific competition in Taiwanese corals with special reference to interactions between alcyonaceans and scleractinians. Marine Ecology Progress Series 60:291-297. Dana TF. 1976. Reef-coral dispersion patterns and environmental variables on a Caribbean coral reef. Bulletin of Marine Science 26:1-13. Dayton PK, Tegner MJ. 1984. The importance of scale in community ecology: A kelp forest example with terrestrial analogs. In Price W, Gaud WS, Slobodchikoff CN, editors. A new ecology: Novel approaches to interactive systems. New York: Wiley, p 457-481. de Laubenfels, MW. 1935. A collection of sponges from Puerto Galera, Oriental Mindoro. Philippine Journal of Science 56:327-337. de Kroon, Plaisier HA, van Groenendael J, Caswell H. 1986. Elasticity: The relative contribution of demographic parameters to population growth rate. Ecology 67(5): 1427-1431. Dennison WC, Barnes DJ. 1988. Effect of water motion on coral photosynthesis and calcification. Journal of Experimental Marine Biology and Ecology 115:67-77. 136 Domantay JS, Roxas HA. 1938. The littoral Asteroidea of Port Galera Bay and adjacent waters. Philippine Journal of Science 65:203-238. Done TJ. 1983. Coral zonation: Its nature and significance. Jo: Barnes DJ, editor. Perspectives on coral reefs. Townsville (Queensland): Australian Institute of Marine Science, p 107-145. Done TJ. 1987. Simulation of the effects of Acanthaster planci on the population structure of massive corals in the genus Porites: evidence of population resilience. Coral Reefs 6(2):75-90. Done TJ. 1988. Simulation of recovery of pre-disturbance size structure in populations of Porites spp. damaged by Crown-of-Thoms starfish Acanthaster planci. Marine Biology 100:51-61. Doty MS. 1971. Measurement of water movement in reference to benthic algal growth. Botanica Marina 14:332-335. Eckman JE. 1983. Hydrodynamic processes affecting benthic recruitment. Limnology and Oceanography 28(2):241-257. English S, Wilkinson C, Baker V. 1994. Survey manual for tropical marine resources. Townsville (Queensland): Australian Institute of Marine Science. 368 p. Forman RTT, Godron M. 1986. Landscape ecology. New York: Wiley. 619 p. Friss L. 1994. [Unpublished handout given to graduate students in a USC dissertation workshop]. Located at: University of Southern California, Graduate School. Fujioka Y, Yamazato K. 1983. Host selection of some Okinawan coral associated gastropods belonging to the genera Drupella, Coralliophila, and Quoyula. Galaxea 2:59-73. Gardner M. 1970. The fantastic combinations of John Conway's new solitaire game "Life." Scientific American 223:120-123. Gardner M. 1971. On cellular automata, self-reproduction, the Garden of Eden and the game "Life." Scientific American 224:112-117. Gleick J. 1987. Chaos. Making a new science. New York: Penguin Books. 352 p. Gomez ED, Alifio PM, Yap HT, Licuanan WY. 1994. A review of the status of Philippine reefs. Marine Pollution Bulletin 29:62-68. Gomez ED, Licuanan WY, Hilomen W . 1988. Reef fish-benthos correlations in the northwestern Philippines. Proceedings of the Sixth International Coral Reef Symposium 3:245-249. Gomez KA, Gomez AA. 1984. Statistical principles for agricultural research. Second Edition. New York: Wiley. 680 p. 137 Grassle JF. 1973. Variety in coral reef environments. In Jones O A , Endean R, editors. Biology and geology of coral reefs 2. Biology 1. New York: Academic Press, p 247-270. Green DG. 1989. Simulated effects of fire, dispersal, and spatial pattern on competition within forest mosaics. Vegetatio 82:139-153. Green DG. 1990. Cellular automata models of crown-of-thoms outbreaks. In Acanthaster and the coral reef. A theoretical perspective. New York: Springer-Verlag. p 157-168. Green DG, Bradbury RH, Reichelt RE. 1987. Patterns of predictability in coral reef community structure. Coral Reefs 6:27-34. Green DG, House APN, House SM. 1985. Stimulating spatial patterns in forest ecosystems. Mathematics and Computers in Simulation 27: 191-198. Grigg RW, Dollar SJ. 1990. Natural and anthropogenic disturbances on coral reefs. In Dubinsky Z, editor. Coral reefs. Amsterdam: Elsevier, p 439-452. Gunthorpe L, Cameron AM. 1990. Widespread but variable toxicity in scleractinian corals. Toxicon 28:1199-1219. Hanski I, Gilpin M . 1991. Metapopulation dynamics: brief history and conceptual domain. Biological Journal of the Linnean Society 42:3-16. Harriott VJ, Fisk WP. 1988. Coral transplantation as a reef management option. Proceedings of the Sixth International Coral Reef Symposium 3:375-379. Hatcher BG, Imberger J, Smith SV. 1987. Scaling analysis of coral reef systems: an approach to problems of scale. Coral Reefs 5:171-181. Highsmith RC. 1982. Reproduction by fragmentation in corals. Marine Ecology Progress Series 27:207-226. Hildemann WH, Dix TG, Collids JD. 1974. Tissue transplantation in diverse marine invertebrates. In: Cooper EL, editor. Contemporary topics in immunobiology 4. New York: Plenum, p 141-150. Hilomen W , Gomez ED. 1988. Distribution patterns offish communities in some Philippine reefs. Proceedings of the Sixth International Coral Reef Symposium 3:257-262. Hughes T. 1984. Population dynamics based on individual size rather than age: a general model with a coral reef example. American Naturalist 123(6):778-795 Hughes T. 1994. Catastrophes, phase shifts, and large-scale degradation of a Caribbean coral reef. Science 265:1547-1551. 138 Huston M, DeAngelis D, Post W. 1988. New computer models unify ecological theory. BioScience 38(10):682-691 Hutchinson GE. 1953. The concept of pattern in ecology. Proceedings of the Academy of Natural Sciences of Philadelphia 105:1-12. Hutchinson GE. 1957. Concluding remarks. Cold Spring Harbor Symposia on Quantitative Biology 82:415-427. Jackson JBC, Buss LW. 1975. Allelopathy and spatial competition among coral reef invertebrates. Proceedings of the National Academy of Sciences, U.S.A. 72:5160-5163. Johnson CR, Preece AL. 1992. Damage, scale and recovery in model coral communities: The importance of system state. Proceedings of the Seventh International Coral Reef Symposium 1:606-615. Jokiel PL, Morrissey J. 1993. Water motion on coral reefs: evaluation of the 'clod card’ technique. Marine Ecology Progress Series 93:175-181. Karlson RH. 1985. Competitive overgrowth interactions among sessile colonial invertebrates: a comparison of stochastic and phenotypic variation. Ecological Modelling 27: 299-312. Karlson RH, Buss LW. 1984. Competition, disturbance and local diversity patterns of substratum-bound clonal organisms. A simulation. Ecological Modelling 23:243-255. Karlson RH, Hurd LE. 1993. Disturbance, coral-reef communities, and changing ecological paradigms. Coral Reefs 12:117-125. Karlson RH, Jackson JBC. 1981. Competitive networks and community structure: A simulation study. Ecology 62(3):670-678. Keddy PA. 1989. Competition. London: Chapman and Hall. 202 p. Kojis BL, Quinn NJ. 1981. Aspects of sexual reproduction and larval development in shallow water hermatypic coral Goniastrea australiensis (Edwards and Haime, 1857). Bulletin of Marine Science 31:558-573. Kolasa J. 1989. Ecological systems in hierarchical perspective: Breaks in community structure and other consequences. Ecology 70(l):36-47. Kolasa J, Pickett STA, editors. 1991. Ecological heterogeneity. New York: Springer-Verlag. 332 p. Lamberts AE. 1978. Coral growth: Alizarin method. In Stoddart DR, Johannes RE, editors. Coral reefs: Research methods. UNESCO Monographs on Oceanographic Methodology 5. Paris: UNESCO, p 523-527. 139 Lang J. 1973. Interspecific aggression by scleractinian corals 2. Why the race is not only to the swift. Bulletin of Marine Science 23:261-279. Lang J, Chomesky EA. 1990. Competition between scleractinian reef corals - A review of mechanisms and effects. In Dubinsky Z, editor. Coral reefs. Amsterdam: Elsevier, p 209-252. Lewis JB. 1970. Some distribution and pattern o f some Atlantic reef corals. Nature 227:1158-1159. Lewis JB. 1992. Recruitment, growth and mortality of a coral-inhabiting barnacle Megabalanus stultus Darwin upon the hydrocoral MUlepora complanata Lamark. Journal of Experimental Marine Biology and Ecology 162:51-64. Licuanan WY. 1988. Analyses of the structural ecology of selected Philippine reefs based on physiognomic attributes. M.Sc. thesis. Quezon City (Philippines): University of the Philippines, Marine Science Institute. Licuanan WY. 1991a. Temporal change in the cover of life-forms in Puerto Galera, Mindoro Island, western Philippines: Preliminary results. In Alcala AC, Chou LM, Miclat R, Kastoro W, Fortes M, Gong W, Sasekumar A, Bifia R, Tridech S, editors. Proceedings of the Regional Symposium on Living Resources in Coastal Areas. Quezon City: University of the Philippines, p 87-96. Licuanan WY. 1991b. The spatial distribution of life-forms in selected coral communities in the northwestern Philippines with notes on the coding method used. In Alcala AC, Chou LM, Miclat R, Kastoro W, Fortes M, Gong W, Sasekumar A, Bifia R, Tridech S, editors. Proceedings of the Regional Symposium in Living Resources in Coastal Areas. Quezon City: University of the Philippines, Marine Science Institute, p 113. (Abstract). Licuanan WY, Alifio PM. In press. Neighborhood effects on settlement of stony corals in loose sediment communities: Early results. Proceedings of the Third National Symposium in Marine Science; University of the Philippines in the Visayas, Miag-ao, Iloilo; 1994 May 23-24. Licuanan WY, Bakus GJ. 1992. Coral spatial distributions. The Ghost of competition past roused? Proceedings of the Seventh International Coral Reef Symposium 1:545-549. Licuanan WY, Gomez ED. 1988. Coral reefs of the northwestern Philippines: A physiognomic-structural approach. Proceedings of the Sixth International Coral Reef Symposium 3:275-280. Lim MES. 1988. Black tiger prawn: The AA-Solid experience. Hongkong: Solid Corporation. 325 p. 140 Logan A. 1984. Interspecific aggression in hermatypic corals from Bermuda. Coral Reefs 3:131-138. Maguire LA, Porter JW. 1977. A spatial model of growth and competition strategies in coral communities. Ecological Modelling 3:249-271. May RM. 1974. Biological populations with non-overlapping generations: stable points, stable cycles, and chaos. Science 186:645-647. Maxwell SE, Delaney HD. 1990. Designing experiments and analyzing data: A model comparison perspective. Belmont (CA): Wadsworth Publishing. 902 p. McFadden CS. 1986. Colony fission increases particle capture rates of a soft coral: advantages of being a small colony. Journal of Experimental Marine Biology and Ecology 103:1-20. McLaughlin JF, Roughgarden J. 1993. Species interactions in space. In Ricklefs RE, Schluter D, editors. Species diversity in ecological communities: Historical and geographical perspectives. Chicago: University of Chicago Press, p 89-98. McManus JW. 1985. Marine speciation, tectonics and sea-level changes in Southeast Asia. Proceedings of the Fifth International Coral Reef Symposium. 4:133- 138. McManus JW. 1988. Coral reefs of the ASEAN region: Status and management. Ambio 17(3): 189-193. Mokady O, Arazi G, Bonar DB, Loya Y. 1992. Settlement and metamorphosis specificity of Lithophaga simplex (Bivalvia, Mytilidae) on Red Sea corals. Journal of Experimental Marine Biology and Ecology 162:243-251. Molofsky J. 1994. Population dynamics and pattern formation in theoretical populations. Ecology 75(l):30-39. Moloney KA. 1986. A generalized algorithm for determining category size. Oecologia 69:176-180. Morse ANC. 1992. Unique patterns of substratum selection by distinct populations of Agaricia humilis contribute to opportunistic distribution within the Caribbean. Proceedings of the Seventh International Coral Reef Symposium 1:501-502. Morse DE, Hooker N, Morse ANC, Jensen RA. 1988. Control of larval metamorphosis and recruitment in sympatric agariciid corals. Journal of Experimental Marine Biology and Ecology 116:193-217. Moyer JT, Emerson WK, Ross M. 1982. Massive destruction of scleractinian corals by the muricid gastropod, Drupella, in Japan and the Philippines. The Nautilus 96:69-82. 141 Munro JL, Williams D McB. 1985. Assessment and management of coral reef fisheries: Biological, environmental and socio-economic aspects. Proceedings of the Fifth International Coral Reef Symposium 4:545- 581. Murdy EO, Ferraris CJ Jr. 1980. The contribution of coral reef fisheries to Philippine fisheries production. International Center for Living Aquatic Resources Management Newsletter 3:21-22. Nelson VM. 1992. Patterns of cover, diversity and spatial arrangement of benthos at Lizard Island, Great Barrier Reef. Proceedings of the Seventh International Coral Reef Symposium 2:827-832 Nemenzo F. 1955a. Systematic studies on Philippine shallow water scleractinians: I. Sub-order Fungiida. Natural and Applied Science Bulletin 15:3-84. Nemenzo F. 1955b. On the scleractinian fauna of the Puerto Galera Bay, Oriental Mindoro and Laguimanoc Bay, Quezon. Natural and Applied Science Bulletin 15:131-137. Nemenzo F. 1959. Systematic studies on Philippine shallow water scleractinians: II. Sub-order Faviida. Natural and Applied Science Bulletin 16:73-135. Nemenzo F. 1960a. Systematic studies on Philippine shallow water scleractinians: III. Sub-order Caiyophyliida. Natural and Applied Science Bulletin 17:207-213. Nemenzo F. 1960b. Systematic studies on Philippine shallow water scleractinians: IV. Sub-order Dendrophyliida. Natural and Applied Science Bulletin 18:1-21. Nemenzo F. 1964. Systematic studies on Philippine shallow water scleractinians V: Suborder Astrocoenida. Natural and Applied Science Bulletin 18(3,4): 193-223. Nemenzo F. 1967. Systematic studies on Philippine shallow water scleractinians VI: Suborder Astrocoenida {Montipora and Acropora). Natural and Applied Science Bulletin 20:1-141. Nemenzo F. 1971. Systematic studies on Philippine shallow water scleractinians VII: Additional forms. Natural and Applied Science Bulletin 23:141-185. Nemenzo F. 1976. Non-scleractinian corals in Philippine reefs. Natural and Applied Science Bulletin 23 :277-297. Nemenzo F. 1981. Studies on the systematics of scleractinian corals in the Philippines. Proceedings of the Fourth International Coral Reef Symposium 1:25-32. O'Neill RV, DeAngelis DL, Waide JB, Allen TFH. 1986. A hierarchical concept of ecosystems. Princeton (NJ): Princeton University Press, 253 p. 142 O'Neill RV, Gardner RH, Milne BT, Turner MG, Jackson B. 1991. Heterogeneity and spatial hierarchies. In Kolasa J, Pickett ST A, editors. Ecological heterogeneity. New York: Springer-Verlag. p 85-96. Pagano M, Gauvreau K. 1993. Principles of biostatistics. Belmont (CA): Wadsworth Publishing. 524 p. Patterson MR. 1984. Patterns of whole colony prey capture in the octocoral Alcyonium siderium. Biological Bulletin 167:613-629. Patterson MR, Sebens KP, Olson RR. 1991. In situ measurements of flow effects on primary production and dark respiration in reef corals. Limnology and Oceanography 36(5):936-948. Pearson RG. 1981. Recovery and recolonization of coral reefs. Marine Ecology Progress Series 4:105-122. Peters RH. 1991. A critique for ecology. Cambridge: Cambridge University Press. 366 P Pielou EC. 1962. Runs of one species with respect to another in transects through plant populations. Biometrics 579-593 Pires A, Hadfield MG. 1993. Responses of isolated vela of nudibranch larvae to inducers of metamorphosis. Journal of Experimental Zoology 266:234-239. Potts D. 1983. Evolutionary disequilibrium among Indo-Pacific corals. Evolution 33(3):619-632. Potts D. 1985. Sea-level fluctuations and speciation in Scleractinia. Proceedings of the Fifth International Coral Reef Congress 4:127-132. Potts D, Done TJ, Isdale PJ, Fisk DA. 1985. Dominance of a coral community by the genus Porites (Scleractinia). Marine Ecology Progress Series 23:79-84. Preece AL, Johnson CR. 1993. Recovery of model coral communities: Complex behaviors from interaction of parameters operating at different spatial scales. In Green DG, Bossomaier T, editors. Complex systems: From biology to computation. Amsterdam: IOS Press, p 69-81. Price W, Gaud WS, Slobodchikoff CN. 1984. A new ecology: Novel approaches to interactive systems. New York: Wiley. 515 p. Reichelt RE, Bainbridge SJ, Green DG. 1988. A stimulation study of crown of thorns starfish outbreaks on the Great Barrier Reef. Mathematics and Computers in Simulation 30: 145-150. 143 Reichelt RE, Bradbury RH. 1984. Spatial patterns in coral reef benthos: Multi-scale analysis of sites from three oceans. Marine Ecology Progress Series 17:251-257. Reichelt RE, Green DG, Bradbury RH. 1985. Discrete simulation of cyclone effects on the spatial pattern and community structure of a coral reef. Proceedings of the 5th International Coral Reef Congress 3:337-342. Reichelt RE, Greve W, Bradbury RH, Moran PJ. 1990. Acanthaster planci outbreak initiation: a starfish-coral site model. Ecological Modelling, 49:153-177. Richmond RH. 1985. Reversible metamorphosis in coral planula larvae. Marine Ecology Progress Series 22:181-185. Rinkevich B, Shashar N, Liberman T. 1992. Nontransitive xenogenic interactions between four common Red Sea sessile invertebrates. Proceedings of the Seventh International Coral Reef Symposium 2:833-839. Romano SL. 1990. Long-term effects of interspecific aggression on growth of the reef-building corals Cyphastrea ocellina (Dana) and Pocillopora damicornis (Linnaeus). Journal of Experimental Marine Biology and Ecology 140:135-141. Roxas HA. 1932. Two new species of Sarcophyton Less from the Philippines. Natural and Applied Science Bulletin 2:73-80. Roxas HA. 1933a. Philippine Alcyonaria: I. The families Comulariidae and Xeniidae. Philippine Journal of Science 50:49-110. Roxas HA. 1933b. Philippine Alcyonaria: II. The families Alcyoniidae and Nephthiidae. Philippine Journal of Science 50:345-470. Sale PF. 1988. What coral reefs can teach us about ecology. Proceedings of the Sixth International Coral Reef Symposium 1:19-27. Sammarco P. 1981. Damselfish territoriality and coral community structure: Reduced grazing, coral recruitment, and effects on coral spat. Proceedings of the Fourth International Coral Reef Symposium 2:525-536. Sammarco P. 1982. Polyp bail-out: An escape response to environmental stress and a new means of reproduction in corals. Marine Ecology Progress Series 10:57-65. Sebens KP, Johnson AS. 1991. Effects of water movement on prey capture and distribution of reef corals. Hydrobiologia 226:91-101. Sheppard CRC. 1979. Interspecific aggression between reef corals with reference to their distribution. Marine Ecology Progress Series 1:237-247. 144 Sheppard CRC. 1982. Coral populations on reef slopes and their major controls. Marine Ecology Progress Series 7:83-1 IS. Silvertown J, Holtier S, Johnson J, Dale P. 1992. Cellular automaton models of interspecific competition-the effect of pattern on process. Journal of Ecology 80:527-534. Soong K, Lang JC. 1992. Reproductive integration in reef corals. Biological Bulletin 183:418-431. Steele JH. 1985. A comparison of terrestrial and marine ecosystems. Science 313:355-358. Stoddart DR. 1969. Ecology and morphology of Recent coral reefs. Biological Reviews 44:433-490. Stoddart JA. 1983. Asexual production of planulae in the coral Pocillopora damicornis. Marine Biology 76:279-284 Sugihara G, May RM. 1990. Applications of fractals in ecology. Trends in Ecology and Evolution 5(3):79-86. Tanner JE. 1992. Experimental analysis of digestive hierarchies in coral assemblages. Proceedings of the Seventh International Coral Reef Symposium 1:569-574. Te FT. 1992. Response to higher sediment loads by Pocillopora damicornis. Coral Reefs 11:131-134. Turner MG, editor. 1987. Landscape heterogeneity and disturbance. New York: Springer-Verlag. 239 p. Turner MG, Gardner RH, editors. 1991. Quantitative methods in landscape ecology. New York: Springer-Verlag. 536 p. van Groenendael J, de Kroon H, Caswell H. 1988. Projection Matrices in Population Biology. Trends in Ecology and Evolution 3(10):264-269. Veron JEN, Wallace CC. 1984. Scleractinia of eastern Australia. Family Acroporidae. Australian Institute of Marine Science Monograph Series Vol. 6. Townsville (Australia): Australian Institute of Marine Science. 485 p. Veron JEN. 1986. Corals of Australia and the Indo-Pacific. North Ryde (New South Wales): Angus and Robertson Publishers. 644 p. Veron JEN. 1992. Hermatypic corals of Japan. Australian Institute of Marine Science Monograph Series Vol. 9. Townsville (Queensland): Australian Institute of Marine Science. 150 p. Veron JEN, Hodgson G. 1989. Annotated checklist of hermatypic corals of the Philippines. Pacific Science 43:234-287. 145 Villa F. 1992. New computer architectures as tools of ecological thought. Trends in Ecology and Evolution 7(6): 179-183. Villanoy CL, Udarbe MJB, Velasquez IB, Natividad GM. In press. Residence times of passive particles as influenced by modelled circulation in Puerto Galera Bay, Oriental Mindoro. Proceedings of the Third National Symposium in Marine Science; 1994 May 23-24; University of the Philippines in the Visayas, Miag-ao, Iloilo. Walker J, Sharpe P, Penridge LK, Wu H. 1989. Ecological field theory: The concept of field tests. Vegetatio 83: 81-95. Wallace CC. 1985. Reproduction, recruitment and fragmentation in nine sympatric species of the coral genus Acropora. Marine Biology 88:217-233. Weinberg JR, Caswell H, Whitlatch RB. 1986. Demographic importance of ecological interactions: How much do statistics tell us? Marine Ecology Progress Series 93:305-310. Wellington GM. 1980. Reversal of digestive interactions between Pacific reef corals: Mediation by sweeper tentacles. Oecologia 47:340-343. Wells J. 1957. Coral Reefs. Memoirs of the Geological Society of America 1(67):609-631 Winer BJ, Brown DR, Michels KM. 1991. Statistical principles in experimental Design. Third Edition. New York: McGraw-Hill. 1057 p. WolframS. 1984. Cellular automata as models of complexity. Nature 311:419-424. Wu H, Sharp P, Walker J, Penridge LK. 1985. Ecological field theory: a special analysis of resource interference among plants. Ecological Modelling 29: 215-243. Yap HT, Alifio PM, Gomez ED. 1992. Trends in growth and mortality of three coral species (Anthozoa: Scleractinia), including effects of transplantation. Marine Ecology Progress Series 83:91-101. Yap HT, Gomez ED. 1981. Growth of Acropora pulchra (Brook) in Bolinao, Pangasinan. Proceedings of the Fourth International Coral Reef Symposium 2:207-213. Yap HT, Gomez ED. 1984. Growth of Acropora pulchra II. Responses of natural and transplanted colonies to temperature and day length. Marine Biology 81:209-215. Yap HT, Gomez ED. 1985a. Coral reef degradation and pollution in the East Asian seas region. In: Dahl AL, Carew-Reid J, editors. Environment and resources in the Pacific. UNEP Regional Seas Report and Studies 69, 294 p. 146 Yap HT, Gomez ED. 1985b. Coral reef degradation and pollution in the East Asian Seas Region. In Dahl AL, Carew-Reid J, editors. Environment and resources in the South Pacific: A regional approach. UNEP Regional Seas Programme. Geneva: UNEP, p 107-207. Yap HT, Gomez ED. 1985c. Growth of Acropora pulchra III. Preliminary observations on the effects of transplantation and sediments on the growth and survival of transplants. Marine Biology 87:203-209. Yap HT, Licuanan WY, Gomez ED. 1990. Studies on coral reef recovery and coral transplantation: aspects relevant to conservation and management. In Southeast Asian marine science and environmental protection. Proceedings of the first ASEAMS symposium on southeast asian marine science and environmental protection. UNEP Regional Seas Reports and Studies. Nairobi: UNEP. 116:117-128. 147 Appendices These appendices contain reports of additional observations, analyses, and experiments which were conducted to clarify, expand, or simply assure the researcher regarding the validity and applicability of the results of the main field experiments and the conclusions drawn from these results. Since most of these studies were made on a time-available basis, sample sizes, sampling frequencies, and study durations were limited. These appendices are not meant to be stand-alone technical papers, but taken in the context of the results of the main experiments, they allow for a greater appreciation of the scope of this research. Appendix 1: Comparisons of branch diameters and spacing in Anacropora puertogalerae Introduction. Observations showing stunting of AP release and some cross-transplants within the first three to six months of transplantation required additional measurements to clarify the nature of these morphological changes. Suspicion that changes in environmental conditions (for example, water flow) may explain most of the results of the field studies led to the measurements of interbranch distance and thickness (described in this appendix) among AP release colonies and various control colonies. Materials and Methods. Three sets of AP colonies were measured for comparison of branch diameters. The first set involved release colonies in Block A, the second set was composed of randomly selected control branches found at the edge of the 148 AP patch in Block A, and the third set consisted of selected control branches about O .S m interior to the patch edge measurements. For each branch, the diameter was measured using a caliper positioned 1 cm from the branch tip (see Figure 1-1). Data were then analyzed as a completely randomized design using analysis of variance (Gomez and Gomez 1984). On a separate occasion in the same block, branch spacing was gauged for AP release and cross transplant colonies by measuring the distance between the bases of secondary branches nearest the tip (see Figure 1-1). Similar measurements were also made for AP transplants and unmanipulated controls. A two sample t-test (Pagano and Gauvreau 1993) was used to compare these sets. Results and Discussion. Significant differences in branch diameters were found for the three sets (p=0.025), with interior branches often being the thickest (Figure 1-2). Multiple comparison tests reveal release branches were significantly different from interior branches (p<0.05), but not from the edge branches. These suggest conditions affecting growth in the release area are similar to those at the patch edges, but not at the interior. These differences are unlikely to be due to age since most release colonies were taken from the patch interior. Upon transplantation, release colonies often lost their longer, thicker branches to eventually take on a low, sprawling form. The bases of these colonies still appeared to be as thick as those in the source patches, however. 149 branch diameter (1 cm from the tip) branch distances Figure 1-1. Diagram showing the positions of the measurements made in Anacropora puertogalerae colonies to compare branch diameters and distances. 150 2.40 2.35 2.30 1 2.25 8 2.20 |». | 2.10 £ 2.05 2.00 1.95 1.90 Figure 1-2. Comparison of branch diameters of Anacropora puertogalerae release colonies and control colonies at the edge of the patch and 0.5 m interior to the edge. Interior colonies were significantly thicker than release colonies (p<0.05). ANOVA Source o f Variation SS d f M S F P-value Between Groups 0.3883 2 0.1942 3.9022 0.0253 Within Groups 3.1345 63 0.0498 Total 3.5228 65 EDGE INTERIOR RELEASE treatm ent 151 It is unclear why branches will be thinner in areas with higher water flow. Done (1983) describes how some finely-branched Acropora species are common in wave-exposed reef crests. Whether this form allows for greater attenuation of water currents for processes, such as feeding, requires more study. With regard to distances of branches, no significant difference was found between the control compared to the release and cross-transplant colonies (p=0.46). Bermas and others (in press) reported increased numbers of branches and decreased distances between them in an experiment involving AP transplanted to the channel near the Third Plateau, an area with higher current speeds. However, the latter analysis is confounded by differences in depth and sedimentation regimes, so it would be premature to draw conclusions at this stage. In conclusion, branch thickness comparisons between source patches and release colonies suggest similarities in conditions between the release area and the outer edge of AP patches. The branch thickness in the former differs significantly from the interior branches, however. This indicates conditions in the patch interior differ from surrounding areas, and thus will have implications on growth and recruitment of corals there. Appendix 2: Competitive pairing experiments Introduction. The cross transplantation experiments were designed mainly to allow detection of many-colonies to single-colony interactions between AP and AS since 152 individual colonies are rarely seen in isolated from conspecifics, and thus such diffuse competition should be the most common type of interaction occurring in the study sites. However, observations of AP branches appearing to grow away from AS (even when the former is numerically dominant and the latter is dying), prompted questions on the outcomes of single colony AS-AP pairings when such avoidance is not an option. This appendix summarizes the results of a preliminary experiment meant to examine one-to-one interactions of same-size fragments of AS and AP. Materials and Methods. Twelve pairs of AS and AP fragments approximately IS cm in length were collected from Block B (First Plateau; 8 pairs) and Block D (Second Plateau, 4 pairs) on March 13, 1994. AS and AP of about the same size were then paired and immediately tied to a steel wire grid using plastic zip ties. The wire grids used in the First Plateau have been in place for about three years, hence encrusting biota had to be cleaned off before use. The colonies of each pair were arranged in such a manner that contact between living tissues of the two is maximized while retaining "natural" orientations. Pairs were separated by at least 15 cm, and the arrangement of each coral of a pair was also randomized. After 26, 53, and 97 days, the coral pairs were examined for any tissue damage, and partial or complete mortality. No measurements were made in all occasions, however. Results and Discussion. Signs of tissue damage in AP were seen even during the first visit, but the outcomes were obvious by the third visit (97 days later). Of the twelve pairs used, ten of the AP fragments showed clear signs of tissue damage, 153 especially the branch tips nearest AS, and at the bases. No sign of damage was seen in AS in any of the three visits. Of the two pairs that did not show any reactions in either species, one pair came loose somehow and thus had a gap between the two species. The other simply did not show any signs of damage. At any rate, these suggest that the redirection of AP away from AS is due to the latter's apparent digestive dominance, a response seen in other corals (Romano 1990). What is unclear is the reason for the tissue death at the bases of the AP colonies, even when these were not in contact with AS. A similar situation was observed in the cross transplantation experiment, where AS branch growth was unaffected even though tissues at the bases of these colonies were getting bleached and were dying. Asexual reproduction, especially in corals often occurs by breakage of branch bases that are usually weakened by other organisms, like sponges boring into the corals' skeleton (Highsmith 1982). These bases are often devoid of coral tissue, hence the rapidity of fouling. It is tempting to speculate that the loss of tissue at the bases is because of energetic limitations, possibly due to the high costs of competition or stressful environmental conditions. This may then be an alternative explanation to the non-intuitive results of the AS cross transplantation of branch growth not being affected by colony mortality. 154 Appendix 3: Density manipulation experiments Introduction. Unlike AP, AS colonies are separated from adjacent conspecifics by gaps. This observation and the possibility that the high density of AP branches is responsible for the high mortalities of AS cross transplants (maybe because of modified water flow) prompted the study of the potential effects of the density of AS colonies on growth and mortality rates. Assuming that intraspecific competition would be stronger than interspecific competition, then the effects of branch densities should be more conspicuous when manipulated among conspecifics in AS patches. Materials and Methods. Two patches, both at 10 m depth (one between Blocks A and B in the First Plateau, and the other near Block D in the Second Plateau; Figure 3), were selected for the study that was started on February 24, 1994 and was last monitored on July 13, 1994. In each patch, three lxl m quadrats were demarcated with steel pegs at the comers and at the centers. Using the central peg as reference, densities of all AS within the each quadrat were changed to form three different treatments per patch. The high density treatment contained corals whose distances (but not orientations) to the central peg were halved until the quadrat was filled (see Figure 3-1). Thus, even colonies from outside the quadrat were included. The low density treatment involved the same procedures except that distances firom the colony centers to the quadrat's central peg were doubled, requiring the removal of some corals in the adjacent areas. The normal density treatment, where colonies inside the quadrat were just moved to the quadrat edge and then immediately returned to their original positions and 155 'High density" treatment 'Low density" treatment Figure 3-1. Diagram of two of three treatments in the Acropora subglabra density manipulation experiment. The high density treatment is where colony distance to the quadrat center is halved, while the low density treatment involved doubling the distance to the quadrat center. 156 orientations, is analogous to the transplant control treatment in the cross-transplantation experiment. Five colonies were haphazardly chosen, tagged, and measured in each quadrat, following the procedures used in the latter experiment (page 31). The three treatment quadrats per plateau were also randomly interspersed and separated by about two meters to minimize interaction effects. Branch growth data were pooled for all colonies per treatment, and analyzed as a completely randomized design (Gomez and Gomez 1984). Results and Discussion. Analyses of branch growth are presented in Figure 3-2. No significant differences in growth rates were seen (p=0.37), indicating that the densities of coral used did not affect branch growth rates. Also, no trends in survival were noted. Only one case of partial mortality was noted, in which the monitored branch died but the rest of the colony remained intact. These results suggest that the high mortality of AS cross-transplants to AP patches was probably not due to changes in availability of planktonic food since such effects should be more apparent with high densities of conspecifics, but are not. It is thus likely that the causes of this mortality are specific to conditions existing in the AP patch, either in the form of stronger interspecific competition, or unique environmental conditions, or both. The lower water flow and sedimentation rates in AP patches (Figure IS and 16) may be the immediate causes of these mortalities. 157 0.70 0.60 HIGH LOW NORMAL density Figure 3-2. Comparison of branch growth of Acropora subglabra in quadrats of different colony densities. No significant differences were found. ANOVA Source o f Variation SS d f M S F P-value Between patches 0.3160 2 0.1580 1.0208 0.3672 Within patches 8.3573 54 0.1548 Total 8.6732 56 158 Appendix 4: The alizarin experiments Introduction. This appendix summarizes the results of a preliminary experiment using an alternative method for measuring branch growth of transplants. The objective of the study was to independently validate the results of the cross-transplantation experiment (Figure 4) using a different method for monitoring growth. Ideally, this would allow the examination of branch growth without any possible trauma introduced by the frequent contact with the observer, and the installation of zip ties. To allow measurements, Alizarin Red S- a stain that is incorporated into the coral skeleton during growth (Lamberts 1978), was used as marker. Unfortunately, the measurements using this method require destructive sampling. Also, alizarin is toxic at high concentrations and thus could have subtle effects of its own. Materials and Methods. On June 16, 1994, five AS and five AP colonies were collected from Block B (First Plateau) for use in this study. These colonies were enclosed in clear plastic bags (301 in volume) containing 10 ppm of alizarin dissolved in seawater (after Lamberts 1978) and left tied together near the collection block at 14 m depth for 24 hours. An equal number of colonies were also collected and treated similarly at Block E (Second Plateau at 6 m depth) a day later. After the 24 h incubation in the alizarin solution, all 20 colonies were removed from the bags, and tagged. Of the five colonies of each species per block, two were relocated to their source patches (the transplant control treatment), two were transplanted to the patch of the other species (cross transplant treatment), and one was 159 moved to the release area, following the same procedures used in the cross transplantation experiment. Results and Discussion. All ten AP colonies had a pinkish tint, especially at the branch tips, when pulled out of the bag after 24 hours. All AS, on the other hand, had tissue sloughing off that immediately attracted fishes that started to pick flesh off the coralla. Colonies were still transplanted as planned, with the hope that the AS colonies would still recover. A day later however, all these AS colonies were completely bleached and cleaned. Still, both AS and AP had a pinkish tinge, suggesting that alizarin was incorporated in the skeleton and did not just settle on the branch surfaces. Thirty-seven days later, some branch tips of the AP treatments were collected and their soft tissues allowed to decay in tap water. Measurements of the distance of the stained skeleton to present branch tips show no obvious differences in growth between the treatments (p=0.18; Figure 4-1), hence the transplants were left in the site longer (where they remain at the time of this writing). Although still inconclusive, the experiment does illustrate that if AS and AP are exposed to similar conditions of elevated temperature (water was noticeably higher in the incubation bags when opened 24 hours later) and reduced water circulation, diminished light (alizarin stained water is deep red/purple), as well as exposure to alizarin itself in 160 0.70 — 0.60 tr a n spl a n t TRANSPLANT CONTROL treatment I r e l e a s e Figure 4-1. Branch growth rates of Anacropora puertogalerae treatment colonies measured using Alizarin Red S marks incorporated in the skeleton. No significant differences were found in the three treatments. ANOVA Source o f Variation SS d f K fS F P-value Between Groups 0.2721 2 0.1361 1.7863 01814 Within Groups 2.8944 38 0.0762 Total 3.1665 40 161 plastic bags, AS is more affected than AP. The experiment was not designed to isolate the effects of these factors, however. Appendix 5: Distribution of AS and AP rubble Introduction. The present coral patches may be relicts of larger, less contagious distributions of AS and AP in the past that were subject to differential mortality, leading to dominance of one species in certain parts of the reef. This differential mortality may have then been amplified by the competitive and recruitment effects described earlier. If so, rubble of both these species should be found in-between the present patches, and not as contagious patches. This study was designed to explore this possibility. Materials and Methods. A rapid, transect-based method for mapping was used to maximize limited field time. On July 14, 1994, a 100 m graduated transect tape that was laid following the 1 1 m depth contour, where both species were most abundant at the Second Plateau. A diver then swam along the line, stopping every 0.5 m to probe the sediments with a knife to about 5 cm depth. All coral rubble unearthed were then identified to the generic level. Boundaries of existing AS and AP patches were also noted. However, no sampling was conducted within these patches to avoid disturbing the then ongoing experiments. Results and Discussion. The resulting map is shown on Figure 5-1. Note that rubble of AS and AP still formed patches, and that these were often associated with living assemblages of the same species. Between the 10 and 20 m marks however, there 162 Anacropora puertogalerae 10 20 30 40 50 60 70 80 a 90 100 0 10 20 30 40 50 60 70 80 90 100 meters Acropora subglabra Figure 5-1. Distribution of live Anacropora puertogalerae (upper bar) and Acropora subglabra (lower bar) patches (black blocks), and their respective rubble (grey blocks). Distributions of both species appear to be more widespread compared to at present, but there remains little mingling between the two species. ON u> was a little mingling of the two species. This suggests that the two species had patches that were (at least) closer to each other than at present. A similar situation could be seen near the 40 m mark. Although coarse and shallow, these results show that colonies of both species still formed patches, rather than mingled with the other species. Depending on what is under the present patches, the results may also mean that the corals were more widespread than they are now. Although the circumstances did not allow sampling within the patch boundaries, sampling for sediment analyses in these patches reveal that rubble underneath were of the same genus as the overlying corals. These results thus raise the possibility that the present patch boundaries define what could be considered refuges, wherein conditions for the survival are better than on other parts of the reef. It is unknown, however, if these hypothesized conditions are due mainly to the corals' presence, or are mainly because of environmental discontinuities in the area (e.g., rock outcrops leading to aggregated settlement), or both. Appendix 6: Predation of recruits on settlement tiles Introduction. The impact of fishes on the cross transplantation results was discounted earlier (page 65). However, observations of fish effects on the coral transplants raised concern that the fishes inhabiting or frequenting the different coral patches may be affecting the settlement of corals in these patches. This appendix summarizes the results of analyses aimed at detecting predation by fishes of recruits in 164 the coral settlement monitoring study. It is assumed here that only the outer tiles of the settlement racks (see Figure 5) are affected by any such activity, while interior tiles are too closely spaced (2 cm) for fish to prey on the recruits there. These two surfaces are compared here. Materials and Methods. Data from the June 1994 sampling of the recruit monitoring study (when recruit total abundances was highest and differences among patches most apparent; see Figure 13), was used for this analysis. Data on counts of live coral recruits (regardless of taxon) on the flat surfaces were extracted for the outer tiles (#1 and #6 in Figure 5), and those immediately interior to these (#2 and #5) from each of the AP, AS, and release treatment racks. Log (x+0.5) transformed data over five blocks (Blocks A,B,C,D, and F) were then summarized, and analyzed as a completely randomized design using a two-way (outer/inner tile versus treatment patch) analysis of variance (Gomez and Gomez 1984). Results and Discussion. Results of the analyses are summarized in Figure 6-1. There was no significant interaction (p=0.58) between the tile number and the patch type, indicating that the trends among patch types (AP, AS, and release) were independent of whether the data came from the inner (protected) or outer (exposed) tile surfaces. The trend described earlier of AP tiles having the lowest settlement rates followed by AS tiles was still obvious (p=0.007), but there was no significant difference between inner and 165 2.00 1.80 1.60 1.40 1.20 1.00 0.80 0.60 0.40 0.20 0.00 □ Outer tiles S inner tiles AP AS RELEASE treatment Figure 6-1. Comparison of coral settlement rates on outer, exposed tiles and on inner, sheltered tiles of racks in Anacropora puertogalerae (AP) patches, Acropora subglabra (AS) patches, and areas devoid of corals (“release” area). Only differences among patches were significant. ANOVA Source o f Variation SS df M S F P-value Tile position 0.0402 1 0.0402 0.2078 0.6526 Patches 2.4024 2 1.2012 6.2121 0.0067 Interaction 0.2123 2 0.1061 0.5489 0.5847 Within 4.6408 24 0.1934 Total 7.2957 29 1 6 6 outer tiles (p=0.65). Therefore grazing activity had no effect on the recruitment patterns observed > n the recruit monitoring study. Appendix 7: Comparison of projection matrix and cellular automaton projections Introduction. The use of parameters from deterministic, population models in a probabilistic, individual-level model of a community has lead to unpredictable results from the community model used in this work. In some cellular automaton simulations, AS or \ p may go extinct in the model reef, even without competition or disturbances. However, the cellular automaton model retains the general properties of its matrix model foundation while allowing greater versatility and more realism. This appendix seeks to demonstrate these common properties. Materials and Methods. Using no-reproduction matrices (with common size class intervals for AS and AP) adjusted for zero growth with asexual reproduction, 100 year projections were made individually for both AS and AP. For each species, two sets of projections were made, a "low-density" projection (with 80 SCI, 80 SC2, and 40 SC3 colonies) and a "high density" projection (with 500 colonies per size class). For comparison, ten runs of a cellular automaton model (with no competition and no disturbance option set) were also made with the two initial densities. The cellular automata were set for aggregated settlement, but since no competition was set, the distributions of the colonies should not affect their projected abundances. 167 Results and Discussion. Results of the comparisons show a close correspondence between cellular automaton and matrix model projections for both AS (Figure 7-1) and AP (Figure 7-2). This is especially in evident in the case of the high density projections for both species (lower graphs; see also correlation coefficients in Table 7-1). This pattern is due mainly to sampling effects leading to random drift in small populations. A small random change in survival of colonies (especially the reproducing SC3 adults) would lead to bigger impacts on small populations than on larger ones. Hence, the chances for extinction are larger in cellular automata runs with small initial densities. Despite these differences, comparisons of stable size-class distributions from projection models with those shown by the automata (on the 100th year) show both model implementations have essentially the same transitions (Table 7-2). Thus, if no neighborhood interactions (e.g., competition) are defined, the cellular automaton model used in this research can be used as probabilistic, individual-level analogs of projection matrix models. Appendix 8: Source code of the cellular automaton model Introduction. The following is a listing of the Turbo Pascal program I wrote to implement the cellular automaton model of the AS-AP community. The current version of this program was not written for distribution, hence it is not user-friendly and interactive. Most options were set by changing certain numbers and lines in the code 168 200 — ♦—MATRIX SCI - ■ — MATRIX SC2 — * —MATRIX SC3 - ■ — AUTOMATA SC1 - « - AUTOMATA SC2 — • — AUTOMATA SC3 150 100 0 10 20 30 40 50 60 70 80 90 100 1500 1000 500 I h 1111111111 m m lit 1111 m i m i m n n 1 1 m 1 1 it h 111111 m 11111 h 111111 'H - m i m i tin i m m 1111111 i 10 20 30 40 50 70 100 0 60 80 years Figure 7-1. Comparison of projected abundances of Acropora subglabra by cellular automata and projection matrix models with low (upper graph) and high (lc wer graph) initial densities. Notice the higher variability of the automata projections with lower initial densities. Legend for both graphs is shown on the upper graph. number o f colonies 200 ISO 100 50 -♦-MATRIX SC1 - • — M ATRIX SC2 — A— M ATRIX SC3 -**-AUTOMATA SCI - • — AUTOMATA SC2 - • — AUTOMATA SC3 0 10 20 30 40 50 60 70 80 90 100 2500 2000 1500 1000 500 0 11111111 m 11111 m 1 1 u 1 1 i i m i m 11111111 m 1111111111111111111 m m 11111 m i m 111111 itw 1 1 1 m 111 0 10 20 30 40 50 60 70 80 90 100 years Figure 7-2. Comparison of projected abundances of Anacropora puertogalerae by cellular automata and projection matrix models with low (upper graph) and high (lower graph) initial densities. Notice the higher variability of the automata projections with lower initial densities. Legend for both graphs is shown on the upper graph. 170 Table 7-1. Correlation coefficients between abundances projected by matrix models and by cellular automata.____________________________________________ SCI SC2 SC3 AP-low density 0.40 0.49 0.60 AP-high density 0.98 0.97 0.94 AS-low density 0.97 0.34 0.40 AS-high density 0.98 0.89 0.90 Table 7-2. Size-class distributions for Acropora subglabra (AS) and Anacropora puertogalerae (AP) from both models. M ATRIX SCI M ATRIX SC2 M ATRIX SC3 AUTOM ATA SCI AUTOMATA SC2 AUTOM ATA SC3 AP-low density 31.0% 59.3% 9.7% 31.4% 59.5% 9.2% AP-high density 31.0% 59.3% 9.7% 30.3% 59.9% 9.9% AS-low density 8.0% 75.6% 16.3% 8.4% 74.3% 17.3% AS-high density 8.0% 75.6% 16.3% 8.0% 75.4% 16.6% 171 itself. Comments have been provided in the appropriate procedures however to facilitate tracing these important steps. PROGRAM AUTOMT92; {simple cellular automaton program of stochastic, 3- stage succession in a Acropora subglabra and Anacropora puertogalerae community. Transition probabilities, recruitment rates, and initial population size structure from Wilfredo Licuanan’ s dissertation research. This version is part of the sequence of models of increasing complexity implementing various parameters to evaluate the role of competition, recruitment patterns, and disturbances on the spatial distributions of corals} uses CRT; const No_Columns= No_Rows= No_Cells= Dist_Low_Col= Dist_Low_Row= Dist_High_Col= Dist_High_Row= Recr_Low_Col= Rec r_Low_Rovv= Recr_High_Col= Recr_High_Row= (80+2); (80+2); No_Rows-2; 2; 2; (No_Columns-l); (NoRows-l); 2; 2; (No_Columns-l); (No_Rows-l); type KindState = 0..8; {state 0= undefined; 1 -space; 2-4 are size classes of AS species; 3-space suitable for AP species (not used in this version); 6-8 - size classes of AP species} Array_Kind_State = array [O..No_Rows, O..No_Columns] of KindState; Transect_Array= array [ 1 . .No_Cells] of integer; File_Name= string[ 15]; var Row, Col, I, J, X, Y, Counter: integer; {Num_State_l} Cover_l, {Num_State_2,} Cover_2, Num_State_3, Cover_3, {Num_State_ll,} Cover_ll, 172 Num_State_12, Cover_12, Num_State_13, Cover_13: integer; No_Iterations, Disturb_Size, No_Disturbances, Recruit Type, Recruit Count, NumlRecruits, Num_2_Recruits, Cell_Type_Base, Cell_Type_Top, NumTrans: Present_State, Next_State: Pattem_File, Stats_File, AbundanceFile: RunArray: Prompt: integer; ArrayKindState; string[21 ]; TransectArray; char; function Rrandom:real; {pseudorandom number generator in the range 0-100} begin {function} RRandom:=Random* 100 end; {function} function Toss_Coin:integer; {pseudorandom number generator producing 0 or 1 } var Silver: real; Heads: integer; begin {function} Silver:=Random* 100; if Silver<50 then Heads :=0 else Heads:=l; Toss_Coin:=Heads; end; {function} procedure Increment(var Present_State:Array_Kind_State; var Next_State:Array_Kind_State; R, C:integer; Num_State_3, Num_State_12, Num_State_13:integer); {procedure to implement probabilities in transitions between states. Species AS is Acropora subglabra, Species AP is Anacropora puertogalerae} var Thumb: real; begin {procedure Increment} case Present_State{R,CJ of {if Present_State[R,C]=2 then is AS-SC 1 } 2: begin if (Num_State_12>8) or (Num_State_13>8) then Next_State[R,C]:=l {die!} else begin {else} Thumb:=Rrandom; if (Thumb <=30.00) then {stay AS-SC 1 } Next_State[R,C] :=Present_State[R,C] else if ((Thumb >30.00) and (Thumb<=40.00)) then {become coral 2} Next_State[R,C] :=Present_State[R,C]+1 elseNext_State[R,C]:=l {die!}; end {else}; end; {case 2} {else if Present_State[R,C]=3 then AS-SC 2} 3: begin if (Num_State_12>8) or (Num_State_13>8) then Next_State[R,C] :=1 {die!} else begin {else} Thumb:=Rrandom; if (Thumb <= 7,41) then {shrink to AS-SC 1} Next_State[R,C] :=Present_State[R,C]-1 else if (((Thumb > 7.41) and (Thumb <=81.48))) then {stay as AS-SC 2 } Next_State[R,C] :=Present_State[R,C] else if ((Thumb > 81.48) and (Thumb <=87.04)) then {become AS-SC 3} Next_State[R,C]:=Present_State[R,C]+l elseNext_State[R,C]:=l {die}; end {else}; end; {else if Present_State[R,C]=4 then AS-SC 3 } 4: begin Thumb:=Rrandom; if Thumb <= 0.00 then {shrink to AS-SC 1 } Next_State[R,C]:=Present_State[R,C]-2 else if ((Thumb > 0.00) and (Thumb <= 22.86)) then {shrink to AS-SC 2} Next_State[R,C]:=Present_State[R,C]-l else if ((Thumb > 22.86) and (Thumb <= 97.14)) then {stay as AS-SC 3} Next_State[R,C]:=Present_State[R,C] elseNext_State[R,C]:=l {die}; {else if Present_State[R,C]=6 then is AP-SC 1 } 6: begin if Num_State_3>8 then Next_State[R,C]:=l {die!} else begin {else} Thumb:=Rrandom; if (Thumb <=60.53) then {stay as AP-SC 1 } Next_State[R,C]:=Present_State[R,C] else if ((Thumb >60.53) and (Thumb<=68.42)) then {becom e AP-SC 2 } Next_State[R,C]:=Present_State[R,C]+l else Next_State[R,C] :=1 {die!}; end {else}; end; {else if Present_State[R,C]=7 then is AP-SC 2} 7: begin Thumb:=Rrandom; if (Thumb <=20.65) then {shrink to AP-SC 1 } Next_State[R,C]: =Present_State[R,C]-1 else if (((Thumb >20.65) and (Thumb <=85.87))) then {stay as AP-SC 2} Next_State[R,C]:=Present_State[R,C]else if ((Thumb > 85.87) and (Thumb <=89.13)) then {becom e AP-SC 3 } Next_State[R,C]:=Present_State[R,C]+1 elseNext_State[R,C]:=l {die}; end; {else 175 if Present_State[R,C]=8 then is AP-SC 3} 8: begin Thumb:=Rrandom; if Thumb <= 0.00 then {shrink to AP-SC 2} Next_State[R,C]:=Present_State[R,C]-2 else if ((Thumb > 0.00) and (Thumb <= 20.00)) then {stay as AP-SC 3} Next_State[R,C] :=Present_State[R,C]-1 else if ((Thumb > 20.00) and (Thumb <=100.00)) then {stay as AP-SC 3} Next_State[R,C]:=Present_State[R,C] else Next_State[R,C]:=l {die}; end; {else if (Present_State[R,C]<l) then} 0: begin Next_State[R,C] := 1 ; sound(700); end; else Next_State[R,C] :=Present_State[R,C]; end; {case} end; {procedure Increm ent} procedure File_Set(Check_File:File_Name; var File_Nick:text); {procedure to prepare disk files for output. This checks if a given file already exists, if which the output is just appended. Otherwise a new file is created} var File_Exists: integer; begin {procedure File S et} assign(File_Nick,Check_File); {$1 * } reset(File_Nick); {$1 +} File_Exists:=ioresult; if File_Exists=0 then append(File_Nick) else rewrite(FileNick); end; {procedure FiIe_Set} procedure Cover(Cover_State:Array_Kind_State; var Cover_l: integer; var Cover_2: integer; var Cover_3: integer; var Cover_l 1 : integer; var Cover_I2: integer; var Cover_13: integer); {procedure to go over the matrix and count the numbers of each category for the whole ’ reef} var R,C: integer; begin; {procedure Cover} Cover_l :=0;Cover_2:=0;Cover_3 :=0; Cover_ll:=0;Cover_12:=0;Cover_13:=0; for R:=2 to No_Rows-l do for C:=2 to NoColumns-1 do case Cover_State[R,C] of 2: Cover_l:=Cover_l+l; 3: Cover_2:=Cover_2+l; 4: Cover_3:=Cover_3+l; 6: Cover_ll:=Cover_Il+l; 7: Cover_12:=Cover_12+l; 8: Cover_13:=Cover_13+l; end; end; {procedure Cover} procedure Screen_Out(var Display_State: Array _Kind_State;Counter:integer; Cover l, Cover_2, Cover_3, Cover_ll, Cover_12, Cover_13:integer); {procedure to display on screen any kind of display state and store each output array on an output file; this is limited by the 80 by 25 matrix allowed by the monitor; the spatial statistics are output on a separate file} var R, C : integer; Filex, Filex3: text; begin {procedure ScreenOut} ClrScr; write('Run # ',Counter:2); textbackground(black);textcolor(yellow); write(' AS-SC l:\Cover_l :4); textbackground(black); w riteC AS-SC2:',Cover_2:4); textbackground(black); w riteC AS-SC3:',Cover_3:3); textbackground(black);textcolor(lightred); write(' AP-SC 1 :',Cover_l 1:4); textbackground(black); 177 write(' AP-SC2:',Cover_12:4); textbackground(black); writelnC AP-SC3:',Cover_13:3); for R:=2 to 24 do begin for C:=2 to 80 do begin highvideo; if D isplay__State[R,C] =1 then begin textbackground(black); textcolor(black); writeC-'); end else if Display __State[R,C) =2 then {AS-SC 1} begin textbackground(black); textcolor(yellow); write('+'); end else if Display_State[R,C] =3 then {AS-SC2} begin textbackground(black); textcolor(yellow); write('_'); end else if Display_State[R,C] =4 then {AS-SC3} begin textbackground(black); textcolor(yellow); write('_’ ); end else if Display__State[R,C] =5 then begin textbackground(black); textcolor(black); write('-'); end else ifDisplay__State[R,C] =6 then (AP-SC 1 } begin textbackground(black); textcolor(lightred); write('+ '); end else if Display_State[R,C] =7 then {AP-SC2} begin textbackground(black); textcolor(lightred); writefj); end else if Display_State[R,C] =8 then {AP-SC3} begin textbackground(black); textcolor(lightred); writefj); end else if Display J5tate[R,C] =0 then begin textbackground(black); textcolor(white); writefO'); end; end; {for columns} textcolor(white); writeln; normvideo; end; {for row s} {the following set of statements deal with the output of reef maps} File_Set(Pattem_File,Filex); writeln(Filex,'»Run # ',Counter:3,' Cover: ',Cover_l :5,Cover_2:5,Cover_3:5,Cover_l 1:5,Cover_12:5,Cover_13:5); writeln(Filex,'Legend: -:space, AS: *:SC1, _:SC2, _:SC3, A P: 1:SC1, 2:SC2, 3:SC3 ’ ); for R:=2 to (No_Rows-l) do begin for C:=2 to (No_Columns-l) do begin ifDisplay_State[R,C] =1 then write(Filex,'-') else if Display_State[R,C] =2 then {A S-SC 1} write(Filex,'*') else if Display_State[R,C] =3 then {AS-SC2} write(Filex,, ') else if Display_State[R,C] =4 then {AS-SC3} write(Filex,'J) else if Display_State[R,C] =5 then 179 write(Filex,'_') else ifDisplay_State[R,C] =6 then {AP-SC 1 } write(Filex,T) else if Display_State[R,C] =7 then {AP-SC2} write(Filex,'2') else if Display_State[R,C] =8 then {AP-SC3} w rite^ex,^) else if Display_State[R,C] =0 then write(Filex,’ 0') end; {ifs; for rows} writeln(Filex) end; {for colum ns} close(Filex); {the following set of statements deal with the output of abundances} File_Set(Abundance_File,Filex3); write(Filex3,'Run # ',Counter:5,' AS #s (SC 1-3) of: ',Cover_l:5,Cover_2:5,Cover_3:5); writeln(Filex3,' AP#s (SC1-3) of: ',Cover_l l:5,Cover_12:5,Cover_13:5); close(Filex3); end; {procedure Screen Out} procedure Count (Count State: Kind_State; var Num_State_X:integer; var Counted_State:Array_Kind_State; Row,Col:integer); {this procedure determines the numbers of all occurrences of a given cell state for the 8 adjoining cells of a given cell.} var R, C : integer; begin {procedure Count} Num_State_X:=0; R:=0; C:=0; for C:= (Col-1) to (Col+1) do for R:= (Row-1) to (Row+1) do if Counted_State[R,C]=Count_State then N umStateX: =N um_S tateX + 1 ; if Counted_State[Row,Col]=Count_State then Num_State_X:=Num_State_X -1; end; {procedure Count} procedure Init_Array(var Clean_State:Array_Kind_State); {procedure to clean reef arrays of contents} 180 var R,C: integer; begin {procedure InitArray} for R:=0 to No_Rows do for C:=0 to NoColumns do Clean_State[R,C]:=1 ; end; {procedure lnit_Array} procedure Init_Integer_Arrayl(var C leanState: T ransectArray); {procedure to clean transect arrays of contents} var R : integer; begin {procedure Init Array} for R:=l to No_Cells do Clean_State[R]:=0; end; {procedure Init Array} procedure Runs(var RunArray :T ransectArray; var Num_Runs:integer; Reef_State: Array_Kind_State; Transect Col, Counter, Cell_Type_Base, Cell_Type_Top:integer); {procedure to compute run lengths of an input range of cell types, store the results in an array for each run, and print the results; since reef is wrapped as a torus, start and end runs are combined if of the same species} var R, Run_Num: integer; Trans: TransectArray; begin {procedure Runs} Init_Integer_Array 1 (RunArray); Init_Integer_Array 1 (Trans); Num_Runs:=0; Run_Num:=0; for R:=2 to (No_Rows-l) do Trans[R-l]:=Reef_State[R,Transect_Col]; R:=l; while R<=(No_Rows-2) do begin {w hile} if ((Trans[R]>=Cell_Type_Base) and (Trans[R]<=Cell_Type_Top)) then begin {w hile} Run_Num:=Run_Num+1 ; end; {w hile} if ((((Trans[R+l]<Cell_Type_Base) or (Trans[R+l]>Cell_Type_Top)) or (R=No_Rows-2)) and ((Trans[R]>=Cell_Type_Base) and (Trans[R]<=Cell_Type_Top))) then begin {if} Num_Runs:=Num_Runs+1 ; Run Array [Num Runs] :=Run_Num; Run_Num:=0; end; {if} R:=R+1; end; {w hile} if ((Trans[No_Rows-2]>=Cell_Type_Base) and (Trans[No_Rows-2]<=Cell_Type_Top)) then if ((Trans[ l]>=Cell_Type_Base) and (Trans[l]<=Cell_Type_Top)) then begin {combination of transect start and end runs} Run_Array[ 1 ]:=Run_Array[ 1 ]+Run_Array [NumRuns]; N umRuns:=Num_Runs-1 ; end; end; {procedure Runs} procedure MetaRuns (Reef_State:Array_Kind_State; Num_Trans:integer; Counter: integer); {procedure for sampling an input reef with an input number of randomly selected transects} const AS_SC1= 2; AP_SC1= 6; AS_SC2= 3; AP_SC2= 7; AS_SC3= 4; AP_SC3= 8; Space= 1 ; var I, T, Transect Col, Num Runs, NumRunsAP, NumRunsAS, NumRunsSpace, Len Runs AP, Len Runs AS, Len_Runs_Space: integer; Filex5: text; begin {procedure Meta Runs} Num_Runs_AP:=0; Num_Runs_AS:=0; Num_Runs_Space:=0; LenRunsAP:=0; Len_Runs_AS:=0; Len_Runs_Space:=0; for I:=l to Num Trans do 182 begin {for number of transects} T ransect_Col:=round(Random(No_Columns-1 )+2); Runs(Run_Array, Num Runs, Reef State, Transect_Col, Counter, AP_SC1, AP_SC3); Num_Runs_AP:=Num_Runs_AP + NumRuns; for T:=l to Num Runs do LenRunsAP:=Len_Runs_AP + Run_Array[T]; Runs(Run_Array, Num Runs, Reef State, TransectCol, Counter, ASSC1, ASSC3); Num_Runs_AS:=Num_Runs_AS + Num_Runs; for T:= 1 to Num Runs do Len_Runs_AS:=Len_Runs_AS + RunArrayfT]; Runs(Run_Array, Num Runs, Reef_State, Transect Col, Counter, Space, Space); Num_Runs_Space:=Num_Runs_Space + NumRuns; for T:=1 to Num Runs do Len_Runs_Space:=Len_Runs_Space + RunArrayfT]; end; {for number of transects} {the following lines concern output of runs results to a file} File_Set(Stats_File,Filex5); if Counter=0 then begin {header} write(Filex5,'*Run # *Run Lengths: **A P** *“ *A S** *Space*'); writeln(Filex5,' *No of Runs: **AP** **A S** *Space*'); end; {header} write(Filex5,' ',Counter:3,' ',Len_Runs_AP:7, LenRunsAS: 7,Len_Runs_Space:7); writeln(Filex5,' ',Num_Runs_AP:7, NumRunsAS: 7, Num_Runs_Space:7); close(Filex5); end; {procedure Meta_Runs} procedure Reef ln (var Present State: Array_Kind_State); {procedure to read in a matrix of numbers representing the states of the cells of the automaton} var fileimp: text; NameOfFile: string] 14]; R,C, Hold: integer; Holder: real; 183 begin {procedure Reef ln} write ('Enter name of file containing initial reef configuration:'); readln(NameOfFile); assign(fileimp,NameOfFile); reset(fileimp); while not eof(fileimp) do begin for R.-2 to (NoRows-l) do begin for C:=2 to (No_Columns-l) do begin read(fileimp,Holder); Hold:=round(Holder); Present_State[R,C]:=Hold; end; readln(fileimp); end; end; close(fileimp); end; {procedure Reef_In} procedure Wrap(var Wrap_Array:Array_Kind_State); {procedure to copy reef row 1 to last empty row and reef column 1 to last empty column, last reef row to first empty row and last reef column to first empty column, this ensures a continuous reef wrapped around as a torus (donut)} var R,C: integer; begin {procedure W rap} for C:=l to (No_Columns) do {vertical wrap} begin Wrap_Array[l,C]:=Wrap_Array[No_Rows-l,C]; Wrap_Array(No_Rows,C] :=Wrap_Array[2,C]; end; for R:=2 to (No_Rows-l) do {horizontal wrap} begin Wrap_Array [R, 1 ] :=Wrap_Array [R,No_Columns-1 ]; Wrap_Array [R,No_Columns]: = W rapArray [ R, 2 ]; end; end; {procedure W rap} procedure Anchor (var Reef_State:Array_Kind_State; R, C:integer); {procedure to simulate anchor impacts by shrinking SC2 and SC3 corals in the impact area to the next smaller size class} begin {procedure Anchor} case Reef_State[R,C] of 3: Reef_State[R,C]:=Reef_State[R,C]-l {shrinks}; 4: Reef_State[R,C]:=Reef_State[R,C]-l {shrinks}; 7: Reef_State[R,C]:=Reef_State[R,C]-l {shrinks}; 8: Reef_State[R,C]:=Reef_State[R,C]-l {shrinks}; end {case}; end; {procedure Anchor} procedure Disturb (var Reef_State:Array_Kind_State; Disturb_Size: integer; DistLowCol, DistLowRow, Dist_High_Col, Dist_High_Row:integer); {procedure to create disturbances at random points of a given size, range of points to be disturbed are randomly chosen from the limits defined} var Range_Col, Range Row, R, C, Ground_Zero_Col, Ground_Zero_Row: integer; begin {procedure Disturb} Range_Col:=Dist_High_Col - Dist Low Col +1; Range_Row:=Dist_High_Row - Dist Low Row +1; {the additional units are to make the range inclusive} Ground_Zero_Col:=round(Random(Range_Col + 1) + Dist_Low_Col); Ground_Zero_Row:=round(Random(Range_Row + 1) + DistLowRow); {the additional units are to correct for the random number generator's inability to produce numbers equalling its maximum} Case Disturb_Size of 1 : Anchor(Reef_State, Ground_Zero_Row, Ground_Zero_Col); 9: for R:= GroundZeroRow-1 to Ground_Zero_Row+l do for C:=Ground_Zero_Col-l to Ground_Zero_Col+l do if ({C > 1) and (C < No_Columns)) then if ((R > 1) and (R < No_Rows)) then Anchor(Reef_State, R, C); end; {case} end; {procedure Disturb} procedure Baby_Num(Parent_Site:Array_Kind_State;R,C:integer; var Recruit_Type:integer; var Baby_Count:integer); { procedure to determine the number of juveniles each colony produces based on a poisson distribution derived from measured or estimated recruitment rates; a potential modification is that the fecundity will depend on the number and identity of the neighbor} var Babe_Num: integer; Pinkie: real; begin {procedure Baby_Num} case Parent_Site[R,C] of 4 {Species AS, SC 3}: begin {case stage 3 } Recruit_Type:=3; Pinkie:=Rrandom; if (Pinkie <=39.78) then {no recruits m ade} Baby_Count:=0 else if ((Pinkie > 39.78) and (Pinkie<=76.45)) then { 1 offspring bom} Baby_Count:=l else if ((Pinkie > 76.45) and (Pinkie <=93.35)) then {2 offspring bom} Baby_Count:=2 else if ((Pinkie > 93.35) and (Pinkie <=98.54)) then {3 offspring bom} Baby_Count:=3 else if ((Pinkie > 98.54) and (Pinkie <=99.74)) then {4 offspring bom} Baby_Count:=4 else if ((Pinkie > 99.74) and (Pinkie <=99.96)) then {5 offspring bom} Baby_Count:=5 else if ((Pinkie > 99.96) and (Pinkie <=99.99)) then {6 offspring bom} Baby_Count:=6 else if ((Pinkie > 99.99) and (Pinkie <=100.00)) then {7 offspring bom} Baby_Count:=7 else Baby_Count:=0 {none}; end {case stage 3}; 8 {AP-SC 3}: begin {Species AP SC 3} Recruit_Type:=7; Pinkie:=Rrandom; if (Pinkie <=18.64) then {no recruits m ade} Baby_Count:=0 else if ((Pinkie > 18.64) and (Pinkie<=49.95)) then { 1 offspring bom} Baby_Count:=l else if ((Pinkie > 49.95) and (Pinkie<=76.25)) then {2 offspring bom} Baby_Count:=2 else if ((Pinkie > 76.25) and (Pinkie <=90.98)) then {3 offspring bom} Baby_Count:=3 else if ((Pinkie > 90.98) and (Pinkie <=97.16)) then {4 offspring bom} Baby_Count:=4 else if ((Pinkie > 97.16) and (Pinkie <=99.24)) then {5 offspring bom} Baby_Count:=5 else if ((Pinkie > 99.24) and (Pinkie <=99.82)) then {6 offspring bom} Baby_Count:=6 else if ((Pinkie > 99.82) and (Pinkie <=99.96)) then {7 offspring bom} Baby_Count:=7 else if ((Pinkie > 99.96) and (Pinkie <=99.99)) then {8 offspring bom} Baby_Count:=8 else if ((Pinkie > 99.99) and (Pinkie <=100.00)) then {9 offspring bom} Baby_Count:=9 else Baby_Count:=0 {none}; end {case AP SC 3}; else Baby_Count:=0 end {case statement}; end; {procedure Baby_Num} procedure Recruit_Sum(Recruit_Type: integer; Recmit_Count:integer; var Num_l_Recruits:integer; var Num_2_Recruits:integer); {procedure to accumulate recruit numbers of die two different types} begin {procedure RecruitSum} case Recruit_Type of 3: Num_l_Recruits:=Num_l_Recruits + RecruitCount; 7: Num_2_Recruits:=Num_2_Recruits + Recruit Count; end; {case} end; {procedure Recruit_Sum} procedure Recruit (var Reef_State:Array_Kind_State; Rec Type, Num Recruits: integer; Recr_Low_Col, Recr L ow Row, Recr_High_Col, Recr_High_Row:integer); {procedure to simulate recruitment of certain coral types, range of points to be recruited on are randomly chosen from the limits defined; each potential site is evaluated in term of numbers and types of neighbors} var Range_Col, Range_Row, I, Touchdown_Col, Touchdown_Row, NumSCO, NumSCl, Num_SC_2, Num_SC_3, NumSCl 1, Num_SC_l 2, Num_SC_l 3: integer; Trials: integer; begin {procedure Recruit} Range_Col:=Recr_High_Col - RecrLowCol +1; 187 Range_Row:=Recr_High_Row - Recr Low Row +1; {the additional units are to make the range inclusive} Randomize; if Num_Recruits>=l then case RecType of 3 {in case of species AS}: begin {Species AS recruit search for settlement site} for 1=1 to NumRecruits do begin {for} {$B+}Trials:=0; repeat begin {repeat} Touchdown_Col:=round(Random(Range_Col) + RecrLowCol); Touchdown_Row:=round(Random(Range_Row) + RecrLowRow); if (Reef_State[Touchdown_Row,Touchdown_Col]=l) then begin {determination of neighborhood} Count(2,Num_SC_l,Reef_State,Touchdown_Row,Touchdown_Col); Count(3,Num_SC_2,Reef_State,Touchdown_Row,Touchdown_Col); Count(4,Num_SC_3,Reef_State,Touchdown_Row,Touchdown_Col); end; {determination of neighborhood} Trials:=Trials+l; end; {repeat} until ((Reef_State[Touchdown_Row,Touchdown_Col]=l) and ( (Num_SC_3>0) or (Num_SC_2>2))); Reef_State[Touchdown_Row,Touchdown_Col]:=Rec_Type; Wrap(Reef_State); end; {for up to number of AS recruits} end; {species AS search} 7 {in case of species AP}: begin {Species AP recruit search for settlement site} for 1=1 to Num Recruits do begin {for} {$B+}Trials:=0; repeat begin {repeat} Touchdown_Col:=round(Random(Range_Col) + Recr_Low_Col); Touchdown_Row:=round(Random(Range_Row) + RecrLo w_Ro w); if (Reef_State[Touchdown_Row,Touchdown_Col]=l) then begin {determination of neighborhood} Count(6,Num_SC_l 1 ,Reef_State,Touchdown_Row,Touchdown_Col); Count(7,Num_SC_12,Reef_State,Touchdown_Row,Touchdown_Col); Count(8,Num_SC_13,Reef_State,Touchdown_Row,Touchdown_Col); end; {determination of neighborhood} Trials:=Trials+l; end; {repeat} 188 until ((Reef_State[Touchdown_Row,Touchdown_Col]=l)) and ( (Num_SC_13>0) or (Num_SC_12>2)); Reef_State[Touchdown_Row,TouchdownCol]:=Rec_Type; Wrap(Reef_State); end; {for up to number of AP recruits} end; {species AP search} end; {case} end; {procedure Recruit} procedure Recruit_Sequence(var Num_l_Recruits:integer; var Num_2_Recruits: integer); {procedure for recruitment, with randomization of the sequence of recruitment, i.e. all of AP or AS may settle first} var Rec_Type, Toss: integer; begin {procedure Recruit_Sequence} Toss:=Toss_Coin; case Toss of 0: begin {recruitment of Species AS comes first} Rec_Type:=3; Recruit(Next_State,Rec_Type, NumlRecmits, Recr Low Col, RecrLowRow, RecrHighCol, Recr High Row); Rec_Type:=7; Recruit(Next_State,Rec_Type, Num_2_Recmits, Recr Low Col, Recr Low Row, Recr_High_Col, Recr_High_Row); end {recruitment of Species AS coming first}; 1 : begin {recruitment of Species AP comes first} Rec_Type:=7; Recruit(Next_State,Rec_Type, Num_2_Recruits, Recr Low Col, RecrLowRow, RecrHighCol, RecrHighRow); Rec_Type:=3; Recruit(Next_State,Rec_Type, Num l Recruits, Recr Low Col, Recr Low Row, Recr High Col, Recr High Row); end {recruitment of Species AP coming first}; end; {case} Num_l_Recruits:=0; Num_2_Recruits:=0; end; {procedure Recmit_Sequence} 189 begin {program} randomize; ClrScr; writeln; writeln; writeln; writeln; writeln (' ******** Automat version 9.2 ♦*******'); writeln ('** CELLULAR AUTOMATON MODEL OF A SIMPLE REEF COMMUNITY**'); writeln (' by WY Licuanan, University of Southern California1 ); writeln; Init_Array(Next_State); Init_Array(Present_State); Reef_In(P resentState); Wrap(Present_State); Num_l_Recruits:=0; Num_2_Recruits:=0; writeln; writeln('Do you want to use the default file names'); writeC f°r output files (Y/N) ?'); readln(Prompt); if ((Prompt=Y') or (Prompt='y')) then begin PattemFile:-c:\MAPout.txt'; Stats_File:='c:\STATout.txt'; Abundance_File:='c:\ABUNDout.txt'; end else begin writeln; write ('Enter drive and name of output file for the reef map:'); readln(Pattem_File); write ('Enter drive and name of output file for the spatial statistics:'); readln(StatsFile); write ('Enter drive and name of output file for the size class abundances:'); readln(Abundance_File); end; writeln; write ('Enter number of iterations to be run:'); readln(No_Iterations); write ('Enter number of disturbances per iteration:'); readln(No_Disturbances); if No_DisturbancesoO then begin {if} write('Enter size of disturbances (1 or 9):'); readln(DisturbSize); end {if} else Disturb_Size:=0; {write('Finally, enter number of transects for runs tests:'); readln(Num_T rans);} Num_Trans:=10; Wrap(Present_State); Row:=2; while Row < (No_Rows) do begin Col:=2; while Col < (No_Colunms) do begin {count procedures can be added here for determing incrementing based on numbers and identities of neighbors} Count(4,Num_State_3,Present_State,Row,Col); Count(7,Num_State_12,Present_State,Row,Col); Count(8,Num_State_13,Present_State,Row,Col); Increment(Present_State, Next_State, Row, Col,Num_State_3,Num_State_l 2,Num_State_l 3); Baby_Num(Present_State, Row, Col, Recruit_Type, Recruit_Count); Recruit_Sum(Recruit_Type, Recruit Count, Num l Recruits, Num_2_Recruits); CoI:=Col+l; end; Row:=Row+l; end; Counter:=0; Meta Runs (Present_State, Num Trans, Counter); Cover(Present_State, Cover_l, Cover_2, Cover_3, Cover_ll, Cover_12, Cover_13); Screen_Out(Present_State,Counter,Cover_ 1 ,Cover_2,Cover_3, Cover ll, Cover_12, Cover_13); for J:= 1 to No_Disturbances do Disturb (Next_State, Disturb Size, Dist_Low_Col, Dist_Low_Row, Dist_High_Col, Dist_High_Row); Counter:=l; Wrap(Next_State); Recruit_Sequence(Num_l_Recruits, Num_2_Recruits); 191 Meta Runs (Next_State, Num Trans, Counter); Cover(Next_State, Cover_l, Cover_2, Cover_3, Cover_l 1 , Cover_12, Cover_13); Screen_Out(Next_State,Counter,Cover_l,Cover_2,Cover_3, Cover_l 1 , Cover_12, Cover_13); Init_Array(Present_State); for I:=2 to (No_Iterations) do begin {Next_State becomes new Present_State} Counter:=Counter+1 ; for X:=l to NoColumns do for Y:=l to No Rows do Present_State[X,Y] :=Next_State[X,Y]; Init_Array(Next_State); Wrap(Present_State); {next cycle of the iterations} Row:=2; while Row < (No_Rows) do begin Col:=2; while Col < (No Columns) do begin Count(4,Num_State_3,Present_State,Row,Col); Count(7,Num_State_12,Present_State,Row,Col); Count(8,Num_State_13,Present_State,Row,Col); Increment(Present_State, Next_State, Row, Col, Num_State_3, Num_State_12, Num_State_13); Baby_Num(Present_State, Row, Col, Recruit Type, Recruit_Count); Recruit_Sum(Recruit_Type, Recruit Count, Num l Recruits, Num_2_Recruits); Col:=Col+l; end; Row:=Row+l; end; Wrap(Next_State); Rec ruit_Sequence(N um_ 1 Recruits, Num_2_Recruits); for J:= 1 to NoDisturbances do Disturb (Next_State, Disturb Size, Dist_Low_Col, Dist Low Row, Dist_High_Col, Dist_High_Row); 192 Meta_Runs (Next_State, NumTrans, Counter); Cover(Next_State, Cover l, Cover_2, Cover_3, Coverll, Cover_12, Cover_13); Screen_Out(Next_State, Counter,Cover_l,Cover_2,Cover_3, Cover_ll, Cover_12, Cover_13); Init_Array(Present_State); end; {for} nosound; end. {program} Appendix 9: Plates 1. A monospecific patch of Anacropora puertogalerae (AP) in Block D. This patch is about 15 m at its widest portion. Note the gaps filled only with rubble of the species. 2. A monospecific patch of Acropora subglabra (AS) in Block E. This patch is about 35 m at its widest portion. 3. An Acropora subglabra (AS) cross transplant in the Anacropora puertogalerae (AP) patch. Note the bleached and the dead, algal-fouled portions. Mortality rate of this treatment is significantly higher compared to the other Acropora subglabra treatments. 4. An Anacropora puertogalerae (AP) cross-transplant encircling a colony of the host Acropora subglabra (AS) patch. 5. A close-up of a 3 month old rack (set in the release area of Block D) showing the level of fouling by algae. Each of the 6 tiles is about 2 cm apart. 6. A close-up of a 3 month old rack set in the middle of an Anacropora puertogalerae (AP) patch (Block D). Note the lighter level of algal fouling compared to the release tiles (Plate 5). 194 Plate 1 195 Plate 2 Plate 3 197 Plate 4 198 Plate 5 Plate 6
Abstract (if available)
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
00001.tif
PDF
00001.tif
PDF
00001.tif
PDF
00001.tif
PDF
00001.tif
PDF
00001.tif
PDF
00001.tif
PDF
00001.tif
PDF
00001.tif
PDF
00001.tif
PDF
00001.tif
PDF
00001.tif
PDF
00001.tif
PDF
00001.tif
PDF
00001.tif
PDF
00001.tif
PDF
00001.tif
PDF
00001.tif
PDF
00001.tif
PDF
00001.tif
Asset Metadata
Core Title
00001.tif
Tag
OAI-PMH Harvest
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-oUC11257222
Unique identifier
UC11257222
Legacy Identifier
9621622