The influence of density-dependent and number-dependent processes on individual demographic rates often results in ecological tradeoffs. Because such processes have important implications for individual level fitness and population regulation, they long have been an important topic in ecological research. I used Dascyllus trimaculatus , a site-attached planktivorous coral reef fish, to determine independently the effects of population density and group size on rates of individual growth and mortality. Somatic growth of D. trimaculatus was positively related to the density of D. trimaculatus outplanted to host anenomes, Heteractis magnifica , with nearly twice as much growth observed in individuals living in the highest density treatment. By contrast, survivorship of D. trimaculatus on H. magnifica exhibited a negative relationship with density. There was no significant effect of group size on either the growth or mortality rates of D. trimaculatus . These relationships suggest a tradeoff between density-dependent growth and survival in this species. My analysis of behavioral data indicated that D. trimaculatus living under conditions of higher local population densities displayed decreased rates of intraspecific aggression because of positive feedback between local population density and the foraging distance from an individual’s host anemone. In turn, lower rates of aggression and feeding higher in the water column resulted in increased somatic growth rates due to: (1) a reduced energy expenditure and (2) an increase in prey consumption. These data are part of a Master's Thesis published by ProQuest: Nielsen, Jessica Anne. Effects of density on behaviorally-mediated tradeoffs between growth and survivorship. MA Thesis, University of California Santa Barbara, 2013. This material is based upon work supported by the U.S. National Science Foundation under Grant No. OCE 16-37396 (and earlier awards) as well as a generous gift from the Gordon and Betty Moore Foundation. Research was completed under permits issued by the French Polynesian Government (Délégation à la Recherche) and the Haut-commissariat de la République en Polynésie Francaise (DTRT) (Protocole d'Accueil 2005-2018). This work represents a contribution of the Moorea Coral Reef (MCR) LTER Site.

In this study, size-based matrix models for the reef-building coral Orbicella annularis at 14-m depth on the Tektite reef in St. John, U.S. Virgin Islands, were used to: (1) explore the demography of changing coral cover over 25 yr, (2) test for spatial homogeneity in demographic properties through a contrast with a previous study (at Yawzi Point, Edmunds and Elahi 2007), and (3) evaluate the potential for future population stability. During three, five year intervals from 1988 to 2002, St. John was affected by hurricanes and bleaching, yet coral cover at Tektite increased from 33% to 49%; from 2002 to 2007, it declined to 27%; and from 2010 to 2013, it stabilized at ∼ 28%. Over a quarter-century, colonies > 50 cm2 became rare, the abundance of colonies ≤ 50 cm2 increased from 58% (1988) to 92% (2013), and population density doubled to 67 colonies m-2 by 2013. Population growth (λ) was greater at Tektite (1.152 ≥ λ ≥ 1.018) than Yawzi Point (0.679 ≥ λ ≥ 0.586), and while population size at Tektite declined due to bleaching and disease in 2005 (λ = 0.753 over 2003–2008), it recovered between 2008 and 2013 (λ = 0.966); the population at Yawzi Point declined from 1988 to 2003 without signs of recovery. Projections suggest a continuation of recent conditions could allow O. annularis at Tektite to retain ∼ 9% cover after 100 yr, but with a return to the rates of growth and survival of 1993–1998, it could attain coverage similar to that of 1988 (33%) in ∼ 15 yr.

These data are evidence of the the long-term dynamics of shallow coral reefs along the south coast of St. John from as early as 1987. These data describe coral reef community structure as percent cover based on the analysis of color photographs. All of these data originate from color images of photoquadrats recorded annually (usually in the summer) from as early as 1987. The data falls into three groups. The two groups that are contained in this data package are (1) Tektite & Yawzi and (2) Random sites. The juvenile coral density is packaged separately. Tektite – this is at 14 m depth on the eastern side of Great Lameshur Bay and is the original site of the Tektite man-in-the sea project in 1969; this project marked the birth of the Virgin Islands Ecological Research Station (later the Virgin Islands Environmental Resource Station) that hosts the field component of the project. The reef in this location consists of a single buttress that has remained dominated by Montastraea anularis since the start of the research (1987). These surveys consist of 30 photoquadrats (1 x 1 m) distributed along three, 10 m transects. Yawzi – this is at 9 m depth and is on the western side of Great Lameshur Bay and has been recorded photographically since 1987. This reef also started the study period dominated by Montastraea annularis, but has degraded much more rapidly that the Tektite site. These surveys consist of 30 photoquadrats (1 x 1 m) distributed along three, 10 m transects. Random sites – were added in 1992 to address the concern that the original sites (Yawzi and Tektite) were selected on “good” areas of reef and, therefore, could only decline in condition. The Random sites were selected using random coordinates in 1992, and consist of 6 sites (at 7-9 m depth) scattered between Cabritte Point and White Point. All lie a little shoreward of Yawzi and Tektite, and have always been characterized by low coral cover (< 10% cover). The surveys consist of 18-40 photoquadrats (0.5 x 0.5 m; with sample size determined by the exposures on a 35 mm cassette versus digital techniques) that are placed at random points along a transect.

A potential consequence of individuals compensating for density-dependent processes is that rare or infrequent events can produce profound and long-term shifts in species abundance. In 1983–1984 a mass mortality event reduced the numbers of the abundant sea urchin Diadema antillarum by 95–99% throughout the caribbean and western atlantic. Following this event, the abundance of macroalgae increased and the few surviving D. antillarum responded by increasing in body size and fecundity. these initial observations suggested that populations of D. antillarum could recover rapidly following release from food limitation. In contrast, published studies of field manipulations indicate that this species had traits making it resistant to density-dependent effects on offspring production and adult mortality; this evidence raises the possibility that density-independent processes might keep populations at a diminished level. Decadal scale (1983–2011) monitoring of recruitment, mortality, population density and size structure of D. antillarum from st John, us Virgin Islands, indicates that population density has remained relatively stable and more than an order of magnitude lower than that before the mortality event of 1983–1984. We detected no evidence of density-dependent mortality or recruitment since this mortality event. In this location, model estimates of equilibrium population density, assuming density-independent processes and based on parameters generated over the first decade following the mortality event, accurately predict the low population density 20 years later (2011). We find no evidence to support the notion that this historically dominant species will rebound from this temporally brief, but spatially widespread, perturbation.

These data are evidence of the the long-term dynamics of shallow coral reefs along the south coast of St. John from as early as 1989. These data come from the long term site at 9 m depth known as Yawzi Point. These data are representative of the Video Transects (VT) starting in 1989 (see Edmunds and Wirman 1991, MEPS 78: 201-204). All images analyzed using 200 randoms dots placed at random on their surface and the substratum beneath each dot annotated (~371,800 decisions by Nov 2017). Hurricanes Irma and Maria struck in September 2017. Because of this, sampling was completed in July and November of 2017. Data are available upon request for 1989 (raw data lost), 1992, 1993, and 1994. These images were analyzed with a coarser resolution that is not consistent with the current format.

This dataset contains the percentage cover of all stony corals (Scleractinia, pooled among genera) and other major groups analyzed from 0.5 x 0.5 m photographic quadrats at the Backreef habitat at the Moorea Coral Reef LTER, French Polynesia. This survey time series began in 2005 and is repeated each year in April. Functional groups counted are: Scleractinian corals, Macroalgae, Crustose Coralline Algae / Bare Space, Soft Corals, Hydrocorals ( Millepora ), Algal Turf and Sand. The coral community was sampled photographically in all represented habitats surrounding the island: Fringing Reef, Lagoon, and Outer Reef. This dataset contains only Lagoon (Backreef) data (see knb-lter-mcr.4 for the other habitats) and is structured in a repeated-measures protocol to allow a statistical contrast of sites, shores and times. Community structure was determined through a coarse analysis of the benthic community, initially completed in situ (2005), but using photoquadrats from 2006. There are quadrats analyzed at each of five areas within each site, and the areas are revisited (but not the quadrats) each year to support the repeated measures design. There are two tables available, providing different views of the same data: a long table having all values in one column and a wide table having a separate column for each observed object. Detailed methods are available in the protocols section.

This dataset contains the percentage cover of the stony corals (Scleractinia) and other major groups analyzed from 0.5 x 0.5 m photographic quadrats in several reef habitats at the Moorea Coral Reef LTER, French Polynesia. This survey has been repeated annually in April since 2005. There are two tables available, providing different views of the same data: a long table having all values in one column and a wide table having a separate column for each dependent variable. Functional groups (i.e., dependent variables) counted are: Scleractinian Corals (by genus where appropriate, see methods), Macroalgae, Crustose Coralline Algae / Bare Space, Soft Corals, Hydrocorals (Millepora), Algal Turf and Sand. The coral community was sampled photographically in all habitats surrounding the island: Fringing Reef, Lagoon (Backreef), and Outer Reef (Forereef.) The sampling regime consists of a repeated-measures protocol in each habitat, and is structured by habitat to allow a statistical contrast of sites, shores, times, and in the case of the outer reef, depths. Detailed methods are available in the protocols section. This material is based upon work supported by the U.S. National Science Foundation under Grant No. OCE 16-37396 (and earlier awards) as well as a generous gift from the Gordon and Betty Moore Foundation. Research was completed under permits issued by the French Polynesian Government (Délégation à la Recherche) and the Haut-commissariat de la République en Polynésie Francaise (DTRT) (Protocole d'Accueil 2005-2022). This work represents a contribution of the Moorea Coral Reef (MCR) LTER Site.

The Moorea Labeled Corals dataset is a subset of the MCR LTER packaged for computer vision research. It contains 2055 images from three habitats IDs: fringing reef outer 10m and outer 17m, from 2008, 2009 and 2010. It also contains random point annotation (row, col, label) for the nine most abundant labels, four non coral labels: (1) Crustose Coralline Algae (CCA), (2) Turf algae, (3) Macroalgae and (4) Sand, and five coral genera: (5) Acropora, (6) Pavona, (7) Montipora, (8) Pocillopora, and (9) Porites. These nine classes account for 96% of the annotations and total to almost 400,000 points. These nine classes are the ones analyzed in (Beijbom, 2012); less-abundant genera not treated in the automation are also present in the dataset. These data were published in Beijbom O., Edmunds P.J., Kline D.I., Mitchell G.B., Kriegman D., 'Automated Annotation of Coral Reef Survey Images', IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Providence, Rhode Island, 2012. [ BibTex ] [ pdf ] These data are a subset of the raw data from which knb-lter-mcr.4 is derived. This material is based upon work supported by the U.S. National Science Foundation under Grant No. OCE 16-37396 (and earlier awards) as well as a generous gift from the Gordon and Betty Moore Foundation. Research was completed under permits issued by the French Polynesian Government (Délégation à la Recherche) and the Haut-commissariat de la République en Polynésie Francaise (DTRT) (Protocole d'Accueil 2005-2018). This work represents a contribution of the Moorea Coral Reef (MCR) LTER Site.

Pacific Labeled Corals is an aggregate dataset containing 2318 coral reef survey images from four Pacific monitoring projects in Moorea (French Polynesia), the northern Line Islands, Nanwan Bay (Taiwan) and Heron Reef (Australia). Pacific Labeled Corals contain a total of 318828 expert annotations across 4 pacific reef locations, and can be used as a benchmark dataset for evaluating object recognition methods and texture descriptors as well as for domain transfer learning research. The images have all been annotated using a random point annotation tool by a coral reef expert. In addition, 200 images from each location have been cross-annotatoed by 6 experts, for a total of 7 sets of annotations for each image. These data will be published in Beijbom O., et al., 'Transforming benthic surveys through automated image annotation' (in submission). These data are a subset of the raw data from which knb-lter-mcr.4 is derived.

This study provides a landscape-scale context to a decadal-scale analysis of community structure on shallow reefs along 4 km of the south shore of St. John, US Virgin Islands. By focusing on 12-14 sites along ~100 km of the shores of St. John and St. Thomas, surveys conducted in 2011 were used to contrast: (1) a local-scale with a landscape-scale analysis on two islands, (2) reefs around St. John and St. Thomas, and (3) reefs on north and south shores. Reefs were censused using photoquadrats that were analyzed for percentage cover first by functional groups (coral, macraolagae and CTB), and then by coral genus. In general, among-site variation for the coarse-resolution analysis eclipsed shore and island effects, but the fine-resolution analysis revealed strong site-specific differences for multiple coral genera that could be the product of priority effects in community succession. Over the next decade these differences probably will create unique community trajectories at each site.

In early years photos were analyzed with less refined taxonomic classsification. This reference table may be used to aggregate taxa to align with coarse analysis. Use this table to define sub-categories and super-categories such as the M. annularis complex. This is a reference dataset containing the taxonomic classifications, groupings of taxonomic classifications, and functional groups. It has been updated in accordance with Budd, et al., 2012.

This data package is formatted as an ecocomDP (Ecological Community Data Pattern). For more information on ecocomDP see https://github.com/EDIorg/ecocomDP. This Level 1 data package was derived from the Level 0 data package found here: https://pasta.lternet.edu/package/metadata/eml/edi/291/2. The abstract below was extracted from the Level 0 data package and is included for context: These data are evidence of the the long-term dynamics of shallow coral reefs along the south coast of St. John from as early as 1987. These data describe coral reef community structure as percent cover based on the analysis of color photographs. All of these data originate from color images of photoquadrats recorded annually (usually in the summer) from as early as 1987. The data falls into three groups. The two groups that are contained in this data package are (1) Tektite & Yawzi and (2) Random sites. The juvenile coral density is packaged separately. Tektite – this is at 14 m depth on the eastern side of Great Lameshur Bay and is the original site of the Tektite man-in-the sea project in 1969; this project marked the birth of the Virgin Islands Ecological Research Station (later the Virgin Islands Environmental Resource Station) that hosts the field component of the project. The reef in this location consists of a single buttress that has remained dominated by Montastraea anularis since the start of the research (1987). These surveys consist of 30 photoquadrats (1 x 1 m) distributed along three, 10 m transects. Yawzi – this is at 9 m depth and is on the western side of Great Lameshur Bay and has been recorded photographically since 1987. This reef also started the study period dominated by Montastraea annularis, but has degraded much more rapidly that the Tektite site. These surveys consist of 30 photoquadrats (1 x 1 m) distributed along three, 10 m transects. Random sites – were added in 1992 to address the concern that the original sites (Yawzi and Tektite) were selected on “good” areas of reef and, therefore, could only decline in condition. The Random sites were selected using random coordinates in 1992, and consist of 6 sites (at 7-9 m depth) scattered between Cabritte Point and White Point. All lie a little shoreward of Yawzi and Tektite, and have always been characterized by low coral cover (< 10% cover). The surveys consist of 18-40 photoquadrats (0.5 x 0.5 m; with sample size determined by the exposures on a 35 mm cassette versus digital techniques) that are placed at random points along a transect. This project focuses on the long-term community dynamics of shallow coral reefs on the south shore of St. John. The project began in 1987 and consists of permanently marked areas that are photographed annually for the analysis of benthic community structure and areas that are censused annually for the density of juvenile corals. All of the work takes place on reefs at less than 14 m depth, and virtually all of the study sites fall between Cabritte Point to the east, and White Point to the west. The project is independent of, but complimentary to, the larger-scale reef monitoring program conducted by the Virgin Islands National Park. The project described here is an independent academic effort led by Dr. Peter Edmunds from California State University, Northridge. The objective of this study is to describe coral reef community dynamics in great detail from a small area to first, elucidate the patterns of change in key components of the benthic coral reefs, and second, to test for processes that drive the changes observed. The core of the project relies on high resolution digital images that have been recorded at least annually since December 1987, and are available here as a invaluable photographic archive for further analysis and to corroborate existing numerical descriptions. In addition to the photographs, the density of juvenile corals has been recorded at multiple sites since 1994.

These data describe the abundance of settler-stage three-spot dascyllus (Dascyllus trimaculatus), a planktivorous damselfish to their juvenile microhabitat, the sea anemone Heteractis magnifica. Surveys were begun in 1993, and, except for 1994, have been conducted daily each year between June and September on a reef adjacent to the UC Berkeley Gump Research Station in Cooks Bay on the north shore of the island of Moorea in French Polynesia. This time series completed in 2012.

This data package is formatted as an ecocomDP (Ecological Community Data Pattern). For more information on ecocomDP see https://github.com/EDIorg/ecocomDP. This Level 1 data package was derived from the Level 0 data package found here: https://pasta.lternet.edu/package/metadata/eml/knb-lter-mcr/1038/10. The abstract below was extracted from the Level 0 data package and is included for context: This dataset contains the percentage cover of all stony corals (Scleractinia, pooled among genera) and other major groups analyzed from 0.5 x 0.5 m photographic quadrats at the Backreef habitat at the Moorea Coral Reef LTER, French Polynesia. This survey time series began in 2005 and is repeated each year in April. Functional groups counted are: Scleractinian corals, Macroalgae, Crustose Coralline Algae / Bare Space, Soft Corals, Hydrocorals (Millepora), Algal Turf and Sand. The coral community was sampled photographically in all represented habitats surrounding the island: Fringing Reef, Lagoon, and Outer Reef. This dataset contains only Lagoon (Backreef) data (see knb-lter-mcr.4 for the other habitats) and is structured in a repeated-measures protocol to allow a statistical contrast of sites, shores and times. Community structure was determined through a coarse analysis of the benthic community, initially completed in situ (2005), but using photoquadrats from 2006. There are quadrats analyzed at each of five areas within each site, and the areas are revisited (but not the quadrats) each year to support the repeated measures design. There are two tables available, providing different views of the same data: a long table having all values in one column and a wide table having a separate column for each observed object. Detailed methods are available in the protocols section. This material is based upon work supported by the U.S. National Science Foundation under Grant No. OCE 16-37396 (and earlier awards) as well as a generous gift from the Gordon and Betty Moore Foundation. Research was completed under permits issued by the French Polynesian Government (Délégation à la Recherche) and the Haut-commissariat de la République en Polynésie Francaise (DTRT) (Protocole d'Accueil 2005-2018). This work represents a contribution of the Moorea Coral Reef (MCR) LTER Site.

In this study, coral recruitment was measured on a kilometer-wide scale on shallow (5–6 m depth) fringing reefs in St. John, US Virgin Islands, with the objective of determining the extent to which variation in recruitment was affected by biophysical coupling involving temperature and flow. Coral recruitment was measured using settlement tiles deployed at 10 sites along 10 km of shore. The tiles were first deployed in August 2006, and thereafter replaced every ≈6 months to sample from either August to January, or January to August over 2 years. Seawater temperature was recorded at the 10 sites using logging thermistors, and flow was quantified using drogues. Overall, corals recruited at a rate equivalent to 76 corals m− 2 6 months− 1, and were represented mostly by poritids (43% of recruits), agaricids (29%), faviids (17%) and siderastreids (7%). Although the density of recruits differed among sites in a pattern that varied among periods and years, there was a consistent trend for mean density to decline from ≈ 4 corals tile− 1 at eastern sites, to ≤ 1 coral tile− 1 at western sites. One aspect of seawater temperature – the daily range – differed among sites and was greater at western compared to eastern sites, and while it was related inversely to recruitment over one of the sampling periods, it was equivocal as a physical process affecting recruitment. Instead, our results are consistent with biophysical coupling involving patch depletion and downstream filtering, whereby patches of coral larvae are delivered to the south shore of St. John and depleted of larvae through settlement as the water progresses westward.

This is a reference dataset containing the site locations, year established, and types of surveys.

This data package is formatted as a Darwin Core Archive (DwC-A, event core). For more information on Darwin Core see https://www.tdwg.org/standards/dwc/. This Level 2 data package was derived from the Level 1 data package found here: https://pasta.lternet.edu/package/metadata/eml/edi/357/2, which was derived from the Level 0 data package found here: https://pasta.lternet.edu/package/metadata/eml/edi/291/2. The abstract below was extracted from the Level 0 data package and is included for context: These data are evidence of the the long-term dynamics of shallow coral reefs along the south coast of St. John from as early as 1987. These data describe coral reef community structure as percent cover based on the analysis of color photographs. All of these data originate from color images of photoquadrats recorded annually (usually in the summer) from as early as 1987. The data falls into three groups. The two groups that are contained in this data package are (1) Tektite & Yawzi and (2) Random sites. The juvenile coral density is packaged separately. Tektite – this is at 14 m depth on the eastern side of Great Lameshur Bay and is the original site of the Tektite man-in-the sea project in 1969; this project marked the birth of the Virgin Islands Ecological Research Station (later the Virgin Islands Environmental Resource Station) that hosts the field component of the project. The reef in this location consists of a single buttress that has remained dominated by Montastraea anularis since the start of the research (1987). These surveys consist of 30 photoquadrats (1 x 1 m) distributed along three, 10 m transects. Yawzi – this is at 9 m depth and is on the western side of Great Lameshur Bay and has been recorded photographically since 1987. This reef also started the study period dominated by Montastraea annularis, but has degraded much more rapidly that the Tektite site. These surveys consist of 30 photoquadrats (1 x 1 m) distributed along three, 10 m transects. Random sites – were added in 1992 to address the concern that the original sites (Yawzi and Tektite) were selected on “good” areas of reef and, therefore, could only decline in condition. The Random sites were selected using random coordinates in 1992, and consist of 6 sites (at 7-9 m depth) scattered between Cabritte Point and White Point. All lie a little shoreward of Yawzi and Tektite, and have always been characterized by low coral cover (< 10% cover). The surveys consist of 18-40 photoquadrats (0.5 x 0.5 m; with sample size determined by the exposures on a 35 mm cassette versus digital techniques) that are placed at random points along a transect. This project focuses on the long-term community dynamics of shallow coral reefs on the south shore of St. John. The project began in 1987 and consists of permanently marked areas that are photographed annually for the analysis of benthic community structure and areas that are censused annually for the density of juvenile corals. All of the work takes place on reefs at less than 14 m depth, and virtually all of the study sites fall between Cabritte Point to the east, and White Point to the west. The project is independent of, but complimentary to, the larger-scale reef monitoring program conducted by the Virgin Islands National Park. The project described here is an independent academic effort led by Dr. Peter Edmunds from California State University, Northridge. The objective of this study is to describe coral reef community dynamics in great detail from a small area to first, elucidate the patterns of change in key components of the benthic coral reefs, and second, to test for processes that drive the changes observed. The core of the project relies on high resolution digital images that have been recorded at least annually since December 1987, and are available here as a invaluable photographic archive for further analysis and to corroborate existing numerical descriptions. In addition to the photographs, the density of juvenile corals has been recorded at multiple sites since 1994.

This data package is formatted as an ecocomDP (Ecological Community Data Pattern). For more information on ecocomDP see https://github.com/EDIorg/ecocomDP. This Level 1 data package was derived from the Level 0 data package found here: https://pasta.lternet.edu/package/metadata/eml/knb-lter-mcr/4/38. The abstract below was extracted from the Level 0 data package and is included for context: This dataset contains the percentage cover of the stony corals (Scleractinia) and other major groups analyzed from 0.5 x 0.5 m photographic quadrats in several reef habitats at the Moorea Coral Reef LTER, French Polynesia. This survey has been repeated annually in April since 2005. There are two tables available, providing different views of the same data: a long table having all values in one column and a wide table having a separate column for each dependent variable. Functional groups (i.e., dependent variables) counted are: Scleractinian Corals (by genus where appropriate, see methods), Macroalgae, Crustose Coralline Algae / Bare Space, Soft Corals, Hydrocorals (Millepora), Algal Turf and Sand. The coral community was sampled photographically in all habitats surrounding the island: Fringing Reef, Lagoon (Backreef), and Outer Reef (Forereef.) The sampling regime consists of a repeated-measures protocol in each habitat, and is structured by habitat to allow a statistical contrast of sites, shores, times, and in the case of the outer reef, depths. Detailed methods are available in the protocols section. This material is based upon work supported by the U.S. National Science Foundation under Grant No. OCE 16-37396 (and earlier awards) as well as a generous gift from the Gordon and Betty Moore Foundation. Research was completed under permits issued by the French Polynesian Government (Délégation à la Recherche) and the Haut-commissariat de la République en Polynésie Francaise (DTRT) (Protocole d'Accueil 2005-2020). This work represents a contribution of the Moorea Coral Reef (MCR) LTER Site.

These data describe the abundance of adult-stage three-spot dascyllus (Dascyllus trimaculatus), a planktivorous damselfish, and their juvenile microhabitat, the sea anemone Heteractis magnifica. Surveys were begun in 1992 and have been conducted two to four times annually by the same observer in the lagoons of the north shore of the island of Moorea in French Polynesia.

The data presented here are the abundances of the major invertebrate herbivores and corallivores on Moorea coral reefs. Abundances are estimated in 4 fixed quadrats along 5 permanent transects at each of 4 habitats at 2 sites on each of the 3 shores of Moorea each year. Counts are made in one-meter-squared quadrats. This material uses data collected by the U.S. National Science Foundation's (NSF) Moorea Coral Reef Long Term Ecological Research (MCR LTER) site under Grant No. OCE 2224354 (and earlier awards). Additional financial support to the MCR LTER site was provided through a generous gift from the Gordon and Betty Moore Foundation. Research was completed under permits issued by the French Polynesian Government (Délégation à la Recherche) and the Haut-commissariat de la République en Polynésie Francaise (DTRT) (Protocole d'Accueil 2005-2023).