Skip to main content
Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

A contemporary baseline of Madagascar’s coral assemblages: Reefs with high coral diversity, abundance, and function associated with marine protected areas

  • Mahery Randrianarivo,

    Roles Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Writing – original draft

    Affiliations Institut Halieutique et des Sciences Marines, Université de Toliara, Toliara, Madagascar, ENTROPIE, Université de La Réunion, IRD, CNRS, IFREMER, Université de la Nouvelle-Calédonie, La Réunion, France

  • François Guilhaumon,

    Roles Formal analysis, Writing – review & editing

    Affiliation ENTROPIE, Université de La Réunion, IRD, CNRS, IFREMER, Université de la Nouvelle-Calédonie, La Réunion, France

  • Johanès Tsilavonarivo,

    Roles Data curation, Investigation, Writing – review & editing

    Affiliation Institut Halieutique et des Sciences Marines, Université de Toliara, Toliara, Madagascar

  • Andriamanjato Razakandrainy,

    Roles Data curation, Investigation, Writing – review & editing

    Affiliation Institut Halieutique et des Sciences Marines, Université de Toliara, Toliara, Madagascar

  • Jacques Philippe,

    Roles Data curation, Investigation, Writing – review & editing

    Affiliation Centre National de Recherches Océanographiques, Nosy-Be, Madagascar

  • Radonirina Lebely Botosoamananto,

    Roles Conceptualization, Writing – review & editing

    Affiliations Institut Halieutique et des Sciences Marines, Université de Toliara, Toliara, Madagascar, ENTROPIE, Université de La Réunion, IRD, CNRS, IFREMER, Université de la Nouvelle-Calédonie, La Réunion, France

  • Lucie Penin,

    Roles Conceptualization, Supervision, Writing – review & editing

    Affiliations ENTROPIE, Université de La Réunion, IRD, CNRS, IFREMER, Université de la Nouvelle-Calédonie, La Réunion, France, Laboratoire d’Excellence “CORAIL”, Paris, France

  • Gildas Todinanahary,

    Roles Conceptualization, Supervision, Writing – review & editing

    Affiliation Institut Halieutique et des Sciences Marines, Université de Toliara, Toliara, Madagascar

  • Mehdi Adjeroud

    Roles Conceptualization, Funding acquisition, Methodology, Resources, Supervision, Writing – original draft

    mehdi.adjeroud@ird.fr

    Affiliations Laboratoire d’Excellence “CORAIL”, Paris, France, ENTROPIE, Université de La Réunion, IRD, CNRS, IFREMER, Université de la Nouvelle-Calédonie, Perpignan, France, UAR 3278 CRIOBE EPHE-UPVD-CNRS, PSL Université Paris, Perpignan, France

Abstract

Madagascar is a major hotspot of biodiversity in the Western Indian Ocean, but, as in many other regions, coral reefs surrounding the island confront large-scale disturbances and human-induced local stressors. Conservation actions have been implemented with encouraging results for fisheries, though their benefit on coral assemblages has never been rigorously addressed. In this context, we analyzed the multiscale spatial variation of the composition, generic richness, abundance, life history strategies, and cover of coral assemblages among 18 stations placed at three regions around the island. The potential influences of marine protected areas (MPAs), algal cover, substrate rugosity, herbivorous fish biomass, and geographic location were also analyzed. Our results highlight the marked spatial variability, with variation at either or both regional and local scales for all coral descriptors. The northeast coastal region of Masoala was characterized by the high abundance of coral colonies, most notably of the competitive Acropora and Pocillopora genera and stress-tolerant taxa at several stations. The southwest station of Salary Nord was distinguished by lower abundances, with depauperate populations of competitive taxa. On the northwest coast, Nosy-Be was characterized by higher diversity and abundance as well as by high coral cover (~42–70%) recorded at unfished stations. Results clearly underline the positive effects of MPAs on all but one of the coral descriptors, particularly at Nosy-Be where the highest contrast between fished and unfished stations was observed. Biomass of herbivorous fishes, crustose coralline algae cover, and substrate rugosity were also positively related to several coral descriptors. The occurrence of reefs with high diversity, abundance, and cover of corals, including the competitive Acropora, is a major finding of this study. Our results strongly support the implementation of locally managed marine areas with strong involvement by primary users, particularly to assist in management in countries with reduced logistic and human resources such as Madagascar.

Introduction

More than 850 million people from over 100 countries rely on the exceptional biodiversity of coral reefs, providing crucial economic, cultural, social, and aesthetic goods and services [1, 2]. These reefs are mostly supported by small colonial, calcifying organisms, the hermatypic scleractinian corals, which create complex three-dimensional habitats offering a variety of shelter and food for thousands of organisms [3, 4]. Coral assemblages often exhibit a marked spatial heterogeneity from local to geographic scales (i.e., from within and among reef habitats to among regions and ocean basins), reflecting the contrasting life history strategies and functional traits of coral species as well as the biological and physical processes that influence their biology and vary in frequency, intensity, and spatial scale [59]. Since corals are particularly sensitive to changes in environmental conditions, reef ecosystems are highly vulnerable to both chronic and acute stressors and may change rapidly in their structure and functioning [1013].

There is concern that the frequency and severity of large-scale disturbances, such as coral bleaching events associated to thermal stress, cyclones, or outbreaks of the coral predator Acanthaster spp. in the Indo-Pacific, have increased over the last four decades [4, 14, 15]. Humans have further contributed to this declining trend by contributing multiple, direct anthropogenic disturbances that kill reef organisms, such as overharvesting of reef organisms, destructive fishing methods, uncontrolled tourism and recreational impacts, and increased sedimentation and pollution associated with dredging, coastal development, deforestation, and intensive agriculture [4, 16, 17]. In response to these threats, coral reefs have been severely impacted by widespread mortalities of foundation species and degraded habitat [1820], in a number of cases undergoing a striking phase shift involving the replacement of corals by macroalgae or other non-reef-building benthic organisms, an undesirable state providing fewer goods and services to human populations [21].

In the context of this “coral reef crisis”, evaluating the vulnerability, adaptability, and resilience of reef communities and their dependent human societies is urgently needed [4, 22, 23]. Encouragingly, management and conservation actions have already been taken to preserve the diversity, structure, and functions of coral reefs and to maintain the goods and services they provide to humans [2426]. The Marine Protected Area (MPA) has been one of the most common tools to support the health of coral reefs and to enhance their resistance and resilience to disturbances [2731]. Although the capacity of MPAs to maintain fisheries productivity and sustainable livelihoods has been demonstrated in many cases [3236] conflicting results and uncertainties regarding their effectiveness in protecting the transformed reefs that are expected in future decades have also been highlighted [6, 26, 32, 3740]. For corals, management of fishing activities within MPAs is intended to maintain a sufficient herbivory rate to remove algae that are competitive with corals, in turn enhancing coral recruitment [31, 4143]. However, this mechanism is not always observed, and a lack of positive effects of MPAs on coral replenishment capacities has been recorded [44, 45]. Among the major drivers of MPA effectiveness are the duration of protection, size, and connectivity, together with the type and compliance of regulation measures [29, 32, 4649].

Madagascar is one of the largest islands in the world, with a coastline extending over ~14° of latitude (11° 47’ to 25° 35’ S) and, with ~2400 km2 of coral reefs, is a major biodiversity hotspot in the Western Indian Ocean [5052]. The high diversity of reef organisms, including ~380 scleractinian coral species, is partly linked to the size, morphological diversity, and contrasting environmental conditions of these reefs [6, 5357]. Studies on the diversity and structure of coral assemblages have taken place since the 1960s with the establishment of the marine research center at Toliara on the southwest coast [58]. However, most of these studies have been restricted to either local [5965] or regional [38, 55] scales, or have focused on documenting the general health status of these reefs [53, 55, 6668]. In contrast, no multiscale analysis of spatial patterns in coral assemblages has been conducted, and the drivers of such variability remain poorly understood.

As most of the world’s coral reefs, those of Madagascar have confronted large-scale disturbances and human-induced local stressors, most notably bleaching events, sedimentation, and overfishing [56, 67, 6971]. Consequently, coral assemblages have declined since the 1980s at several reefs around the island, as documented along the Great Reef of Toliara [6163], though a return to a healthier coral community has recently been documented for this reef [65]. In response to the overall degradation of coral reefs, the Malagasy authorities have implemented mitigation actions, resolutely investing in the establishment of MPAs since 1989, with the creation of the first one, the Mananara Nord Biosphere Reserve on the east coast. Twenty-two MPAs currently cover an area of 14,451 km², representing 1.26% of the Exclusive Economic Zone [72, 73]. Involvement of primary users, most notably fishermen and sea farmers, have been encouraged through the implantation of Locally Managed Marine Areas (LMMA) that have shown some success in effectively managing fisheries [35, 74, 75]. However, the effects of these MPAs on coral assemblages have not been rigorously examined, leaving the benefit and effects of such a management tool on the overall reef biodiversity, functions, and services difficult to assess in Madagascar [35, 75].

In this context, the aim of this study was to analyze the spatial variation of coral assemblages at multiple scales around Madagascar. Composition, generic richness, abundance, life history strategies, and cover of coral assemblages were compared among 18 stations located at three regions around the island. We analyzed the potential influence of several environmental factors, including algal cover, substrate rugosity, herbivorous fish biomass, and geographic location on the spatial patterns of coral variables. By comparing stations in unfished areas (MPAs) and stations where fishing is allowed, we also investigated the effect of fishing protection level on coral community structure. Although this study is a snapshot in a highly dynamic system, the data set and questions addressed in this study provide a valuable contemporary baseline to understand the drivers of coral community structure and to monitor future changes of these reefs, and may also help identify effective conservation and management actions.

Materials and methods

Study area and sampling strategy

Fieldwork was conducted from March to October 2020 at three regions around Madagascar (Fig 1). Masoala, on the northeast coast (15°59’08” S, 50°09’27” E), is composed of fringing reefs of 0.5 to 3 km wide with several passes to the open ocean. This region includes the Masoala Marine Park, created in March 1997 (S1 Table). Nosy-Be, in the northwest (13°24’21” S, 48°16’31” E), is composed by several small islands with narrow (< 1 km) fringing reefs. This region comprises the Lokobe Marine Park and the Nosy Tanihely Marine Park, both created in September 2011. Salary Nord, along the southwest coast (22°33’17” S, 43°17’10” E), is a region with fringing reefs of 1.7 to 4.9 km wide, with a channel (~5 m depth) that allows the circulation of boats. These reefs are integrated within the Soariaka MPA, created in April 2015. In recent decades, these reefs have been impacted by the bleaching event of 1998 [70, 76], and 2016 [77] (S1 Table). The major cyclones that had a potential impact on coral assemblages are Haruna at Salary Nord in 2013, Enawo 2017 at Masoala and Nosy-Be in 2017, and Belan at Nosy-Be in 2019 [78]. However, the impact of these disturbances on coral assemblages in our three regions has not been quantified. At each region, six sampling stations were surveyed, including three stations in unfished areas and three stations in areas where fishing is allowed (Fig 1 and S1 Table). The location of our stations took into account logistical considerations (accessibility of sites), meteorological constraints during sampling, and the directives of local authorities (Madagascar National Park, Wildlife Conservation Society, and local managers). All 18 stations were located on the outer reef slope at ~10 m depth, where direct anthropogenic disturbances are lower, and where diversity and abundance of coral assemblages is generally higher compared to other reef habitats [65]. Stations codes are abbreviated as follow: the first two letters indicate the region (NE for Masoala in the northeast, NW for Nosy-Be in the northwest, and SW for Salary Nord in the southwest), the number (1 to 6) differentiates the six stations from each region, and NTZ (“No Take Zone”) is for stations in unfished areas. Field work was carried out with a research permit granted by the Malagasy Ministry of Environment and Sustainable Development (75/20/MEDD/SG/DGEF/DGNRE).

thumbnail
Fig 1. Location of the 18 sampling stations in the three regions around Madagascar.

Masoala (A) is located in the northeast coast, Nosy-Be (B) in the northwest, and Salary Nord (C) in the southwest. See Materials and Methods section for station codes.

https://doi.org/10.1371/journal.pone.0275017.g001

At each station, generic richness (GR) and abundance of coral assemblages (scleractinian corals and the calcareous hydrocoral Millepora) were estimated using three randomly replicated belt-transects of 10 m2 (10 × 1 m), laid parallel to depth contours and separated by ~2 m [79]. To complement the characterization of the diversity of coral assemblage composition, the Shannon diversity index (H’) was calculated at each station using log2 and colony abundance data for each coral genus. In order to estimate the functional trait diversity of coral assemblages, we assigned one of the four life history strategies, as defined by Darling et al. [80], to each coral colony encountered in the belt-transects: competitive, generalist, stress-tolerant, and weedy. For those coral genera that encompass species spanning several of these life history strategies, we have assigned proportional values based on life history strategy of the species typically found in the region ([81]; for example, Pocillopora was estimated to include 75% generalist and 25% weedy species). The percent cover of corals, macroalgae, turf algae, and crustose coralline algae (CCA) was estimated at each station using three line intercept transects (LIT) of 10 m, placed in the middle of the belt-transects. Biomass of herbivorous fishes was estimated at each station using underwater visual censuses on three randomly replicated belt-transects of 250 m² (50 × 5 m) laid parallel to depth contours and separated by ~2 m. All herbivorous fishes were identified at the species level, counted, and measured according to their total length. Abundance values were converted to kilograms of biomass per unit area of reef (kg.ha-1) using species-specific length-weight equations: W = a⋅Lb, where W is the weight (in g), L is the total length (in cm), and parameters a and b are species-specific constants that have been extracted from FishBase [82]. At each station, we estimated the substrate rugosity with a visual assessment of reef topography, graded from 0 to 5 (0 = no vertical relief; 1 = low and sparse relief; 2 = low but widespread relief; 3 = moderately complex; 4 = very complex with numerous fissures and caves; 5 = exceptionally complex with numerous caves and overhangs), following Polunin & Roberts [83].

Cover of algae (macroalgae, turf, and CCA), rugosity, herbivorous fish biomass, fishing protection level (fished vs. unfished areas; i.e., we did not measure fishing pressure at each station), and geographic location (belonging to one of the three regions) have been recorded to estimate their potential influence on the spatial variation of coral assemblages, and will be referred to as “explanatory variables” hereafter.

Data analysis

The spatial variation in the composition of coral assemblages was analyzed using nonmetric multidimensional scaling (nMDS), based on the Bray-Curtis dissimilarity index of the abundance of coral genera recorded at each station. The nMDS was performed using the “metaMDS” function in the “vegan” R package [84]. The potential influence of explanatory variables was analyzed with PERMANOVA on distance matrices using the “Adonis” function in the “vegan” R package [85]. Additionally, pairwise post-hoc multilevel comparisons were conducted to test for significant differences in the composition of coral assemblages between the three regions, using the “pairwise.Adonis2” function in the “pairwiseAdonis” R package [86].

We used linear mixed effect models to determine which explanatory variables were influencing the spatial variation of the coral assemblage descriptors (GR, H’, total abundance, abundance of the four life history strategies, and overall coral cover). Prior to analysis, Spearman rank correlations were calculated between all pairs of explanatory variables (cover of macroalgae, turf algae, CCA, herbivorous fish biomass, and rugosity) to identify potential collinearity between these variables. As no strong correlations (-0.7 < overall ρ < 0.7) were recorded, all our explanatory variables were considered as independent and included in the models (S1 Fig). All explanatory variables were centered to a mean of zero and scaled to a standard deviation of 1 in order to allow for direct comparisons of their effects. Linear mixed effect models were selected as they take into account the hierarchical structure of the data set: here, the non-independence of data points (transects) belonging to particular stations, in turn belonging to the same region [87]. We used a multi-model information-theoretic approach to compare the explanatory variable contributions and their most important interactions in describing the variation in coral descriptors. To ensure that all model assumptions were met and to take into account overdispersion, we modeled GR, H’, abundance and cover data using a Gaussian distribution, while abundance of the four life history strategies were modeled using a Poisson distribution. Models were fitted by maximum likelihood estimation using the “lmer” and “glmer” functions in the “lme4” R package [88]. We estimated how much of the variation in coral descriptors was explained by the variables included in the models using the marginal R² (“r.squaredGLMM” function in the “MuMIn” R package; [89]). For each coral variable, we selected the most parsimonious combinations of fixed effects by comparing models with all possible combinations of the predictor variables using the corrected Akaike’s information criterion (AICc; [90]). The subset of best models was selected as the ones with ΔAICc value < 2 (the difference between each model’s AICc and the lowest AICc) using the “dredge” function in the “MuMIn” R package. All models with a ΔAICc < 2 were considered as having similar levels of support from the data, thus belonging to the group of best models (i.e., equally parsimonious; [90]). We used Akaike weights (wAICc), derived from the AICc, to evaluate the relative likelihood of each model given the data set and the set of models considered as well as to estimate the relative importance of each explanatory variable by summing these wAICc across the models in which they were included. Akaike weights were directly interpreted as each model’s probability of being the best at explaining the data [90]. Subsequently, the effect sizes (lm coefficients) of the predictions of the individual models selected at the previous step were averaged using the Akaike weights as weighting coefficients. Finally, we investigated the differences among factor modalities in the mixed models using Tukey post-hoc tests (“emmeans” function in the “emmeans” R package; [91]). All statistical analyses were performed using R 4.1.0 [92].

Results

Composition of coral assemblages

A total of 7,638 coral colonies, representing 16 families and 40 genera (30 at Masoala, 38 at Nosy-Be, and 29 at Salary Nord), were recorded at the 18 stations. The nMDS ordination showed a substantial geographical clustering in the composition of coral assemblages (Fig 2). Among the seven potential explanatory variables examined (Fig 3), geographic location (belonging to one of the three regions) was the strongest contributor to the overall dissimilarity in the composition of coral assemblages, explaining ~27% of the variance (PERMANOVA, F = 3.825, p = 0.001; R² = 0.27) and followed by rugosity, though to a lesser degree (PERMANOVA, F = 2.444, p = 0.021; R² = 0.08; Table 1). In fact, the differences in the composition of coral assemblages between regions were all significant (pairwise post-hoc tests, Masoala × Nosy-Be, F = 2.384, p = 0.013; Masoala × Salary Nord, F = 2.817, p = 0.029; Nosy-Be × Salary Nord, F = 3.380, p = 0.006). The first two axes of the nMDS differentiated the stations of Masoala, characterized by higher abundances of Acropora, Pavona, Hydnophora, and Leptastrea, from the stations of the two other regions. An exception is station NE2-NTZ, where lower abundance of Acropora was recorded and which was more similar to Salary Nord stations (S2 Table). Most Nosy-Be stations were located in the upper part of the nMDS plot, and were characterized by the presence of Diploastrea, Euphyllia, Isopora, and Tubastrea that were recorded only in this region (S2 Table). Station NW4 at Nosy-Be was however more similar in its composition to several Masoala and Salary Nord stations. Stations from Salary Nord, in the right portion of the nMDS plot, were characterized by higher abundance of Leptoseris, Pachyseris, and Stylophora at several stations (S2 Table).

thumbnail
Fig 2. Non-metric multidimensional scaling (nMDS) showing the spatial variation in the composition and abundance of coral assemblages.

Analyses are based on the Bray-Curtis dissimilarity index among the 18 stations located in the three regions. Position of stations and coral genera is given along the first two axes.

https://doi.org/10.1371/journal.pone.0275017.g002

thumbnail
Fig 3. Spatial variation of explanatory variables that may influence the spatial variation of coral assemblages.

Data are given for the 18 sampling stations located at the three regions. Errors bars represent standard error.

https://doi.org/10.1371/journal.pone.0275017.g003

thumbnail
Table 1. Summary of the PERMANOVA to examine the influence of explanatory variables in the variation of composition of coral assemblages.

Analyses are based on the Bray-Curtis dissimilarity index of the abundance of coral genera recorded at the 18 stations of the three regions. Significant P-values (<0.05) are highlighted in bold.

https://doi.org/10.1371/journal.pone.0275017.t001

Generic richness

Generic richness was highly variable among stations within the regions, but the range of variation was similar among the three regions, with values from 10.00 ± 1.00 genera.10 m-2 (mean ± se) to 21.66 ± 0.33 at Masoala, from 10.66 ± 1.20 to 19.00 ± 2.08 at Nosy-Be, and from 11.66 ± 1.76 to 20.66 ± 0.66 at Salary Nord (Fig 4 and S3 Table). The selection procedure identified four models to explain the variation in GR (ΔAIC < 2; Tables 2 and S4 and S5 and S2 and S6 Figs), with fishing protection level and herbivorous fish biomass having the strongest contribution. The biomass of herbivorous fish was included in all four models and had the maximum value of relative importance (1.000). Models indicated that the number of coral genera per 10 m2 increases by 1.33-fold with an increase of 1 kg.ha-1 in herbivorous fish biomass. Fishing protection level also had a strong influence on the variation of GR (relative importance: 0.736), with 2.55 times more coral genera per 10 m2 in unfished stations compared to fished ones. Cover of CCA and turf algae weakly predicted the variation in GR (relative importance: 0.191 and 0.154, respectively), with a positive coefficient for CCA (0.39) and a negative one for turf algae (-0.28). All other explanatory variables were not selected by the models to account for the variation in GR (Table 2).

thumbnail
Fig 4. Spatial variation in generic richness, Shannon diversity index, abundance, and percent cover of coral assemblages.

Data are given for the 18 sampling stations located at the three regions. Errors bars represent standard error.

https://doi.org/10.1371/journal.pone.0275017.g004

thumbnail
Table 2. Summary of best linear mixed models using a Gaussian distribution to describe variation of generic richness, Shannon diversity index, coral abundance, and coral cover.

FP: fishing protection level, MA: macroalgae, CCA: crustose coralline algae, HFB: herbivorous fish biomass.

https://doi.org/10.1371/journal.pone.0275017.t002

Shannon diversity index

The Shannon diversity index was also variable among stations within regions, particularly at Masoala, with H’ values from 1.54 ± 0.03 (mean ± se) to 2.85 ± 0.00, compared to Nosy-Be (1.89 ± 0.06 to 2.49 ± 0.06) and Salary Nord (1.98 ± 0.08 to 2.76 ± 0.02; Fig 4 and S6 Table). Five models were selected as the most parsimonious to explain variation of H’ (Tables 2 and S7 and S8 and S3 and S7 Figs). CCA cover was the most important predictor (relative importance: 1.000), followed by herbivorous fish biomass (0.581) and fishing protection level (0.424), with all three variables having a positive contribution (coefficient of 0.08, 0.07, and 0.15, respectively). Rugosity was also selected in some models, but its contribution to the spatial variation in H’ was lower (relative importance: 0.118; coefficient: 0.03).

Coral abundance

Abundance of coral colonies was highly variable among stations within region, notably for Masoala, with 81.70 ± 8.09 to 345.66 ± 25.01 colonies.10 m-2 (mean ± se), compared to Nosy-Be (96.70 ± 2.03 to 205.33 ± 31.40 colonies.10 m-2) and Salary Nord (75.00 ± 7.77 to 144.33 ± 7.26 colonies.10 m-2; Fig 4 and S9 Table). Three models were selected to explain the variation in coral abundance (Tables 2 and S10 and S11 and S4 and S8 Figs). Herbivorous fish biomass had the highest relative importance (1.000), followed to a lesser degree by fishing protection level (0.250) and CCA cover (0.205). The effect of herbivorous fish biomass on coral abundance was strong and positive, with a coral abundance increase by 26.81-fold with an increase of 1 kg.ha-1 in herbivorous fish biomass. The models also suggested that 30.68 times more coral colonies were recorded in unfished stations, whereas the effect of CCA cover on variation in coral abundance was negative (coefficient: -3.35).

The abundance of the 10 major coral genera at each region, as well as other, less represented genera (“others”), was highly variable among stations and regions (Fig 5). Acropora was the most abundant coral at eight of the 18 stations, with particularly high values at NE1-NTZ and NE4 located at Masoala and, to a lesser degree, at the unfished stations of Nosy-Be (NW1-NTZ, NW2-NTZ, NW3-NTZ). Several genera were frequently recorded at certain regions while greatly less abundant in others. This is the case for Galaxea, which was abundant at Masoala and Nosy-Be stations but less represented at Salary Nord, and Pocillopora, which was abundant at Masoala and Salary Nord but not at Nosy-Be. Porites and Seriatopora were recorded at almost all stations, being among the dominant corals at several stations (NW6 and SW2-NTZ for Porites, NW5, SW1-NTZ, and SW5 for Seriatopora). Several genera were part of the 10 dominant corals at only one of the three regions, as Favia at Masoala, Leptoria, Fungia and Merulina at Nosy-Be, and Leptoseris and Stylophora at Salary Nord (Fig 5).

thumbnail
Fig 5. Spatial variation in the abundance of the 10 major coral genera and other less represented genera (“others”).

Data are given for 18 sampling stations located at the three regions. Errors bars represent standard error.

https://doi.org/10.1371/journal.pone.0275017.g005

Coral assemblages were dominated by stress-tolerant taxa at 11 of the 18 stations, whereas competitive taxa were the most represented life history strategy at NE1-NTZ and NE4 in Masoala and, to a lesser degree, at NW1-NTZ in Nosy-Be (Fig 6 and S12 and S13 Tables). Generalist and weedy taxa were generally less abundant except at stations NE5 in Masoala and SW1-NTZ and SW2-NTZ in Salary Nord. Five models were selected as the most parsimonious to explain the variation in the abundance of the four life history strategies (Tables 3 and S14 and S15 and S9 Fig). Region, fishing protection level, life history strategies, herbivorous fish biomass, and the interactions of region × fishing protection level, region × life history strategies, fishing protection level × life history strategies, and region × fishing protection level × life history strategies had the maximum value of relative importance (1.000). Rugosity (0.322), macroalgae cover (0.526), turf (0.169), and CCA (0.169) were also selected by the models, but their relative importance was lower. The models underlined the higher abundance of competitive corals at some Masoala stations, the higher abundance of competitive corals in unfished stations at Nosy-Be, and the higher abundance of weedy corals in unfished stations at Salary Nord.

thumbnail
Fig 6. Spatial variation in the abundance of the four coral life history strategies.

Data are given for the 18 sampling stations located at the three regions. Errors bars represent standard error.

https://doi.org/10.1371/journal.pone.0275017.g006

thumbnail
Table 3. Summary of the best linear mixed models using a Poisson distribution to describe variation in abundance of coral life history strategies.

FP: fishing protection level, LHS: life history strategies, MA: macroalgae, CCA: crustose coralline algae, HFB: herbivorous fish biomass.

https://doi.org/10.1371/journal.pone.0275017.t003

Coral cover

The overall coral cover was highly variable within and among regions. Higher cover values were recorded at Nosy-Be (from 42.6% ± 2.9 to 69.8% ± 4.2, mean ± se), compared to Salary Nord (from 25.8% ± 1.3 to 49.7% ± 3.0) and Masoala (from 26.5% ± 0.5 to 46.2% ± 5.6; Fig 4 and S16 Table). The spatial variation in coral cover was supported by seven models (Tables 2 and S17 and S18 and S5 and S10 Figs). Fishing protection level and rugosity had the maximum relative importance (1.000), followed by region (0.603). CCA, turf, and the interaction region × fishing protection level had a lower relative importance (0.474, 0.200 and 0.094, respectively). The models suggested that coral cover increases by 7.04% with an increase of rugosity by 1 unit, and by 8.04% in unfished areas compared to fished ones, this effect being more pronounced at Nosy-Be (12.16% increase; Table 2).

Discussion

In the context of increasing degradation of coral communities around the globe, as well as the localized threats that Madagascar is facing, our results demonstrate the occurrence of reefs with high diversity, abundance, and cover of corals, including the competitive Acropora, which has decreased among other coral reefs. Our study also underlines a positive effect of MPAs on most coral variables, but with varying intensity among regions. These outcomes support Madagascar as a biodiversity hotspot and offer a strong argument for the need to maintain and strengthen conservation and management actions.

Spatial heterogeneity of coral assemblages

Our survey highlights the marked spatial variability of coral assemblages in Madagascar, with variation at either or both regional and local scales for all coral descriptors. This multiscale spatial heterogeneity is consistent with several previous studies on coral reefs worldwide, and thus represents one of the major characteristics of this ecosystem [6, 9, 57, 93, 94]. As in previous studies in the Western Indian Ocean [6, 38, 57], we found a marked regional-scale variation in the composition of coral assemblages, whereas changes at the local scale (among stations within regions) were less pronounced. Despite this regional dissimilarity in species composition, the local variation of generic richness showed a similar amplitude among regions, with up to two-fold differences among the six stations at each region. This local heterogeneity in generic richness was followed by a similar pattern for the Shannon diversity index, with a particularly high variability at Masoala. Our results also underline a marked heterogeneity at both regional and local scales in the abundance and cover of coral assemblages, with variable contribution of the four life history strategies.

Masoala was characterized by the high abundance of coral colonies (81–345 colonies.10 m-2), most notably of the competitive Acropora and Pocillopora as well as the stress-tolerant taxa Porites and Galaxea at some stations. The highest abundance values ever recorded at Masoala are greater than the one recorded during this study at Salary Nord, Nosy-Be, or at Toliara on the southwest coast (150–220 colonies.10 m-2; [65]). The range of coral cover (~26–46%) that we recorded at Masoala was higher than the one reported by previous studies in this region (~20–35% in 1998, [76]; ~13% in 2005, [68]). Our results support the outcomes of McClanahan [59], who suggested that good ecological conditions of these reefs, with a diverse and abundant coral assemblage and a high dominance of Acropora reported in 1996, is probably related to the moderate human impacts in this region. Moreover, the high hydrodynamic and water circulation in the northeast coast [9598] likely facilitates the larval dispersal of broadcast spawning corals such as Acropora and their subsequent regional recruitment. However, this hypothesis should be investigated by a connectivity survey among reefs in this region, which is currently lacking. The good ecological state of the reefs at Masoala seem nevertheless conditioned to the fact that sedimentation and nutrients from rivers do not increase in the coming years [59, 99].

In contrast, Salary Nord was distinguished by lower abundances of coral assemblages, notably with a depauperate population of competitive taxa, mainly Acropora. In contrast to the two other regions, a higher proportion of stress-tolerant taxa, such as massive Porites, and weedy corals, such as Seriatopora and Stylophora, were recorded at most stations. The abundance of these taxa, known to colonize and often dominate disturbed environments [5, 100], supports the general consensus that coral reefs of the southwest coast of Madagascar are the most threatened of the island [101]. This lower diversity and abundance of southwestern coral assemblages has been attributed to high-resource extraction and sedimentation [38, 60]. However, the range of coral cover that we recorded at Salary Nord in 2020 (~25–49%) was higher compared to those found by Randriamanantsoa et al. [102] in 2008 (~21–24%). Thus, though comparisons between studies with different sampling methodologies should be taken with caution, our results suggest that Salary Nord reefs have not shown a clear declining trend in recent years.

Coral assemblages at Nosy-Be were characterized by a higher diversity (38 coral genera), higher abundances of several taxa, including Diploastrea and Merulina, that were much less abundant or not even recorded at the other two regions, and also by the high coral cover (~42–70%) recorded at unfished stations. These high coral cover values, among the highest ever recorded at Madagascar [53, 65], are similar to what was found by Bigot et al. [103] in 1999 (~68% on the outer slope of Nosy Tanihely), but higher than those recorded by Webster & McMahon [66] in the same year (~28–51%). Even if they are confronted by several threats, coral reefs of Nosy-Be are generally considered to be healthier compared to most other reefs around Madagascar, with a high capacity to recover from disturbances [6, 55, 69, 104, 105]. In fact, Nosy-Be is located within the high diversity center in the Northern Mozambique Channel [6, 38, 106, 107]. The high diversity and abundance of coral assemblages in this region is generally associated with the high connectivity among reefs generated by meso-scale eddies [108], and to less frequent storms [6]. This region is therefore considered a priority area for conservation [6].

The present study has focused on the regional and local scales and has provided a valuable contemporary baseline to determine temporal trajectories of coral assemblages, which is crucial for implementing effective management and conservation actions. Given the size of Madagascar and its diversity in environmental conditions, our survey should be reinforced with, for example, the addition of other reef habitats such as shallow back reefs and mesophotic depths, as well as other reefs around the island, notably those from less studied regions such as those in the Kirindy-Mite and Barren Isles in the west, Ambodivahibe in the northeast, and Anosy in the southeast.

Influence of management status and environmental conditions

Our results clearly underline the positive effects of fishing protection levels on most descriptors of coral assemblages. Stations localized in unfished areas (MPAs) of the three regions showed higher values of generic richness, Shannon diversity index, colony abundance, and cover. The relative contribution of the four life history strategies was also influenced by fishing protection level with, as an example, higher abundance of competitive coral taxa in unfished areas. In fact, species composition was the only component of the coral assemblages that was not related to fishing protection level and rather linked to geographic location. This “MPA effect” was particularly evident at Nosy-Be, with the highest contrast between fished and unfished stations being in terms of abundance and cover of coral colonies, with notably higher abundance of encrusting and massive coral growth forms and reduced macroalgae cover at unfished stations. Size, age, depth and connectivity are major factors in the effectiveness and success of MPAs [31, 32, 36, 4648]. Here, we suggest that the relatively small size [31, 42, 109] and accessibility of the MPAs are key aspects of the MPA effect at Nosy-Be, facilitating management in terms of monitoring, control, and awareness-raising. Moreover, MPAs at Nosy-Be are the oldest in this study, and this greater time of protection may also explain the highest MPA effect recorded here, consistent with other coral reefs worldwide [46, 47]. Moreover, with several small, scattered islands, and favorable current conditions, the connectivity in the Nosy-Be region is likely a factor that contributes to this positive MPA effect on the higher diversity and abundance of coral assemblages, as recently documented [36, 46, 48].

The biomass of herbivorous fishes was also linked to the spatial variation in generic richness, Shannon diversity index, life history strategies, and abundance of coral assemblages, with higher values of these coral descriptors at stations with high biomass of herbivorous fishes. Herbivory on coral reefs is considered a key function because it often mediates coral-algal interactions in favor of reef-building corals [110, 111]. This expected negative correlation between herbivorous fish biomass and cover of turf and macroalgae was recorded at Masoala and Nosy-Be, but not at Salary Nord. This herbivorous-algal interaction probably explains the negative link that we also recorded between algal cover (turf and macroalgae) and the generic richness, cover, and life history strategies of coral assemblages. These outcomes reinforce the demonstrated mechanism of a high biomass of herbivorous taxa limiting the cover of algae, in turn benefitting coral assemblages, as fleshy algae compete for space and use allelopathy to reduce recruitment, growth, and survivorship of corals [30, 112115]. This mechanism has been effectively demonstrated in several MPAs [30, 31, 116118] though could not be confirmed in others [75, 119, 120].

Crustose coralline algae (CCA) cover was positively related to spatial variation of the Shannon diversity index and, to a lesser degree, of generic richness and cover. These relationships could be explained by the fact that some species of CCA induce the metamorphosis and settlement of larvae of several coral species [121123], and also tend to prevent the establishment and growth of macroalgae by producing antifouling chemical signals [124]. The fact that CCA is a composite metric with not all species inducing coral settlement likely explains the relatively weak, although statistically significant, relationship that we found here, notably for generic richness and cover [125, 126]. Furthermore, the negative link between CCA cover and abundance of coral colonies that we recorded is difficult to interpret. Due to the correlative nature of our analysis, this relationship does not necessarily imply a strong underlying ecological mechanism. Biochemical and habitat-specific cues not considered in this study likely influence the density of coral populations at local scales.

Substrate rugosity was strongly and positively linked with coral cover and composition as well as, at a reduced level, the Shannon diversity index and the relative abundance of the four life history strategies. High rugosity typically provides a larger surface available for sessile organisms as along with more protection from environmental stressors and ecological niche space for marine organisms [127130]. For several reef invertebrates, including corals, high substrate micro-topography promotes larval settlement and recruitment by reducing mortality by predation up to a point where this factor represents a bottleneck in the replenishment of local populations [122, 131]. Consequently, the reduction of substrate rugosity, notably after storms, cyclones, or gleaning activities, is associated with reef degradation [132134]. The outcomes of our survey thus reinforce the positive link between rugosity and the diversity and abundance of reef invertebrates and fishes previously recorded [132, 133, 135137]. However, because the substrate rugosity was visually assessed in the present study, it should be interesting to examine the relationship between rugosity and the structure and dynamics of coral assemblages with the recently developed underwater 3D photogrammetry approach, which allows a much more precise measurement of the structural complexity of reef habitats [130, 138, 139].

Since the interactive effects of multiple biotic and abiotic drivers govern the structure and dynamics of coral assemblages [59], it is evident that other environmental factors not selected in the present study could also have a major influence on the spatial patterns revealed here. Sedimentation, light penetration, nutrients dynamics, and temperature regime, all of which have been identified as structuring factors for other coral reefs in the SWIO [57] and around Madagascar [65, 67, 140], should be considered in a future survey.

Conclusions and perspectives

One of the major results of this study is the identification of reefs with high diversity, abundance, and coverage of corals, including the competitive Acropora. These original outcomes are particularly unexpected in the current context of increasing degradation of many coral reefs around the world. However, these optimistic results should be tempered as our survey is only a snapshot in a highly dynamic system and especially as large-scale disturbances, such as thermally induced bleaching events and local stressors associated with human growth, are expected to increase in the near future, with possible shifts in coral assemblage distribution, diversity, and structure [141, 142]. Nevertheless, our survey represents a valuable contemporary baseline to document changes of these reefs and may help to determine if they will provide refuge from future environmental changes.

The positive effects of MPAs on coral assemblages have been clearly demonstrated here. The outcomes of our study strongly support the implementation of MPAs where fishing should be reduced or banned, notably for herbivorous fishes, for which we demonstrated a positive role for the abundance and diversity of coral assemblages, in line with previous studies [28, 30, 42, 143]. Conservation actions should also focus on reducing all fishing activities, such as gleaning and using destructive gears, that strongly alter the reef habitat rugosity and structural complexity that is critical for the health of reef communities [132, 133, 138]. Alternative sources of income, such sea cucumber and seaweed farming, that may reduce the negative effects of overfishing on coral communities, should be also encouraged [140, 144]. The strong spatial variability this study reveals on coral assemblages at the local scale suggests that MPAs should integrate a sufficiently larger scale to capture such heterogeneity [38] but that, in contrast, small reserves are easier to manage in Madagascar given the reduced logistic and human resources generally allotted to them. In fact, priority should be given to the implementation of locally managed marine areas with strong primary user involvement, as has been successfully done for fisheries in the Toliara region [35, 71, 74, 75] and elsewhere, such as in Fiji [42]. The outcomes of our study also confirm the need to maintain and strengthen existing MPAs, as older MPAs are known to be more effective in preventing coral decline [32]. This effort should notably focus on the MPAs of Nosy-Be, as this high diversity area may act as a larval source for other “sink” reefs and may favor their replenishment and resilience [48]. It is also timely to establish long-term interannual monitoring of the reef communities, environmental conditions, and threats at several reefs of various protection levels to precisely examine and improve the effectiveness of coral reef conservation at Madagascar.

Supporting information

S1 Table. Main characteristics of the 18 stations surveyed at the three regions around Madagascar, and major disturbances that were reported in recent decades.

https://doi.org/10.1371/journal.pone.0275017.s001

(PDF)

S2 Table. Composition and abundance (mean number of colonies per 10 m-2) of coral assemblages at the 18 stations used for nonmetric multidimensional scaling (nMDS).

Standard errors (SE) in brackets.

https://doi.org/10.1371/journal.pone.0275017.s002

(PDF)

S3 Table. Summary of post-hoc tests to examine differences of coral generic richness between the three regions.

https://doi.org/10.1371/journal.pone.0275017.s003

(PDF)

S4 Table. t-test from the linear mixed models to examine differences of coral generic richness between fished and unfished stations.

https://doi.org/10.1371/journal.pone.0275017.s004

(PDF)

S5 Table. Summary of post-hoc tests to examine differences of coral generic richness according to fishing protection level (fished vs. unfished areas) at each of the three regions.

https://doi.org/10.1371/journal.pone.0275017.s005

(PDF)

S6 Table. Summary of post-hoc tests to examine differences of coral Shannon diversity index between the three regions.

https://doi.org/10.1371/journal.pone.0275017.s006

(PDF)

S7 Table. t-test from the linear mixed models to examine differences of the Shannon diversity index between fished and unfished stations.

https://doi.org/10.1371/journal.pone.0275017.s007

(PDF)

S8 Table. Summary of post-hoc tests to examine differences of the Shannon diversity index according to fishing protection level at each of the three regions.

https://doi.org/10.1371/journal.pone.0275017.s008

(PDF)

S9 Table. Summary of post-hoc tests to examine differences of coral abundance between the three regions.

https://doi.org/10.1371/journal.pone.0275017.s009

(PDF)

S10 Table. t-test from the linear mixed models to examine differences of coral abundance between fished and unfished stations.

https://doi.org/10.1371/journal.pone.0275017.s010

(PDF)

S11 Table. Summary of post-hoc tests to examine differences of coral abundance according to fishing protection level at each of the three regions.

https://doi.org/10.1371/journal.pone.0275017.s011

(PDF)

S12 Table. Summary of post-hoc tests to examine differences of abundance of coral life history strategies.

https://doi.org/10.1371/journal.pone.0275017.s012

(PDF)

S13 Table. Summary of post-hoc tests to examine differences of abundance of coral life history strategies between the three regions.

https://doi.org/10.1371/journal.pone.0275017.s013

(PDF)

S14 Table. Summary of post-hoc tests to examine differences of abundance of coral life history strategies according to fishing protection level.

https://doi.org/10.1371/journal.pone.0275017.s014

(PDF)

S15 Table. Summary of post-hoc tests to examine differences of abundance of coral life history strategies according to fishing protection level at each of the three regions.

https://doi.org/10.1371/journal.pone.0275017.s015

(PDF)

S16 Table. Summary of post-hoc tests to examine differences of coral cover between the three regions.

https://doi.org/10.1371/journal.pone.0275017.s016

(PDF)

S17 Table. t-test from the linear mixed models to examine differences of coral cover between fished and unfished stations.

https://doi.org/10.1371/journal.pone.0275017.s017

(PDF)

S18 Table. Summary of post-hoc tests to examine differences of coral cover according to fishing protection level at each of the three regions.

https://doi.org/10.1371/journal.pone.0275017.s018

(PDF)

S19 Table. Summary of post-hoc tests to examine differences in macroalgal cover between the three regions.

https://doi.org/10.1371/journal.pone.0275017.s019

(PDF)

S20 Table. Generalized linear mixed models χ2-test to examine differences of macroalgal cover between fished and unfished stations.

https://doi.org/10.1371/journal.pone.0275017.s020

(PDF)

S21 Table. Summary of post-hoc tests to examine differences in turf cover between the three regions.

https://doi.org/10.1371/journal.pone.0275017.s021

(PDF)

S22 Table. Generalized linear mixed models χ2-test to examine differences of turf cover between fished and unfished stations.

https://doi.org/10.1371/journal.pone.0275017.s022

(PDF)

S23 Table. Summary of post-hoc tests to examine differences in crustose coralline algae (CCA) cover between the three regions.

https://doi.org/10.1371/journal.pone.0275017.s023

(PDF)

S24 Table. Generalized linear mixed models χ2-test to examine differences of crustose coralline algae (CCA) cover between fished and unfished stations.

https://doi.org/10.1371/journal.pone.0275017.s024

(PDF)

S25 Table. Summary of post-hoc tests to examine differences in herbivorous fish biomass between the three regions.

https://doi.org/10.1371/journal.pone.0275017.s025

(PDF)

S26 Table. Generalized linear mixed models χ2-test to examine differences of herbivorous fish biomass between fished and unfished stations.

https://doi.org/10.1371/journal.pone.0275017.s026

(PDF)

S27 Table. Summary of post-hoc tests to examine differences in rugosity index between the three regions.

https://doi.org/10.1371/journal.pone.0275017.s027

(PDF)

S28 Table. Generalized linear mixed models χ2-test to examine differences of rugosity index between fished and unfished stations.

https://doi.org/10.1371/journal.pone.0275017.s028

(PDF)

S1 Fig. Correlation analysis between explanatory variables.

Spearman correlation coefficients (ρ) were calculated for all stations/regions (overall) and for each of the three regions (above diagonal). Scatterplots of each pair of explanatory variables (below diagonal), and the distribution of variable (diagonal) are also given.

https://doi.org/10.1371/journal.pone.0275017.s029

(PDF)

S2 Fig. Relationships between coral generic richness and the four explanatory variables (herbivorous fish biomass, fishing protection level, CCA cover, and turf cover) selected by the linear mixed models.

Blue lines are linear model fits and red lines are LOESS (locally weighted scatterplot smoothing), with standard error for each (dark grey for linear model fits, and light grey for LOESS). RI: relative importance of each variable.

https://doi.org/10.1371/journal.pone.0275017.s030

(PDF)

S3 Fig. Relationships between Shannon diversity index and the four explanatory variables (CCA cover, herbivorous fish biomass, fishing protection level, and rugosity) selected by the linear mixed models.

Blue lines are linear model fits and red lines are LOESS (locally weighted scatterplot smoothing), with standard error for each (dark grey for linear model fits, and light grey for LOESS). RI: relative importance of each variable.

https://doi.org/10.1371/journal.pone.0275017.s031

(PDF)

S4 Fig. Relationships between coral abundance and the three explanatory variables (herbivorous fish biomass, fishing protection level, and CCA cover) selected by the linear mixed models.

Blue lines are linear model fits and red lines are LOESS (locally weighted scatterplot smoothing), with standard error for each (dark grey for linear model fits, and light grey for LOESS). RI: relative importance of each variable.

https://doi.org/10.1371/journal.pone.0275017.s032

(PDF)

S5 Fig. Relationships between coral cover and the five explanatory variables (fishing protection level, rugosity, region, CCA cover, and turf cover) selected by the linear mixed models.

Blue lines are linear model fits and red lines are LOESS (locally weighted scatterplot smoothing), with standard error for each (dark grey for linear model fits, and light grey for LOESS). RI: relative importance of each variable.

https://doi.org/10.1371/journal.pone.0275017.s033

(PDF)

S6 Fig. Mean effects of explanatory variables on the spatial variation in coral generic richness.

Values have been standardized as effect sizes; circles represent mean parameter estimates and lines represent 95% confidence intervals. Filled dots indicate significant mean values (i.e., different from zero).

https://doi.org/10.1371/journal.pone.0275017.s034

(PDF)

S7 Fig. Mean effects of explanatory variables on the spatial variation in Shannon diversity index.

Values have been standardized as effect sizes; circles represent mean parameter estimates and lines represent 95% confidence intervals. Filled dots indicate significant mean values (i.e., different from zero).

https://doi.org/10.1371/journal.pone.0275017.s035

(PDF)

S8 Fig. Mean effects of explanatory variables on the spatial variation in coral abundance.

Values have been standardized as effect sizes; circles represent mean parameter estimates and lines represent 95% confidence intervals. Filled dots indicate significant mean values (i.e., different from zero).

https://doi.org/10.1371/journal.pone.0275017.s036

(PDF)

S9 Fig. Mean effects of explanatory variables on the spatial variation in abundance of the four life history strategies.

Values have been standardized as effect sizes; circles represent mean parameter estimates, and lines represent 95% confidence intervals. Filled dots indicate significant mean values (i.e., different from zero).

https://doi.org/10.1371/journal.pone.0275017.s037

(PDF)

S10 Fig. Mean effects of explanatory variables on the spatial variation in coral cover.

Values have been standardized as effect sizes, circles represent mean parameter estimates and lines represent 95% confidence intervals. Filled dots indicate significant mean values (i.e., different from zero).

https://doi.org/10.1371/journal.pone.0275017.s038

(PDF)

Acknowledgments

We thank the Centre National de Recherches Océanographiques, the Madagascar National Park, and the Wildlife Conservation Society Madagascar for their technical and logistical supports in the field. We would also thank the Malagasy Ministry of Environment and Sustainable Development, and the local managers for providing us with the necessary authorizations for field work. We gratefully acknowledge critical comments made by Jane Ballard, and two anonymous reviewers. This research is product of the JEAI ACOM and LMI Mikaroka.

References

  1. 1. Kittinger JN, Finkbeiner EM, Glazier EW, Crowder LB. Human dimensions of coral reef social-ecological systems. Ecology and Society. 2012; 17. https://doi.org/10.5751/ES-05115-170417
  2. 2. Woodhead AJ, Hicks CC, Norström AV, Williams GJ, Graham NA. Coral reef ecosystem services in the Anthropocene. Functional Ecology. 2019; 33: 1023–1034. https://doi.org/10.1111/1365-2435.13331
  3. 3. Bellwood DR, Hoey AS, Ackerman JL, Depczynski M. Coral bleaching, reef fish community phase shifts and the resilience of coral reefs. Global Change Biology. 2006; 12: 1587–1594. https://doi.org/10.1111/j.1365-2486.2006.01204.x
  4. 4. Hughes TP, Barnes ML, Bellwood DR, Cinner JE, Cumming GS, Jackson JB, et al. Coral reefs in the Anthropocene. Nature. 2017; 546: 82–90. pmid:28569801
  5. 5. Darling E, McClanahan T, Côté I. Life histories predict coral community disassembly under multiple stressors. Global Change Biology. 2013; 19. pmid:23504982
  6. 6. McClanahan TR, Ateweberhan M, Darling E, Graham N, Muthiga N. Biogeography and change among regional coral communities across the Western Indian Ocean. PLoS ONE. 2014; 9: e93385. pmid:24718371
  7. 7. Veron J, Stafford-Smith M, DeVantier L, Turak E. Overview of distribution patterns of zooxanthellate Scleractinia. Frontiers in Marine Science. 2015; 1: 1–19. https://doi.org/10.3389/fmars.2014.00081
  8. 8. Zinke J, Gilmour JP, Fisher R, Puotinen M, Maina J, Darling E, et al. Gradients of disturbance and environmental conditions shape coral community structure for south-eastern Indian Ocean reefs. Diversity and Distributions. 2018; 24: 605–620. https://doi.org/10.1111/ddi.12714
  9. 9. Adjeroud M, Poisson E, Peignon C, Penin L, Kayal M. Spatial patterns and short-term changes of coral assemblages along a cross-shelf gradient in the southwestern lagoon of New Caledonia. Diversity. 2019; 11: 21. https://doi.org/10.3390/d11020021
  10. 10. Hughes TP, Baird AH, Bellwood DR, Card M, Connolly SR, Folke C, et al. Climate change, human impacts, and the resilience of coral reefs. Science. 2003; 301: 929–33. pmid:12920289
  11. 11. Hughes TP, Anderson KD, Connolly SR, Heron SF, Kerry JT, Lough JM, et al. Spatial and temporal patterns of mass bleaching of corals in the Anthropocene. Science. 2018a; 359: 80–83. https://doi.org/10.1126/science.aan8048
  12. 12. Bellwood DR, Hughes TP, Folke C, Nyström M. Confronting the coral reef crisis. Nature. 2004; 429: 827–833. pmid:15215854
  13. 13. Tebbett SB, Morais RA, Goatley CHR, Bellwood DR. Collapsing ecosystem functions on an inshore coral reef. Journal of Environmental Management. 2021; 289: 112471. pmid:33812145
  14. 14. Hoegh-Guldberg O, Mumby PJ, Hooten AJ, Steneck RS, Greenfield P, Gomez E, et al. Coral reefs under rapid climate change and ocean acidification. Science. 2007; 318: 1737–42. pmid:18079392
  15. 15. Hughes TP, Kerry JT, Baird AH, Connolly SR, Dietzel A, Eakin CM, et al. Global warming transforms coral reef assemblages. Nature. 2018b; 556: 492–496. https://doi.org/10.1038/s41586-018-0041-2
  16. 16. Grigg RW, Dollar SJ. Natural and anthropogenic disturbance on coral reef ecology. In: Dubinsky Z, editors. Ecosystems of the world 25: Coral reefs. Amsterdam: Elsevier; 1990. pp. 439–452.
  17. 17. Hoegh-Guldberg O. The adaptation of coral reefs to climate change: Is the Red Queen being outpaced? Scientia Marina. 2012; 76: 403–408. https://doi.org/10.3989/scimar.03660.29A
  18. 18. Bruno JF, Selig ER. Regional decline of coral cover in the Indo-Pacific: timing, extent, and subregional comparisons. PLoS ONE. 2007; 2: e711. pmid:17684557
  19. 19. McClanahan TR, Ateweberhan M, Graham NAJ, Wilson SK, Sebastián CR, Guillaume MM, et al. Western Indian Ocean coral communities: bleaching responses and susceptibility to extinction. Marine Ecology Progress Series. 2007; 337: 1–13.
  20. 20. Díaz S, Settele J, Brondízio ES, Ngo HT, Agard J, Arneth A, et al. Pervasive human-driven decline of life on Earth points to the need for transformative change. Science. 2019; 366: eaax3100. https://doi.org/10.1126/science.aax3100
  21. 21. McManus JW, Polsenberg JF. Coral–algal phase shifts on coral reefs: Ecological and environmental aspects. Progress in Oceanography. 2004; 60: 263–279. https://doi.org/10.1016/j.pocean.2004.02.014
  22. 22. Roff G, Mumby PJ. Global disparity in the resilience of coral reefs. Trends in Ecology & Evolution. 2012; 27: 404–413. pmid:22658876
  23. 23. Comte A, Pendleton LH. Management strategies for coral reefs and people under global environmental change: 25 years of scientific research. Journal of Environmental Management. 2018; 209: 462–474. pmid:29309969
  24. 24. Bellwood DR, Streit RP, Brandl SJ, Tebbett SB. The meaning of the term ‘function’ in ecology: A coral reef perspective. Functional Ecology. 2019; 33: 948–961. https://doi.org/10.1111/1365-2435.13265
  25. 25. Darling ES, McClanahan TR, Maina J, Guerneys GG, Graham NAJ, Januchowski-Hartley F et al. Social–environmental drivers inform strategic management of coral reefs in the Anthropocene. Nature Ecology and Evolution 2019; 3. pp. 1341–1350. pmid:31406279
  26. 26. Graham NAJ, Robinson JPW, Smith SE, Govinden R, Gendron G, Wilson SK. Changing role of coral reef marine reserves in a warming climate. Nature Communications. 2020; 11: 2000. pmid:32332721
  27. 27. Mumby PJ, Elliott IA, Eakin CM, Skirving W, Paris CB, Edwards HJ, et al. Reserve design for uncertain responses of coral reefs to climate change. Ecology Letters. 2011; 14: 132–140. pmid:21105980
  28. 28. Mumby PJ, Steneck RS, Roff G, Paul VJ. Marine reserves, fisheries ban, and 20 years of positive change in a coral reef ecosystem. Conservation Biology. 2021; 35: 1473–1483. pmid:33909928
  29. 29. Edgar GJ, Stuart-Smith RD, Willis TJ, Kininmonth S, Baker SC, Banks S, et al. Global conservation outcomes depend on marine protected areas with five key features. Nature. 2014; 506: 216–220. pmid:24499817
  30. 30. Steneck RS, Mumby PJ, MacDonald C, Rasher DB, Stoyle G. Attenuating effects of ecosystem management on coral reefs. Science Advances. 2018; 4: eaao5493. pmid:29750192
  31. 31. Strain EMA, Edgar GJ, Ceccarelli D, Stuart-Smith RD, Hosack GR, Thomson RJ. A global assessment of the direct and indirect benefits of marine protected areas for coral reef conservation. Diversity and Distributions. 2019; 25: 9–20. https://doi.org/10.1111/ddi.12838
  32. 32. Selig ER, Bruno JF. A Global analysis of the effectiveness of marine protected areas in preventing coral loss. PLoS ONE. 2010; 5: e9278. pmid:20174644
  33. 33. Cinner J, Fuentes MMPB, Randriamahazo H. Exploring Social Resilience in Madagascar’s Marine Protected Areas. Ecology and Society. 2009; 14(1):41
  34. 34. Benbow S, Humber F, Oliver T, Oleson K, Raberinary D, Nadon M, et al. Lessons learnt from experimental temporary octopus fishing closures in south-west Madagascar: benefits of concurrent closures. African Journal of Marine Science. 2014; 36 (1): 31–37. https://doi.org/10.2989/1814232X.2014.893256
  35. 35. Ratsimbazafy H, Lavitra T, Kochzius M, Huge J. Emergence and diversity of marine protected areas in Madagascar. Marine Policy. 2019; 105: 91–108. https://doi.org/10.1016/j.marpol.2019.03.008
  36. 36. Goetze JS, Wilson S, Radford B, Fisher R, Langlois TJ, Monk J, et al. Increased connectivity and depth improve the effectiveness of marine reserves. Global Change Biology. 2021;27: 3432–3447. pmid:34015863
  37. 37. Selig ER, Casey KS, Bruno JF. Temperature-driven coral decline: the role of marine protected areas. Global Change Biology. 2012; 18: 1561–1570. https://doi.org/10.1111/j.1365-2486.2012.02658.x
  38. 38. Ateweberhan M, McClanahan TR. Partitioning scleractinian coral diversity across reef sites and regions in the Western Indian Ocean. Ecosphere. 2016; 7: e01243. https://doi.org/10.1002/ecs2.1243
  39. 39. Bruno JF, Côté IM, Toth LT. Climate change, coral loss, and the curious case of the parrotfish paradigm: why don’t marine protected areas improve reef resilience? Annual Review of Marine Science. 2019; 11: 307–334. pmid:30606097
  40. 40. Díaz-Osorio AC, Schmitter-Soto JJ, Vega-Zepeda A, Espinoza-Tenorio A. How effective are marine parks in protecting their coral reef ecosystem? A study case in the Mexican Caribbean. Aquatic Conservation: Marine and Freshwater Ecosystems. 2022;32: 1126–1140. https://doi.org/10.1002/aqc.3820
  41. 41. Nash KL, Graham NA, Jennings S, Wilson SK, Bellwood DR. Herbivore cross-scale redundancy supports response diversity and promotes coral reef resilience. Journal of Applied Ecology. 2016;53: 646–655
  42. 42. Bonaldo RM, Pires MM, Guimarães PR, Hoey AS, Hay ME. Small marine protected areas in Fiji provide refuge for reef fish assemblages, feeding groups, and corals. PLoS ONE. 2017; 12: e0170638. pmid:28122006
  43. 43. Chung AE, Wedding LM, Green AL, Friedlander AM, Goldberg G, Meadows A, et al. Building Coral Reef Resilience Through Spatial Herbivore Management. Frontiers in Marine Science. 2019;6: 98 https://doi.org/10.3389/fmars.2019.00098
  44. 44. Page C, Baker D, Harvell C, Golbuu Y, Raymundo L, Neale S, et al. Influence of marine reserves on coral disease prevalence. Disease Aquatic Organism. 2009;87: 135–150. pmid:20095248
  45. 45. Jouval F, Latreille AC, Bureau S, Adjeroud M, Penin L. Multiscale variability in coral recruitment in the Mascarene Islands: From centimetric to geographical scale. PLoS ONE. 2019; 14(3): e0214163. pmid:30901355
  46. 46. Ferreira HM, Magris RA, Floeter SR, Ferreira CE. Drivers of ecological effectiveness of marine protected areas: A meta-analytic approach from the Southwestern Atlantic Ocean (Brazil). Journal of Environmental Management. 2022;301: 113889. pmid:34610560
  47. 47. Friedlander AM, Golbuu Y, Ballesteros E, Caselle JE, Gouezo M, Olsudong D, et al. Size, age, and habitat determine effectiveness of Palau’s Marine Protected Areas. PLoS ONE. 2017; 12: e0174787. pmid:28358910
  48. 48. Fontoura L, D’Agata S, Gamoyo M, Barneche DR, Luiz OJ, Madin EMP, et al. Protecting connectivity promotes successful biodiversity and fisheries conservation. Science. 2022; 375: 336–340. pmid:35050678
  49. 49. Guabiroba HC, Vilar CC, Pinheiro HT, Joyeux JC. Limited human access is linked to higher effectiveness in a marine sanctuary. Journal of Environmental Management. 2022;311: 114838. pmid:35279488
  50. 50. Myers N, Mittermeier RA, Mittermeier CG, Da Fonseca GA, Kent J. Biodiversity hotspots for conservation priorities. Nature. 2000; 403: 853–858. pmid:10706275
  51. 51. Brooks TM, Mittermeier RA, Da Fonseca GA, Gerlach J, Hoffmann M, Lamoreux JF, et al. Global biodiversity conservation priorities. Science. 2006; 313: 58–61. pmid:16825561
  52. 52. Cooke A. Madagascar: a guide to marine biodiversity. Antananarivo, Madagascar: Wildlife Conservation Society; 2012. 176pp.
  53. 53. Veron JEN, Turak E. Reef corals of Northwest Madagascar. In: McKenna SA, Allen GR, editors. A rapid marine biodiversity assessment of Northwest Madagascar. Washington, DC, USA: Bulletin of the Rapid Assessment Program 31, Conservation International; 2003. pp. 26–31.
  54. 54. Ahamada S, Bijoux J, Cauvin B, Hagan A, Harris A, Koonjul M, et al. Status of coral reefs of the South-West Indian Ocean islands states: Comoros, Madagascar, Mauritius, Reunion, Seychelles. In: Wilkinson CR, editors. Status of coral reefs of the world. GCRMN Report. Townsville, Australia: Australian Institute of Marine Science; 2008. pp. 105–118.
  55. 55. McClanahan TR, Ateweberhan M, Omukoto J, Pearson L. Recent seawater temperature histories, status, and predictions for Madagascar’s coral reefs. Marine Ecology Progress Series. 2009; 380: 117–128.
  56. 56. Obura D, Di Carlo G, Rabearisoa A, Oliver T. A Rapid Marine Biodiversity Assessment of the coral reefs of northeast Madagascar. RAP Bulletin of Biological Assessment. 2011; 61.
  57. 57. Ateweberhan M, McClanahan TR, Maina J, Sheppard C. Thermal energy and stress properties as the main drivers of regional distribution of coral species richness in the Indian Ocean. Journal of Biogeography. 2018; 45: 1355–1366. https://doi.org/10.1111/jbi.13224
  58. 58. Pichon M. Recherches sur les peuplements à dominance d’anthozoaires dans les récifs coralliens de Tuléar. Atoll Research Bulletin. 1978; 222, 1–447.
  59. 59. McClanahan TR. Testing for correspondence between coral reef invertebrate diversity and marine park designation on the Masoala Peninsula of eastern Madagascar. Marine and Freshwater Ecosystem. 2007; 17: 409–419. https://doi.org/10.1002/aqc.775
  60. 60. Nadon MO, Griffiths D, Doherty E, Harris A. The status of coral reefs in the remote region of Andavadoaka, southwest Madagascar. Western Indian Ocean Journal of Marine Science. 2008; 6. https://doi.org/10.4314/wiojms.v6i2.48244
  61. 61. Harris A, Manahira G, Sheppard A, Gouch C, Sheppard C. Demise of Madagascar’s once great barrier reef: changes in coral reef conditions over 40 years. Atoll Research Bulletin. 2010; 574.
  62. 62. Bruggemann J, Rodier M, Guillaume M, Andréfouët S, Arfi R, Cinner J, et al. Wicked social-ecological problems forcing unprecedented change on the latitudinal margins of coral reefs: the case of southwest Madagascar. Ecology and Society. 2012; 17(4): 47. https://doi.org/10.5751/ES-05300-170447
  63. 63. Andréfouët S, Guillaume MM, Delval A, Rasoamanendrika FMA, Blanchot J, Bruggemann JH. Fifty years of changes in reef flat habitats of the Grand Récif of Toliara (SW Madagascar) and the impact of gleaning. Coral Reefs. 2013; 32: 757–768. https://doi.org/10.1007/s00338-013-1026-0
  64. 64. Todinanahary GG, Refoty ME, Terrana L, Lavitra T, Eeckhaut I. Previously unlisted scleractinian species recorded from the Great Reef of Toliara, southwest Madagascar. Western Indian Ocean Journal of Marin Science. 2018; 17: 67.
  65. 65. Botosoamananto RL, Todinanahary G, Razakandrainy A, Randrianarivo M, Penin L, Adjeroud M. Spatial patterns of coral community structure in the Toliara region of Southwest Madagascar and implications for conservation and management. Diversity. 2021; 13: 486. https://doi.org/10.3390/d13100486
  66. 66. Webster FJ, McMahon K. An assessment of coral reefs in northwest Madagascar. In: Lindén O, Souter D, Wilhelmsson D, Obura D, editors. Coral reef degradation in the Indian ocean: Status report; 2002. 190–201.
  67. 67. Harding S, Randriamanantsoa B, Hardy T, Curd A. Coral reef monitoring and biodiversity assessment to support the planning of a proposed MPA at Andavadoaka. London Blue: Ventures Conservation; Antananarivo: the Wildlife Conservation Society; 2006. pp. 1–100
  68. 68. Harding S, Randriamanantsoa B. Coral reef monitoring in marine reserves of northern Madagascar ten years after bleaching–facing the consequences of climate change in the Indian Ocean. CORDIO Status Report; 2008. pp. 93–106.
  69. 69. Gabrié C, Vasseur P, Randriamiarana H, Maharavo J, Mara E. The coral reefs of Madagascar. In: McClanahan TR, Sheppard CRC, Obura DO, editors. Coral reefs of the Indian Ocean their ecology and conservation; 2000. pp. 411–444.
  70. 70. Ahamada S, Bigot L, Bijoux J, Maharavo J, Meunier S, Moyne-Picard M, et al. Status of coral reefs in the south west Indian Ocean Island node: Comoros, Madagascar, Mauritius, Reunion and Seychelles. In: Wilkinson CR, editors. Status of coral reefs of the world. GCRMN Report. Townsville, Australia: Australian Institute of Marine Science; 2002. pp. 79–100.
  71. 71. Cochrane KL, Rakotondrazafy H, Aswani S, Chaigneau T, Downey-Breedt N, Lemahieu A, et al. Tools to enrich vulnerability assessment and adaptation planning for coastal communities in data-poor regions: application to a case study in Madagascar. Frontiers in Marine Science. 2019; 505. https://doi.org/10.3389/fmars.2018.00505
  72. 72. Ratsimbazafy RE. L’océan au cœur de la Grande Île: Les aires marines protégées, un outil de développement durable pour Madagascar. In Madagascar Conservation and Development. Institute and Museum of Anthropology University of Zurich, Switzerland. 2011; 6:5.
  73. 73. UNEP-Nairobi Convention and WIOMSA. Western Indian Ocean Marine Protected Areas Outlook: Towards achievement of the Global Biodiversity Framework Targets. Nairobi, Kenya: UNEP and WIOMSA; 2021. 298 pp.
  74. 74. Mahafina J. Perception et comportements des pêcheurs pour une gestion durable de la biodiversité et de la pêcherie récifale: Application au niveau des réserves marines temporaires du Sud-Ouest de Madagascar. Ph.D. Thesis, University of Toliara, Toliara, Madagascar. 2011.
  75. 75. McClanahan TR, Jadot C. Managing coral reef fish community biomass is a priority for biodiversity conservation in Madagascar. Marine Ecology Progress Series. 2017; 580: 169–190. https://doi.org/10.3354/meps12267
  76. 76. McClanahan TR, Obura D. Monitoring, training and assessment of the coral reefs of the Masoala Peninsula. Project Report to Care-Masoala; 1998. 49 pp.
  77. 77. Gudka M, Obura D, Mbugua J, Ahamada S, Kloiber U, Holter T. Participatory reporting of the 2016 bleaching event in the Western Indian Ocean. Coral Reefs. 2020; 39: 1–11. https://doi.org/10.1007/s00338-019-01851-3
  78. 78. Carter AL, Gilchrist H, Dexter KG, Gardner CJ, Gough C, Rocliffe S, et al. Cyclone Impacts on Coral Reef Communities in Southwest Madagascar. Frontier in Marine Science. 2022; 9: 753325. https://doi.org/10.3389/fmars.2022.753325
  79. 79. Penin L, Adjeroud M, Pratchett MS, Hughes TP. Spatial distribution of juvenile and adult corals around Moorea (French Polynesia): implications for population regulation. Bulletin of Marine Science. 2007; 80: 379–389.
  80. 80. Darling ES, Alvarez‐Filip L, Oliver TA, McClanahan TR, Côté IM. Evaluating life-history strategies of reef corals from species traits. Ecology Letters. 2012; 15: 1378–1386. pmid:22938190
  81. 81. Keith SA, Baird AH, Hughes TP, Madin JS, Connolly SR. Faunal breaks and species composition of Indo-Pacific corals: the role of plate tectonics, environment and habitat distribution. Proceedings of the Royal Society B: Biological Sciences. 2013; 280: 20130818. pmid:23698011
  82. 82. Froese R, Pauly D. Fish Base. Worldwide Web Electronic Publication. 2021. URL https://www.fishbase.org
  83. 83. Polunin NVC, Roberts CM. Greater biomass and value of target coral-reef fishes in two small Caribbean marine reserves. Marine Ecology Progress Series. 1993; 100: 167–176.
  84. 84. Oksanen J, Blanchet FG, Friendly M, Kindt R, Legendre P, McGlinn D, et al. Vegan: Community Ecology Package. R Package version 2020; 2.5–7.
  85. 85. Anderson MJ. A new method for non-parametric multivariate analysis of variance. Austral Ecology. 2001; 26: 32–46. https://doi.org/10.1111/j.1442-9993.2001.01070.pp.x
  86. 86. Martinez AP. Pairwise Adonis: Pairwise multilevel comparison using ADONIS. R Package version 2020; 0.4.
  87. 87. Zuur AF, Ieno EN, Walker N, Saveliev AA, Smith GM. Mixed effects models and extensions in ecology with R. New York: Springer; 2009. https://doi.org/10.1007/978-0-387-87458-6
  88. 88. Bates D, Maechler M, Bolker B, Walker S, Christensen RH, Singmann H, et al. The lme4 package. R Package version 2021; 1.1–27.1
  89. 89. Barton K. MuMIn: Multi-model inference. R Package version 2020; 1.43.17.
  90. 90. Burnham KP, Anderson D. Model selection and multimodel inference: a practical information-theoretic approach. New York: Springer; 2002. 515 pp.
  91. 91. Lenth R. Emmeans: estimated marginal means, aka Least-Squares Means. R Package version 2020; 1.4.5.
  92. 92. R Core Team. R: A language and environment for statistical computing. R Foundation for Statistical Computing: Vienna, Austria. 2021. URL https://www.Rproject.org/
  93. 93. Zvuloni A, van Woesik R, Loya Y. Diversity partitioning of stony corals across multiple spatial scales around Zanzibar Island, Tanzania. PLoS ONE. 2010; 5: e9941. pmid:20360956
  94. 94. Karisa JF, Obura DO, Chen CA. Spatial heterogeneity of coral reef benthic communities in Kenya. PLoS ONE. 2020; 15: e0237397. pmid:32845883
  95. 95. Kremen C, Razafimahatratra V, Guillery RP, Rakotomalala J, Weiss A, Ratsisompatrarivo JS. Designing the Masoala National Park in Madagascar based on biological and socioeconomic data. Conservation Biology. 1999; 13: 1055–1068. https://doi.org/10.1046/j.1523-1739.1999.98374.x
  96. 96. Chen Z, Wu L, Qiu B, Sun S, Jia F. Seasonal variation of the South Equatorial Current bifurcation off Madagascar. Journal of Physical Oceanography. 2014; 44: 618–631. https://doi.org/10.1175/JPO-D-13-0147.1
  97. 97. Crochelet E, Roberts J, Lagabrielle E, Obura D, Petit M, Chabanet P. A model-based assessment of reef larvae dispersal in the Western Indian Ocean reveals regional connectivity patterns—Potential implications for conservation policies. Regional Studies in Marine Science. 2016; 7: 159–167. https://doi.org/10.1016/j.rsma.2016.06.007
  98. 98. Gamoyo M, Obura D, Reason C. Estimating connectivity through larval dispersal in the Western Indian Ocean. Journal of Geophysical Research: Biogeosciences. 2019; 124. https://doi.org/10.1029/2019JG005128
  99. 99. Maina J, De Moel H, Zinke J, Madin J, McClanahan T, Vermaat JE. Human deforestation outweighs future climate change impacts of sedimentation on coral reefs. Nature communications. 2013; 4: 1986. pmid:23736941
  100. 100. Bento R, Hoey AS, Bauman AG, Feary DA, Burt JA. The implications of recurrent disturbances within the world’s hottest coral reef. Marine Pollution Bulletin. 2016; 105: 466–472. pmid:26478453
  101. 101. Obura D, Gudka M, Rabi FA, Gian SB, Bijoux J, Freed S, et al. Coral Reef Status Report for the Western Indian Ocean. Nairobi, Kenya: Global Coral Reef Monitoring Network (GCRMN)/International Coral Reef Initiative; 2017. 144pp.
  102. 102. Randriamanantsoa B, Zavatra JB, Razanakoto I, Razafindratondra F, Lope JC, Bemanaja E. Etude bio-écologique relative à la mise en place des réserves marines et de l’aire marine protégée dans la zone de Salary Nord, région Sud-Ouest de Madagascar. Antananarivo, Madagascar: Wildlife Conservation Society / World Wildlife Fund; 2010. 50 pp.
  103. 103. Bigot L, Charpy L, Maharavo J, Rabi FA, Paupiah N, Aumeeruddy R, et al. Status of coral reefs of the southern Indian Ocean: the Indian Ocean commission node for Comoros, Madagascar, Mauritius, Réunion and Seychelles. In: Wilkinson CR, editors. Status of Coral Reefs of the World. GCRMN report. Queensland, Australia: Australian Institute of Marine Science; 2000. pp. 77–94.
  104. 104. McKenna SA, Allen GR. The condition of coral reefs in northwest Madagascar. A Rapid Marine Biodiversity Assessment of the Coral Reefs of Northwest Madagascar. Bulletin of the Rapid Assessment Program. 2003; 31: 54–67.
  105. 105. Obura D. Coral reef resilience assessment of the Nosy Hara marine protected area, Northwest Madagascar. Gland, Switzerland: IUCN; 2009. 70pp.
  106. 106. McClanahan TR, Maina JM, Muthiga NA. Associations between climate stress and coral reef diversity in the western Indian Ocean. Global Change Biology. 2011; 17: 2023–2032. https://doi.org/10.1111/j.1365-2486.2011.02395.x
  107. 107. Obura D. The Diversity and biogeography of Western Indian Ocean reef-building corals. PLoS ONE. 2012; 7: e45013. pmid:23028737
  108. 108. Shouten MW, de Ruijter WPM, van Leeuwen PJ, Ridderinkhof H. Eddies and variability in the Mozambique Channel. Deep-Sea Research II. 2003; 50: 1987–2003. https://doi:10.1016/S0967-0645(03)00042-0
  109. 109. Bayley DT, Mogg AO, Purvis A, Koldewey HJ. Evaluating the efficacy of small-scale marine protected areas for preserving reef health: A case study applying emerging monitoring technology. Aquatic Conservation: Marine and Freshwater Ecosystems. 2019a; 1–9. https://doi.org/10.1002/aqc.3215
  110. 110. Burkepile DE, Hay ME. Impact of herbivore identity on algal succession and coral growth on a Caribbean reef. PLoS ONE. 2010; 5: e8963. pmid:20126450
  111. 111. Vincent IV, Hincksman CM, Tibbetts IR, Harris A. Biomass and abundance of herbivorous fishes on coral reefs off Andavadoaka, Western Madagascar. Western Indian Ocean Journal of Marine Science. 2011; 10: 83–99–99.
  112. 112. Box SJ, Mumby PJ. Effect of macroalgal competition on growth and survival of juvenile Caribbean corals. Marine Ecology Progress Series. 2007; 342: 139–149. https://doi.org/10.3354/meps342139
  113. 113. Birrell CL, McCook LJ, Willis BL, Diaz-Pulido GA. Effects of benthic algae on the replenishment of corals and the implications for the resilience of coral reefs. Oceanography and Marine Biology. 2008; pp. 25–63.
  114. 114. Penin L, Michonneau F, Carroll A, Adjeroud M. Effects of predators and grazers exclusion on early post-settlement coral mortality. Hydrobiologia. 2011; 663: 259–264. https://doi.org/10.1007/s10750-010-0569-0
  115. 115. Ditzel P, König S, Musembi P, Peters MK. Correlation between coral reef condition and the diversity and abundance of fishes and sea urchins on an East African Coral Reef. Oceans. 2022; 3(1): 1–14. https://doi.org/10.3390/oceans3010001
  116. 116. Mumby PJ, Dahlgren CP, Harborne AR, Kappel CV, Micheli F, Brumbaugh DR, et al. Fishing, trophic cascades, and the process of grazing on coral reefs. Science. 2006; 311: 98–101. pmid:16400152
  117. 117. Mumby PJ, Harborne AR. Marine reserves enhance the recovery of corals on Caribbean reefs. PLoS ONE. 2010; 5: e8657. pmid:20066158
  118. 118. Mellin C, MacNeil AM, Cheal AJ, Emslie MJ, Caley JM. Marine protected areas increase resilience among coral reef communities. Ecology Letters. 2016; 19: 629–637. pmid:27038889
  119. 119. Kramer KL, Heck KL Jr. Top-down trophic shifts in Florida Keys patch reef marine protected areas. Marine Ecology Progress Series. 2007; 349: 111–123. https://doi.org/10.3354/meps07083
  120. 120. Graham NAJ, McClanahan TR, MacNeil MA, Wilson SK, Polunin NVC, Jennings S, et al. Climate warming, marine protected areas and the ocean-scale integrity of coral reef ecosystems. PLoS ONE. 2008; 3: e3039. pmid:18728776
  121. 121. Harrington L, Fabricius K, De’Ath G, Negri A. Recognition and selection of settlement substrata determine post-settlement survival in corals. Ecology. 2004; 85: 3428–3437. https://doi.org/10.1890/04-0298
  122. 122. Ritson-Williams R, Arnold SN, Fogarty ND, Steneck RS, Vermeij MJ, Paul VJ. New perspectives on ecological mechanisms affecting coral recruitment on reefs. Smithsonian Contributions to the Marine Sciences. 2009; 38: 437. https://doi.org/10.5479/si.01960768.38.437
  123. 123. Whitman TN, Negri AP, Bourne DG, Randall CJ. Settlement of larvae from four families of corals in response to a crustose coralline alga and its biochemical morphogens. Scientific Reports. 2020; 10: 16397. pmid:33009428
  124. 124. Vermeij MJA, Dailer ML, Smith CM. Crustose coralline algae can suppress macroalgal growth and recruitment on Hawaiian coral reefs. Marine Ecology Progress Series. 2011; 422: 1–7. https://doi.org/10.3354/meps08964
  125. 125. Tebben J, Motti CA, Siboni N, Tapiolas DM, Negri AP, Schupp PJ, et al. Chemical mediation of coral larval settlement by crustose coralline algae. Scientific Reports. 2015; 5: 1–11. pmid:26042834
  126. 126. Smith JN, Mongin M, Thompson A, Jonker MJ, De’ath G, Fabricius KE. Shifts in coralline algae, macroalgae, and coral juveniles in the Great Barrier Reef associated with present-day ocean acidification. Global Change Biology. 2020; 26: 2149–2160. https://doi.org/10.1111/gcb.14985
  127. 127. Newman SP, Meesters EH, Dryden CS, Williams SM, Sanchez C, Mumby PJ, et al. Reef flattening effects on total richness and species responses in the Caribbean. Journal of Animal Ecology. 2015; 84: 1678–1689. pmid:26344713
  128. 128. Ferrari R, McKinnon D, He H, Smith RN, Corke P, González-Rivero M, et al. Quantifying multiscale habitat structural complexity: a cost-effective framework for underwater 3D modelling. Remote Sensing. 2016; 8: 113. https://doi.org/10.3390/rs8020113
  129. 129. Ford AK, Eich A, McAndrews RS, Mangubhai S, Nugues MM, Bejarano S, et al. Evaluation of coral reef management effectiveness using conventional versus resilience-based metrics. Ecological Indicators. 2018; 8(5): 308–317. https://doi.org/10.1016/j.ecolind.2017.10.002
  130. 130. Urbina-Barreto I, Chiroleu F, Pinel R, Fréchon L, Mahamadaly V, Elise S, et al. Quantifying the shelter capacity of coral reefs using photogrammetric 3D modeling: From colonies to reefscapes. Ecological Indicators. 2021; 121: 107151. https://doi.org/10.1016/j.ecolind.2020.107151
  131. 131. Adjeroud M, Kayal M, Penin L. Importance of recruitment processes in the dynamics and resilience of coral reef assemblages. In: Rossi S, Bramanti L, Gori A, Orejas C, editors. Marine Animal Forests. Cham: Springer; 2017. pp. 549–569. https://doi.org/10.1007/978-3-319-21012-4_12
  132. 132. Graham NAJ, Nash KL. The importance of structural complexity in coral reef ecosystems. Coral Reefs. 2013; 32: 315–326. https://doi.org/10.1007/s00338-012-0984-y
  133. 133. Darling ES, Graham NA, Januchowski-Hartley FA, Nash KL, Pratchett MS, Wilson SK. Relationships between structural complexity, coral traits, and reef fish assemblages. Coral Reefs. 2017; 36: 561–575. https://doi.org/10.1007/s00338-017-1539-z
  134. 134. Pascoe KH, Fukunaga A, Kosaki RK, Burns JH. 3D assessment of a coral reef at Lalo Atoll reveals varying responses of habitat metrics following a catastrophic hurricane. Scientific Reports. 2021; 11: 1–9. https://doi.org/10.1038/s41598-021-91509-4
  135. 135. Alvarez-Filip L, Dulvy NK, Gill JA, Côté IM, Watkinson AR. Flattening of Caribbean coral reefs: region-wide declines in architectural complexity. Proceedings of the Royal Society B. 2009; 2 76: 3019–3025. pmid:19515663
  136. 136. Alvarez-Filip L, Carricart-Ganivet JP, Horta-Puga G, Iglesias-Prieto R. Shifts in coral-assemblage composition do not ensure persistence of reef functionality. Scientific Reports. 2013; 3: 3486. pmid:24336631
  137. 137. Urbina-Barreto I, Elise S, Guilhaumon F, Bruggemann JH, Pinel R, Kulbicki M, et al. Underwater photogrammetry reveals new links between coral reefscape traits and fishes that ensure key functions. Ecosphere. 2022; 13: e3934. https://doi.org/10.1002/ecs2.3934
  138. 138. Bayley DT, Mogg AO, Koldewey H, Purvis A. Capturing complexity: field-testing the use of ‘structure from motion’derived virtual models to replicate standard measures of reef physical structure. PeerJ. 2019a; 7: e6540. https://doi.org/10.7717/peerj.6540
  139. 139. Fukunaga A, Pascoe KH, Pugh AR, Kosaki RK, Burns JH. Underwater photogrammetry captures the initial recovery of a coral reef at Lalo Atoll. Diversity. 2022;14: 39. https://doi.org/10.3390/d14010039
  140. 140. Gough CLA, Dewar KM, Godley BJ, Zafindranosy E, Broderick AC. Evidence of overfishing in small-scale fisheries in Madagascar. Frontiers in Marine Science. 2020; 7: 317. https://doi.org/10.3389/fmars.2020.00317
  141. 141. McClanahan TR, Azali MK. Environmental variability and threshold model’s predictions for coral reefs. Frontiers in Marine Science. 2021; 1774. https://doi.org/10.3389/fmars.2021.778121
  142. 142. Obura D, Gudka M, Samoilys M, Osuka K, Mbugua J, Keith DA, et al. Vulnerability to collapse of coral reef ecosystems in the Western Indian Ocean. Nature Sustainability. 2021. https://doi.org/10.1038/s41893-021-00817-0
  143. 143. Mumby PJ, Hastings A, Edwards HJ. Thresholds and the resilience of Caribbean coral reefs. Nature. 2007; 450: 98–101. pmid:17972885
  144. 144. Todinanahary GGB, Behivoke F, Nomenisoa AL, Ravelojaona DK, Rakotoson AT, Tatangirafeno S, et al. Inventaire et étude de faisabilité de sites propices à l’algoculture, l’holothuriculture, la gestion de l’exploitation de poulpes et de crabes dans la région Atsimo Andrefana. MHSA—PRU (Contrat n 166/C/PIC2/2016) Report. Toliara, Madagascar; 2016.