When depth fails to prevent bleaching but limits coral death: insights from the 2019 heatwave in Mo’orea

When depth fails to prevent bleaching but limits coral death: insights from the 2019 heatwave in Mo’orea | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article When depth fails to prevent bleaching but limits coral death: insights from the 2019 heatwave in Mo’orea Claire Boitel, Yann Lacube, Alexandre Mercière, Caroline Dubé, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8725130/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 08 Mar, 2026 Read the published version in Coral Reefs → Version 1 posted 7 You are reading this latest preprint version Abstract Marine heatwaves (MHWs) are intensifying and increasingly threatening coral reef persistence by inducing widespread bleaching and mortality. Although depth is often proposed as a natural refuge, its protective role, especially in reducing post-bleaching coral loss, remains debated. This study quantified how depth (5, 12 and 20m) shaped bleaching and post-bleaching mortality and evaluated drivers of depth-mediated protection during the unprecedented severe 2019 MHW in Mo’orea (French Polynesia), across 6 sites. Despite thermal stress levels considered moderate according to global standards (3°C-weeks), the event caused unexpectedly severe impacts, with widespread bleaching resulting in the loss of half the coral cover. Bleaching was only weakly mitigated by depth with heightened protection (< 30% reduction) from 12 to 20m in sensitive taxa such as Acropora , Montipora and Pocillopora , not present at all in others, indicating refuge from bleaching was not universal. Mortality was concentrated in branching Acropora and Pocillopora , while other heavily bleached taxa showed no mortality. Relative total coral cover loss fell 3.5-fold from 45% at 5m to 13% at 20m, driving 3-fold weaker community shifts, revealing that depth more strongly limited post-bleaching mortality. This effect was partly mediated by differences in pre-bleaching community composition and coral cover, yet ~ 74% of the protective effect remained unexplained, suggesting that additional depth-related environmental or physiological mechanisms, such as light attenuation or colony size, were at play. These findings position deeper reefs as potential natural buffers, even in the absence of effective thermal stratification, and priorities for proactive conservation under accelerating climate change. Coral bleaching Depth Coral mortality Reef resilience Marine heatwave Community composition Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Introduction Mass coral bleaching events are becoming increasingly frequent and severe as marine heatwaves (MHWs) intensify due to climate change, posing a growing threat to the ecological integrity of coral reef ecosystems worldwide (Eakin et al. 2018 ; Hughes et al. 2018a ; Sully et al. 2019 ). Anomalous increases in temperatures, especially during summer months when irradiance is high, can readily break down the symbiosis between corals and their photosynthetic microalgae ( Symbiodiniaceae ), resulting in visible paling (Helgoe et al. 2024 ). Unless conditions improve, prolonged bleaching can result in extensive mortality, causing dramatic declines in coral cover, which can drive profound and long-lasting shifts in reef assemblages (Hughes et al. 2018b ; Watt-Pringle et al. 2022 ). Because branching corals can be especially thermally sensitive (Loya et al. 2001 ; Muir et al. 2017a ), reefs tend to taxonomically homogenize, losing important structural complexity and functional diversity (Graham et al. 2015 ; Denis et al. 2017 ; Pisapia et al. 2019 ). This can have cascading consequences on the associated biodiversity relying on corals for recruitment, shelter or food (e.g. fish), threatening the ecosystem services reefs provide (Lamy et al. 2016 ; Riegl et al. 2019 ; Robinson et al. 2019 ). When bleaching interacts with local stressors (e.g. pollution, cyclones, disease or predator outbreaks), recovery capacity of reefs can further decline, accelerating transitions toward degraded states (Kayal et al. 2012 ; Riegl and Purkis 2015 ). Over the past three decades, the Indo-Pacific has been particularly affected by repeated large-scale bleaching events causing substantial losses in coral cover and reef integrity (Bruno and Selig 2007 ; Souter et al. 2020 ). Bleaching, however, does not affect all coral communities equally. As coral species can have markedly different sensitivities to bleaching, vulnerability of a reef system during MHWs is highly reliant on its taxonomic composition (Loya et al. 2001 ; Van Woesik et al. 2011 ; Muir et al. 2017a , 2021 ). Even within a single reef or stress event, spatial variation in environmental conditions can lead to considerable differences in bleaching susceptibility and outcomes (McClanahan et al. 2020 ; Suggett and Smith 2020 ; Asner et al. 2022 ; McClanahan 2023 ). Among these environmental gradients modulating responses, depth consistently emerges as a key predictor influencing bleaching severity (Hoogenboom et al. 2017 ; Muir et al. 2017; Baird et al. 2018a ; Winston et al. 2022 ; Muñiz-Castillo et al. 2024 ). Along reef slopes, corals can experience rapid changes in light and temperature over short vertical distances, which can structure community composition (Schramek et al. 2018 ; Tamir et al. 2019 ). Because deeper sites typically offer cooler and dimmer conditions, bleaching often declines with depth for most species (Mumby et al. 2001 ; Bridge et al. 2014 ; Muir et al. 2017a ; Baird et al. 2018a ; Crosbie et al. 2019 ; Riegl et al. 2019 ; Pérez-Rosales et al. 2021 ). As such, mesophotic reefs (> 30m) have received growing attention for their hypothesized role in buffering shallow coral populations from climate change by helping prevent local extinction and supporting recovery through larval supply (the “deep reef refuge hypothesis”; Glynn 1996 ; Bongaerts et al. 2010 ; Smith et al. 2014 ). Nevertheless, this protection remains debated (Eakin et al. 2019 ). Some studies reported limited or inconsistent depth refuge during severe or prolonged heatwaves, especially when cooling mechanisms (e.g. upwelling, internal waves) did not coincide (Chollett et al. 2010 ; Neal et al. 2014 ; Frade et al. 2018 ), or local adaptation enhanced vulnerability at depth (Smith et al. 2016 ; Morais and Santos 2018 ; Eyal et al. 2022 ). Additionally, most studies have inferred depth refuge from patterns of initial bleaching prevalence alone (Muir et al. 2017a ; Baird et al. 2018a ; Frade et al. 2018 ; Pérez-Rosales et al. 2021 ; Winston et al. 2022 ; Diaz et al. 2023 ), while comparatively few have concurrently assessed post-bleaching mortality (Cowburn et al. 2019 ; Crosbie et al. 2019 ). Crucially, bleaching not always translate into mass mortality depending on past thermal exposure, local species assemblages or the magnitude of stressors experienced. Some reefs may extensively bleach and mostly recover, with minimal losses (e.g. >70% bleached but 50% mortality; Doherty et al. 2025 ). Therefore, bleaching prevalence alone may not reliably predict the ecological impacts of MHWs, while it is coral mortality that ultimately determines structural and functional loss, shaping the long-term trajectory of coral communities. Understanding not only the susceptibility of coral assemblages and taxa to bleaching, but also the variability in their subsequent survival, is essential to improve forecasting of reef futures (Cantin and Spalding 2018 ). Assessing the degree to which depth mitigates bleaching and its consequences is essential to efficiently evaluate the potential of deeper habitats to act as refuges, and to guide effective conservation strategies in the face of accelerating climate change. Surrounded by a steep outer slope that spans over mesophotic depths, the high volcanic island of Mo’orea in French Polynesia offers a valuable natural laboratory to investigate these dynamics. Over the past decades, the island’s reefs have been repeatedly affected by multiple large-scale disturbances, including severe bleaching events (1987, 1991, 1994, 2003), a major outbreak of crown-of-thorns starfish (2007–2009), and cyclones (2010; Gleason 1993 ; Hoegh-Guldberg and Salvat 1995 ; Penin et al. 2007 ; Adjeroud et al. 2009 ; Kayal et al. 2012 ; Pratchett et al. 2013 ; Lamy et al. 2016 ; Moritz et al. 2021 ). These successive and overlapping events nearly eradicated coral cover on the outer reef slope, reaching historically low levels by 2011, with most sites around the island approaching zero coral cover (Moritz et al. 2021 ). Despite the severity of this ecological collapse, coral cover steadily recovered over the following years, dominated by fast-growing branching Pocillopora taxa (Moritz et al. 2021 ). By 2019, reefs were still in their recovery phase and had not yet experienced acute thermal stress (Hédouin et al. 2020 ), when the island experienced a widespread and severe MHW, providing a unique opportunity to assess coral responses across depth gradients. In this study, we investigated the interplay between depth, taxonomic sensitivity, coral community composition, and environmental conditions in shaping coral susceptibility to bleaching and post-bleaching mortality. Spanning three depths (5, 12, and 20m), we quantified bleaching prevalence at the peak of the event and coral cover loss six months later. We specifically aimed to (i) quantify the prevalence and severity of coral bleaching (post-bleaching mortality) across depths, (ii) examine depth-related changes in coral communities, and (iii) assess potential mediators of the effect of depth, during the 2019 marine heatwave that struck Mo’orea. By integrating ecological surveys with thermal context, we assessed to which extent depth can provide functional refuge from bleaching and explored the ecological and environmental mechanisms underpinning these patterns. Materials and methods Study site and sampling design Mo’orea is a high volcanic island (17°30′S, 149°50′W) in the Society Islands archipelago of French Polynesia, enclosing a shallow lagoon bordered by a steep outer reef slope that drops off rapidly beyond the crest. The island’s reef system includes multiple bays and oceanic passes, generating strong hydrodynamic and ecological gradients around it (Adjeroud 1997 ). To monitor spatial and depth-related patterns of coral bleaching during the 2019 thermal anomaly, six sites were selected around the island based on prior knowledge of spatial variation in disturbance impacts and exposure regimes. One site was located on the north coast (Entre 2 Baies – E2B), two on the west coast (Gendron and Haapiti), and three on the east coast (Temae, Afareitu, and Maatea) (Fig. 1 ). These sites span a gradient of seasonal oceanic swell exposure: sites on the west coast are highly exposed to southwestern swells that dominate much of the year, while the northern and eastern sites are more sheltered from such exposure but receive low-energy northern swells between November and April (Laurent 2019 ). All sites are situated on the outer reef slope and share a similar geomorphology with spur-and-groove formations extending from the reef crest to beyond 30m depth. At each site, three depths (5, 12, and 20m) were surveyed, corresponding to upper- and mid-slope habitats. Prior to the 2019 event, all six sites were considered healthy, with coral cover typically exceeding 30% (Fig. S1 ). Temperature Satellite-derived sea surface temperature (SST) data were obtained from the NOAA Coral Reef Watch (CRW) 5-km resolution program for each of the six study sites during the 2019 bleaching event (Liu et al. 2017 ). The site-specific maximum monthly mean (MMM) temperature was extracted directly from CRW climatologies, ranging from 28.79°C to 28.81°C. Degree Heating Weeks (DHW 1°C ) were extracted for each site as the accumulated weekly temperature anomalies exceeding 1°C above the site-specific MMM (Liu et al. 2017 ). To assess the low-magnitude stress, we also calculated DHW with the same MMM climatology but with a cutoff at 0°C (when SST exceeds MMM) instead of the conventional 1°C cutoff (referred to as DHW 0°C ; Kim et al. 2019 ; Szereday et al. 2024 ). To place the 2019 thermal anomaly in historical context, one representative site was selected per coast – Haapiti (West), E2B (North), and Temae (East) – and the 2019 DHW values were compared to those recorded during previous bleaching years in Mo’orea (1987, 1991, 1994, 2003, 2016). Bleaching was observed in 2002 (Penin et al. 2007 ), yet satellite data from NOAA CRW did not register thermal stress (0 DHW). Daily SST values were averaged, and DHW was computed using the same threshold (1°C above MMM = 28.80°C). In addition to SST data, in-situ seawater temperature was also analysed from the Moorea Coral Reef Long Term Ecological Research program (Moorea Coral Reef LTER et al. 2025 ) to explore how thermal exposure varied along the depth gradient. Bottom-mounted temperature loggers deployed along the outer reef slope recorded continuous temperature every 20 minutes at depths of 10, 20, 30, and 40m. These records cover five of the six sites used in this study (Maatea was not included in the LTER network), and two of the 3 depths studied (10 and 20m). Due to instrument failure during the 2019 MHW, temperatures at Temae at 10 and 20m and at Gendron at 20m are not available. Unlike SST data, these in-situ records provide insight into thermal conditions and daily variability directly within reef habitats at and beyond the surveyed depths. From these data, two complementary thermal metrics were derived. Accumulated heat stress (AHS) was calculated using the same method as DHW but based on in-situ daily mean temperatures rather than SST (Claar et al. 2019 ). Weekly temperature means exceeding 1°C above the site-specific maximum monthly mean (MMM) were summed over a 12-week moving window to capture the cumulative stress experienced underwater. In-situ MMMs were computed for each depth and site as the average of daily means during the warmest month, based on the full temperature record from 2005 to 2023 (Frade et al. 2018 ). Site- and depth-specific MMM values are detailed in Table S1 . Additionally, we calculated the mean daily temperature fluctuation (MDTF), defined as the average daily temperature range over the 30 days preceding the onset of bleaching, which can have a strong mitigating effect on bleaching (Safaie et al. 2018 ). The bleaching onset was identified as March 28, 2019, corresponding to the first day when DHW exceeded zero. To assess whether thermal stress and daily variability differed across depths, we performed Kruskal-Wallis tests followed by pairwise Dunn’s post-hoc tests on AHS and MDTF values at 10, 20, 30, and 40m. Coral bleaching assessment To quantify coral bleaching during the peak of the 2019 thermal anomaly, surveys were conducted in mid-April. At each site and depth, 40 photo-quadrats (75cm x 75cm frame) were randomly photographed with an underwater camera (Sony RX100 with lights mounted on each side). Twenty quadrats were randomly selected at each depth and sites for analysis. Species of the same genera can be hard to distinguish in-situ , especially when they are bleached (Speare et al. 2022 ), and some, like species of pocilloporids, can show high phenotypic plasticity (Johnston and Burgess 2023 ). As such, individual coral colonies within each photo-quadrat were identified to genus following Bosserelle et al. ( 2014 ), except for fungiid corals. Accurate genus-level identification within the family Fungiidae requires access to both the upper and lower surfaces of the corallum as well as examination of septal dentition and costal spines for instance (Bosserelle et al. 2014 ). As these features were hardly discernible from our photographs, and to avoid misclassification, fungiid colonies were conservatively grouped at the family level (under the category Fungiidae ) and excluded from genus-specific analyses. Similarly, colonies of Napopora irregularis (the only species of Napopora recorded in French Polynesia) were not considered as a Porites species, despite having been synonymized with Porites irregularis (Hoeksema and Cairns, 2025). To this date, this species remains listed as taxon inquirendum , and showed clear morphological and ecological differences (e.g. branching growth form and high bleaching susceptibility) compared to the massive Porites species which were recorded under the Porites category (Bosserelle et al. 2014 ; Carlot et al. 2022 ). As such, Napopora colonies were kept as a distinct genus in our dataset (Meistertzheim et al. 2019 ; Carlot et al. 2022 ). Following identification, each colony was visually classified based on its health status into three categories: healthy (no visible paling), pale (partial bleaching, 10–60% of colony surface area presenting loss of pigmentation), and bleached (> 60%). The number of colonies per genus and bleaching category was recorded, totalling 9,535 colonies. Recently dead colonies in April (less than 2% of total), mostly Acropora , were classified as bleached. Post-bleaching mortality To assess depth-related coral mortality following the 2019 heatwave, a second photographic survey was conducted 6 months later, in October 2019 following the same design as in April. Due to logistical constraints, the exact same quadrats could not be relocated, and instead, 40 new photo-quadrats were randomly taken within the same sites at each depth, and 25 were randomly selected for analysis. For each quadrat, following recommendations (Van Rein et al. 2011 ), 75 random points were generated on PhotoQuad (Trygonis and Sini 2012 ), and benthic organisms beneath each point were identified. Categories included rubble, sand, bare substrate, calcareous coralline algae, macroalgae and coral. Corals were classified as either living (< 80% of colony partially dead), recently dead or old dead. Skeletal structure can remain distinguishable for more than 6 months post-mortem (Molina-Hernández et al. 2022 ; Morais et al. 2022 ). Thus, corals were considered recently dead when corallite architecture was identifiable and skeletal structure was preserved, with minimal algae overgrowth (Kramer 2003 ; Lirman et al. 2014 ; Molina‐Hernández et al. 2022; Morais et al. 2022 ). Conversely, old dead corals showed signs of full colonisation by encrusting organisms (e.g. coralline algae, macroalgae) and advanced structural degradation (Molina‐Hernández et al. 2022). Corals, whether dead or alive, were identified, when possible, to the genus level using the same methodology described for April quadrats. To maximise statistical power and enable quadrat-level resolution (n = 354), we chose to estimate changes in coral cover and composition using only post-bleaching quadrats, rather than comparing average values from independent April and October quadrats (which would have limited the analysis to n = 18 site x depth combinations). We estimated coral cover and community composition post-bleaching as the proportion of points falling on living corals over the cloud of 75 points. Pre-bleaching community and coral cover were inferred by summing the points of live and recently dead coral. Relative coral cover loss between April and October was computed at the community and genus level, for each quadrat, as a proxy of bleaching-induced mortality (Mumby et al. 2001 ; Lirman et al. 2014 ; Cantin and Spalding 2018 ; Cowburn et al. 2019 ; Hédouin et al. 2020 ). No major disturbances (e.g. crown-of-thorn predation scars, diseases, or storm damage) were reported at the study sites that were frequently visited between April and October. While we cannot fully exclude the possibility that some recent mortality was unrelated to bleaching, any such cases likely represent a very small fraction given the timing, scale (amount of colonies), and context of the surveys. Statistical analyses Data analyses were carried out using R v 4.4.2 (R Core Team 2024 ). The relationship between coral bleaching prevalence and depth was tested using generalized linear mixed models (GLMMs) with a binomial error distribution, in which the response variable was binary (0 = healthy; 1 = pale or bleached, i.e., a bleaching response). Depth was included as a fixed effect (categorical variable, to allow for flexible non-linear relationships), and site was included as a random effect to account for the sampling design. Analyses were conducted at two levels: (i) the community level, where genus was modelled as a random effect to account for differences in thermal sensitivity among coral genera; and (ii) the genus level (only genera with > 10 observations per depth were retained) where taxa was included as a fixed effect interacting with depth. Post-bleaching mortality was assessed through the relative loss in coral cover compared to pre-bleaching levels. Relative coral cover loss was modelled using GLMMs with a beta error distribution with a logit link (glmmTMB package; McGillycuddy et al. 2025 ), using the same fixed and random effect structure described above. One model was run at the community level (all genera combined), and a second set of models at the taxa level focused on genera studied in their bleaching response that exhibited mortality responses (> 5 observations). For each model, the best-fitting random effect structure was selected based on likelihood ratio tests compared to when random effects were removed (e.g., site or genus). To examine how depth-dependent sensitivity to bleaching translated into shifts in coral community structure 6 months after bleaching, we compared the composition of assemblages before and after the bleaching event using non-metric multidimensional scaling (nMDS) based on Bray-Curtis dissimilarity matrices (package vegan; Oksanen et al. 2025 ). The magnitude of community change was quantified for each assemblage (i.e. quadrat) as the Euclidean distance between its pre- and post-bleaching positions in nMDS space. The influence of depth on this shift was tested fitting a GLMM with a Gamma distribution with a log link (glmmTMB package), with depth as a fixed effect and site as a random effect; significance of site was tested via a likelihood ratio test. To test whether community composition differed significantly among depths and between time points (before vs. after bleaching), we used PERMANOVA with site as a stratification factor (to constrain permutations within sites). Pairwise differences in composition were then explored using pairwiseAdonis (Martinez Arbizu 2020 ), both between depths and between time points at each depth. Type II Wald F tests with Kenward-Roger degrees of freedom approximations were run on each model to quantify the relative importance (significance) of each main effect (depth) and its interaction with taxa when included. Post-hoc pairwise Tukey-adjusted tests were performed with the emmeans package (Lenth 2025 ), between depths, and taxa when pertinent at the population-level (when site as random effect was significant). When random effects were retained in GLMMs, bias adjustment was applied to predicted values and marginal effects to account for Jensen's inequality, which can otherwise lead to biased estimates due to the non-linearity of link functions in generalized models. Finally, to determine which variables mediate the effect of depth on overall relative coral cover loss, we used a multiple mediation analysis (MMA) framework implemented in the mma R package (Yu and Li 2017 ). This method allows for the decomposition of the total effect of a predictor (here, depth as a continuous variable) into a direct effect (the part of the effect not explained by other variables) and indirect effects (the part mediated through other variables (VanderWeele and Vansteelandt 2014 ; Correia et al. 2025 ). We tested to what extent the effect of depth on coral mortality was mediated by depth-related changes in (1) pre-bleaching coral community composition and (2) initial coral cover prior to bleaching. Effect of community composition was quantified as the independent but also joint effect of the relative abundance of dominant taxa, which represented > 5% of total coral cover. To account for potential non-linear relationships, we allowed quadratic terms for the predictor and mediators. The model estimated the relative contribution of each indirect pathway (i.e. through community composition or initial cover) to the total effect of depth on coral cover loss. Confidence intervals (95%) for all estimates were obtained via bootstrapping (sampling was repeated 5,000 times). Results Thermal stress patterns and intensity during the 2019 bleaching event The 2019 marine heatwave (MHW) impacted all sites around Mo’orea. Satellite-derived Degree Heating Weeks (DHW) reached values ranging from 2.9°C-weeks at Haapiti (southwest coast) to 3.4°C-weeks at Entre 2 baies (E2B, northern coast), with southern sites generally experiencing slightly lower thermal stress (Fig. S2a). Sea surface temperatures (SSTs) exceeded the maximum monthly mean (MMM; 28.8°C) mid-December 2018 and hovered between MMM and MMM + 1°C until late March (Fig. S2b). A sharp thermal spike in early April pushed SSTs above bleaching thresholds (MMM + 1°C), triggering rapid DHW accumulation (> 2.5°C-weeks within 2 weeks) before plateauing through mid-June (Fig. S2a). SSTs thus remained above MMM for nearly 5 months (Fig. S2b). Thermal stress was higher than all previous bleaching years to date, reaching 3.2°C-weeks, representing the most intense MHW ever recorded on the island (Fig. 2 b). It surpassed the last two major bleaching events in 1994 and 2003 (both ~ 2.8°C-weeks) and was far above the last event in 2016 (< 1°C-week; Fig. 2 b). April 2019 was particularly anomalously warm, with SSTs frequently exceeding 30°C, unlike previous bleaching years (Fig. 2 a). In-situ temperature data recorded along the outer reef slope at depths of 10 to 40m confirmed strong vertical thermal gradients (Fig. 3 ; S3). While upper- and mid-reef slopes (10-20m) experienced intense and sustained heat stress, deeper zones (30-40m) were partially thermally decoupled and exposed to cooler, more variable conditions following the thermal spike. Accumulated heat stress (AHS), assessed against long-term temperatures at each depth, reached 3.14 ± 0.37°C-weeks (mean ± sd) at 10m and 2.84 ± 0.42°C-weeks at 20m (no significant difference, p = 0.381; Table S1 ). Beyond 20m, AHS declined sharply by more than half (p < 0.05), dropping to 1.23 ± 0.54°C-weeks at 30m and to similar levels at 40m (0.96 ± 0.49°C-weeks, p = 0.397). Conversely, daily temperature variability over the 30 days preceding bleaching onset (MDTF) increased with depth: from averaging 0.41 ± 0.09°C at 10m to 0.55 ± 0.14°C at 20m (no significant differences, p = 0.325), and then more than doubled to 1.05 ± 0.19°C at 30m and 1.82 ± 0.19°C at 40m (p < 0.05; Table S1 ). This increase in MDTF was primarily driven by increasingly cooler daily minima with depth, a pattern observed across all sites and extending from October to early May at 30m, and until early June at 40m (Fig. S3). These fluctuations coincided with the timing of bleaching onset and confirm patterns observed in SST data, notably the sharp increase in temperature in early April across all depths. Bleaching susceptibility Half of coral colonies were bleached (i.e. across all coral genera) at both 5m and 12m depths (50.0 ± 0.7% – emmeans ± se – and 50.2 ± 0.9%, respectively; p = 0.120; Fig. 3 a-b). Although bleaching prevalence at 20m was statistically lower (p < 0.001), the decrease was marginal (48.9 ± 2.4%; Fig. 4 a, S5). Overall, bleaching was primarly structured by site-level variability and genus-specific differences in susceptibility. Depth alone explained < 1% of the observed variance in bleaching prevalence (marginal R 2 ), while the inclusion of site and taxa as random effects increased the explained variance to 53% (conditional R 2 ; Table S3). Bleaching happened island-wide, with substantial bleaching recorded at all sites (> 45%, averaged across depths; Fig. S4). Bleaching susceptibility varied markedly among taxa (genus effect: \(\:\chi\:\) 2 (9) = 1371.94, p < 0.001). Eighteen coral genera were recorded across all quadrats, though 4 were only observed at 20m and in very low abundance (< 10 individuals, Pachyseris , Goniastrea , Leptoseris , and Cyphastrea ; Fig. S5). Only eight were retained for taxa-level susceptibility analysis, each with at least 10 observations per depth: Acropora , Pocillopora , Montipora , Leptastrea , Millepora , Porites , Pavona , and Astrea (Fig. 4 b). Acropora , Astrea , and Millepora were the most susceptible to bleach, with bleaching prevalence of 93.8 ± 1.1% (emmeans ± se), 91.3 ± 1.6%, and 87.6 ± 3.9% respectively (averaged across all depths; no significant differences among them, p > 0.05; Fig. S7). Montipora showed slightly lower prevalence (87.9 ± 1.9%) than Acropora (p = 0.002), but did not differ from Astrea and Millepora (Fig. S7). Leptastrea (63.8 ± 5.4%) was significantly less susceptible than these four taxa, and bleached more than Pocillopora , which exhibited an intermediate and milder response (45.0 ± 3.3%, all pairwise < 0.01; Fig. S7). Pavona and Porites were the most resistant taxa, exhibiting minimal and mostly partial bleaching (0.1 ± 0.1% and 7.2 ± 1.70% of colonies were bleached respectively, all pairwise < 0.05; Fig. 4 b). Bleaching severity also varied among taxa (genus effect: \(\:\chi\:\) 2 (11) = 33.91, p 60% loss of pigmentation). In Astrea and Millepora , bleaching was milder as only 40–50% of bleached colonies were severely depigmented. Bleaching was rather partial in Leptastrea (< 20%). Reponses to depth differed substantially between taxa (effect depth x genus: \(\:\chi\:\) 2 (18) = 207.46; p < 0.001; Fig. 4 ). Both Acropora and Astrea displayed no difference in bleaching susceptibility between 5m and 12m (94 to 97%; p = 0.688 and p = 0.855). Bleaching prevalence significantly decreased at 20m for both, though by only 11–15%, remaining high overall (85.4 ± 3.2% and 79.6 ± 3.4%, respectively at 20m, both p < 0.001; Fig. S6). Likewise, bleaching in Montipora was only slightly mitigated by 6% from 5 to 12m (from 95.5 ± 1.3% at 5m to 89.7 ± 2.0% at 12m, p = 0.011), but was more strongly reduced (-24%) from 12 to 20m (68.4 ± 4.1% at 20m, p < 0.001; Fig. S6). Bleaching in Pocillopora slightly peaked at intermediate depths (12m: 51.7 ± 3.5%; +5% compared to 5m, p = 0.008) before decreasing by 29% to 36.8 ± 3.4% at 20m (p < 0.001). Conversely, bleaching prevalence increased with depth for Porites and Pavona . Porites bleaching increased 8-fold from 2.5 ± 1.4% at 5m to 20.2 ± 3.3% at 20m (p < 0.001), while Pavona did not bleach at 5m but reached 17.9 ± 2.8% at 20m (p < 0.001; Fig. 4 b, S6). Pavona also tended to bleach more severely beyond 12m (Table S4). Bleaching prevalence remained high across all depths in Millepora (~ 50%) and Leptastrea (~ 90%; no significant differences, all pairwise > 0.05; Fig. 4 b, S6). Finally, the relative proportion of partially vs severely bleached colonies remained broadly stable with depth (Fig. 4 , Table S4). Depth-dependent coral cover loss The effect of depth on coral mortality was notably more pronounced than its effect on bleaching susceptibility (depth explained 41% of coral cover loss variability; Table S3). Among the most abundant taxa analysed for bleaching, only Pocillopora and Acropora had experienced significant and severe mortality by October (p 0.05; Table S5). Together they accounted for 96% of the coral loss at 5m (85% at 20m; Fig. 6 ). The taxa-level mortality analysis thus focused on these two genera, which exhibited different depth responses (genus effect: \(\:\chi\:\) 2 (1) = 43.980; p < 0.0001; depth x genus interaction: \(\:\chi\:\) 2 (2) = 16.944; p 35% at 5m to ~ 22% at 20m, while Acropora was rarer (Fig. 5 ), with cover decreasing from ~ 7% at 5m to 4% at 20m. Following the bleaching event, Acropora experienced the highest losses, with cover reduced by 73.9 ± 3.5% at 5m (relative to pre-bleaching levels, p < 0.001; Fig. S8). Mortality was less severe for Pocillopora at the same depth, with coral cover declining by 46.5 ± 3.8% (p < 0.001; Fig. S8, Table S7). While both taxa exhibited reduced mortality with depth, the protective effect of depth was stronger for Acropora (Table S7). Between 5 and 12m, coral mortality significantly declined by 21–27% for both taxa (12m: 54.5 ± 4.2% for Acropora , p < 0.001; 35.4 ± 3.4% for Pocillopora , p = 0.022; Fig. 5 , S8). From 12 and 20m, however, Acropora relative coral cover loss decreased by 49% (to 29.4 ± 4.1% i.e. -63% relative to 5m; p < 0.001), while it declined by only 23% in Pocillopora (to 27.7 ± 3.1% i.e. -39% relative to 5m; p < 0.001; Fig. 4 ). In October, Pocillopora cover had dropped below 20% at all depths. As Pocillopora dominated the reef, representing on average 68 ± 25% of the total pre-bleaching live cover (mean ± sd across all depths; Fig. 6 ), these losses contributed disproportionately to the overall decline in coral cover (Fig. 5 ). Before bleaching, coral cover was high, exceeding 30% across all depths, declining from 45.9 ± 3.2% (emmeans ± se) at 5m, to 39.8 ± 3.1% at 12m and to 30.8 ± 2.8% at 20m (all pairwise before < 0.001; Table S5, S6). Following the bleaching event, roughly half of the coral cover was lost at 5m (45.0 ± 2.4% relatively to pre-bleaching levels; emmeans ± se), which was nearly twice the loss recorded at 12m (26.3 ± 2.0%, p < 0.001), and 3.5 times the loss at 20m (12.6 ± 1.2% i.e. 52% less loss relative to 12m, p < 0.001; Fig. S8, Table S4). Post-bleaching coral cover thus ranged from 23 to 27% across depths (post-bleaching pairwise all significant, p < 0.03; Table S5). Coral cover was especially high in E2B at 5m, towering to 75% prior to bleaching, while most sites did not exceed 50% (Fig. S1 , S9). It was also the most dominated by pocilloporids (> 93%; Fig S10) and the most affected by coral mortality, losing on average 61% of its cover, reducing coral cover to about 30% (Fig. S9). Despite such an important decline, E2B remained the site with the highest live coral cover following bleaching (Fig. S9). Changes in coral assemblages Before the bleaching event, coral assemblages along the reef slope and across sites were broadly similar in composition, dominated by Pocillopora , with variable but lower contributions from Acropora (7 ± 12% – mean ± sd), Porites (7 ± 15%), and Montipora (6 ± 12%), which together accounted for nearly 90% of overall coral cover (Fig. 6 , S10). Pocillopora accounted for 75 ± 21% at 5m to 60 ± 29% at 20m of the coral assemblage. Acropora slightly peaked at 12m (9 ± 13%) compared to 5m and 20m (7 ± 12% and 6 ± 11%, respectively). Montipora was more abundant at 5m (8 ± 13%) than at 12m or 20m (both ~ 5%), while Porites increased markedly with depth, from 3 ± 5% at 5m to nearly four times as much at 20m (11 ± 20% at 20m, and 8 ± 14% at 12m; Fig. 6 ). Consequently, depth was found to be a strong predictor of pre-bleaching assemblages as they differed significantly among depths (PERMANOVA pairwise comparisons before bleaching, all p = 0.001 in Table S4). Table 1 PERMANOVA results on coral assemblages: effects of depth and bleaching combined Variable Df SumOfSqs R 2 F p-value Bleaching 1 5.79 0.04 36.79 0.001 *** Depth 2 4.49 0.03 14.28 0.001 *** Bleaching:Depth 2 1.92 0.01 6.11 0.001 *** Residual 862 135.59 0.92 Total 867 147.80 1.00 These initial differences, coupled with genus-specific sensitivities to bleaching and mortality, led to depth-dependent shifts in community structure. Bleaching significantly altered community composition across depths (PERMANOVA depth x time = 5.85, p = 0.001; Table 1 ), although it accounted for a small fraction of the variance (R 2 ~ 4%; Table 1 ). Changes in reef composition were most pronounced at 5m and least at 20m, with intermediate impact at 12m. This pattern was supported by both PERMANOVA effect sizes (R 2 from pairwise PERMANOVA after-before; Table S8) and changes in community structure in nMDS space (Fig. 7 ). The average distance in nMDS space crossed by assemblages because of bleaching was more than 3 times greater at 5m (0.62 ± 0.11; emmeans ± se) than at 20m (0.18 ± 0.04, p < 0.001) and twice as high at 12m (0.32 ± 0.06, p = 0.028; Table S9). Several quadrats at 20m showed little to no change in composition (Fig. 7 ). Changes were largely driven by declines in Pocillopora , which was strongly negatively correlated with the primary axis of compositional change across depths (R 2 = 74%; Fig. S11). Drivers of depth-mediated protection To identify processes underlying the attenuation of coral mortality with depth, we conducted a multiple mediation analysis (MMA) to test whether the effect of depth on relative coral cover loss (of the assemblage) was mediated by initial coral community composition (i.e. the relative abundance of bleaching-susceptible versus bleaching-tolerant taxa) and pre-bleaching coral cover, both of which declined along the depth gradient (Fig. 5 , 6 , S1, Table S6). Pre-bleaching community composition was represented by the joint effect of the relative abundance of dominant taxa, Pocillopora, Acropora, Montipora, and Porites , each accounting for > 5% of the total coral cover. The total effect of depth on relative coral cover loss was significant and negative (-2.074 [95% CI: -2.445 ; -1.696]; Table 2 ), confirming that coral mortality decreased with increasing depth. This attenuation was partly mediated by the two ecological variables tested. Initial coral community composition accounted for 13.0% [0.4 ; 24.4] of the effect of depth (indirect effect = -0.270 [-0.516 ; -0.007]), while pre-bleaching coral cover explained 13.4% [5.0 ; 20.7] (indirect effect = -0.279 [-0.443 ; -0.098]). Combined, these two mediators significantly explained 26.5% [11.5 ; 38.6] of the reduction in coral mortality with depth (Table 2 ). However, the majority of the effect (73.5% [61.4 ; 88.5]) remained unexplained by these variables (Table 2 ). Table 2 Summary of the multiple mediation analysis looking at the effect of depth (as numeric) mediated by coral composition and coral cover before bleaching on the relative coral cover loss (in percentage). Variables in bold have significant indirect/direct effect. CI intervals (in brackets) come from 5000 bootstrap samples. Variable Effect (95% CI) Relative effect (%) Indirect Joint effect pre-bleaching composition -0.270 (-0.516 ; -0.007) 13.0 (0.4 ; 24.4) Initial coral cover -0.279 (-0.443 ; -0.098) 13.4 (5.0 ; 20.7) All mediators -0.549 (-0.829 ; -0.219) 26.5 (11.5 ; 38.6) Direct Depth -1.525 (-1.930 ; -1.172) 73.5 (61.4 ; 88.5) Total Depth -2.074 (-2.445 ; -1.696) Discussion An unprecedented heatwave resulting into widespread bleaching and mortality The 2019 MHW was the most intense thermal stress event ever recorded in Mo’orea, surpassing all previously documented bleaching-inducing MHWs in 1987, 1991, 1994, and 2003. It was marked by an abrupt onset, with ~ 3°C-weeks of thermal anomalies accumulating in just two to three weeks. This acute heat stress triggered widespread coral bleaching across the island, with roughly 50% of all coral colonies showing visible signs of bleaching, and a resulting loss of half of the coral cover on shallow reefs. These figures rank among the highest levels of bleaching and mortality ever recorded on the island (Gleason 1993 ; Hoegh-Guldberg and Salvat 1995 ; Penin et al. 2007 ; Carroll et al. 2017 ; Hédouin et al. 2020 ; Speare et al. 2022 ). Bleaching was even so severe that reefs shifted from carbon sinks to carbon sources (Seabrook et al. 2023 ). What makes this event particularly striking is that such severe ecological impacts were observed under what would conventionally be considered “moderate” thermal stress levels (≤ 4°C-weeks; Liu et al. 2003 ). With a DHW barely exceeding 3°C-weeks, the 2019 event caused coral mortality levels that are typically only seen under much higher DHW values. For instance, on the Great Barrier Reef in 2016, coral cover remained mostly stable for DHW values below 3°C-weeks and only declined by 40% at 4°C-weeks, with > 80% losses only occurring at DHW ≥ 9°C-weeks (Hughes et al. 2018b ). Similarly, aerial surveys in Hawai’i reported 50% coral mortality only above 10°C-weeks (Asner et al. 2022 ), while mortality of 42% of Acropora required 11°C-weeks in Japan (Sakai et al. 2019 ). In Little Cayman, a DHW of 17 in 2023–2024 led to 80% of colonies bleaching and 54% mortality, mostly in sensitive corals (Doherty et al. 2025 ). In contrast, Mo’orea’s 2019 heatwave caused similar or greater bleaching and mortality with a much lower heat exposure, strongly suggesting a heightened vulnerability of coral communities. This amplified sensitivity likely reflects the thermal naivety of Mo’orea’s coral assemblages. The reef was still recovering from the massive ecological collapse caused by the 2007–2009 crown-of-thorn outbreak and Cyclone Oli in 2010, which had reduced coral cover on the outer slope to near zero by 2011 (Moritz et al. 2021 ). While coral communities had steadily rebuilt in the following years, they had not experienced any significant thermal stress during this recovery. The only recent heat event prior to 2019 was the 2016 bleaching, which, although thermally mild (DHW < 1°C-week), still triggered bleaching in 77% of coral colonies – particularly in Astrea and Pocillopora , while Acropora remained largely unaffected (Hédouin et al. 2020 ). In Mo’orea, the 2019 heatwave thus acted on a reef that had not yet experienced acute thermal stress, resulting in disproportionately high mortality for a moderate DHW. These findings highlight the limitations of DHW as a standalone predictor of bleaching severity and support the need to refine how thermal stress is quantified, accounting for local ecological context, including disturbance history, recovery dynamics and species composition. Incorporating metrics that account for prior thermal exposure, low-magnitude but persistent heat stress, and the influence of additional global and local stressors may significantly improve predictive accuracy, as suggested by recent studies (Heron et al. 2016 ; Pereira et al. 2022 ; Szereday et al. 2024 ; Whitaker and DeCarlo 2024 ). In our case, lowering the DHW threshold from + 1°C to + 0°C allowed us to better capture the period of oscillating thermal anomalies between the MMM and MMM + 1°C in the weeks leading up to bleaching onset. As a result, cumulative heat stress measured using DHW 0°C was more than twice that measured using the standard 1°C cutoff (DHW 1°C ), aligning more closely with the severity of observed bleaching patterns. Long-standing patterns of bleaching sensitivity The taxonomic hierarchy of bleaching susceptibility observed during the 2019 event closely mirrors long-standing patterns previously documented in Mo’orea and across the Pacific and Indian Oceans (Loya et al. 2001 ; Penin et al. 2007 ; Van Woesik et al. 2011 ; Guest et al. 2012 ; Pratchett et al. 2013 ; Carroll et al. 2017 ; Muir et al. 2017a ; Baird et al. 2018a ; Crosbie et al. 2019 ; Winston et al. 2022 ). Consistent with earlier studies (Penin et al. 2007 ; Carroll et al. 2017 ; Hédouin et al. 2020 ), Acropora , Montipora , and Astrea emerged as the most sensitive genera, each exhibiting bleaching prevalence exceeding 90%. Pocillopora showed intermediate susceptibility, while Porites and Pavona remained the most resistant, with minimal to no signs of bleaching. These findings suggest that the characteristic gradient of taxon-specific sensitivity has re-established on Mo’orea’s reefs, despite the major ecological reset and subsequent community rebuilding that followed the crown-of-thorns outbreak and cyclone in the late 2000s (Hédouin et al. 2020 ; Moritz et al. 2021 ). Interestingly, some genera appeared more vulnerable in 2019 than during previous bleaching events. For instance, on average ~ 50% of Pocillopora colonies showed signs of bleaching in this study, compared to only ~ 20–30% during the successive 2002–2003 mass bleaching events, while Acropora not only reached higher bleaching levels (> 93%) but also showed a more severe response (mostly fully bleached compared to ~ 80%, primarily partially bleached colonies in 2003; Carroll et al. 2017 ). These heightened responses in 2019 compared with 2002–2003, even though marine heatwaves appeared similar in force (DHW ~ 3°C-weeks), likely reflect the lack of recent thermal exposure prior to 2019 and support the hypothesis that recurrent and successive bleaching events seemingly drive acclimation or the selective removal of more susceptible genotypes (Guest et al. 2012 ; Pratchett et al. 2013 ) – mechanisms which had likely not yet acted at full strength in Mo’orea’s recovering coral assemblages. While bleaching susceptibility followed a well-defined taxonomic hierarchy, post-bleaching mortality did not scale proportionally with bleaching prevalence. Indeed, although at least six genera exhibited widespread and often severe bleaching, only two experienced substantial post-bleaching mortality. Averaged across depths, Acropora lost half of its coral cover, indicating a particularly high risk of bleaching-induced death. Although Pocillopora showed intermediate susceptibility, with roughly 45% of colonies bleaching, it experienced disproportionately higher mortality, losing on average ~ 37% of its coral cover, more closely tracking bleaching prevalence than Acropora did. In contrast, Montipora , Astrea , Millepora and Leptastrea that experienced widespread bleaching recorded little to no mortality. Coral cover of these taxa, as well as that of Porites and Pavona – despite enhanced bleaching with depth – remained largely unchanged, indicating high resilience. As such, high bleaching prevalence does not always correlate with high mortality, even in highly susceptible taxa (Sampayo et al. 2008 ; Crosbie et al. 2019 ; Banha et al. 2020 ; Hédouin et al. 2020 ; Page et al. 2023 ), highlighting instead strongly taxon- and context-dependent recovery potential. Documenting local coral assemblage composition and historic patterns of bleaching susceptibility of taxa, even species when possible, thus appears highly important to be able to robustly predict demographic outcomes following increasingly intense and frequent MHWs (Hughes et al. 2018b ). Depth weakly mitigated bleaching but strongly limited mortality Across the 2019 marine heatwave, depth up to 20m provided only limited protection against bleaching initiation but instead strongly reduced bleaching-induced mortality. Bleaching prevalence showed little to no reduction between 5 and 12m (-6% to + 5%) across dominant and sensitive coral genera, with a more pronounced, yet still moderate, decline observed between 12 and 20m for Acropora , Astrea , Montipora and Pocillopora . This attenuation ranged from − 11% in Acropora to -24-29% in Montipora and Pocillopora . Despite this decline, bleaching prevalence remained high at all depths, exceeding 70% for Acropora , Astrea and Montipora , and remaining above 35% for Pocillopora . Physiological thresholds for bleaching were thus widely exceeded throughout the depth range surveyed and bleaching-inducing stress was hardly alleviated for most sensitive taxa (Helgoe et al. 2024 ), if at all in Leptastrea and Millepora , which showed no reduction in bleaching probability with depth. As bleaching likelihood increased with depth in Porites and Pavona , bleaching prevalence at the community level was not substantially reduced from 5 to 20m, with half of corals bleaching overall, more than three-quarters of which were severely affected. Conversely, the modest reduction in bleaching prevalence was mirrored by a far more substantial and linear decline in post-bleaching mortality with increasing depth. Acropora suffered near-total mortality at 5m (-74% coral cover) and still exhibited substantial loss at 20m, though divided by more than half (-29%), and mortality was reduced by 39% in Pocillopora (from − 47% to -28% between 5 and 20m). Reduction of bleaching and mortality with depth were unlikely to have resulted from depth-related thermal relief along the 5-20m gradient. In-situ temperature records revealed no significant differences in thermal intensity or variability between 10 and 20m during the 2019 heatwave. This absence of thermal stratification is consistent with the anticyclonic conditions that prevailed during the austral summer, which substantially reduced internal wave cooling (IWC; Wyatt et al. 2023 ). While internal waves can normally cool waters as shallow as ~ 10m (e.g. average depth of the 29.8°C bleaching isotherm was 15m in 2016), this threshold deepened to ~ 28m in 2019, effectively confining thermal relief to depths below those surveyed (Wyatt et al. 2023 ). Under these conditions, depth-related mitigation of bleaching, and especially of mortality, likely stemmed primarily from light attenuation (Baird et al. 2018b ). Reduced irradiance can limit photosystem saturation and bleaching-inducing oxidative stress, particularly under thermal stress when repair mechanisms are impaired (Baird et al. 2018b ; Helgoe et al. 2024 ). Notably, most of the reduction in bleaching associated with light attenuation is known to occur at very shallow depths, typically between <7m, where irradiance declines most steeply (Baird et al. 2018a , 2018b ; Laverick et al. 2020 ; Asner et al. 2022 ). By starting at 5m, our surveys likely missed the portion of the depth gradient where light-driven mitigation of bleaching is strongest. At this depth, UV radiation were probably already strongly attenuated (Dunne and Brown 1996 ; Kuwahara et al. 2010 ; Downs et al. 2013b ), suggesting depth-related refuge within our study range primarily reflected reductions in photosynthetically active radiation (PAR). Therefore, consistent with limited thermal relief, studies from the northern Great Barrier Reef during the 2016 mass bleaching reported similarly limited depth refuges because of limited upwelling activity. Bleaching declined modestly from 69% at 5-10m to 60% at 25m (Frade et al. 2018 ), and for Acropora specifically, little change was observed between 5 and 10m, followed by a ~ 24% decline between 10 and 20m (Crosbie et al. 2019 ). On the contrary, regions where IWC extended into shallower waters during MHW, such as the Maldives in 2016, exhibited much stronger depth refuge, with bleaching risk declining by ~ 23% between 3–5 m and 10m and by up to 60% between 3–5 m and 24–30 m (Muir et al. 2017b ; Cowburn et al. 2019 ). Together, these regional differences suggest that strong depth refugia from bleaching emerge most clearly when light attenuation and thermal relief act synergistically, typically beyond the upper 20m of the water column. It is therefore likely that surveying greater depths would have revealed stronger mitigation of both bleaching and mortality, potentially leading to negligible mortality for Acropora and Pocillopora below 20m (Muir et al. 2017a ; Crosbie et al. 2019 ). Beyond physical drivers, depth zonation of species likely modulated how taxa responded to depth-related gradients. While our analyses were conducted at the genus level, we acknowledge that such taxonomic resolution can obscure important mechanisms underlying depth-related reductions in bleaching impacts. Substantial variability in bleaching susceptibility and survival has been documented between species from the same genera, even within highly sensitive taxa like Acropora (Marshall and Baird 2000 ; Muir et al. 2017a , 2021 ). Depth-related shifts in species composition, such as differences in the relative abundance of shallow specialists versus depth generalists (Bridge et al. 2014 ; Roberts et al. 2015 ; Pérez-Rosales et al. 2022 ), can therefore influence both the probability of bleaching and subsequent mortality within genera. In reefs where depth-sensitive species have restricted vertical distributions, sharp mortality thresholds have been observed, as illustrated by the near-complete disappearance of Acropora mortality beyond 8m during the 2010 heatwave in Pulau Weh (Bridge et al. 2014 ). In Mo’orea, the more gradual decline in mortality observed here, with no clear inflexion point, suggests broader vertical distributions of vulnerable species, resulting in smoother depth-related patterns. Although depth-related alleviation of bleaching-inducing stressors was insufficient to strongly prevent bleaching onset in most taxa, the steep decline in mortality suggests that depth primarily limited the severity of physiological damage after bleaching occurred. Light attenuation with depth likely constrained further accumulation of oxidative damage following bleaching, thereby enhancing recovery and survival (Brown et al. 2000 ; Downs et al. 2013a ; Coles et al. 2018 ). Photo-physiological acclimation with depth may have further contributed to both reduced bleaching severity and enhanced survival at depth, particularly through changes in endosymbiont community composition (Bongaerts et al. 2015 ; Ezzat et al. 2017 ; Wall et al. 2020 ). An increasing dominance of Cladocopium symbionts with depth has been linked to higher carbon translocation and assimilation by the coral host (Ezzat et al. 2017 ; Wall et al. 2020 ) and to higher heterotrophic plasticity, even in corals from mesophotic depths (~ 50m; Ezzat et al. 2017 ). Such traits may have conferred a competitive advantage to deeper corals, possibly better energetically equipped to withstand stress, especially if it was concurrently reduced with increasing depth. Given that depth was identified as the strongest environmental driver of Symbiodiniaceae composition in Montipora capitata across a much shallower gradient (0.5–3.5 m; De Souza et al. 2023 ), its influence is likely even more pronounced for some taxa across the broader depth range examined here. Despite trophic plasticity, depth can strongly slow growth, particularly in branching corals, usually resulting in smaller colonies with increasing depth for a given age (Kramer et al. 2020 ). Colony size has repeatedly been shown to influence both bleaching susceptibility and mortality, with larger colonies of Acropora and Pocillopora bleaching more frequently and experiencing higher mortality (Speare et al. 2022 ; Winslow et al. 2024 ). Notably, both studies analysed the same 2019 bleaching event in Mo’orea, indicating that if colony size was effectively reduced at depth, it likely contributed jointly to the observed reduction in bleaching susceptibility and increased resilience in these taxa. Ultimately, differences in resilience among coral genera, combined with their abundance and depth distributions, mediated overall coral cover loss. Depth can strongly structure coral community composition through associated thermal and light gradients (Tamir et al. 2019 ; Laverick et al. 2020 ; Pérez-Rosales et al. 2022 ), a pattern that was also evident here prior to bleaching. Resistant genera such as Porites and Pavona increased in relative abundance with depth, whereas highly sensitive taxa ( Acropora , Pocillopora , Montipora ) declined. Such depth distributions concurrently with the strong depth-related reduction in mortality of dominant Pocillopora resulted in a 42% lower coral cover loss at 12m compared to 5m, followed by a further 52% decline between 12 and 20m. Changes in pre-bleaching assemblage structure alone explained 13% of this protective effect, increasing to 31% when combined with initial coral cover. To our knowledge, this provides one of the first quantitative estimates of the indirect influence of ecological context on post-bleaching mortality. Our results are consistent with earlier findings from the 2002 bleaching event in Mo’orea, where the composition of coral assemblages was hypothesized to drive patterns in bleaching across depths (Penin et al. 2007 ). At that time, higher bleaching incidence at 18m compared to 6m was attributed to the greater relative abundance of sensitive taxa at depth. However, the ecological reset following the 2007–2010 disturbances profoundly restructured coral assemblages, with sensitive taxa now decreasing and resistant taxa increasing with depth, reversing pre-disturbance patterns. As a result, assemblage structure alone can no longer account for the depth-related bleaching and mortality patterns observed in Mo’orea during the 2019 heatwave. Instead, while thermal refuge was constrained here, contemporary depth responses were more likely driven by light gradients, that may have limited bleaching onset and more thoroughly ensuing mortality. Implications for coral reef monitoring and conservation The taxon- and depth-specific mortality patterns identified in this study have important implications for coral reef resilience and management. In Mo’orea, where pocilloporids were a dominant component of reef assemblages, abrupt losses of these ecologically important branching corals are likely to have disproportionately large consequences for reef architecture, habitat availability and associated biodiversity. Such impacts were probably most pronounced on the northern coast, particularly at E2B, where exceptionally high coral cover (> 75%) might have correlated with higher reef fish abundance and richness (Darling et al. 2017 ; Russ et al. 2021 ). In this context, reduced mortality observed at depth may contribute to preserving reef fish assemblages over the long term (Crosbie et al. 2019 ). The persistence of relatively undisturbed deeper coral assemblages – three times less affected at 20m than at 5m – also highlight their potential role as sources of larvae to replenish impacted shallow zones (Leinbach et al. 2021 ), provided sufficient vertical connectivity exists. The absence of an abrupt inflection point in depth-related mortality and important turnover of generic dominance between 5 and 20m could support the existence of at least partial demographic connectivity across depths (Bridge et al. 2014 ; Laverick et al. 2016 ; Pérez-Rosales et al. 2022 ). These findings also add to previous evidence on the limits of depth refugia as its benefits are highly dependent on local ecological and environmental context, especially sensitive to large-scale oceanographic processes (Frade et al. 2018 ; Wyatt et al. 2023 ). Had typical IWC occurred, reefs as shallow as 15m may have experienced thermal buffering (Wyatt et al. 2020 , 2023 ). Therefore, the increasing intensity and frequency of MHWs may undermine the reliability of the “deep reef refuge” hypothesis may be increasingly uncertain given its dependence on transient cooling processes (Hughes et al. 2018a ; Eakin et al. 2019 ). Together, these patterns emphasize the need for future bleaching surveys and monitoring programs to systematically incorporate multiple depths to accurately quantify reef-wide impacts, recovery potential and connectivity pathways. From a management perspective, this further reinforces the importance of protecting deeper reef habitats as integral components of reef resilience. Additionally, reef habitats below 30m may offer more consistent protection from extreme thermal stress. In our study, these zones experienced low heat accumulation and exhibited high daily thermal variability (MDTF), conditions previously linked to enhanced coral tolerance (Oliver and Palumbi 2011 ; Safaie et al. 2018 ; Wyatt et al. 2020 ). These deeper areas could represent promising targets for active conservation, such as assisted reef restoration or the temporary relocation of coral nurseries during heat stress events (Tavakoli-Kolour et al. 2024 ; Henry et al. 2025 ). Expanding technical diving and improving monitoring of deeper reefs will be critical to assess the feasibility, risks and long-term benefits of these approaches. Conclusion By combining in-situ assessments of bleaching, mortality and community structure with depth-resolved thermal data, this study clarifies how depth can modulate coral responses to extreme thermal stress during the unprecedented 2019 marine heatwave in Mo’orea. Bleaching affected over half of coral colonies across all depths, with only limited attenuation along the reef slope in highly sensitive taxa, indicating widespread exposure to acute stress even at 20m. In contrast, post-bleaching mortality declined more sharply with depth, with coral cover loss reduced by nearly 3.5-fold between 5m and 20m. Although bleaching and mortality still occurred at 20m, deeper assemblages were less vulnerable to disturbance than shallow reefs. Mediation analyses showed that differences in coral cover and composition prior to bleaching explained only a quarter of the depth effect on mortality, pointing to additional depth-related protective mechanisms. As temperature gradients were absent, our results highlight the importance light reduction with depth can have on mitigating bleaching and its impacts. Together, these findings contribute to the evaluation of the deep reef refuge hypothesis and highlight the need to incorporate vertical habitat gradients into bleaching assessments and conservation strategies as marine heatwaves intensify. Declarations Conflict of interest The authors declare that they have no conflict of interest. Funding This research was supported by the PolyBleach project “Assessment of coral bleaching in French Polynesia”, funded by the French Ministry for the Ecological Transition and Territorial Cohesion ( Ministère de la Transition écologique et de la Cohésion des territoires ) through the IFRECOR program (French Initiative for Coral Reefs). Author Contribution CB, YL, AM, CD and LH contributed to the study conception and design. YL, AM, CD and LH planned and carried out the fieldwork. CB conducted the data analysis and GS helped in retrieving the multiple mediation analysis results. LH secured funding for and supervised this research. The manuscript was written by CB. All authors read, reviewed and approved the final manuscript. Acknowledgement We extend our thanks to everyone involved in the collection of the data, whether it is the dive support team, students and engineers at CRIOBE research station. We also thank Andreas Eich for his help in cleaning the data. Data Availability All scripts and data necessary to replicate the analysis can be publicly accessed here: [https://github.com/clr-btl/Bleaching-2019](https:/github.com/clr-btl/Bleaching-2019) . References Adjeroud M (1997) Factors influencing spatial patterns on coral reefs around Moorea, French Polynesia. Mar Ecol Prog Ser 159:105–119. https://doi.org/10.3354/meps159105 Adjeroud M, Michonneau F, Edmunds PJ, Chancerelle Y, De Loma TL, Penin L, Thibaut L, Vidal-Dupiol J, Salvat B, Galzin R (2009) Recurrent disturbances, recovery trajectories, and resilience of coral assemblages on a South Central Pacific reef. Coral Reefs 28:775–780. https://doi.org/10.1007/s00338-009-0515-7 Asner GP, Vaughn NR, Martin RE, Foo SA, Heckler J, Neilson BJ, Gove JM (2022) Mapped coral mortality and refugia in an archipelago-scale marine heat wave. Proc Natl Acad Sci 119:e2123331119. https://doi.org/10.1073/pnas.2123331119 Baird A, Madin J, Álvarez-Noriega M, Fontoura L, Kerry J, Kuo C, Precoda K, Torres-Pulliza D, Woods R, Zawada K, Hughes T (2018a) A decline in bleaching suggests that depth can provide a refuge from global warming in most coral taxa. Mar Ecol Prog Ser 603:257–264. https://doi.org/10.3354/meps12732 Baird ME, Mongin M, Rizwi F, Bay LK, Cantin NE, Soja-Woźniak M, Skerratt J (2018b) A mechanistic model of coral bleaching due to temperature-mediated light-driven reactive oxygen build-up in zooxanthellae. Ecol Model 386:20–37. https://doi.org/10.1016/j.ecolmodel.2018.07.013 Banha TNS, Capel KCC, Kitahara MV, Francini-Filho RB, Francini CLB, Sumida PYG, Mies M (2020) Low coral mortality during the most intense bleaching event ever recorded in subtropical Southwestern Atlantic reefs. Coral Reefs 39:515–521. https://doi.org/10.1007/s00338-019-01856-y Bongaerts P, Carmichael M, Hay KB, Tonk L, Frade PR, Hoegh-Guldberg O (2015) Prevalent endosymbiont zonation shapes the depth distributions of scleractinian coral species. R Soc Open Sci 2:140297. https://doi.org/10.1098/rsos.140297 Bongaerts P, Ridgway T, Sampayo EM, Hoegh-Guldberg O (2010) Assessing the ‘deep reef refugia’ hypothesis: focus on Caribbean reefs. Coral Reefs 29:309–327. https://doi.org/10.1007/s00338-009-0581-x Bosserelle P, Berteaux-Lecellier V, Chancerelle Y, Hédouin L, Nugues M, Wallace C, Pichon M (2014) Guide d’identification des coraux de Moorea. CRIOBE, Papetoai, Moorea (Polynésie française) Bridge TCL, Hoey AS, Campbell SJ, Muttaqin E, Rudi E, Fadli N, Baird AH (2014) Depth-dependent mortality of reef corals following a severe bleaching event: implications for thermal refuges and population recovery. F1000Research 2:187. https://doi.org/10.12688/f1000research.2-187.v3 Brown B, Dunne R, Warner M, Ambarsari I, Fitt W, Gibb S, Cummings D (2000) Damage and recovery of Photosystem II during a manipulative field experiment on solar bleaching in the coral Goniastrea aspera. Mar Ecol Prog Ser 195:117–124. Bruno JF, Selig ER (2007) Regional Decline of Coral Cover in the Indo-Pacific: Timing, Extent, and Subregional Comparisons. PLoS ONE 2:e711. https://doi.org/10.1371/journal.pone.0000711 Cantin N, Spalding M (2018) Detecting and monitoring coral bleaching events. Coral Bleaching: Patterns, Processes, Causes and Consequences. Springer, pp 85–110 Carlot J, Rouzé H, Barneche DR, Mercière A, Espiau B, Cardini U, Brandl SJ, Casey JM, Pérez-Rosales G, Adjeroud M, Hédouin L, Parravicini V (2022) Scaling up calcification, respiration, and photosynthesis rates of six prominent coral taxa. Ecol Evol 12:e8613. https://doi.org/10.1002/ece3.8613 Carroll AG, Harrison PL, Adjeroud M (2017) Susceptibility of coral assemblages to successive bleaching events at Moorea, French Polynesia. Mar Freshw Res 68:760. https://doi.org/10.1071/MF15134 Chollett I, Mumby P, Cortés J (2010) Upwelling areas do not guarantee refuge for coral reefs in a warming ocean. Mar Ecol Prog Ser 416:47–56. https://doi.org/10.3354/meps08775 Claar DC, Cobb KM, Baum JK (2019) In situ and remotely sensed temperature comparisons on a Central Pacific atoll. Coral Reefs 38:1343–1349. https://doi.org/10.1007/s00338-019-01850-4 Coles SL, Bahr KD, Rodgers KS, May SL, McGowan AE, Tsang A, Bumgarner J, Han JH (2018) Evidence of acclimatization or adaptation in Hawaiian corals to higher ocean temperatures. PeerJ 6:e5347. https://doi.org/10.7717/peerj.5347 Correia HE, Dee LE, Ferraro PJ (2025) Designing causal mediation analyses to quantify intermediary processes in ecology. Biol Rev 100:1512–1533. https://doi.org/10.1111/brv.70011 Cowburn B, Moritz C, Grimsditch G, Solandt J (2019) Evidence of coral bleaching avoidance, resistance and recovery in the Maldives during the 2016 mass-bleaching event. Mar Ecol Prog Ser 626:53–67. https://doi.org/10.3354/meps13044 Crosbie A, Bridge T, Jones G, Baird A (2019) Response of reef corals and fish at Osprey Reef to a thermal anomaly across a 30 m depth gradient. Mar Ecol Prog Ser 622:93–102. https://doi.org/10.3354/meps13015 Darling ES, Graham NAJ, Januchowski-Hartley FA, Nash KL, Pratchett MS, Wilson SK (2017) Relationships between structural complexity, coral traits, and reef fish assemblages. Coral Reefs 36:561–575. https://doi.org/10.1007/s00338-017-1539-z De Souza MR, Caruso C, Ruiz-Jones L, Drury C, Gates RD, Toonen RJ (2023) Importance of depth and temperature variability as drivers of coral symbiont composition despite a mass bleaching event. Sci Rep 13:8957. https://doi.org/10.1038/s41598-023-35425-9 Denis V, Ribas-Deulofeu L, Sturaro N, Kuo C-Y, Chen CA (2017) A functional approach to the structural complexity of coral assemblages based on colony morphological features. Sci Rep 7:9849. https://doi.org/10.1038/s41598-017-10334-w Diaz C, Foster NL, Attrill MJ, Bolton A, Ganderton P, Howell KL, Robinson E, Hosegood P (2023) Mesophotic coral bleaching associated with changes in thermocline depth. Nat Commun 14:6528. https://doi.org/10.1038/s41467-023-42279-2 Doherty ML, Johnson JV, Goodbody-Gringley G (2025) Widespread coral bleaching and mass mortality during the 2023–2024 marine heatwave in Little Cayman. PLOS One 20:e0322636. https://doi.org/10.1371/journal.pone.0322636 Downs CA, McDougall KE, Woodley CM, Fauth JE, Richmond RH, Kushmaro A, Gibb SW, Loya Y, Ostrander GK, Kramarsky-Winter E (2013a) Heat-Stress and Light-Stress Induce Different Cellular Pathologies in the Symbiotic Dinoflagellate during Coral Bleaching. PLoS ONE 8:e77173. https://doi.org/10.1371/journal.pone.0077173 Downs NJ, Schouten PW, Parisi AV (2013b) Seasonal Variations in the Subsurface Ultraviolet-B on an Inshore Pacific Coral Reef Ecosystem. Photochem Photobiol 89:1234–1243. https://doi.org/10.1111/php.12101 Dunne RP, Brown BE (1996) Penetration of solar UVB radiation in shallow tropical waters and its potential biological effects on coral reefs; results from the central Indian Ocean and Andaman Sea. Mar Ecol Prog Ser 144:109–118. Eakin C, Lough J, Heron S, Liu G (2018) Climate variability and change: monitoring data and evidence for increased coral bleaching stress. Coral Bleach Patterns Process Causes Consequences 51–84. https://doi.org/10.1007/s00338-019-01844-2 Eakin CM, Sweatman HPA, Brainard RE (2019) The 2014–2017 global-scale coral bleaching event: insights and impacts. Coral Reefs 38:539–545. https://doi.org/10.1007/s00338-019-01844-2 Eyal G, Laverick JH, Ben-Zvi O, Brown KT, Kramer N, Tamir R, Lindemann Y, Levy O, Pandolfi JM (2022) Selective deep water coral bleaching occurs through depth isolation. Sci Total Environ 844:157180. https://doi.org/10.1016/j.scitotenv.2022.157180 Ezzat L, Fine M, Maguer J-F, Grover R, Ferrier-Pagès C (2017) Carbon and Nitrogen Acquisition in Shallow and Deep Holobionts of the Scleractinian Coral S. pistillata. Front Mar Sci 4: https://doi.org/10.3389/fmars.2017.00102 Frade PR, Bongaerts P, Englebert N, Rogers A, Gonzalez-Rivero M, Hoegh-Guldberg O (2018) Deep reefs of the Great Barrier Reef offer limited thermal refuge during mass coral bleaching. Nat Commun 9:3447. https://doi.org/10.1038/s41467-018-05741-0 Gleason M (1993) Effects of disturbance on coral communities: bleaching in Moorea, French Polynesia. Coral Reefs 12:193–201. https://doi.org/10.1007/bf00334479 Glynn PW (1996) Coral reef bleaching: facts, hypotheses and implications. Glob Change Biol 2:495–509. Graham NAJ, Jennings S, MacNeil MA, Mouillot D, Wilson SK (2015) Predicting climate-driven regime shifts versus rebound potential in coral reefs. Nature 518:94–97. https://doi.org/10.1038/nature14140 Guest JR, Baird AH, Maynard JA, Muttaqin E, Edwards AJ, Campbell SJ, Yewdall K, Affendi YA, Chou LM (2012) Contrasting Patterns of Coral Bleaching Susceptibility in 2010 Suggest an Adaptive Response to Thermal Stress. PLoS ONE 7:e33353. https://doi.org/10.1371/journal.pone.0033353 Hédouin L, Rouzé H, Berthe C, Perez-Rosales G, Martinez E, Chancerelle Y, Galand PE, Lerouvreur F, Nugues MM, Pochon X (2020) Contrasting patterns of mortality in Polynesian coral reefs following the third global coral bleaching event in 2016. Coral Reefs 39:939–952. https://doi.org/10.1007/s00338-020-01914-w Helgoe J, Davy SK, Weis VM, Rodriguez-Lanetty M (2024) Triggers, cascades, and endpoints: connecting the dots of coral bleaching mechanisms. Biol Rev 99:715–752. https://doi.org/10.1111/brv.13042 Henry JA, Szereday S, Bernard GGR, Patterson JT (2025) Improved coral nursery production through contingent heat stress events via depth manipulation. Aquaculture 605:742558. https://doi.org/10.1016/j.aquaculture.2025.742558 Heron S, Johnston L, Liu G, Geiger E, Maynard J, De La Cour J, Johnson S, Okano R, Benavente D, Burgess T, Iguel J, Perez D, Skirving W, Strong A, Tirak K, Eakin C (2016) Validation of Reef-Scale Thermal Stress Satellite Products for Coral Bleaching Monitoring. Remote Sens 8:59. https://doi.org/10.3390/rs8010059 Hoegh-Guldberg O, Salvat B (1995) Periodic mass-bleaching and elevated sea temperatures: bleaching of outer reef slope communities in Moorea, French Polynesia. Mar Ecol Prog Ser 121:181–190. https://doi.org/10.3354/meps121181 Hoogenboom MO, Frank GE, Chase TJ, Jurriaans S, Álvarez-Noriega M, Peterson K, Critchell K, Berry KLE, Nicolet KJ, Ramsby B, Paley AS (2017) Environmental Drivers of Variation in Bleaching Severity of Acropora Species during an Extreme Thermal Anomaly. Front Mar Sci 4:376. https://doi.org/10.3389/fmars.2017.00376 Hughes TP, Anderson KD, Connolly SR, Heron SF, Kerry JT, Lough JM, Baird AH, Baum JK, Berumen ML, Bridge TC, Claar DC, Eakin CM, Gilmour JP, Graham NAJ, Harrison H, Hobbs J-PA, Hoey AS, Hoogenboom M, Lowe RJ, McCulloch MT, Pandolfi JM, Pratchett M, Schoepf V, Torda G, Wilson SK (2018a) Spatial and temporal patterns of mass bleaching of corals in the Anthropocene. Science 359:80–83. https://doi.org/10.1126/science.aan8048 Hughes TP, Kerry JT, Baird AH, Connolly SR, Dietzel A, Eakin CM, Heron SF, Hoey AS, Hoogenboom MO, Liu G, McWilliam MJ, Pears RJ, Pratchett MS, Skirving WJ, Stella JS, Torda G (2018b) Global warming transforms coral reef assemblages. Nature 556:492–496. https://doi.org/10.1038/s41586-018-0041-2 Johnston EC, Burgess SC (2023) Pocillopora tuahiniensis: a new species of scleractinian coral (Scleractinia, Pocilloporidae) from French Polynesia. Zootaxa 5369:117–124. https://doi.org/10.11646/zootaxa.5369.1.5 Kayal M, Vercelloni J, Lison De Loma T, Bosserelle P, Chancerelle Y, Geoffroy S, Stievenart C, Michonneau F, Penin L, Planes S, Adjeroud M (2012) Predator Crown-of-Thorns Starfish (Acanthaster planci) Outbreak, Mass Mortality of Corals, and Cascading Effects on Reef Fish and Benthic Communities. PLoS ONE 7:e47363. https://doi.org/10.1371/journal.pone.0047363 Kim SW, Sampayo EM, Sommer B, Sims CA, Gómez-Cabrera MDC, Dalton SJ, Beger M, Malcolm HA, Ferrari R, Fraser N, Figueira WF, Smith SDA, Heron SF, Baird AH, Byrne M, Eakin CM, Edgar R, Hughes TP, Kyriacou N, Liu G, Matis PA, Skirving WJ, Pandolfi JM (2019) Refugia under threat: Mass bleaching of coral assemblages in high‐latitude eastern Australia. Glob Change Biol 25:3918–3931. https://doi.org/10.1111/gcb.14772 Kramer N, Tamir R, Eyal G, Loya Y (2020) Coral Morphology Portrays the Spatial Distribution and Population Size-Structure Along a 5–100 m Depth Gradient. Front Mar Sci 7:615. https://doi.org/10.3389/fmars.2020.00615 Kramer PA (2003) Synthesis of coral reef health indicators for the western Atlantic: Results of the AGRRA program(1997–2000). Atoll Res Bull 496:1–57. Kuwahara V, Nakajima R, Othman B, Kushairi M, Toda T (2010) Spatial variability of UVR attenuation and bio-optical factors in shallow coral-reef waters of Malaysia. Coral Reefs 29:693–704. Lamy T, Galzin R, Kulbicki M, Lison de Loma T, Claudet J (2016) Three decades of recurrent declines and recoveries in corals belie ongoing change in fish assemblages. Coral Reefs 35:293–302. https://doi.org/10.1007/s00338-015-1371-2 Laurent V (2019) Atlas climatologique de la Polynésie française. Météo France, Délégation interrégionale de Polynésie française, Laverick JH, Andradi-Brown DA, Exton DA, Bongaerts P, Bridge TCL, Lesser MP, Pyle RL, Slattery M, Wagner D, Rogers AD (2016) To what extent do mesophotic coral ecosystems and shallow reefs share species of conservation interest? Environ Evid 5:16. https://doi.org/10.1186/s13750-016-0068-5 Laverick JH, Tamir R, Eyal G, Loya Y (2020) A generalized light-driven model of community transitions along coral reef depth gradients. Glob Ecol Biogeogr 29:1554–1564. https://doi.org/10.1111/geb.13140 Leinbach SE, Speare KE, Rossin AM, Holstein DM, Strader ME (2021) Energetic and reproductive costs of coral recovery in divergent bleaching responses. Sci Rep 11:23546. https://doi.org/10.1038/s41598-021-02807-w Lenth RV (2025) emmeans: Estimated Marginal Means, aka Least-Squares Means. Lirman D, Formel N, Schopmeyer S, Ault JS, Smith SG, Gilliam D, Riegl B (2014) Percent recent mortality (PRM) of stony corals as an ecological indicator of coral reef condition. Ecol Indic 44:120–127. https://doi.org/10.1016/j.ecolind.2013.10.021 Liu G, Skirving WJ, Geiger EF, De La Cour JL, Marsh BL, Heron SF, Tirak KV, Strong AE, Eakin CM (2017) NOAA Coral Reef Watch’s 5km satellite coral bleaching heat stress monitoring product suite version 3 and four-month outlook version 4. Reef Encount 32:39–45. Liu G, Strong AE, Skirving W (2003) Remote sensing of sea surface temperatures during 2002 Barrier Reef coral bleaching. Eos Trans Am Geophys Union 84:137–141. https://doi.org/10.1029/2003EO150001 Loya Y, Sakai K, Yamazato K, Nakano Y, Sambali H, van Woesik R (2001) Coral bleaching: the winners and the losers. Ecol Lett 4:122–131. https://doi.org/10.1046/j.1461-0248.2001.00203.x Marshall PA, Baird AH (2000) Bleaching of corals on the Great Barrier Reef: differential susceptibilities among taxa. Coral Reefs 19:155–163. https://doi.org/10.1007/s003380000086 Martinez Arbizu P (2020) pairwiseAdonis: Pairwise multilevel comparison using adonis. R Package Version 04 1: McClanahan TR (2023) Local heterogeneity of coral reef diversity and environmental stress provides opportunities for small-scale conservation. Divers Distrib 29:1324–1340. https://doi.org/10.1111/ddi.13770 McClanahan TR, Maina JM, Darling ES, Guillaume MMM, Muthiga NA, D’agata S, Leblond J, Arthur R, Jupiter SD, Wilson SK, Mangubhai S, Ussi AM, Humphries AT, Patankar V, Shedrawi G, Julius P, Ndagala J, Grimsditch G (2020) Large geographic variability in the resistance of corals to thermal stress. Glob Ecol Biogeogr 29:2229–2247. https://doi.org/10.1111/geb.13191 McGillycuddy M, Warton DI, Popovic G, Bolker BM (2025) Parsimoniously Fitting Large Multivariate Random Effects in glmmTMB. J Stat Softw 112:1–19. https://doi.org/10.18637/jss.v112.i01 Meistertzheim A-L, Pochon X, Wood SA, Ghiglione J-F, Hédouin L (2019) Development of a quantitative PCR–high-resolution melting assay for absolute measurement of coral-Symbiodiniaceae associations and its application to investigating variability at three spatial scales. Mar Biol 166:13. https://doi.org/10.1007/s00227-018-3458-0 Molina-Hernández A, Medellín‐Maldonado F, Lange ID, Perry CT, Álvarez‐Filip L (2022) Coral reef erosion: In situ measurement on different dead coral substrates on a Caribbean reef. Limnol Oceanogr 67:2734–2749. https://doi.org/10.1002/lno.12234 Moorea Coral Reef LTER, Leichter J, Adam T, Krumbholz H (2025) MCR LTER: Coral Reef: Benthic Water Temperature, ongoing since 2005. https://doi.org/10.6073/PASTA/02E0FA99C6FCA29A1BDD26C46013F0F7 Morais J, Morais R, Tebbett SB, Bellwood DR (2022) On the fate of dead coral colonies. Funct Ecol 36:3148–3160. https://doi.org/10.1111/1365-2435.14182 Morais J, Santos BA (2018) Limited potential of deep reefs to serve as refuges for tropical Southwestern Atlantic corals. Ecosphere 9:e02281. https://doi.org/10.1002/ecs2.2281 Moritz C, Brandl S, Rouzé H, Vii J, Pérez-Rosales G, Bosserelle P, Chancerelle Y, Galzin R, Liao V, Siu G, Taiarui M, Nugues M, Hédouin L (2021) Long-term monitoring of benthic communities reveals spatial determinants of disturbance and recovery dynamics on coral reefs. Mar Ecol Prog Ser 672:141–152. https://doi.org/10.3354/meps13807 Muir PR, Done T, Aguirre JD, Keith S (2021) High regional and intrageneric variation in susceptibility to mass bleaching in Indo-Pacific coral species. Glob Ecol Biogeogr 30:1889–1898. https://doi.org/10.1111/geb.13353 Muir PR, Marshall PA, Abdulla A, Aguirre JD (2017a) Species identity and depth predict bleaching severity in reef-building corals: shall the deep inherit the reef? Proc R Soc B Biol Sci 284:20171551. https://doi.org/10.1098/rspb.2017.1551 Muir PR, Marshall PA, Abdulla A, Aguirre JD (2017b) Species identity and depth predict bleaching severity in reef-building corals: shall the deep inherit the reef? Proc R Soc B Biol Sci 284:20171551. https://doi.org/10.1098/rspb.2017.1551 Mumby P, Chisholm J, Edwards A, Clark C, Roark E, Andrefouet S, Jaubert J (2001) Unprecedented bleaching-induced mortality in Porites spp. at Rangiroa Atoll, French Polynesia. Mar Biol 139:183–189. Muñiz-Castillo AI, Rivera-Sosa A, McField M, Chollett I, Eakin CM, Enríquez S, Giró A, Drysdale I, Rueda M, Soto M, Craig N, Arias-González JE (2024) Underlying drivers of coral reef vulnerability to bleaching in the Mesoamerican Reef. Commun Biol 7:1452. https://doi.org/10.1038/s42003-024-07128-y Neal BP, Condit C, Liu G, Dos Santos S, Kahru M, Mitchell BG, Kline DI (2014) When depth is no refuge: cumulative thermal stress increases with depth in Bocas del Toro, Panama. Coral Reefs 33:193–205. https://doi.org/10.1007/s00338-013-1081-6 Oksanen J, Simpson GL, Blanchet FG, Kindt R, Legendre P, Minchin PR, O’Hara RB, Solymos P, Stevens MHH, Szoecs E, Wagner H, Barbour M, Bedward M, Bolker B, Borcard D, Borman T, Carvalho G, Chirico M, Caceres MD, Durand S, Evangelista HBA, FitzJohn R, Friendly M, Furneaux B, Hannigan G, Hill MO, Lahti L, McGlinn D, Ouellette M-H, Cunha ER, Smith T, Stier A, Braak CJFT, Weedon J (2025) vegan: Community Ecology Package. Oliver TA, Palumbi SR (2011) Do fluctuating temperature environments elevate coral thermal tolerance? Coral Reefs 30:429–440. https://doi.org/10.1007/s00338-011-0721-y Page CA, Giuliano C, Bay LK, Randall CJ (2023) High survival following bleaching underscores the resilience of a frequently disturbed region of the Great Barrier Reef. Ecosphere 14:e4280. https://doi.org/10.1002/ecs2.4280 Penin L, Adjeroud M, Schrimm M, Lenihan HS (2007) High spatial variability in coral bleaching around Moorea (French Polynesia): patterns across locations and water depths. C R Biol 330:171–181. https://doi.org/10.1016/j.crvi.2006.12.003 Pereira PHC, Lima GV, Pontes AVF, Côrtes LGF, Gomes E, Sampaio CLS, Pinto TK, Miranda RJ, Cardoso ATC, Araujo JC, Seoane JCS (2022) Unprecedented Coral Mortality on Southwestern Atlantic Coral Reefs Following Major Thermal Stress. Front Mar Sci 9:725778. https://doi.org/10.3389/fmars.2022.725778 Pérez-Rosales G, Pichon M, Rouzé H, Villéger S, Torda G, Bongaerts P, Carlot J, Under The Pole Consortium, Parravicini V, Hédouin L (2022) Mesophotic coral ecosystems of French Polynesia are hotspots of alpha and beta generic diversity for scleractinian assemblages. Divers Distrib 28:1391–1403. https://doi.org/10.1111/ddi.13549 Pérez-Rosales G, Rouzé H, Torda G, Bongaerts P, Pichon M, Under The Pole Consortium, Parravicini V, Hédouin L (2021) Mesophotic coral communities escape thermal coral bleaching in French Polynesia. R Soc Open Sci 8:210139. https://doi.org/10.1098/rsos.210139 Pisapia C, Burn D, Pratchett MS (2019) Changes in the population and community structure of corals during recent disturbances (February 2016-October 2017) on Maldivian coral reefs. Sci Rep 9:8402. https://doi.org/10.1038/s41598-019-44809-9 Pratchett MS, McCowan D, Maynard JA, Heron SF (2013) Changes in Bleaching Susceptibility among Corals Subject to Ocean Warming and Recurrent Bleaching in Moorea, French Polynesia. PLoS ONE 8:e70443. https://doi.org/10.1371/journal.pone.0070443 R Core Team (2024) R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria Riegl B, Glynn PW, Banks S, Keith I, Rivera F, Vera-Zambrano M, D’Angelo C, Wiedenmann J (2019) Heat attenuation and nutrient delivery by localized upwelling avoided coral bleaching mortality in northern Galapagos during 2015/2016 ENSO. Coral Reefs 38:773–785. https://doi.org/10.1007/s00338-019-01787-8 Riegl B, Purkis S (2015) Coral population dynamics across consecutive mass mortality events. Glob Change Biol 21:3995–4005. https://doi.org/10.1111/gcb.13014 Roberts TE, Moloney JM, Sweatman HPA, Bridge TCL (2015) Benthic community composition on submerged reefs in the central Great Barrier Reef. Coral Reefs 34:569–580. https://doi.org/10.1007/s00338-015-1261-7 Robinson JPW, Wilson SK, Jennings S, Graham NAJ (2019) Thermal stress induces persistently altered coral reef fish assemblages. Glob Change Biol 25:2739–2750. https://doi.org/10.1111/gcb.14704 Russ GR, Rizzari JR, Abesamis RA, Alcala AC (2021) Coral cover a stronger driver of reef fish trophic biomass than fishing. Ecol Appl 31:e02224. https://doi.org/10.1002/eap.2224 Safaie A, Silbiger NJ, McClanahan TR, Pawlak G, Barshis DJ, Hench JL, Rogers JS, Williams GJ, Davis KA (2018) High frequency temperature variability reduces the risk of coral bleaching. Nat Commun 9:1671. https://doi.org/10.1038/s41467-018-04074-2 Sakai K, Singh T, Iguchi A (2019) Bleaching and post-bleaching mortality of Acropora corals on a heat-susceptible reef in 2016. PeerJ 7:e8138. https://doi.org/10.7717/peerj.8138 Sampayo EM, Ridgway T, Bongaerts P, Hoegh-Guldberg O (2008) Bleaching susceptibility and mortality of corals are determined by fine-scale differences in symbiont type. Proc Natl Acad Sci 105:10444–10449. https://doi.org/10.1073/pnas.0708049105 Schramek TA, Colin PL, Merrifield MA, Terrill EJ (2018) Depth-Dependent Thermal Stress Around Corals in the Tropical Pacific Ocean. Geophys Res Lett 45:9739–9747. https://doi.org/10.1029/2018GL078782 Seabrook S, Correa AM, Conetta D, Grupstra CG, Howe-Kerr LI, McConnell K, Schmeltzer ER, Vega Thurber R, Thurber AR (2023) Coral Bleaching Induced Mortality Transforms Local and Global Carbon Cycles: An Unrecognized Feedback Loop That May Accelerate Reef Decline. EGUsphere 2023:1–24. Smith TB, Glynn PW, Maté JL, Toth LT, Gyory J (2014) A depth refugium from catastrophic coral bleaching prevents regional extinction. Ecology 95:1663–1673. https://doi.org/10.1890/13-0468.1 Smith TB, Gyory J, Brandt ME, Miller WJ, Jossart J, Nemeth RS (2016) Caribbean mesophotic coral ecosystems are unlikely climate change refugia. Glob Change Biol 22:2756–2765. https://doi.org/10.1111/gcb.13175 Souter D, Planes S, Wicquart J, Logan M, Obura D, Staub F (2020) Status of Coral Reefs of the World: 2020. 20. Speare KE, Adam TC, Winslow EM, Lenihan HS, Burkepile DE (2022) Size-dependent mortality of corals during marine heatwave erodes recovery capacity of a coral reef. Glob Change Biol 28:1342–1358. https://doi.org/10.1111/gcb.16000 Suggett DJ, Smith DJ (2020) Coral bleaching patterns are the outcome of complex biological and environmental networking. Glob Change Biol 26:68–79. https://doi.org/10.1111/gcb.14871 Sully S, Burkepile DE, Donovan MK, Hodgson G, van Woesik R (2019) A global analysis of coral bleaching over the past two decades. Nat Commun 10:1264. https://doi.org/10.1038/s41467-019-09238-2 Szereday S, Voolstra CR, Amri AY (2024) Back-to-back bleaching events in Peninsular Malaysia (2019–2020) selectively affect hard coral taxa across- and within-reef scales. Mar Biol 171:183. https://doi.org/10.1007/s00227-024-04495-2 Tamir R, Eyal G, Kramer N, Laverick JH, Loya Y (2019) Light environment drives the shallow-to‐mesophotic coral community transition. Ecosphere 10:e02839. https://doi.org/10.1002/ecs2.2839 Tavakoli-Kolour P, Sinniger F, Morita M, Hazraty-Kari S, Nakamura T, Harii S (2024) Shallow corals acclimate to mesophotic depths while maintaining their heat tolerance against ongoing climate change. Mar Pollut Bull 209:117277. https://doi.org/10.1016/j.marpolbul.2024.117277 Trygonis V, Sini M (2012) photoQuad: a dedicated seabed image processing software, and a comparative error analysis of four photoquadrat methods. J Exp Mar Biol Ecol 424:99–108. https://doi.org/10.1016/j.jembe.2012.04.018 Van Rein H, Schoeman DS, Brown C, Quinn R, Breen J (2011) Development of benthic monitoring methods using photoquadrats and scuba on heterogeneous hard-substrata: a boulder-slope community case study. Aquat Conserv Mar Freshw Ecosyst 21:676–689. https://doi.org/10.1002/aqc.1224 Van Woesik R, Sakai K, Ganase A, Loya Y (2011) Revisiting the winners and the losers a decade after coral bleaching. Mar Ecol Prog Ser 434:67–76. https://doi.org/10.3354/meps09203 VanderWeele T, Vansteelandt S (2014) Mediation Analysis with Multiple Mediators. Epidemiol Methods 2: https://doi.org/10.1515/em-2012-0010 Wall CB, Kaluhiokalani M, Popp BN, Donahue MJ, Gates RD (2020) Divergent symbiont communities determine the physiology and nutrition of a reef coral across a light-availability gradient. ISME J 14:945–958. https://doi.org/10.1038/s41396-019-0570-1 Watt-Pringle R, Smith DJ, Ambo-Rappe R, Lamont TAC, Jompa J (2022) Suppressed recovery of functionally important branching Acropora drives coral community composition changes following mass bleaching in Indonesia. Coral Reefs 41:1337–1350. https://doi.org/10.1007/s00338-022-02275-2 Whitaker H, DeCarlo T (2024) Re(de)fining degree-heating week: coral bleaching variability necessitates regional and temporal optimization of global forecast model stress metrics. Coral Reefs 43:969–984. https://doi.org/10.1007/s00338-024-02512-w Winslow EM, Speare KE, Adam TC, Burkepile DE, Hench JL, Lenihan HS (2024) Corals survive severe bleaching event in refuges related to taxa, colony size, and water depth. Sci Rep 14:9006. https://doi.org/10.1038/s41598-024-58980-1 Winston M, Oliver T, Couch C, Donovan MK, Asner GP, Conklin E, Fuller K, Grady BW, Huntington B, Kageyama K, Kindinger TL, Kozar K, Kramer L, Martinez T, McCutcheon A, McKenna S, Rodgers K, Shayler CK, Vargas-Angel B, Zgliczynski B (2022) Coral taxonomy and local stressors drive bleaching prevalence across the Hawaiian Archipelago in 2019. PLOS ONE 17:e0269068. https://doi.org/10.1371/journal.pone.0269068 Wyatt ASJ, Leichter JJ, Toth LT, Miyajima T, Aronson RB, Nagata T (2020) Heat accumulation on coral reefs mitigated by internal waves. Nat Geosci 13:28–34. https://doi.org/10.1038/s41561-019-0486-4 Wyatt ASJ, Leichter JJ, Washburn L, Kui L, Edmunds PJ, Burgess SC (2023) Hidden heatwaves and severe coral bleaching linked to mesoscale eddies and thermocline dynamics. Nat Commun 14:25. https://doi.org/10.1038/s41467-022-35550-5 Yu Q, Li B (2017) mma: an R package for mediation analysis with multiple mediators. J Open Res Softw 5:11–11. Additional Declarations No competing interests reported. Supplementary Files BoiteletalSupplementary.pdf Cite Share Download PDF Status: Published Journal Publication published 08 Mar, 2026 Read the published version in Coral Reefs → Version 1 posted Editorial decision: Revision requested 07 Feb, 2026 Reviews received at journal 05 Feb, 2026 Reviewers agreed at journal 04 Feb, 2026 Reviewers invited by journal 04 Feb, 2026 Editor assigned by journal 04 Feb, 2026 Submission checks completed at journal 03 Feb, 2026 First submitted to journal 28 Jan, 2026 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8725130","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":587572530,"identity":"968ca438-318d-441c-b9fc-21f8f4acc7c8","order_by":0,"name":"Claire Boitel","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA+ElEQVRIiWNgGAWjYBACAwbGBgaGAgYGPjCjAijEzNxASEtjA5BkYANrOQPSwkhIC0gpWAsQMLaBSfxazNmb2x98MGCQZ2M/3Pbh47zaaP52oJYfFdtwarHsOdjYOMOAwbCNJ7F55sxtx3NnHGZsYOw5cxu3w24kNjbzGDAksDEkNjPzbjuW2wDUwszYRkDLH5AW/ofNzH/nHMudT5QWBpAWCaAtjA01uRsIaQH5ZWaPgYRhm8TDZsaeYwdyNwK1HMTnF3N2YID9qLCR5+dPf8zwo6Yud975wwcf/KjArQUKJGCMw2DyACH1yKCOFMWjYBSMglEwQgAAWE5aGomh7G4AAAAASUVORK5CYII=","orcid":"","institution":"Centre de Recherches Insulaires et Observatoire de l'Environnement","correspondingAuthor":true,"prefix":"","firstName":"Claire","middleName":"","lastName":"Boitel","suffix":""},{"id":587572532,"identity":"87cf6efd-8509-481e-a8ef-71b59bf00252","order_by":1,"name":"Yann Lacube","email":"","orcid":"","institution":"Centre de Recherches Insulaires et Observatoire de l'Environnement","correspondingAuthor":false,"prefix":"","firstName":"Yann","middleName":"","lastName":"Lacube","suffix":""},{"id":587572535,"identity":"feec57f9-3927-406e-9091-a96e52c23a0d","order_by":2,"name":"Alexandre Mercière","email":"","orcid":"","institution":"Centre de Recherches Insulaires et Observatoire de l'Environnement","correspondingAuthor":false,"prefix":"","firstName":"Alexandre","middleName":"","lastName":"Mercière","suffix":""},{"id":587572537,"identity":"0956d1a6-932f-42d5-bc92-cd2d41e5de9b","order_by":3,"name":"Caroline Dubé","email":"","orcid":"","institution":"Centre de Recherches Insulaires et Observatoire de l'Environnement","correspondingAuthor":false,"prefix":"","firstName":"Caroline","middleName":"","lastName":"Dubé","suffix":""},{"id":587572541,"identity":"89afeec9-aafc-461c-ad48-e63848347663","order_by":4,"name":"Gilles Siu","email":"","orcid":"","institution":"Centre de Recherches Insulaires et Observatoire de l'Environnement","correspondingAuthor":false,"prefix":"","firstName":"Gilles","middleName":"","lastName":"Siu","suffix":""},{"id":587572545,"identity":"3a11ae0b-9e71-4417-b2b5-e0269603834b","order_by":5,"name":"Laetitia Hédouin","email":"","orcid":"","institution":"Centre de Recherches Insulaires et Observatoire de l'Environnement","correspondingAuthor":false,"prefix":"","firstName":"Laetitia","middleName":"","lastName":"Hédouin","suffix":""}],"badges":[],"createdAt":"2026-01-28 21:08:25","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8725130/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8725130/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1007/s00338-026-02839-6","type":"published","date":"2026-03-08T15:58:25+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":102187357,"identity":"c18de9b4-fd7d-4c1f-826b-a6c12707f0db","added_by":"auto","created_at":"2026-02-09 08:34:50","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":156124,"visible":true,"origin":"","legend":"\u003cp\u003e(a) Six sites were surveyed around the island of Mo’orea (French Polynesia) on the outer-reef slope at 5, 12 and 20m during and after the 2019 mass bleaching event. (b) The reef, mostly dominated by \u003cem\u003ePocillopora\u003c/em\u003e coral colonies, was severely bleached at 10m (photo credit Alexis Rosenfeld).\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-8725130/v1/7095e20aca94e5af725897f6.png"},{"id":102187359,"identity":"39070009-ee3b-4d2a-a3c2-e583c483d837","added_by":"auto","created_at":"2026-02-09 08:34:51","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":33812,"visible":true,"origin":"","legend":"\u003cp\u003eSeverity of the 2019 thermal stress in Mo’orea. (a) Sea surface temperatures recorded in April 2019 compared with April SST recorded during past thermal anomalies (1987, 1991, 1994, 2003, 2016) which led to (b) the most stressful thermal anomaly yet. SSTs are averaged from 3 different sites around Mo’orea (one on each coast). DHWs are computed from the averaged SSTs based on an averaged MMM of 28,8°C.\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-8725130/v1/a4a4b944117a31177c8bf24e.png"},{"id":102297036,"identity":"7f289dd5-d8b9-47e7-b7fe-313ac43f6793","added_by":"auto","created_at":"2026-02-10 10:25:09","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":9671,"visible":true,"origin":"","legend":"\u003cp\u003eDecreasing maximum accumulated heat stress (AHS in °C-weeks; top) and increasing mean daily temperature fluctuations over the month leading up to bleaching (MDTF in °C; bottom) with increasing depth during the 2019 bleaching event (from Octobre 2018 to September 2019). Both were computed from \u003cem\u003ein-situ\u003c/em\u003e temperature timeseries from 3-5 sites around Mo’orea, extracted from Moorea Coral Reef LTER et al. (2025). AHS was calculated from daily \u003cem\u003ein-situ\u003c/em\u003etemperature averages. Letters on the right are pairwise differences (Dunn’s post-hoc test, see Table S2).\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-8725130/v1/811057b7730f8a380fec6ed8.png"},{"id":102187363,"identity":"0c704613-b98d-4a9b-9786-1c61d9e7ee1b","added_by":"auto","created_at":"2026-02-09 08:34:51","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":40069,"visible":true,"origin":"","legend":"\u003cp\u003eBleaching susceptibility across depths of (a) coral assemblages and (b) the most abundant coral taxa (with \u0026gt; 10 observations for each depth). All sites were combined to give relative abundance of healthy, partially bleached (10-60% of colony surface area bleached) and severely bleached (\u0026gt; 60%) colonies at 5, 12 and 20m. Taxa are ranked from most (top) to less (bottom) bleaching susceptible susceptible. Sample sizes (n) are indicated for taxon and depth. Rare taxa (\u0026lt; 10 obs. per depth) are presented Fig.\u003cstrong\u003e S5.\u003c/strong\u003e\u003cem\u003e \u003c/em\u003eEstimated bleaching probabilities across depths over the assemblage and across taxa are provided in Fig. \u003cstrong\u003eS6.\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-8725130/v1/e461506abd505b678fd19ca9.png"},{"id":102745508,"identity":"68366054-7d21-42e5-a4ab-e4794303bd2b","added_by":"auto","created_at":"2026-02-16 08:51:19","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":38199,"visible":true,"origin":"","legend":"\u003cp\u003eLive coral cover (± se) before and after the bleaching event for (a) the assemblage and (b) across the most abundant taxa (\u0026gt; 10 observations per depth), inferred from recently dead and living corals in October 2019. Bleaching induced mortality especially in dominant \u003cem\u003eAcropora\u003c/em\u003e and \u003cem\u003ePocillopora\u003c/em\u003e, resulting in a substantial loss of coral cover. Relative loss of coral cover across the assemblage and these two taxa with depth is detailed Fig. S8.\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-8725130/v1/5de57ebedc8ac9e716a6be7d.png"},{"id":102296913,"identity":"ae0df3e1-76bf-4dd3-9e2f-98fea8dd5cc0","added_by":"auto","created_at":"2026-02-10 10:22:43","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":18003,"visible":true,"origin":"","legend":"\u003cp\u003eCoral assemblages before (accounting for living and recently dead corals in October 2019) and after (accounting for living only) the bleaching event and relative abundance of dead corals over depth\u003c/p\u003e","description":"","filename":"6.png","url":"https://assets-eu.researchsquare.com/files/rs-8725130/v1/e98284567e810de7ddee5462.png"},{"id":102296741,"identity":"3f4e5de3-8bb4-4235-a6f0-e8da6014e2c6","added_by":"auto","created_at":"2026-02-10 10:21:14","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":49784,"visible":true,"origin":"","legend":"\u003cp\u003eShift in composition of coral assemblages during the 2019 bleaching event. (a) Change in coral composition at 5, 12 and 20m following bleaching is represented in a two-dimensional nMDS. Pre-bleaching assemblages are in grey vs post-bleaching are colored according to depth. Arrows show how much each assemblage changed with bleaching. (b) The Euclidian distance travelled in the nMDS is a proxy of the magnitude of change.\u003c/p\u003e","description":"","filename":"7.png","url":"https://assets-eu.researchsquare.com/files/rs-8725130/v1/d20962f75d1ddb755e2e1868.png"},{"id":104251878,"identity":"716c6638-f9e7-4aca-9148-92419e68d1c6","added_by":"auto","created_at":"2026-03-09 16:15:44","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1470438,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8725130/v1/576d8222-9f3e-4d1b-919e-a1224e12faa7.pdf"},{"id":102187364,"identity":"c4338b9c-5caa-45b2-b811-fe4d5af5f4b2","added_by":"auto","created_at":"2026-02-09 08:34:51","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":11108735,"visible":true,"origin":"","legend":"","description":"","filename":"BoiteletalSupplementary.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8725130/v1/7238544450548981d9f17b23.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"When depth fails to prevent bleaching but limits coral death: insights from the 2019 heatwave in Mo’orea","fulltext":[{"header":"Introduction","content":"\u003cp\u003eMass coral bleaching events are becoming increasingly frequent and severe as marine heatwaves (MHWs) intensify due to climate change, posing a growing threat to the ecological integrity of coral reef ecosystems worldwide (Eakin et al. \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Hughes et al. \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e2018a\u003c/span\u003e; Sully et al. \u003cspan citationid=\"CR105\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). Anomalous increases in temperatures, especially during summer months when irradiance is high, can readily break down the symbiosis between corals and their photosynthetic microalgae (\u003cem\u003eSymbiodiniaceae\u003c/em\u003e), resulting in visible paling (Helgoe et al. \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Unless conditions improve, prolonged bleaching can result in extensive mortality, causing dramatic declines in coral cover, which can drive profound and long-lasting shifts in reef assemblages (Hughes et al. \u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e2018b\u003c/span\u003e; Watt-Pringle et al. \u003cspan citationid=\"CR114\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Because branching corals can be especially thermally sensitive (Loya et al. \u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e2001\u003c/span\u003e; Muir et al. \u003cspan citationid=\"CR75\" class=\"CitationRef\"\u003e2017a\u003c/span\u003e), reefs tend to taxonomically homogenize, losing important structural complexity and functional diversity (Graham et al. \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Denis et al. \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Pisapia et al. \u003cspan citationid=\"CR87\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). This can have cascading consequences on the associated biodiversity relying on corals for recruitment, shelter or food (e.g. fish), threatening the ecosystem services reefs provide (Lamy et al. \u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Riegl et al. \u003cspan citationid=\"CR90\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Robinson et al. \u003cspan citationid=\"CR93\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). When bleaching interacts with local stressors (e.g. pollution, cyclones, disease or predator outbreaks), recovery capacity of reefs can further decline, accelerating transitions toward degraded states (Kayal et al. \u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e2012\u003c/span\u003e; Riegl and Purkis \u003cspan citationid=\"CR91\" class=\"CitationRef\"\u003e2015\u003c/span\u003e). Over the past three decades, the Indo-Pacific has been particularly affected by repeated large-scale bleaching events causing substantial losses in coral cover and reef integrity (Bruno and Selig \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2007\u003c/span\u003e; Souter et al. \u003cspan citationid=\"CR102\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eBleaching, however, does not affect all coral communities equally. As coral species can have markedly different sensitivities to bleaching, vulnerability of a reef system during MHWs is highly reliant on its taxonomic composition (Loya et al. \u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e2001\u003c/span\u003e; Van Woesik et al. \u003cspan citationid=\"CR111\" class=\"CitationRef\"\u003e2011\u003c/span\u003e; Muir et al. \u003cspan citationid=\"CR75\" class=\"CitationRef\"\u003e2017a\u003c/span\u003e, \u003cspan citationid=\"CR74\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Even within a single reef or stress event, spatial variation in environmental conditions can lead to considerable differences in bleaching susceptibility and outcomes (McClanahan et al. \u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Suggett and Smith \u003cspan citationid=\"CR104\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Asner et al. \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; McClanahan \u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Among these environmental gradients modulating responses, depth consistently emerges as a key predictor influencing bleaching severity (Hoogenboom et al. \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Muir et al. 2017; Baird et al. \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2018a\u003c/span\u003e; Winston et al. \u003cspan citationid=\"CR117\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Mu\u0026ntilde;iz-Castillo et al. \u003cspan citationid=\"CR78\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Along reef slopes, corals can experience rapid changes in light and temperature over short vertical distances, which can structure community composition (Schramek et al. \u003cspan citationid=\"CR98\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Tamir et al. \u003cspan citationid=\"CR107\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). Because deeper sites typically offer cooler and dimmer conditions, bleaching often declines with depth for most species (Mumby et al. \u003cspan citationid=\"CR77\" class=\"CitationRef\"\u003e2001\u003c/span\u003e; Bridge et al. \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Muir et al. \u003cspan citationid=\"CR75\" class=\"CitationRef\"\u003e2017a\u003c/span\u003e; Baird et al. \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2018a\u003c/span\u003e; Crosbie et al. \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Riegl et al. \u003cspan citationid=\"CR90\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; P\u0026eacute;rez-Rosales et al. \u003cspan citationid=\"CR86\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). As such, mesophotic reefs (\u0026gt;\u0026thinsp;30m) have received growing attention for their hypothesized role in buffering shallow coral populations from climate change by helping prevent local extinction and supporting recovery through larval supply (the \u0026ldquo;deep reef refuge hypothesis\u0026rdquo;; Glynn \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e1996\u003c/span\u003e; Bongaerts et al. \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2010\u003c/span\u003e; Smith et al. \u003cspan citationid=\"CR100\" class=\"CitationRef\"\u003e2014\u003c/span\u003e). Nevertheless, this protection remains debated (Eakin et al. \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). Some studies reported limited or inconsistent depth refuge during severe or prolonged heatwaves, especially when cooling mechanisms (e.g. upwelling, internal waves) did not coincide (Chollett et al. \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2010\u003c/span\u003e; Neal et al. \u003cspan citationid=\"CR79\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Frade et al. \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2018\u003c/span\u003e), or local adaptation enhanced vulnerability at depth (Smith et al. \u003cspan citationid=\"CR101\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Morais and Santos \u003cspan citationid=\"CR72\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Eyal et al. \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2022\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eAdditionally, most studies have inferred depth refuge from patterns of initial bleaching prevalence alone (Muir et al. \u003cspan citationid=\"CR75\" class=\"CitationRef\"\u003e2017a\u003c/span\u003e; Baird et al. \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2018a\u003c/span\u003e; Frade et al. \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; P\u0026eacute;rez-Rosales et al. \u003cspan citationid=\"CR86\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Winston et al. \u003cspan citationid=\"CR117\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Diaz et al. \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2023\u003c/span\u003e), while comparatively few have concurrently assessed post-bleaching mortality (Cowburn et al. \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Crosbie et al. \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). Crucially, bleaching not always translate into mass mortality depending on past thermal exposure, local species assemblages or the magnitude of stressors experienced. Some reefs may extensively bleach and mostly recover, with minimal losses (e.g. \u0026gt;70% bleached but \u0026lt;\u0026thinsp;1% mortality; H\u0026eacute;douin et al. \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2020\u003c/span\u003e), whereas others may suffer mass coral loss, even under comparable or lower bleaching intensity (e.g. 80% bleached with \u0026gt;\u0026thinsp;50% mortality; Doherty et al. \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). Therefore, bleaching prevalence alone may not reliably predict the ecological impacts of MHWs, while it is coral mortality that ultimately determines structural and functional loss, shaping the long-term trajectory of coral communities. Understanding not only the susceptibility of coral assemblages and taxa to bleaching, but also the variability in their subsequent survival, is essential to improve forecasting of reef futures (Cantin and Spalding \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). Assessing the degree to which depth mitigates bleaching and its consequences is essential to efficiently evaluate the potential of deeper habitats to act as refuges, and to guide effective conservation strategies in the face of accelerating climate change.\u003c/p\u003e \u003cp\u003eSurrounded by a steep outer slope that spans over mesophotic depths, the high volcanic island of Mo\u0026rsquo;orea in French Polynesia offers a valuable natural laboratory to investigate these dynamics. Over the past decades, the island\u0026rsquo;s reefs have been repeatedly affected by multiple large-scale disturbances, including severe bleaching events (1987, 1991, 1994, 2003), a major outbreak of crown-of-thorns starfish (2007\u0026ndash;2009), and cyclones (2010; Gleason \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e1993\u003c/span\u003e; Hoegh-Guldberg and Salvat \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e1995\u003c/span\u003e; Penin et al. \u003cspan citationid=\"CR83\" class=\"CitationRef\"\u003e2007\u003c/span\u003e; Adjeroud et al. \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2009\u003c/span\u003e; Kayal et al. \u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e2012\u003c/span\u003e; Pratchett et al. \u003cspan citationid=\"CR88\" class=\"CitationRef\"\u003e2013\u003c/span\u003e; Lamy et al. \u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Moritz et al. \u003cspan citationid=\"CR73\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). These successive and overlapping events nearly eradicated coral cover on the outer reef slope, reaching historically low levels by 2011, with most sites around the island approaching zero coral cover (Moritz et al. \u003cspan citationid=\"CR73\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Despite the severity of this ecological collapse, coral cover steadily recovered over the following years, dominated by fast-growing branching Pocillopora taxa (Moritz et al. \u003cspan citationid=\"CR73\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). By 2019, reefs were still in their recovery phase and had not yet experienced acute thermal stress (H\u0026eacute;douin et al. \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2020\u003c/span\u003e), when the island experienced a widespread and severe MHW, providing a unique opportunity to assess coral responses across depth gradients.\u003c/p\u003e \u003cp\u003eIn this study, we investigated the interplay between depth, taxonomic sensitivity, coral community composition, and environmental conditions in shaping coral susceptibility to bleaching and post-bleaching mortality. Spanning three depths (5, 12, and 20m), we quantified bleaching prevalence at the peak of the event and coral cover loss six months later. We specifically aimed to (i) quantify the prevalence and severity of coral bleaching (post-bleaching mortality) across depths, (ii) examine depth-related changes in coral communities, and (iii) assess potential mediators of the effect of depth, during the 2019 marine heatwave that struck Mo\u0026rsquo;orea. By integrating ecological surveys with thermal context, we assessed to which extent depth can provide functional refuge from bleaching and explored the ecological and environmental mechanisms underpinning these patterns.\u003c/p\u003e"},{"header":"Materials and methods","content":"\u003cp\u003e \u003c/p\u003e \u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eStudy site and sampling design\u003c/h2\u003e \u003cp\u003eMo\u0026rsquo;orea is a high volcanic island (17\u0026deg;30\u0026prime;S, 149\u0026deg;50\u0026prime;W) in the Society Islands archipelago of French Polynesia, enclosing a shallow lagoon bordered by a steep outer reef slope that drops off rapidly beyond the crest. The island\u0026rsquo;s reef system includes multiple bays and oceanic passes, generating strong hydrodynamic and ecological gradients around it (Adjeroud \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1997\u003c/span\u003e). To monitor spatial and depth-related patterns of coral bleaching during the 2019 thermal anomaly, six sites were selected around the island based on prior knowledge of spatial variation in disturbance impacts and exposure regimes. One site was located on the north coast (Entre 2 Baies \u0026ndash; E2B), two on the west coast (Gendron and Haapiti), and three on the east coast (Temae, Afareitu, and Maatea) (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). These sites span a gradient of seasonal oceanic swell exposure: sites on the west coast are highly exposed to southwestern swells that dominate much of the year, while the northern and eastern sites are more sheltered from such exposure but receive low-energy northern swells between November and April (Laurent \u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). All sites are situated on the outer reef slope and share a similar geomorphology with spur-and-groove formations extending from the reef crest to beyond 30m depth. At each site, three depths (5, 12, and 20m) were surveyed, corresponding to upper- and mid-slope habitats. Prior to the 2019 event, all six sites were considered healthy, with coral cover typically exceeding 30% (Fig. \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eTemperature\u003c/h3\u003e\n\u003cp\u003eSatellite-derived sea surface temperature (SST) data were obtained from the NOAA Coral Reef Watch (CRW) 5-km resolution program for each of the six study sites during the 2019 bleaching event (Liu et al. \u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). The site-specific maximum monthly mean (MMM) temperature was extracted directly from CRW climatologies, ranging from 28.79\u0026deg;C to 28.81\u0026deg;C. Degree Heating Weeks (DHW\u003csub\u003e1\u0026deg;C\u003c/sub\u003e) were extracted for each site as the accumulated weekly temperature anomalies exceeding 1\u0026deg;C above the site-specific MMM (Liu et al. \u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). To assess the low-magnitude stress, we also calculated DHW with the same MMM climatology but with a cutoff at 0\u0026deg;C (when SST exceeds MMM) instead of the conventional 1\u0026deg;C cutoff (referred to as DHW\u003csub\u003e0\u0026deg;C\u003c/sub\u003e; Kim et al. \u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Szereday et al. \u003cspan citationid=\"CR106\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). To place the 2019 thermal anomaly in historical context, one representative site was selected per coast \u0026ndash; Haapiti (West), E2B (North), and Temae (East) \u0026ndash; and the 2019 DHW values were compared to those recorded during previous bleaching years in Mo\u0026rsquo;orea (1987, 1991, 1994, 2003, 2016). Bleaching was observed in 2002 (Penin et al. \u003cspan citationid=\"CR83\" class=\"CitationRef\"\u003e2007\u003c/span\u003e), yet satellite data from NOAA CRW did not register thermal stress (0 DHW). Daily SST values were averaged, and DHW was computed using the same threshold (1\u0026deg;C above MMM\u0026thinsp;=\u0026thinsp;28.80\u0026deg;C).\u003c/p\u003e \u003cp\u003eIn addition to SST data, \u003cem\u003ein-situ\u003c/em\u003e seawater temperature was also analysed from the Moorea Coral Reef Long Term Ecological Research program (Moorea Coral Reef LTER et al. \u003cspan citationid=\"CR70\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) to explore how thermal exposure varied along the depth gradient. Bottom-mounted temperature loggers deployed along the outer reef slope recorded continuous temperature every 20 minutes at depths of 10, 20, 30, and 40m. These records cover five of the six sites used in this study (Maatea was not included in the LTER network), and two of the 3 depths studied (10 and 20m). Due to instrument failure during the 2019 MHW, temperatures at Temae at 10 and 20m and at Gendron at 20m are not available. Unlike SST data, these \u003cem\u003ein-situ\u003c/em\u003e records provide insight into thermal conditions and daily variability directly within reef habitats at and beyond the surveyed depths. From these data, two complementary thermal metrics were derived. Accumulated heat stress (AHS) was calculated using the same method as DHW but based on \u003cem\u003ein-situ\u003c/em\u003e daily mean temperatures rather than SST (Claar et al. \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). Weekly temperature means exceeding 1\u0026deg;C above the site-specific maximum monthly mean (MMM) were summed over a 12-week moving window to capture the cumulative stress experienced underwater. \u003cem\u003eIn-situ\u003c/em\u003e MMMs were computed for each depth and site as the average of daily means during the warmest month, based on the full temperature record from 2005 to 2023 (Frade et al. \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). Site- and depth-specific MMM values are detailed in Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e. Additionally, we calculated the mean daily temperature fluctuation (MDTF), defined as the average daily temperature range over the 30 days preceding the onset of bleaching, which can have a strong mitigating effect on bleaching (Safaie et al. \u003cspan citationid=\"CR95\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). The bleaching onset was identified as March 28, 2019, corresponding to the first day when DHW exceeded zero. To assess whether thermal stress and daily variability differed across depths, we performed Kruskal-Wallis tests followed by pairwise Dunn\u0026rsquo;s post-hoc tests on AHS and MDTF values at 10, 20, 30, and 40m.\u003c/p\u003e\n\u003ch3\u003eCoral bleaching assessment\u003c/h3\u003e\n\u003cp\u003eTo quantify coral bleaching during the peak of the 2019 thermal anomaly, surveys were conducted in mid-April. At each site and depth, 40 photo-quadrats (75cm x 75cm frame) were randomly photographed with an underwater camera (Sony RX100 with lights mounted on each side). Twenty quadrats were randomly selected at each depth and sites for analysis. Species of the same genera can be hard to distinguish \u003cem\u003ein-situ\u003c/em\u003e, especially when they are bleached (Speare et al. \u003cspan citationid=\"CR103\" class=\"CitationRef\"\u003e2022\u003c/span\u003e), and some, like species of pocilloporids, can show high phenotypic plasticity (Johnston and Burgess \u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). As such, individual coral colonies within each photo-quadrat were identified to genus following Bosserelle et al. (\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2014\u003c/span\u003e), except for fungiid corals. Accurate genus-level identification within the family \u003cem\u003eFungiidae\u003c/em\u003e requires access to both the upper and lower surfaces of the corallum as well as examination of septal dentition and costal spines for instance (Bosserelle et al. \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2014\u003c/span\u003e). As these features were hardly discernible from our photographs, and to avoid misclassification, fungiid colonies were conservatively grouped at the family level (under the category \u003cem\u003eFungiidae\u003c/em\u003e) and excluded from genus-specific analyses. Similarly, colonies of \u003cem\u003eNapopora irregularis\u003c/em\u003e (the only species of \u003cem\u003eNapopora\u003c/em\u003e recorded in French Polynesia) were not considered as a \u003cem\u003ePorites\u003c/em\u003e species, despite having been synonymized with \u003cem\u003ePorites irregularis\u003c/em\u003e (Hoeksema and Cairns, 2025). To this date, this species remains listed as \u003cem\u003etaxon inquirendum\u003c/em\u003e, and showed clear morphological and ecological differences (e.g. branching growth form and high bleaching susceptibility) compared to the massive \u003cem\u003ePorites\u003c/em\u003e species which were recorded under the \u003cem\u003ePorites\u003c/em\u003e category (Bosserelle et al. \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Carlot et al. \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). As such, \u003cem\u003eNapopora\u003c/em\u003e colonies were kept as a distinct genus in our dataset (Meistertzheim et al. \u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Carlot et al. \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Following identification, each colony was visually classified based on its health status into three categories: healthy (no visible paling), pale (partial bleaching, 10\u0026ndash;60% of colony surface area presenting loss of pigmentation), and bleached (\u0026gt;\u0026thinsp;60%). The number of colonies per genus and bleaching category was recorded, totalling 9,535 colonies. Recently dead colonies in April (less than 2% of total), mostly \u003cem\u003eAcropora\u003c/em\u003e, were classified as bleached.\u003c/p\u003e\n\u003ch3\u003ePost-bleaching mortality\u003c/h3\u003e\n\u003cp\u003eTo assess depth-related coral mortality following the 2019 heatwave, a second photographic survey was conducted 6 months later, in October 2019 following the same design as in April. Due to logistical constraints, the exact same quadrats could not be relocated, and instead, 40 new photo-quadrats were randomly taken within the same sites at each depth, and 25 were randomly selected for analysis. For each quadrat, following recommendations (Van Rein et al. \u003cspan citationid=\"CR110\" class=\"CitationRef\"\u003e2011\u003c/span\u003e), 75 random points were generated on PhotoQuad (Trygonis and Sini \u003cspan citationid=\"CR109\" class=\"CitationRef\"\u003e2012\u003c/span\u003e), and benthic organisms beneath each point were identified. Categories included rubble, sand, bare substrate, calcareous coralline algae, macroalgae and coral. Corals were classified as either living (\u0026lt;\u0026thinsp;80% of colony partially dead), recently dead or old dead. Skeletal structure can remain distinguishable for more than 6 months post-mortem (Molina-Hern\u0026aacute;ndez et al. \u003cspan citationid=\"CR69\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Morais et al. \u003cspan citationid=\"CR71\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Thus, corals were considered recently dead when corallite architecture was identifiable and skeletal structure was preserved, with minimal algae overgrowth (Kramer \u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e2003\u003c/span\u003e; Lirman et al. \u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Molina‐Hern\u0026aacute;ndez et al. 2022; Morais et al. \u003cspan citationid=\"CR71\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Conversely, old dead corals showed signs of full colonisation by encrusting organisms (e.g. coralline algae, macroalgae) and advanced structural degradation (Molina‐Hern\u0026aacute;ndez et al. 2022). Corals, whether dead or alive, were identified, when possible, to the genus level using the same methodology described for April quadrats.\u003c/p\u003e \u003cp\u003eTo maximise statistical power and enable quadrat-level resolution (n\u0026thinsp;=\u0026thinsp;354), we chose to estimate changes in coral cover and composition using only post-bleaching quadrats, rather than comparing average values from independent April and October quadrats (which would have limited the analysis to n\u0026thinsp;=\u0026thinsp;18 site x depth combinations). We estimated coral cover and community composition post-bleaching as the proportion of points falling on living corals over the cloud of 75 points. Pre-bleaching community and coral cover were inferred by summing the points of live and recently dead coral. Relative coral cover loss between April and October was computed at the community and genus level, for each quadrat, as a proxy of bleaching-induced mortality (Mumby et al. \u003cspan citationid=\"CR77\" class=\"CitationRef\"\u003e2001\u003c/span\u003e; Lirman et al. \u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Cantin and Spalding \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Cowburn et al. \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; H\u0026eacute;douin et al. \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). No major disturbances (e.g. crown-of-thorn predation scars, diseases, or storm damage) were reported at the study sites that were frequently visited between April and October. While we cannot fully exclude the possibility that some recent mortality was unrelated to bleaching, any such cases likely represent a very small fraction given the timing, scale (amount of colonies), and context of the surveys.\u003c/p\u003e\n\u003ch3\u003eStatistical analyses\u003c/h3\u003e\n\u003cp\u003eData analyses were carried out using R v 4.4.2 (R Core Team \u003cspan citationid=\"CR89\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). The relationship between coral bleaching prevalence and depth was tested using generalized linear mixed models (GLMMs) with a binomial error distribution, in which the response variable was binary (0\u0026thinsp;=\u0026thinsp;healthy; 1\u0026thinsp;=\u0026thinsp;pale or bleached, i.e., a bleaching response). Depth was included as a fixed effect (categorical variable, to allow for flexible non-linear relationships), and site was included as a random effect to account for the sampling design. Analyses were conducted at two levels: (i) the community level, where genus was modelled as a random effect to account for differences in thermal sensitivity among coral genera; and (ii) the genus level (only genera with \u0026gt;\u0026thinsp;10 observations per depth were retained) where taxa was included as a fixed effect interacting with depth. Post-bleaching mortality was assessed through the relative loss in coral cover compared to pre-bleaching levels. Relative coral cover loss was modelled using GLMMs with a beta error distribution with a logit link (glmmTMB package; McGillycuddy et al. \u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e2025\u003c/span\u003e), using the same fixed and random effect structure described above. One model was run at the community level (all genera combined), and a second set of models at the taxa level focused on genera studied in their bleaching response that exhibited mortality responses (\u0026gt;\u0026thinsp;5 observations). For each model, the best-fitting random effect structure was selected based on likelihood ratio tests compared to when random effects were removed (e.g., site or genus). To examine how depth-dependent sensitivity to bleaching translated into shifts in coral community structure 6 months after bleaching, we compared the composition of assemblages before and after the bleaching event using non-metric multidimensional scaling (nMDS) based on Bray-Curtis dissimilarity matrices (package vegan; Oksanen et al. \u003cspan citationid=\"CR80\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). The magnitude of community change was quantified for each assemblage (i.e. quadrat) as the Euclidean distance between its pre- and post-bleaching positions in nMDS space. The influence of depth on this shift was tested fitting a GLMM with a Gamma distribution with a log link (glmmTMB package), with depth as a fixed effect and site as a random effect; significance of site was tested via a likelihood ratio test. To test whether community composition differed significantly among depths and between time points (before vs. after bleaching), we used PERMANOVA with site as a stratification factor (to constrain permutations within sites). Pairwise differences in composition were then explored using pairwiseAdonis (Martinez Arbizu \u003cspan citationid=\"CR64\" class=\"CitationRef\"\u003e2020\u003c/span\u003e), both between depths and between time points at each depth.\u003c/p\u003e \u003cp\u003eType II Wald F tests with Kenward-Roger degrees of freedom approximations were run on each model to quantify the relative importance (significance) of each main effect (depth) and its interaction with taxa when included. Post-hoc pairwise Tukey-adjusted tests were performed with the emmeans package (Lenth \u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e2025\u003c/span\u003e), between depths, and taxa when pertinent at the population-level (when site as random effect was significant). When random effects were retained in GLMMs, bias adjustment was applied to predicted values and marginal effects to account for Jensen's inequality, which can otherwise lead to biased estimates due to the non-linearity of link functions in generalized models.\u003c/p\u003e \u003cp\u003eFinally, to determine which variables mediate the effect of depth on overall relative coral cover loss, we used a multiple mediation analysis (MMA) framework implemented in the mma R package (Yu and Li \u003cspan citationid=\"CR120\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). This method allows for the decomposition of the total effect of a predictor (here, depth as a continuous variable) into a direct effect (the part of the effect not explained by other variables) and indirect effects (the part mediated through other variables (VanderWeele and Vansteelandt \u003cspan citationid=\"CR112\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Correia et al. \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). We tested to what extent the effect of depth on coral mortality was mediated by depth-related changes in (1) pre-bleaching coral community composition and (2) initial coral cover prior to bleaching. Effect of community composition was quantified as the independent but also joint effect of the relative abundance of dominant taxa, which represented\u0026thinsp;\u0026gt;\u0026thinsp;5% of total coral cover. To account for potential non-linear relationships, we allowed quadratic terms for the predictor and mediators. The model estimated the relative contribution of each indirect pathway (i.e. through community composition or initial cover) to the total effect of depth on coral cover loss. Confidence intervals (95%) for all estimates were obtained via bootstrapping (sampling was repeated 5,000 times).\u003c/p\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003eThermal stress patterns and intensity during the 2019 bleaching event\u003c/h2\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe 2019 marine heatwave (MHW) impacted all sites around Mo\u0026rsquo;orea. Satellite-derived Degree Heating Weeks (DHW) reached values ranging from 2.9\u0026deg;C-weeks at Haapiti (southwest coast) to 3.4\u0026deg;C-weeks at Entre 2 baies (E2B, northern coast), with southern sites generally experiencing slightly lower thermal stress (Fig. S2a). Sea surface temperatures (SSTs) exceeded the maximum monthly mean (MMM; 28.8\u0026deg;C) mid-December 2018 and hovered between MMM and MMM\u0026thinsp;+\u0026thinsp;1\u0026deg;C until late March (Fig. S2b). A sharp thermal spike in early April pushed SSTs above bleaching thresholds (MMM\u0026thinsp;+\u0026thinsp;1\u0026deg;C), triggering rapid DHW accumulation (\u0026gt;\u0026thinsp;2.5\u0026deg;C-weeks within 2 weeks) before plateauing through mid-June (Fig. S2a). SSTs thus remained above MMM for nearly 5 months (Fig. S2b).\u003c/p\u003e \u003cp\u003eThermal stress was higher than all previous bleaching years to date, reaching 3.2\u0026deg;C-weeks, representing the most intense MHW ever recorded on the island (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eb). It surpassed the last two major bleaching events in 1994 and 2003 (both ~\u0026thinsp;2.8\u0026deg;C-weeks) and was far above the last event in 2016 (\u0026lt;\u0026thinsp;1\u0026deg;C-week; Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eb). April 2019 was particularly anomalously warm, with SSTs frequently exceeding 30\u0026deg;C, unlike previous bleaching years (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ea).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cem\u003eIn-situ\u003c/em\u003e temperature data recorded along the outer reef slope at depths of 10 to 40m confirmed strong vertical thermal gradients (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e; S3). While upper- and mid-reef slopes (10-20m) experienced intense and sustained heat stress, deeper zones (30-40m) were partially thermally decoupled and exposed to cooler, more variable conditions following the thermal spike. Accumulated heat stress (AHS), assessed against long-term temperatures at each depth, reached 3.14\u0026thinsp;\u0026plusmn;\u0026thinsp;0.37\u0026deg;C-weeks (mean\u0026thinsp;\u0026plusmn;\u0026thinsp;sd) at 10m and 2.84\u0026thinsp;\u0026plusmn;\u0026thinsp;0.42\u0026deg;C-weeks at 20m (no significant difference, p\u0026thinsp;=\u0026thinsp;0.381; Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e). Beyond 20m, AHS declined sharply by more than half (p\u0026thinsp;\u0026lt;\u0026thinsp;0.05), dropping to 1.23\u0026thinsp;\u0026plusmn;\u0026thinsp;0.54\u0026deg;C-weeks at 30m and to similar levels at 40m (0.96\u0026thinsp;\u0026plusmn;\u0026thinsp;0.49\u0026deg;C-weeks, p\u0026thinsp;=\u0026thinsp;0.397). Conversely, daily temperature variability over the 30 days preceding bleaching onset (MDTF) increased with depth: from averaging 0.41\u0026thinsp;\u0026plusmn;\u0026thinsp;0.09\u0026deg;C at 10m to 0.55\u0026thinsp;\u0026plusmn;\u0026thinsp;0.14\u0026deg;C at 20m (no significant differences, p\u0026thinsp;=\u0026thinsp;0.325), and then more than doubled to 1.05\u0026thinsp;\u0026plusmn;\u0026thinsp;0.19\u0026deg;C at 30m and 1.82\u0026thinsp;\u0026plusmn;\u0026thinsp;0.19\u0026deg;C at 40m (p\u0026thinsp;\u0026lt;\u0026thinsp;0.05; Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e). This increase in MDTF was primarily driven by increasingly cooler daily minima with depth, a pattern observed across all sites and extending from October to early May at 30m, and until early June at 40m (Fig. S3). These fluctuations coincided with the timing of bleaching onset and confirm patterns observed in SST data, notably the sharp increase in temperature in early April across all depths.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eBleaching susceptibility\u003c/h3\u003e\n\u003cp\u003e \u003c/p\u003e \u003cp\u003eHalf of coral colonies were bleached (i.e. across all coral genera) at both 5m and 12m depths (50.0\u0026thinsp;\u0026plusmn;\u0026thinsp;0.7% \u0026ndash; emmeans\u0026thinsp;\u0026plusmn;\u0026thinsp;se \u0026ndash; and 50.2\u0026thinsp;\u0026plusmn;\u0026thinsp;0.9%, respectively; p\u0026thinsp;=\u0026thinsp;0.120; Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ea-b). Although bleaching prevalence at 20m was statistically lower (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), the decrease was marginal (48.9\u0026thinsp;\u0026plusmn;\u0026thinsp;2.4%; Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ea, S5). Overall, bleaching was primarly structured by site-level variability and genus-specific differences in susceptibility. Depth alone explained\u0026thinsp;\u0026lt;\u0026thinsp;1% of the observed variance in bleaching prevalence (marginal R\u003csup\u003e2\u003c/sup\u003e), while the inclusion of site and taxa as random effects increased the explained variance to 53% (conditional R\u003csup\u003e2\u003c/sup\u003e; Table S3). Bleaching happened island-wide, with substantial bleaching recorded at all sites (\u0026gt;\u0026thinsp;45%, averaged across depths; Fig. S4).\u003c/p\u003e \u003cp\u003eBleaching susceptibility varied markedly among taxa (genus effect: \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\chi\\:\\)\u003c/span\u003e\u003c/span\u003e\u003csup\u003e2\u003c/sup\u003e(9)\u0026thinsp;=\u0026thinsp;1371.94, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001). Eighteen coral genera were recorded across all quadrats, though 4 were only observed at 20m and in very low abundance (\u0026lt;\u0026thinsp;10 individuals, \u003cem\u003ePachyseris\u003c/em\u003e, \u003cem\u003eGoniastrea\u003c/em\u003e, \u003cem\u003eLeptoseris\u003c/em\u003e, and \u003cem\u003eCyphastrea\u003c/em\u003e; Fig. S5). Only eight were retained for taxa-level susceptibility analysis, each with at least 10 observations per depth: \u003cem\u003eAcropora\u003c/em\u003e, \u003cem\u003ePocillopora\u003c/em\u003e, \u003cem\u003eMontipora\u003c/em\u003e, \u003cem\u003eLeptastrea\u003c/em\u003e, \u003cem\u003eMillepora\u003c/em\u003e, \u003cem\u003ePorites\u003c/em\u003e, \u003cem\u003ePavona\u003c/em\u003e, and \u003cem\u003eAstrea\u003c/em\u003e (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eb). \u003cem\u003eAcropora\u003c/em\u003e, \u003cem\u003eAstrea\u003c/em\u003e, and \u003cem\u003eMillepora\u003c/em\u003e were the most susceptible to bleach, with bleaching prevalence of 93.8\u0026thinsp;\u0026plusmn;\u0026thinsp;1.1% (emmeans\u0026thinsp;\u0026plusmn;\u0026thinsp;se), 91.3\u0026thinsp;\u0026plusmn;\u0026thinsp;1.6%, and 87.6\u0026thinsp;\u0026plusmn;\u0026thinsp;3.9% respectively (averaged across all depths; no significant differences among them, p\u0026thinsp;\u0026gt;\u0026thinsp;0.05; Fig. S7). \u003cem\u003eMontipora\u003c/em\u003e showed slightly lower prevalence (87.9\u0026thinsp;\u0026plusmn;\u0026thinsp;1.9%) than \u003cem\u003eAcropora\u003c/em\u003e (p\u0026thinsp;=\u0026thinsp;0.002), but did not differ from \u003cem\u003eAstrea\u003c/em\u003e and \u003cem\u003eMillepora\u003c/em\u003e (Fig. S7). \u003cem\u003eLeptastrea\u003c/em\u003e (63.8\u0026thinsp;\u0026plusmn;\u0026thinsp;5.4%) was significantly less susceptible than these four taxa, and bleached more than \u003cem\u003ePocillopora\u003c/em\u003e, which exhibited an intermediate and milder response (45.0\u0026thinsp;\u0026plusmn;\u0026thinsp;3.3%, all pairwise\u0026thinsp;\u0026lt;\u0026thinsp;0.01; Fig. S7). \u003cem\u003ePavona\u003c/em\u003e and \u003cem\u003ePorites\u003c/em\u003e were the most resistant taxa, exhibiting minimal and mostly partial bleaching (0.1\u0026thinsp;\u0026plusmn;\u0026thinsp;0.1% and 7.2\u0026thinsp;\u0026plusmn;\u0026thinsp;1.70% of colonies were bleached respectively, all pairwise\u0026thinsp;\u0026lt;\u0026thinsp;0.05; Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eb). Bleaching severity also varied among taxa (genus effect: \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\chi\\:\\)\u003c/span\u003e\u003c/span\u003e\u003csup\u003e2\u003c/sup\u003e(11)\u0026thinsp;=\u0026thinsp;33.91, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001; Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eb). Nearly all bleached \u003cem\u003eAcropora\u003c/em\u003e colonies, 84% of bleached \u003cem\u003eMontipora\u003c/em\u003e and 74% of bleached \u003cem\u003ePocillopora\u003c/em\u003e were entirely and severely bleached (\u0026gt;\u0026thinsp;60% loss of pigmentation). In \u003cem\u003eAstrea\u003c/em\u003e and \u003cem\u003eMillepora\u003c/em\u003e, bleaching was milder as only 40\u0026ndash;50% of bleached colonies were severely depigmented. Bleaching was rather partial in \u003cem\u003eLeptastrea\u003c/em\u003e (\u0026lt;\u0026thinsp;20%).\u003c/p\u003e \u003cp\u003eReponses to depth differed substantially between taxa (effect depth x genus: \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\chi\\:\\)\u003c/span\u003e\u003c/span\u003e\u003csup\u003e2\u003c/sup\u003e(18)\u0026thinsp;=\u0026thinsp;207.46; p\u0026thinsp;\u0026lt;\u0026thinsp;0.001; Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e). Both \u003cem\u003eAcropora\u003c/em\u003e and \u003cem\u003eAstrea\u003c/em\u003e displayed no difference in bleaching susceptibility between 5m and 12m (94 to 97%; p\u0026thinsp;=\u0026thinsp;0.688 and p\u0026thinsp;=\u0026thinsp;0.855). Bleaching prevalence significantly decreased at 20m for both, though by only 11\u0026ndash;15%, remaining high overall (85.4\u0026thinsp;\u0026plusmn;\u0026thinsp;3.2% and 79.6\u0026thinsp;\u0026plusmn;\u0026thinsp;3.4%, respectively at 20m, both p\u0026thinsp;\u0026lt;\u0026thinsp;0.001; Fig. S6). Likewise, bleaching in \u003cem\u003eMontipora\u003c/em\u003e was only slightly mitigated by 6% from 5 to 12m (from 95.5\u0026thinsp;\u0026plusmn;\u0026thinsp;1.3% at 5m to 89.7\u0026thinsp;\u0026plusmn;\u0026thinsp;2.0% at 12m, p\u0026thinsp;=\u0026thinsp;0.011), but was more strongly reduced (-24%) from 12 to 20m (68.4\u0026thinsp;\u0026plusmn;\u0026thinsp;4.1% at 20m, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001; Fig. S6). Bleaching in \u003cem\u003ePocillopora\u003c/em\u003e slightly peaked at intermediate depths (12m: 51.7\u0026thinsp;\u0026plusmn;\u0026thinsp;3.5%; +5% compared to 5m, p\u0026thinsp;=\u0026thinsp;0.008) before decreasing by 29% to 36.8\u0026thinsp;\u0026plusmn;\u0026thinsp;3.4% at 20m (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001). Conversely, bleaching prevalence increased with depth for \u003cem\u003ePorites\u003c/em\u003e and \u003cem\u003ePavona\u003c/em\u003e. \u003cem\u003ePorites\u003c/em\u003e bleaching increased 8-fold from 2.5\u0026thinsp;\u0026plusmn;\u0026thinsp;1.4% at 5m to 20.2\u0026thinsp;\u0026plusmn;\u0026thinsp;3.3% at 20m (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), while \u003cem\u003ePavona\u003c/em\u003e did not bleach at 5m but reached 17.9\u0026thinsp;\u0026plusmn;\u0026thinsp;2.8% at 20m (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001; Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eb, S6). \u003cem\u003ePavona\u003c/em\u003e also tended to bleach more severely beyond 12m (Table S4). Bleaching prevalence remained high across all depths in \u003cem\u003eMillepora\u003c/em\u003e (~\u0026thinsp;50%) and \u003cem\u003eLeptastrea\u003c/em\u003e (~\u0026thinsp;90%; no significant differences, all pairwise\u0026thinsp;\u0026gt;\u0026thinsp;0.05; Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eb, S6). Finally, the relative proportion of partially vs severely bleached colonies remained broadly stable with depth (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e, Table S4).\u003c/p\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eDepth-dependent coral cover loss\u003c/h2\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe effect of depth on coral mortality was notably more pronounced than its effect on bleaching susceptibility (depth explained 41% of coral cover loss variability; Table S3). Among the most abundant taxa analysed for bleaching, only \u003cem\u003ePocillopora\u003c/em\u003e and \u003cem\u003eAcropora\u003c/em\u003e had experienced significant and severe mortality by October (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001 for all pairwise comparisons between coral cover before and after; all other genera, p\u0026thinsp;\u0026gt;\u0026thinsp;0.05; Table S5). Together they accounted for 96% of the coral loss at 5m (85% at 20m; Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e). The taxa-level mortality analysis thus focused on these two genera, which exhibited different depth responses (genus effect: \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\chi\\:\\)\u003c/span\u003e\u003c/span\u003e\u003csup\u003e2\u003c/sup\u003e(1)\u0026thinsp;=\u0026thinsp;43.980; p\u0026thinsp;\u0026lt;\u0026thinsp;0.0001; depth x genus interaction: \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\chi\\:\\)\u003c/span\u003e\u003c/span\u003e\u003csup\u003e2\u003c/sup\u003e(2)\u0026thinsp;=\u0026thinsp;16.944; p\u0026thinsp;\u0026lt;\u0026thinsp;0.0001). Before bleaching, coral cover of \u003cem\u003ePocillopora\u003c/em\u003e was high and declined linearly from \u0026gt;\u0026thinsp;35% at 5m to ~\u0026thinsp;22% at 20m, while \u003cem\u003eAcropora\u003c/em\u003e was rarer (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e), with cover decreasing from ~\u0026thinsp;7% at 5m to 4% at 20m. Following the bleaching event, \u003cem\u003eAcropora\u003c/em\u003e experienced the highest losses, with cover reduced by 73.9\u0026thinsp;\u0026plusmn;\u0026thinsp;3.5% at 5m (relative to pre-bleaching levels, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001; Fig. S8). Mortality was less severe for \u003cem\u003ePocillopora\u003c/em\u003e at the same depth, with coral cover declining by 46.5\u0026thinsp;\u0026plusmn;\u0026thinsp;3.8% (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001; Fig. S8, Table S7). While both taxa exhibited reduced mortality with depth, the protective effect of depth was stronger for \u003cem\u003eAcropora\u003c/em\u003e (Table S7). Between 5 and 12m, coral mortality significantly declined by 21\u0026ndash;27% for both taxa (12m: 54.5\u0026thinsp;\u0026plusmn;\u0026thinsp;4.2% for \u003cem\u003eAcropora\u003c/em\u003e, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001; 35.4\u0026thinsp;\u0026plusmn;\u0026thinsp;3.4% for \u003cem\u003ePocillopora\u003c/em\u003e, p\u0026thinsp;=\u0026thinsp;0.022; Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e, S8). From 12 and 20m, however, \u003cem\u003eAcropora\u003c/em\u003e relative coral cover loss decreased by 49% (to 29.4\u0026thinsp;\u0026plusmn;\u0026thinsp;4.1% i.e. -63% relative to 5m; p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), while it declined by only 23% in \u003cem\u003ePocillopora\u003c/em\u003e (to 27.7\u0026thinsp;\u0026plusmn;\u0026thinsp;3.1% i.e. -39% relative to 5m; p\u0026thinsp;\u0026lt;\u0026thinsp;0.001; Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e). In October, \u003cem\u003ePocillopora\u003c/em\u003e cover had dropped below 20% at all depths.\u003c/p\u003e \u003cp\u003eAs \u003cem\u003ePocillopora\u003c/em\u003e dominated the reef, representing on average 68\u0026thinsp;\u0026plusmn;\u0026thinsp;25% of the total pre-bleaching live cover (mean\u0026thinsp;\u0026plusmn;\u0026thinsp;sd across all depths; Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e), these losses contributed disproportionately to the overall decline in coral cover (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e). Before bleaching, coral cover was high, exceeding 30% across all depths, declining from 45.9\u0026thinsp;\u0026plusmn;\u0026thinsp;3.2% (emmeans\u0026thinsp;\u0026plusmn;\u0026thinsp;se) at 5m, to 39.8\u0026thinsp;\u0026plusmn;\u0026thinsp;3.1% at 12m and to 30.8\u0026thinsp;\u0026plusmn;\u0026thinsp;2.8% at 20m (all pairwise before \u0026lt;\u0026thinsp;0.001; Table S5, S6). Following the bleaching event, roughly half of the coral cover was lost at 5m (45.0\u0026thinsp;\u0026plusmn;\u0026thinsp;2.4% relatively to pre-bleaching levels; emmeans\u0026thinsp;\u0026plusmn;\u0026thinsp;se), which was nearly twice the loss recorded at 12m (26.3\u0026thinsp;\u0026plusmn;\u0026thinsp;2.0%, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), and 3.5 times the loss at 20m (12.6\u0026thinsp;\u0026plusmn;\u0026thinsp;1.2% i.e. 52% less loss relative to 12m, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001; Fig. S8, Table S4). Post-bleaching coral cover thus ranged from 23 to 27% across depths (post-bleaching pairwise all significant, p\u0026thinsp;\u0026lt;\u0026thinsp;0.03; Table S5). Coral cover was especially high in E2B at 5m, towering to 75% prior to bleaching, while most sites did not exceed 50% (Fig. \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e, S9). It was also the most dominated by pocilloporids (\u0026gt;\u0026thinsp;93%; Fig S10) and the most affected by coral mortality, losing on average 61% of its cover, reducing coral cover to about 30% (Fig. S9). Despite such an important decline, E2B remained the site with the highest live coral cover following bleaching (Fig. S9).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eChanges in coral assemblages\u003c/h2\u003e \u003cp\u003eBefore the bleaching event, coral assemblages along the reef slope and across sites were broadly similar in composition, dominated by \u003cem\u003ePocillopora\u003c/em\u003e, with variable but lower contributions from \u003cem\u003eAcropora\u003c/em\u003e (7\u0026thinsp;\u0026plusmn;\u0026thinsp;12% \u0026ndash; mean\u0026thinsp;\u0026plusmn;\u0026thinsp;sd), \u003cem\u003ePorites\u003c/em\u003e (7\u0026thinsp;\u0026plusmn;\u0026thinsp;15%), and \u003cem\u003eMontipora\u003c/em\u003e (6\u0026thinsp;\u0026plusmn;\u0026thinsp;12%), which together accounted for nearly 90% of overall coral cover (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e, S10). \u003cem\u003ePocillopora\u003c/em\u003e accounted for 75\u0026thinsp;\u0026plusmn;\u0026thinsp;21% at 5m to 60\u0026thinsp;\u0026plusmn;\u0026thinsp;29% at 20m of the coral assemblage. \u003cem\u003eAcropora\u003c/em\u003e slightly peaked at 12m (9\u0026thinsp;\u0026plusmn;\u0026thinsp;13%) compared to 5m and 20m (7\u0026thinsp;\u0026plusmn;\u0026thinsp;12% and 6\u0026thinsp;\u0026plusmn;\u0026thinsp;11%, respectively). \u003cem\u003eMontipora\u003c/em\u003e was more abundant at 5m (8\u0026thinsp;\u0026plusmn;\u0026thinsp;13%) than at 12m or 20m (both ~\u0026thinsp;5%), while \u003cem\u003ePorites\u003c/em\u003e increased markedly with depth, from 3\u0026thinsp;\u0026plusmn;\u0026thinsp;5% at 5m to nearly four times as much at 20m (11\u0026thinsp;\u0026plusmn;\u0026thinsp;20% at 20m, and 8\u0026thinsp;\u0026plusmn;\u0026thinsp;14% at 12m; Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e). Consequently, depth was found to be a strong predictor of pre-bleaching assemblages as they differed significantly among depths (PERMANOVA pairwise comparisons before bleaching, all p\u0026thinsp;=\u0026thinsp;0.001 in Table S4).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003ePERMANOVA results on coral assemblages: effects of depth and bleaching combined\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"7\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVariable\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDf\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eSumOfSqs\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eR\u003csup\u003e2\u003c/sup\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eF\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003ep-value\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBleaching\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e5.79\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.04\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e36.79\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDepth\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e4.49\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.03\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e14.28\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBleaching:Depth\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.92\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e6.11\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eResidual\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e862\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e135.59\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.92\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTotal\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e867\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e147.80\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eThese initial differences, coupled with genus-specific sensitivities to bleaching and mortality, led to depth-dependent shifts in community structure. Bleaching significantly altered community composition across depths (PERMANOVA depth x time\u0026thinsp;=\u0026thinsp;5.85, p\u0026thinsp;=\u0026thinsp;0.001; Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e), although it accounted for a small fraction of the variance (R\u003csup\u003e2\u003c/sup\u003e\u0026thinsp;~\u0026thinsp;4%; Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Changes in reef composition were most pronounced at 5m and least at 20m, with intermediate impact at 12m. This pattern was supported by both PERMANOVA effect sizes (R\u003csup\u003e2\u003c/sup\u003e from pairwise PERMANOVA after-before; Table S8) and changes in community structure in nMDS space (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003e). The average distance in nMDS space crossed by assemblages because of bleaching was more than 3 times greater at 5m (0.62\u0026thinsp;\u0026plusmn;\u0026thinsp;0.11; emmeans\u0026thinsp;\u0026plusmn;\u0026thinsp;se) than at 20m (0.18\u0026thinsp;\u0026plusmn;\u0026thinsp;0.04, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001) and twice as high at 12m (0.32\u0026thinsp;\u0026plusmn;\u0026thinsp;0.06, p\u0026thinsp;=\u0026thinsp;0.028; Table S9). Several quadrats at 20m showed little to no change in composition (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003e). Changes were largely driven by declines in \u003cem\u003ePocillopora\u003c/em\u003e, which was strongly negatively correlated with the primary axis of compositional change across depths (R\u003csup\u003e2\u003c/sup\u003e\u0026thinsp;=\u0026thinsp;74%; Fig. S11).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eDrivers of depth-mediated protection\u003c/h2\u003e \u003cp\u003eTo identify processes underlying the attenuation of coral mortality with depth, we conducted a multiple mediation analysis (MMA) to test whether the effect of depth on relative coral cover loss (of the assemblage) was mediated by initial coral community composition (i.e. the relative abundance of bleaching-susceptible versus bleaching-tolerant taxa) and pre-bleaching coral cover, both of which declined along the depth gradient (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e, \u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e, S1, Table S6). Pre-bleaching community composition was\u003c/p\u003e \u003cp\u003erepresented by the joint effect of the relative abundance of dominant taxa, \u003cem\u003ePocillopora, Acropora, Montipora, and Porites\u003c/em\u003e, each accounting for \u0026gt;\u0026thinsp;5% of the total coral cover.\u003c/p\u003e \u003cp\u003eThe total effect of depth on relative coral cover loss was significant and negative (-2.074 [95% CI: -2.445 ; -1.696]; Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e), confirming that coral mortality decreased with increasing depth. This attenuation was partly mediated by the two ecological variables tested. Initial coral community composition accounted for 13.0% [0.4 ; 24.4] of the effect of depth (indirect effect = -0.270 [-0.516 ; -0.007]), while pre-bleaching coral cover explained 13.4% [5.0 ; 20.7] (indirect effect = -0.279 [-0.443 ; -0.098]). Combined, these two mediators significantly explained 26.5% [11.5 ; 38.6] of the reduction in coral mortality with depth (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). However, the majority of the effect (73.5% [61.4 ; 88.5]) remained unexplained by these variables (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eSummary of the multiple mediation analysis looking at the effect of depth (as numeric) mediated by coral composition and coral cover before bleaching on the relative coral cover loss (in percentage). Variables in bold have significant indirect/direct effect. CI intervals (in brackets) come from 5000 bootstrap samples.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\"\u0026minus;\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eVariable\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eEffect (95% CI)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRelative effect (%)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003eIndirect\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003eJoint effect\u003c/b\u003e\u003c/p\u003e \u003cp\u003e\u003cb\u003epre-bleaching composition\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026minus;\" colname=\"c3\"\u003e \u003cp\u003e-0.270 (-0.516 ; -0.007)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e13.0 (0.4 ; 24.4)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003eInitial coral cover\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026minus;\" colname=\"c3\"\u003e \u003cp\u003e-0.279 (-0.443 ; -0.098)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e13.4 (5.0 ; 20.7)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003eAll mediators\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026minus;\" colname=\"c3\"\u003e \u003cp\u003e-0.549 (-0.829 ; -0.219)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e26.5 (11.5 ; 38.6)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDirect\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003eDepth\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026minus;\" colname=\"c3\"\u003e \u003cp\u003e-1.525 (-1.930 ; -1.172)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e73.5 (61.4 ; 88.5)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTotal\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003eDepth\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026minus;\" colname=\"c3\"\u003e \u003cp\u003e-2.074 (-2.445 ; -1.696)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003eAn unprecedented heatwave resulting into widespread bleaching and mortality\u003c/h2\u003e \u003cp\u003eThe 2019 MHW was the most intense thermal stress event ever recorded in Mo\u0026rsquo;orea, surpassing all previously documented bleaching-inducing MHWs in 1987, 1991, 1994, and 2003. It was marked by an abrupt onset, with ~\u0026thinsp;3\u0026deg;C-weeks of thermal anomalies accumulating in just two to three weeks. This acute heat stress triggered widespread coral bleaching across the island, with roughly 50% of all coral colonies showing visible signs of bleaching, and a resulting loss of half of the coral cover on shallow reefs. These figures rank among the highest levels of bleaching and mortality ever recorded on the island (Gleason \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e1993\u003c/span\u003e; Hoegh-Guldberg and Salvat \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e1995\u003c/span\u003e; Penin et al. \u003cspan citationid=\"CR83\" class=\"CitationRef\"\u003e2007\u003c/span\u003e; Carroll et al. \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; H\u0026eacute;douin et al. \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Speare et al. \u003cspan citationid=\"CR103\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Bleaching was even so severe that reefs shifted from carbon sinks to carbon sources (Seabrook et al. \u003cspan citationid=\"CR99\" class=\"CitationRef\"\u003e2023\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eWhat makes this event particularly striking is that such severe ecological impacts were observed under what would conventionally be considered \u0026ldquo;moderate\u0026rdquo; thermal stress levels (\u0026le;\u0026thinsp;4\u0026deg;C-weeks; Liu et al. \u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e2003\u003c/span\u003e). With a DHW barely exceeding 3\u0026deg;C-weeks, the 2019 event caused coral mortality levels that are typically only seen under much higher DHW values. For instance, on the Great Barrier Reef in 2016, coral cover remained mostly stable for DHW values below 3\u0026deg;C-weeks and only declined by 40% at 4\u0026deg;C-weeks, with \u0026gt;\u0026thinsp;80% losses only occurring at DHW\u0026thinsp;\u0026ge;\u0026thinsp;9\u0026deg;C-weeks (Hughes et al. \u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e2018b\u003c/span\u003e). Similarly, aerial surveys in Hawai\u0026rsquo;i reported 50% coral mortality only above 10\u0026deg;C-weeks (Asner et al. \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2022\u003c/span\u003e), while mortality of 42% of \u003cem\u003eAcropora\u003c/em\u003e required 11\u0026deg;C-weeks in Japan (Sakai et al. \u003cspan citationid=\"CR96\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). In Little Cayman, a DHW of 17 in 2023\u0026ndash;2024 led to 80% of colonies bleaching and 54% mortality, mostly in sensitive corals (Doherty et al. \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). In contrast, Mo\u0026rsquo;orea\u0026rsquo;s 2019 heatwave caused similar or greater bleaching and mortality with a much lower heat exposure, strongly suggesting a heightened vulnerability of coral communities.\u003c/p\u003e \u003cp\u003eThis amplified sensitivity likely reflects the thermal naivety of Mo\u0026rsquo;orea\u0026rsquo;s coral assemblages. The reef was still recovering from the massive ecological collapse caused by the 2007\u0026ndash;2009 crown-of-thorn outbreak and Cyclone Oli in 2010, which had reduced coral cover on the outer slope to near zero by 2011 (Moritz et al. \u003cspan citationid=\"CR73\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). While coral communities had steadily rebuilt in the following years, they had not experienced any significant thermal stress during this recovery. The only recent heat event prior to 2019 was the 2016 bleaching, which, although thermally mild (DHW\u0026thinsp;\u0026lt;\u0026thinsp;1\u0026deg;C-week), still triggered bleaching in 77% of coral colonies \u0026ndash; particularly in \u003cem\u003eAstrea\u003c/em\u003e and \u003cem\u003ePocillopora\u003c/em\u003e, while \u003cem\u003eAcropora\u003c/em\u003e remained largely unaffected (H\u0026eacute;douin et al. \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). In Mo\u0026rsquo;orea, the 2019 heatwave thus acted on a reef that had not yet experienced acute thermal stress, resulting in disproportionately high mortality for a moderate DHW. These findings highlight the limitations of DHW as a standalone predictor of bleaching severity and support the need to refine how thermal stress is quantified, accounting for local ecological context, including disturbance history, recovery dynamics and species composition. Incorporating metrics that account for prior thermal exposure, low-magnitude but persistent heat stress, and the influence of additional global and local stressors may significantly improve predictive accuracy, as suggested by recent studies (Heron et al. \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Pereira et al. \u003cspan citationid=\"CR84\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Szereday et al. \u003cspan citationid=\"CR106\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Whitaker and DeCarlo \u003cspan citationid=\"CR115\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). In our case, lowering the DHW threshold from +\u0026thinsp;1\u0026deg;C to +\u0026thinsp;0\u0026deg;C allowed us to better capture the period of oscillating thermal anomalies between the MMM and MMM\u0026thinsp;+\u0026thinsp;1\u0026deg;C in the weeks leading up to bleaching onset. As a result, cumulative heat stress measured using DHW\u003csub\u003e0\u0026deg;C\u003c/sub\u003e was more than twice that measured using the standard 1\u0026deg;C cutoff (DHW\u003csub\u003e1\u0026deg;C\u003c/sub\u003e), aligning more closely with the severity of observed bleaching patterns.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003eLong-standing patterns of bleaching sensitivity\u003c/h2\u003e \u003cp\u003eThe taxonomic hierarchy of bleaching susceptibility observed during the 2019 event closely mirrors long-standing patterns previously documented in Mo\u0026rsquo;orea and across the Pacific and Indian Oceans (Loya et al. \u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e2001\u003c/span\u003e; Penin et al. \u003cspan citationid=\"CR83\" class=\"CitationRef\"\u003e2007\u003c/span\u003e; Van Woesik et al. \u003cspan citationid=\"CR111\" class=\"CitationRef\"\u003e2011\u003c/span\u003e; Guest et al. \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2012\u003c/span\u003e; Pratchett et al. \u003cspan citationid=\"CR88\" class=\"CitationRef\"\u003e2013\u003c/span\u003e; Carroll et al. \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Muir et al. \u003cspan citationid=\"CR75\" class=\"CitationRef\"\u003e2017a\u003c/span\u003e; Baird et al. \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2018a\u003c/span\u003e; Crosbie et al. \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Winston et al. \u003cspan citationid=\"CR117\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Consistent with earlier studies (Penin et al. \u003cspan citationid=\"CR83\" class=\"CitationRef\"\u003e2007\u003c/span\u003e; Carroll et al. \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; H\u0026eacute;douin et al. \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2020\u003c/span\u003e), \u003cem\u003eAcropora\u003c/em\u003e, \u003cem\u003eMontipora\u003c/em\u003e, and \u003cem\u003eAstrea\u003c/em\u003e emerged as the most sensitive genera, each exhibiting bleaching prevalence exceeding 90%. \u003cem\u003ePocillopora\u003c/em\u003e showed intermediate susceptibility, while \u003cem\u003ePorites\u003c/em\u003e and \u003cem\u003ePavona\u003c/em\u003e remained the most resistant, with minimal to no signs of bleaching. These findings suggest that the characteristic gradient of taxon-specific sensitivity has re-established on Mo\u0026rsquo;orea\u0026rsquo;s reefs, despite the major ecological reset and subsequent community rebuilding that followed the crown-of-thorns outbreak and cyclone in the late 2000s (H\u0026eacute;douin et al. \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Moritz et al. \u003cspan citationid=\"CR73\" class=\"CitationRef\"\u003e2021\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eInterestingly, some genera appeared more vulnerable in 2019 than during previous bleaching events. For instance, on average\u0026thinsp;~\u0026thinsp;50% of \u003cem\u003ePocillopora\u003c/em\u003e colonies showed signs of bleaching in this study, compared to only\u0026thinsp;~\u0026thinsp;20\u0026ndash;30% during the successive 2002\u0026ndash;2003 mass bleaching events, while \u003cem\u003eAcropora\u003c/em\u003e not only reached higher bleaching levels (\u0026gt;\u0026thinsp;93%) but also showed a more severe response (mostly fully bleached compared to ~\u0026thinsp;80%, primarily partially bleached colonies in 2003; Carroll et al. \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). These heightened responses in 2019 compared with 2002\u0026ndash;2003, even though marine heatwaves appeared similar in force (DHW\u0026thinsp;~\u0026thinsp;3\u0026deg;C-weeks), likely reflect the lack of recent thermal exposure prior to 2019 and support the hypothesis that recurrent and successive bleaching events seemingly drive acclimation or the selective removal of more susceptible genotypes (Guest et al. \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2012\u003c/span\u003e; Pratchett et al. \u003cspan citationid=\"CR88\" class=\"CitationRef\"\u003e2013\u003c/span\u003e) \u0026ndash; mechanisms which had likely not yet acted at full strength in Mo\u0026rsquo;orea\u0026rsquo;s recovering coral assemblages.\u003c/p\u003e \u003cp\u003eWhile bleaching susceptibility followed a well-defined taxonomic hierarchy, post-bleaching mortality did not scale proportionally with bleaching prevalence. Indeed, although at least six genera exhibited widespread and often severe bleaching, only two experienced substantial post-bleaching mortality. Averaged across depths, \u003cem\u003eAcropora\u003c/em\u003e lost half of its coral cover, indicating a particularly high risk of bleaching-induced death. Although \u003cem\u003ePocillopora\u003c/em\u003e showed intermediate susceptibility, with roughly 45% of colonies bleaching, it experienced disproportionately higher mortality, losing on average\u0026thinsp;~\u0026thinsp;37% of its coral cover, more closely tracking bleaching prevalence than \u003cem\u003eAcropora\u003c/em\u003e did. In contrast, \u003cem\u003eMontipora\u003c/em\u003e, \u003cem\u003eAstrea\u003c/em\u003e, \u003cem\u003eMillepora\u003c/em\u003e and \u003cem\u003eLeptastrea\u003c/em\u003e that experienced widespread bleaching recorded little to no mortality. Coral cover of these taxa, as well as that of \u003cem\u003ePorites\u003c/em\u003e and \u003cem\u003ePavona\u003c/em\u003e \u0026ndash; despite enhanced bleaching with depth \u0026ndash; remained largely unchanged, indicating high resilience. As such, high bleaching prevalence does not always correlate with high mortality, even in highly susceptible taxa (Sampayo et al. \u003cspan citationid=\"CR97\" class=\"CitationRef\"\u003e2008\u003c/span\u003e; Crosbie et al. \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Banha et al. \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; H\u0026eacute;douin et al. \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Page et al. \u003cspan citationid=\"CR82\" class=\"CitationRef\"\u003e2023\u003c/span\u003e), highlighting instead strongly taxon- and context-dependent recovery potential. Documenting local coral assemblage composition and historic patterns of bleaching susceptibility of taxa, even species when possible, thus appears highly important to be able to robustly predict demographic outcomes following increasingly intense and frequent MHWs (Hughes et al. \u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e2018b\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003eDepth weakly mitigated bleaching but strongly limited mortality\u003c/h2\u003e \u003cp\u003eAcross the 2019 marine heatwave, depth up to 20m provided only limited protection against bleaching initiation but instead strongly reduced bleaching-induced mortality. Bleaching prevalence showed little to no reduction between 5 and 12m (-6% to +\u0026thinsp;5%) across dominant and sensitive coral genera, with a more pronounced, yet still moderate, decline observed between 12 and 20m for \u003cem\u003eAcropora\u003c/em\u003e, \u003cem\u003eAstrea\u003c/em\u003e, \u003cem\u003eMontipora\u003c/em\u003e and \u003cem\u003ePocillopora\u003c/em\u003e. This attenuation ranged from \u0026minus;\u0026thinsp;11% in \u003cem\u003eAcropora\u003c/em\u003e to -24-29% in \u003cem\u003eMontipora\u003c/em\u003e and \u003cem\u003ePocillopora\u003c/em\u003e. Despite this decline, bleaching prevalence remained high at all depths, exceeding 70% for \u003cem\u003eAcropora\u003c/em\u003e, \u003cem\u003eAstrea\u003c/em\u003e and \u003cem\u003eMontipora\u003c/em\u003e, and remaining above 35% for \u003cem\u003ePocillopora\u003c/em\u003e. Physiological thresholds for bleaching were thus widely exceeded throughout the depth range surveyed and bleaching-inducing stress was hardly alleviated for most sensitive taxa (Helgoe et al. \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2024\u003c/span\u003e), if at all in \u003cem\u003eLeptastrea\u003c/em\u003e and \u003cem\u003eMillepora\u003c/em\u003e, which showed no reduction in bleaching probability with depth. As bleaching likelihood increased with depth in \u003cem\u003ePorites\u003c/em\u003e and \u003cem\u003ePavona\u003c/em\u003e, bleaching prevalence at the community level was not substantially reduced from 5 to 20m, with half of corals bleaching overall, more than three-quarters of which were severely affected. Conversely, the modest reduction in bleaching prevalence was mirrored by a far more substantial and linear decline in post-bleaching mortality with increasing depth. \u003cem\u003eAcropora\u003c/em\u003e suffered near-total mortality at 5m (-74% coral cover) and still exhibited substantial loss at 20m, though divided by more than half (-29%), and mortality was reduced by 39% in \u003cem\u003ePocillopora\u003c/em\u003e (from \u0026minus;\u0026thinsp;47% to -28% between 5 and 20m).\u003c/p\u003e \u003cp\u003eReduction of bleaching and mortality with depth were unlikely to have resulted from depth-related thermal relief along the 5-20m gradient. \u003cem\u003eIn-situ\u003c/em\u003e temperature records revealed no significant differences in thermal intensity or variability between 10 and 20m during the 2019 heatwave. This absence of thermal stratification is consistent with the anticyclonic conditions that prevailed during the austral summer, which substantially reduced internal wave cooling (IWC; Wyatt et al. \u003cspan citationid=\"CR119\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). While internal waves can normally cool waters as shallow as ~\u0026thinsp;10m (e.g. average depth of the 29.8\u0026deg;C bleaching isotherm was 15m in 2016), this threshold deepened to ~\u0026thinsp;28m in 2019, effectively confining thermal relief to depths below those surveyed (Wyatt et al. \u003cspan citationid=\"CR119\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Under these conditions, depth-related mitigation of bleaching, and especially of mortality, likely stemmed primarily from light attenuation (Baird et al. \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2018b\u003c/span\u003e). Reduced irradiance can limit photosystem saturation and bleaching-inducing oxidative stress, particularly under thermal stress when repair mechanisms are impaired (Baird et al. \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2018b\u003c/span\u003e; Helgoe et al. \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Notably, most of the reduction in bleaching associated with light attenuation is known to occur at very shallow depths, typically between \u0026lt;7m, where irradiance declines most steeply (Baird et al. \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2018a\u003c/span\u003e, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2018b\u003c/span\u003e; Laverick et al. \u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Asner et al. \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). By starting at 5m, our surveys likely missed the portion of the depth gradient where light-driven mitigation of bleaching is strongest. At this depth, UV radiation were probably already strongly attenuated (Dunne and Brown \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e1996\u003c/span\u003e; Kuwahara et al. \u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e2010\u003c/span\u003e; Downs et al. \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2013b\u003c/span\u003e), suggesting depth-related refuge within our study range primarily reflected reductions in photosynthetically active radiation (PAR). Therefore, consistent with limited thermal relief, studies from the northern Great Barrier Reef during the 2016 mass bleaching reported similarly limited depth refuges because of limited upwelling activity. Bleaching declined modestly from 69% at 5-10m to 60% at 25m (Frade et al. \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2018\u003c/span\u003e), and for \u003cem\u003eAcropora\u003c/em\u003e specifically, little change was observed between 5 and 10m, followed by a\u0026thinsp;~\u0026thinsp;24% decline between 10 and 20m (Crosbie et al. \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). On the contrary, regions where IWC extended into shallower waters during MHW, such as the Maldives in 2016, exhibited much stronger depth refuge, with bleaching risk declining by ~\u0026thinsp;23% between 3\u0026ndash;5 m and 10m and by up to 60% between 3\u0026ndash;5 m and 24\u0026ndash;30 m (Muir et al. \u003cspan citationid=\"CR76\" class=\"CitationRef\"\u003e2017b\u003c/span\u003e; Cowburn et al. \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). Together, these regional differences suggest that strong depth refugia from bleaching emerge most clearly when light attenuation and thermal relief act synergistically, typically beyond the upper 20m of the water column. It is therefore likely that surveying greater depths would have revealed stronger mitigation of both bleaching and mortality, potentially leading to negligible mortality for \u003cem\u003eAcropora\u003c/em\u003e and \u003cem\u003ePocillopora\u003c/em\u003e below 20m (Muir et al. \u003cspan citationid=\"CR75\" class=\"CitationRef\"\u003e2017a\u003c/span\u003e; Crosbie et al. \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2019\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eBeyond physical drivers, depth zonation of species likely modulated how taxa responded to depth-related gradients. While our analyses were conducted at the genus level, we acknowledge that such taxonomic resolution can obscure important mechanisms underlying depth-related reductions in bleaching impacts. Substantial variability in bleaching susceptibility and survival has been documented between species from the same genera, even within highly sensitive taxa like \u003cem\u003eAcropora\u003c/em\u003e (Marshall and Baird \u003cspan citationid=\"CR63\" class=\"CitationRef\"\u003e2000\u003c/span\u003e; Muir et al. \u003cspan citationid=\"CR75\" class=\"CitationRef\"\u003e2017a\u003c/span\u003e, \u003cspan citationid=\"CR74\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Depth-related shifts in species composition, such as differences in the relative abundance of shallow specialists versus depth generalists (Bridge et al. \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Roberts et al. \u003cspan citationid=\"CR92\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; P\u0026eacute;rez-Rosales et al. \u003cspan citationid=\"CR85\" class=\"CitationRef\"\u003e2022\u003c/span\u003e), can therefore influence both the probability of bleaching and subsequent mortality within genera. In reefs where depth-sensitive species have restricted vertical distributions, sharp mortality thresholds have been observed, as illustrated by the near-complete disappearance of \u003cem\u003eAcropora\u003c/em\u003e mortality beyond 8m during the 2010 heatwave in Pulau Weh (Bridge et al. \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2014\u003c/span\u003e). In Mo\u0026rsquo;orea, the more gradual decline in mortality observed here, with no clear inflexion point, suggests broader vertical distributions of vulnerable species, resulting in smoother depth-related patterns.\u003c/p\u003e \u003cp\u003eAlthough depth-related alleviation of bleaching-inducing stressors was insufficient to strongly prevent bleaching onset in most taxa, the steep decline in mortality suggests that depth primarily limited the severity of physiological damage after bleaching occurred. Light attenuation with depth likely constrained further accumulation of oxidative damage following bleaching, thereby enhancing recovery and survival (Brown et al. \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2000\u003c/span\u003e; Downs et al. \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2013a\u003c/span\u003e; Coles et al. \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). Photo-physiological acclimation with depth may have further contributed to both reduced bleaching severity and enhanced survival at depth, particularly through changes in endosymbiont community composition (Bongaerts et al. \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Ezzat et al. \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Wall et al. \u003cspan citationid=\"CR113\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). An increasing dominance of \u003cem\u003eCladocopium\u003c/em\u003e symbionts with depth has been linked to higher carbon translocation and assimilation by the coral host (Ezzat et al. \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Wall et al. \u003cspan citationid=\"CR113\" class=\"CitationRef\"\u003e2020\u003c/span\u003e) and to higher heterotrophic plasticity, even in corals from mesophotic depths (~\u0026thinsp;50m; Ezzat et al. \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). Such traits may have conferred a competitive advantage to deeper corals, possibly better energetically equipped to withstand stress, especially if it was concurrently reduced with increasing depth. Given that depth was identified as the strongest environmental driver of \u003cem\u003eSymbiodiniaceae\u003c/em\u003e composition in \u003cem\u003eMontipora capitata\u003c/em\u003e across a much shallower gradient (0.5\u0026ndash;3.5 m; De Souza et al. \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2023\u003c/span\u003e), its influence is likely even more pronounced for some taxa across the broader depth range examined here.\u003c/p\u003e \u003cp\u003eDespite trophic plasticity, depth can strongly slow growth, particularly in branching corals, usually resulting in smaller colonies with increasing depth for a given age (Kramer et al. \u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Colony size has repeatedly been shown to influence both bleaching susceptibility and mortality, with larger colonies of \u003cem\u003eAcropora\u003c/em\u003e and \u003cem\u003ePocillopora\u003c/em\u003e bleaching more frequently and experiencing higher mortality (Speare et al. \u003cspan citationid=\"CR103\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Winslow et al. \u003cspan citationid=\"CR116\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Notably, both studies analysed the same 2019 bleaching event in Mo\u0026rsquo;orea, indicating that if colony size was effectively reduced at depth, it likely contributed jointly to the observed reduction in bleaching susceptibility and increased resilience in these taxa.\u003c/p\u003e \u003cp\u003eUltimately, differences in resilience among coral genera, combined with their abundance and depth distributions, mediated overall coral cover loss. Depth can strongly structure coral community composition through associated thermal and light gradients (Tamir et al. \u003cspan citationid=\"CR107\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Laverick et al. \u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; P\u0026eacute;rez-Rosales et al. \u003cspan citationid=\"CR85\" class=\"CitationRef\"\u003e2022\u003c/span\u003e), a pattern that was also evident here prior to bleaching. Resistant genera such as \u003cem\u003ePorites\u003c/em\u003e and \u003cem\u003ePavona\u003c/em\u003e increased in relative abundance with depth, whereas highly sensitive taxa (\u003cem\u003eAcropora\u003c/em\u003e, \u003cem\u003ePocillopora\u003c/em\u003e, \u003cem\u003eMontipora\u003c/em\u003e) declined. Such depth distributions concurrently with the strong depth-related reduction in mortality of dominant \u003cem\u003ePocillopora\u003c/em\u003e resulted in a 42% lower coral cover loss at 12m compared to 5m, followed by a further 52% decline between 12 and 20m. Changes in pre-bleaching assemblage structure alone explained 13% of this protective effect, increasing to 31% when combined with initial coral cover. To our knowledge, this provides one of the first quantitative estimates of the indirect influence of ecological context on post-bleaching mortality. Our results are consistent with earlier findings from the 2002 bleaching event in Mo\u0026rsquo;orea, where the composition of coral assemblages was hypothesized to drive patterns in bleaching across depths (Penin et al. \u003cspan citationid=\"CR83\" class=\"CitationRef\"\u003e2007\u003c/span\u003e). At that time, higher bleaching incidence at 18m compared to 6m was attributed to the greater relative abundance of sensitive taxa at depth. However, the ecological reset following the 2007\u0026ndash;2010 disturbances profoundly restructured coral assemblages, with sensitive taxa now decreasing and resistant taxa increasing with depth, reversing pre-disturbance patterns. As a result, assemblage structure alone can no longer account for the depth-related bleaching and mortality patterns observed in Mo\u0026rsquo;orea during the 2019 heatwave. Instead, while thermal refuge was constrained here, contemporary depth responses were more likely driven by light gradients, that may have limited bleaching onset and more thoroughly ensuing mortality.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003eImplications for coral reef monitoring and conservation\u003c/h2\u003e \u003cp\u003eThe taxon- and depth-specific mortality patterns identified in this study have important implications for coral reef resilience and management. In Mo\u0026rsquo;orea, where pocilloporids were a dominant component of reef assemblages, abrupt losses of these ecologically important branching corals are likely to have disproportionately large consequences for reef architecture, habitat availability and associated biodiversity. Such impacts were probably most pronounced on the northern coast, particularly at E2B, where exceptionally high coral cover (\u0026gt;\u0026thinsp;75%) might have correlated with higher reef fish abundance and richness (Darling et al. \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Russ et al. \u003cspan citationid=\"CR94\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). In this context, reduced mortality observed at depth may contribute to preserving reef fish assemblages over the long term (Crosbie et al. \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2019\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe persistence of relatively undisturbed deeper coral assemblages \u0026ndash; three times less affected at 20m than at 5m \u0026ndash; also highlight their potential role as sources of larvae to replenish impacted shallow zones (Leinbach et al. \u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e2021\u003c/span\u003e), provided sufficient vertical connectivity exists. The absence of an abrupt inflection point in depth-related mortality and important turnover of generic dominance between 5 and 20m could support the existence of at least partial demographic connectivity across depths (Bridge et al. \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Laverick et al. \u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; P\u0026eacute;rez-Rosales et al. \u003cspan citationid=\"CR85\" class=\"CitationRef\"\u003e2022\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThese findings also add to previous evidence on the limits of depth refugia as its benefits are highly dependent on local ecological and environmental context, especially sensitive to large-scale oceanographic processes (Frade et al. \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Wyatt et al. \u003cspan citationid=\"CR119\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Had typical IWC occurred, reefs as shallow as 15m may have experienced thermal buffering (Wyatt et al. \u003cspan citationid=\"CR118\" class=\"CitationRef\"\u003e2020\u003c/span\u003e, \u003cspan citationid=\"CR119\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Therefore, the increasing intensity and frequency of MHWs may undermine the reliability of the \u0026ldquo;deep reef refuge\u0026rdquo; hypothesis may be increasingly uncertain given its dependence on transient cooling processes (Hughes et al. \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e2018a\u003c/span\u003e; Eakin et al. \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). Together, these patterns emphasize the need for future bleaching surveys and monitoring programs to systematically incorporate multiple depths to accurately quantify reef-wide impacts, recovery potential and connectivity pathways. From a management perspective, this further reinforces the importance of protecting deeper reef habitats as integral components of reef resilience.\u003c/p\u003e \u003cp\u003eAdditionally, reef habitats below 30m may offer more consistent protection from extreme thermal stress. In our study, these zones experienced low heat accumulation and exhibited high daily thermal variability (MDTF), conditions previously linked to enhanced coral tolerance (Oliver and Palumbi \u003cspan citationid=\"CR81\" class=\"CitationRef\"\u003e2011\u003c/span\u003e; Safaie et al. \u003cspan citationid=\"CR95\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Wyatt et al. \u003cspan citationid=\"CR118\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). These deeper areas could represent promising targets for active conservation, such as assisted reef restoration or the temporary relocation of coral nurseries during heat stress events (Tavakoli-Kolour et al. \u003cspan citationid=\"CR108\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Henry et al. \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). Expanding technical diving and improving monitoring of deeper reefs will be critical to assess the feasibility, risks and long-term benefits of these approaches.\u003c/p\u003e \u003c/div\u003e"},{"header":"Conclusion","content":"\u003cp\u003eBy combining \u003cem\u003ein-situ\u003c/em\u003e assessments of bleaching, mortality and community structure with depth-resolved thermal data, this study clarifies how depth can modulate coral responses to extreme thermal stress during the unprecedented 2019 marine heatwave in Mo\u0026rsquo;orea. Bleaching affected over half of coral colonies across all depths, with only limited attenuation along the reef slope in highly sensitive taxa, indicating widespread exposure to acute stress even at 20m. In contrast, post-bleaching mortality declined more sharply with depth, with coral cover loss reduced by nearly 3.5-fold between 5m and 20m. Although bleaching and mortality still occurred at 20m, deeper assemblages were less vulnerable to disturbance than shallow reefs. Mediation analyses showed that differences in coral cover and composition prior to bleaching explained only a quarter of the depth effect on mortality, pointing to additional depth-related protective mechanisms. As temperature gradients were absent, our results highlight the importance light reduction with depth can have on mitigating bleaching and its impacts. Together, these findings contribute to the evaluation of the deep reef refuge hypothesis and highlight the need to incorporate vertical habitat gradients into bleaching assessments and conservation strategies as marine heatwaves intensify.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e \u003ch2\u003eConflict of interest\u003c/h2\u003e \u003cp\u003eThe authors declare that they have no conflict of interest.\u003c/p\u003e \u003c/p\u003e\u003ch2\u003eFunding\u003c/h2\u003e \u003cp\u003eThis research was supported by the \u003cem\u003ePolyBleach\u003c/em\u003e project \u0026ldquo;Assessment of coral bleaching in French Polynesia\u0026rdquo;, funded by the French Ministry for the Ecological Transition and Territorial Cohesion (\u003cem\u003eMinist\u0026egrave;re de la Transition \u0026eacute;cologique et de la Coh\u0026eacute;sion des territoires\u003c/em\u003e) through the IFRECOR program (French Initiative for Coral Reefs).\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eCB, YL, AM, CD and LH contributed to the study conception and design. YL, AM, CD and LH planned and carried out the fieldwork. CB conducted the data analysis and GS helped in retrieving the multiple mediation analysis results. LH secured funding for and supervised this research. The manuscript was written by CB. All authors read, reviewed and approved the final manuscript.\u003c/p\u003e\u003ch2\u003eAcknowledgement\u003c/h2\u003e\u003cp\u003eWe extend our thanks to everyone involved in the collection of the data, whether it is the dive support team, students and engineers at CRIOBE research station. We also thank Andreas Eich for his help in cleaning the data.\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003eAll scripts and data necessary to replicate the analysis can be publicly accessed here: [https://github.com/clr-btl/Bleaching-2019](https:/github.com/clr-btl/Bleaching-2019) .\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eAdjeroud M (1997) Factors influencing spatial patterns on coral reefs around Moorea, French Polynesia. Mar Ecol Prog Ser 159:105\u0026ndash;119. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3354/meps159105\u003c/span\u003e\u003cspan address=\"10.3354/meps159105\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAdjeroud M, Michonneau F, Edmunds PJ, Chancerelle Y, De Loma TL, Penin L, Thibaut L, Vidal-Dupiol J, Salvat B, Galzin R (2009) Recurrent disturbances, recovery trajectories, and resilience of coral assemblages on a South Central Pacific reef. Coral Reefs 28:775\u0026ndash;780. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00338-009-0515-7\u003c/span\u003e\u003cspan address=\"10.1007/s00338-009-0515-7\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAsner GP, Vaughn NR, Martin RE, Foo SA, Heckler J, Neilson BJ, Gove JM (2022) Mapped coral mortality and refugia in an archipelago-scale marine heat wave. Proc Natl Acad Sci 119:e2123331119. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1073/pnas.2123331119\u003c/span\u003e\u003cspan address=\"10.1073/pnas.2123331119\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBaird A, Madin J, \u0026Aacute;lvarez-Noriega M, Fontoura L, Kerry J, Kuo C, Precoda K, Torres-Pulliza D, Woods R, Zawada K, Hughes T (2018a) A decline in bleaching suggests that depth can provide a refuge from global warming in most coral taxa. Mar Ecol Prog Ser 603:257\u0026ndash;264. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3354/meps12732\u003c/span\u003e\u003cspan address=\"10.3354/meps12732\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBaird ME, Mongin M, Rizwi F, Bay LK, Cantin NE, Soja-Woźniak M, Skerratt J (2018b) A mechanistic model of coral bleaching due to temperature-mediated light-driven reactive oxygen build-up in zooxanthellae. Ecol Model 386:20\u0026ndash;37. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.ecolmodel.2018.07.013\u003c/span\u003e\u003cspan address=\"10.1016/j.ecolmodel.2018.07.013\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBanha TNS, Capel KCC, Kitahara MV, Francini-Filho RB, Francini CLB, Sumida PYG, Mies M (2020) Low coral mortality during the most intense bleaching event ever recorded in subtropical Southwestern Atlantic reefs. Coral Reefs 39:515\u0026ndash;521. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00338-019-01856-y\u003c/span\u003e\u003cspan address=\"10.1007/s00338-019-01856-y\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBongaerts P, Carmichael M, Hay KB, Tonk L, Frade PR, Hoegh-Guldberg O (2015) Prevalent endosymbiont zonation shapes the depth distributions of scleractinian coral species. R Soc Open Sci 2:140297. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1098/rsos.140297\u003c/span\u003e\u003cspan address=\"10.1098/rsos.140297\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBongaerts P, Ridgway T, Sampayo EM, Hoegh-Guldberg O (2010) Assessing the \u0026lsquo;deep reef refugia\u0026rsquo; hypothesis: focus on Caribbean reefs. Coral Reefs 29:309\u0026ndash;327. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00338-009-0581-x\u003c/span\u003e\u003cspan address=\"10.1007/s00338-009-0581-x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBosserelle P, Berteaux-Lecellier V, Chancerelle Y, H\u0026eacute;douin L, Nugues M, Wallace C, Pichon M (2014) Guide d\u0026rsquo;identification des coraux de Moorea. CRIOBE, Papetoai, Moorea (Polyn\u0026eacute;sie fran\u0026ccedil;aise)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBridge TCL, Hoey AS, Campbell SJ, Muttaqin E, Rudi E, Fadli N, Baird AH (2014) Depth-dependent mortality of reef corals following a severe bleaching event: implications for thermal refuges and population recovery. F1000Research 2:187. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.12688/f1000research.2-187.v3\u003c/span\u003e\u003cspan address=\"10.12688/f1000research.2-187.v3\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBrown B, Dunne R, Warner M, Ambarsari I, Fitt W, Gibb S, Cummings D (2000) Damage and recovery of Photosystem II during a manipulative field experiment on solar bleaching in the coral Goniastrea aspera. Mar Ecol Prog Ser 195:117\u0026ndash;124.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBruno JF, Selig ER (2007) Regional Decline of Coral Cover in the Indo-Pacific: Timing, Extent, and Subregional Comparisons. PLoS ONE 2:e711. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1371/journal.pone.0000711\u003c/span\u003e\u003cspan address=\"10.1371/journal.pone.0000711\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCantin N, Spalding M (2018) Detecting and monitoring coral bleaching events. Coral Bleaching: Patterns, Processes, Causes and Consequences. Springer, pp 85\u0026ndash;110\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCarlot J, Rouz\u0026eacute; H, Barneche DR, Merci\u0026egrave;re A, Espiau B, Cardini U, Brandl SJ, Casey JM, P\u0026eacute;rez-Rosales G, Adjeroud M, H\u0026eacute;douin L, Parravicini V (2022) Scaling up calcification, respiration, and photosynthesis rates of six prominent coral taxa. Ecol Evol 12:e8613. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/ece3.8613\u003c/span\u003e\u003cspan address=\"10.1002/ece3.8613\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCarroll AG, Harrison PL, Adjeroud M (2017) Susceptibility of coral assemblages to successive bleaching events at Moorea, French Polynesia. Mar Freshw Res 68:760. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1071/MF15134\u003c/span\u003e\u003cspan address=\"10.1071/MF15134\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChollett I, Mumby P, Cort\u0026eacute;s J (2010) Upwelling areas do not guarantee refuge for coral reefs in a warming ocean. Mar Ecol Prog Ser 416:47\u0026ndash;56. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3354/meps08775\u003c/span\u003e\u003cspan address=\"10.3354/meps08775\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eClaar DC, Cobb KM, Baum JK (2019) In situ and remotely sensed temperature comparisons on a Central Pacific atoll. Coral Reefs 38:1343\u0026ndash;1349. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00338-019-01850-4\u003c/span\u003e\u003cspan address=\"10.1007/s00338-019-01850-4\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eColes SL, Bahr KD, Rodgers KS, May SL, McGowan AE, Tsang A, Bumgarner J, Han JH (2018) Evidence of acclimatization or adaptation in Hawaiian corals to higher ocean temperatures. PeerJ 6:e5347. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.7717/peerj.5347\u003c/span\u003e\u003cspan address=\"10.7717/peerj.5347\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCorreia HE, Dee LE, Ferraro PJ (2025) Designing causal mediation analyses to quantify intermediary processes in ecology. Biol Rev 100:1512\u0026ndash;1533. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/brv.70011\u003c/span\u003e\u003cspan address=\"10.1111/brv.70011\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCowburn B, Moritz C, Grimsditch G, Solandt J (2019) Evidence of coral bleaching avoidance, resistance and recovery in the Maldives during the 2016 mass-bleaching event. Mar Ecol Prog Ser 626:53\u0026ndash;67. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3354/meps13044\u003c/span\u003e\u003cspan address=\"10.3354/meps13044\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCrosbie A, Bridge T, Jones G, Baird A (2019) Response of reef corals and fish at Osprey Reef to a thermal anomaly across a 30 m depth gradient. Mar Ecol Prog Ser 622:93\u0026ndash;102. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3354/meps13015\u003c/span\u003e\u003cspan address=\"10.3354/meps13015\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDarling ES, Graham NAJ, Januchowski-Hartley FA, Nash KL, Pratchett MS, Wilson SK (2017) Relationships between structural complexity, coral traits, and reef fish assemblages. Coral Reefs 36:561\u0026ndash;575. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00338-017-1539-z\u003c/span\u003e\u003cspan address=\"10.1007/s00338-017-1539-z\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDe Souza MR, Caruso C, Ruiz-Jones L, Drury C, Gates RD, Toonen RJ (2023) Importance of depth and temperature variability as drivers of coral symbiont composition despite a mass bleaching event. Sci Rep 13:8957. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41598-023-35425-9\u003c/span\u003e\u003cspan address=\"10.1038/s41598-023-35425-9\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDenis V, Ribas-Deulofeu L, Sturaro N, Kuo C-Y, Chen CA (2017) A functional approach to the structural complexity of coral assemblages based on colony morphological features. Sci Rep 7:9849. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41598-017-10334-w\u003c/span\u003e\u003cspan address=\"10.1038/s41598-017-10334-w\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDiaz C, Foster NL, Attrill MJ, Bolton A, Ganderton P, Howell KL, Robinson E, Hosegood P (2023) Mesophotic coral bleaching associated with changes in thermocline depth. Nat Commun 14:6528. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41467-023-42279-2\u003c/span\u003e\u003cspan address=\"10.1038/s41467-023-42279-2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDoherty ML, Johnson JV, Goodbody-Gringley G (2025) Widespread coral bleaching and mass mortality during the 2023\u0026ndash;2024 marine heatwave in Little Cayman. PLOS One 20:e0322636. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1371/journal.pone.0322636\u003c/span\u003e\u003cspan address=\"10.1371/journal.pone.0322636\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDowns CA, McDougall KE, Woodley CM, Fauth JE, Richmond RH, Kushmaro A, Gibb SW, Loya Y, Ostrander GK, Kramarsky-Winter E (2013a) Heat-Stress and Light-Stress Induce Different Cellular Pathologies in the Symbiotic Dinoflagellate during Coral Bleaching. PLoS ONE 8:e77173. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1371/journal.pone.0077173\u003c/span\u003e\u003cspan address=\"10.1371/journal.pone.0077173\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDowns NJ, Schouten PW, Parisi AV (2013b) Seasonal Variations in the Subsurface Ultraviolet-B on an Inshore Pacific Coral Reef Ecosystem. Photochem Photobiol 89:1234\u0026ndash;1243. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/php.12101\u003c/span\u003e\u003cspan address=\"10.1111/php.12101\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDunne RP, Brown BE (1996) Penetration of solar UVB radiation in shallow tropical waters and its potential biological effects on coral reefs; results from the central Indian Ocean and Andaman Sea. Mar Ecol Prog Ser 144:109\u0026ndash;118.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEakin C, Lough J, Heron S, Liu G (2018) Climate variability and change: monitoring data and evidence for increased coral bleaching stress. Coral Bleach Patterns Process Causes Consequences 51\u0026ndash;84. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00338-019-01844-2\u003c/span\u003e\u003cspan address=\"10.1007/s00338-019-01844-2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEakin CM, Sweatman HPA, Brainard RE (2019) The 2014\u0026ndash;2017 global-scale coral bleaching event: insights and impacts. Coral Reefs 38:539\u0026ndash;545. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00338-019-01844-2\u003c/span\u003e\u003cspan address=\"10.1007/s00338-019-01844-2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEyal G, Laverick JH, Ben-Zvi O, Brown KT, Kramer N, Tamir R, Lindemann Y, Levy O, Pandolfi JM (2022) Selective deep water coral bleaching occurs through depth isolation. Sci Total Environ 844:157180. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.scitotenv.2022.157180\u003c/span\u003e\u003cspan address=\"10.1016/j.scitotenv.2022.157180\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEzzat L, Fine M, Maguer J-F, Grover R, Ferrier-Pag\u0026egrave;s C (2017) Carbon and Nitrogen Acquisition in Shallow and Deep Holobionts of the Scleractinian Coral S. pistillata. Front Mar Sci 4: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3389/fmars.2017.00102\u003c/span\u003e\u003cspan address=\"10.3389/fmars.2017.00102\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFrade PR, Bongaerts P, Englebert N, Rogers A, Gonzalez-Rivero M, Hoegh-Guldberg O (2018) Deep reefs of the Great Barrier Reef offer limited thermal refuge during mass coral bleaching. Nat Commun 9:3447. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41467-018-05741-0\u003c/span\u003e\u003cspan address=\"10.1038/s41467-018-05741-0\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGleason M (1993) Effects of disturbance on coral communities: bleaching in Moorea, French Polynesia. Coral Reefs 12:193\u0026ndash;201. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/bf00334479\u003c/span\u003e\u003cspan address=\"10.1007/bf00334479\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGlynn PW (1996) Coral reef bleaching: facts, hypotheses and implications. Glob Change Biol 2:495\u0026ndash;509.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGraham NAJ, Jennings S, MacNeil MA, Mouillot D, Wilson SK (2015) Predicting climate-driven regime shifts versus rebound potential in coral reefs. Nature 518:94\u0026ndash;97. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/nature14140\u003c/span\u003e\u003cspan address=\"10.1038/nature14140\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGuest JR, Baird AH, Maynard JA, Muttaqin E, Edwards AJ, Campbell SJ, Yewdall K, Affendi YA, Chou LM (2012) Contrasting Patterns of Coral Bleaching Susceptibility in 2010 Suggest an Adaptive Response to Thermal Stress. PLoS ONE 7:e33353. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1371/journal.pone.0033353\u003c/span\u003e\u003cspan address=\"10.1371/journal.pone.0033353\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eH\u0026eacute;douin L, Rouz\u0026eacute; H, Berthe C, Perez-Rosales G, Martinez E, Chancerelle Y, Galand PE, Lerouvreur F, Nugues MM, Pochon X (2020) Contrasting patterns of mortality in Polynesian coral reefs following the third global coral bleaching event in 2016. Coral Reefs 39:939\u0026ndash;952. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00338-020-01914-w\u003c/span\u003e\u003cspan address=\"10.1007/s00338-020-01914-w\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHelgoe J, Davy SK, Weis VM, Rodriguez-Lanetty M (2024) Triggers, cascades, and endpoints: connecting the dots of coral bleaching mechanisms. Biol Rev 99:715\u0026ndash;752. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/brv.13042\u003c/span\u003e\u003cspan address=\"10.1111/brv.13042\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHenry JA, Szereday S, Bernard GGR, Patterson JT (2025) Improved coral nursery production through contingent heat stress events via depth manipulation. Aquaculture 605:742558. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.aquaculture.2025.742558\u003c/span\u003e\u003cspan address=\"10.1016/j.aquaculture.2025.742558\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHeron S, Johnston L, Liu G, Geiger E, Maynard J, De La Cour J, Johnson S, Okano R, Benavente D, Burgess T, Iguel J, Perez D, Skirving W, Strong A, Tirak K, Eakin C (2016) Validation of Reef-Scale Thermal Stress Satellite Products for Coral Bleaching Monitoring. Remote Sens 8:59. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/rs8010059\u003c/span\u003e\u003cspan address=\"10.3390/rs8010059\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHoegh-Guldberg O, Salvat B (1995) Periodic mass-bleaching and elevated sea temperatures: bleaching of outer reef slope communities in Moorea, French Polynesia. Mar Ecol Prog Ser 121:181\u0026ndash;190. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3354/meps121181\u003c/span\u003e\u003cspan address=\"10.3354/meps121181\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHoogenboom MO, Frank GE, Chase TJ, Jurriaans S, \u0026Aacute;lvarez-Noriega M, Peterson K, Critchell K, Berry KLE, Nicolet KJ, Ramsby B, Paley AS (2017) Environmental Drivers of Variation in Bleaching Severity of Acropora Species during an Extreme Thermal Anomaly. Front Mar Sci 4:376. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3389/fmars.2017.00376\u003c/span\u003e\u003cspan address=\"10.3389/fmars.2017.00376\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHughes TP, Anderson KD, Connolly SR, Heron SF, Kerry JT, Lough JM, Baird AH, Baum JK, Berumen ML, Bridge TC, Claar DC, Eakin CM, Gilmour JP, Graham NAJ, Harrison H, Hobbs J-PA, Hoey AS, Hoogenboom M, Lowe RJ, McCulloch MT, Pandolfi JM, Pratchett M, Schoepf V, Torda G, Wilson SK (2018a) Spatial and temporal patterns of mass bleaching of corals in the Anthropocene. Science 359:80\u0026ndash;83. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1126/science.aan8048\u003c/span\u003e\u003cspan address=\"10.1126/science.aan8048\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHughes TP, Kerry JT, Baird AH, Connolly SR, Dietzel A, Eakin CM, Heron SF, Hoey AS, Hoogenboom MO, Liu G, McWilliam MJ, Pears RJ, Pratchett MS, Skirving WJ, Stella JS, Torda G (2018b) Global warming transforms coral reef assemblages. Nature 556:492\u0026ndash;496. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41586-018-0041-2\u003c/span\u003e\u003cspan address=\"10.1038/s41586-018-0041-2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJohnston EC, Burgess SC (2023) Pocillopora tuahiniensis: a new species of scleractinian coral (Scleractinia, Pocilloporidae) from French Polynesia. Zootaxa 5369:117\u0026ndash;124. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.11646/zootaxa.5369.1.5\u003c/span\u003e\u003cspan address=\"10.11646/zootaxa.5369.1.5\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKayal M, Vercelloni J, Lison De Loma T, Bosserelle P, Chancerelle Y, Geoffroy S, Stievenart C, Michonneau F, Penin L, Planes S, Adjeroud M (2012) Predator Crown-of-Thorns Starfish (Acanthaster planci) Outbreak, Mass Mortality of Corals, and Cascading Effects on Reef Fish and Benthic Communities. PLoS ONE 7:e47363. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1371/journal.pone.0047363\u003c/span\u003e\u003cspan address=\"10.1371/journal.pone.0047363\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKim SW, Sampayo EM, Sommer B, Sims CA, G\u0026oacute;mez-Cabrera MDC, Dalton SJ, Beger M, Malcolm HA, Ferrari R, Fraser N, Figueira WF, Smith SDA, Heron SF, Baird AH, Byrne M, Eakin CM, Edgar R, Hughes TP, Kyriacou N, Liu G, Matis PA, Skirving WJ, Pandolfi JM (2019) Refugia under threat: Mass bleaching of coral assemblages in high‐latitude eastern Australia. Glob Change Biol 25:3918\u0026ndash;3931. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/gcb.14772\u003c/span\u003e\u003cspan address=\"10.1111/gcb.14772\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKramer N, Tamir R, Eyal G, Loya Y (2020) Coral Morphology Portrays the Spatial Distribution and Population Size-Structure Along a 5\u0026ndash;100 m Depth Gradient. Front Mar Sci 7:615. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3389/fmars.2020.00615\u003c/span\u003e\u003cspan address=\"10.3389/fmars.2020.00615\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKramer PA (2003) Synthesis of coral reef health indicators for the western Atlantic: Results of the AGRRA program(1997\u0026ndash;2000). Atoll Res Bull 496:1\u0026ndash;57.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKuwahara V, Nakajima R, Othman B, Kushairi M, Toda T (2010) Spatial variability of UVR attenuation and bio-optical factors in shallow coral-reef waters of Malaysia. Coral Reefs 29:693\u0026ndash;704.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLamy T, Galzin R, Kulbicki M, Lison de Loma T, Claudet J (2016) Three decades of recurrent declines and recoveries in corals belie ongoing change in fish assemblages. Coral Reefs 35:293\u0026ndash;302. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00338-015-1371-2\u003c/span\u003e\u003cspan address=\"10.1007/s00338-015-1371-2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLaurent V (2019) Atlas climatologique de la Polyn\u0026eacute;sie fran\u0026ccedil;aise. M\u0026eacute;t\u0026eacute;o France, D\u0026eacute;l\u0026eacute;gation interr\u0026eacute;gionale de Polyn\u0026eacute;sie fran\u0026ccedil;aise,\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLaverick JH, Andradi-Brown DA, Exton DA, Bongaerts P, Bridge TCL, Lesser MP, Pyle RL, Slattery M, Wagner D, Rogers AD (2016) To what extent do mesophotic coral ecosystems and shallow reefs share species of conservation interest? Environ Evid 5:16. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/s13750-016-0068-5\u003c/span\u003e\u003cspan address=\"10.1186/s13750-016-0068-5\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLaverick JH, Tamir R, Eyal G, Loya Y (2020) A generalized light-driven model of community transitions along coral reef depth gradients. Glob Ecol Biogeogr 29:1554\u0026ndash;1564. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/geb.13140\u003c/span\u003e\u003cspan address=\"10.1111/geb.13140\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLeinbach SE, Speare KE, Rossin AM, Holstein DM, Strader ME (2021) Energetic and reproductive costs of coral recovery in divergent bleaching responses. Sci Rep 11:23546. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41598-021-02807-w\u003c/span\u003e\u003cspan address=\"10.1038/s41598-021-02807-w\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLenth RV (2025) emmeans: Estimated Marginal Means, aka Least-Squares Means.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLirman D, Formel N, Schopmeyer S, Ault JS, Smith SG, Gilliam D, Riegl B (2014) Percent recent mortality (PRM) of stony corals as an ecological indicator of coral reef condition. Ecol Indic 44:120\u0026ndash;127. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.ecolind.2013.10.021\u003c/span\u003e\u003cspan address=\"10.1016/j.ecolind.2013.10.021\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu G, Skirving WJ, Geiger EF, De La Cour JL, Marsh BL, Heron SF, Tirak KV, Strong AE, Eakin CM (2017) NOAA Coral Reef Watch\u0026rsquo;s 5km satellite coral bleaching heat stress monitoring product suite version 3 and four-month outlook version 4. Reef Encount 32:39\u0026ndash;45.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu G, Strong AE, Skirving W (2003) Remote sensing of sea surface temperatures during 2002 Barrier Reef coral bleaching. Eos Trans Am Geophys Union 84:137\u0026ndash;141. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1029/2003EO150001\u003c/span\u003e\u003cspan address=\"10.1029/2003EO150001\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLoya Y, Sakai K, Yamazato K, Nakano Y, Sambali H, van Woesik R (2001) Coral bleaching: the winners and the losers. Ecol Lett 4:122\u0026ndash;131. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1046/j.1461-0248.2001.00203.x\u003c/span\u003e\u003cspan address=\"10.1046/j.1461-0248.2001.00203.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMarshall PA, Baird AH (2000) Bleaching of corals on the Great Barrier Reef: differential susceptibilities among taxa. Coral Reefs 19:155\u0026ndash;163. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s003380000086\u003c/span\u003e\u003cspan address=\"10.1007/s003380000086\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMartinez Arbizu P (2020) pairwiseAdonis: Pairwise multilevel comparison using adonis. R Package Version 04 1:\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMcClanahan TR (2023) Local heterogeneity of coral reef diversity and environmental stress provides opportunities for small-scale conservation. Divers Distrib 29:1324\u0026ndash;1340. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/ddi.13770\u003c/span\u003e\u003cspan address=\"10.1111/ddi.13770\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMcClanahan TR, Maina JM, Darling ES, Guillaume MMM, Muthiga NA, D\u0026rsquo;agata S, Leblond J, Arthur R, Jupiter SD, Wilson SK, Mangubhai S, Ussi AM, Humphries AT, Patankar V, Shedrawi G, Julius P, Ndagala J, Grimsditch G (2020) Large geographic variability in the resistance of corals to thermal stress. Glob Ecol Biogeogr 29:2229\u0026ndash;2247. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/geb.13191\u003c/span\u003e\u003cspan address=\"10.1111/geb.13191\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMcGillycuddy M, Warton DI, Popovic G, Bolker BM (2025) Parsimoniously Fitting Large Multivariate Random Effects in glmmTMB. J Stat Softw 112:1\u0026ndash;19. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.18637/jss.v112.i01\u003c/span\u003e\u003cspan address=\"10.18637/jss.v112.i01\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMeistertzheim A-L, Pochon X, Wood SA, Ghiglione J-F, H\u0026eacute;douin L (2019) Development of a quantitative PCR\u0026ndash;high-resolution melting assay for absolute measurement of coral-Symbiodiniaceae associations and its application to investigating variability at three spatial scales. Mar Biol 166:13. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00227-018-3458-0\u003c/span\u003e\u003cspan address=\"10.1007/s00227-018-3458-0\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMolina-Hern\u0026aacute;ndez A, Medell\u0026iacute;n‐Maldonado F, Lange ID, Perry CT, \u0026Aacute;lvarez‐Filip L (2022) Coral reef erosion: In situ measurement on different dead coral substrates on a Caribbean reef. Limnol Oceanogr 67:2734\u0026ndash;2749. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/lno.12234\u003c/span\u003e\u003cspan address=\"10.1002/lno.12234\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMoorea Coral Reef LTER, Leichter J, Adam T, Krumbholz H (2025) MCR LTER: Coral Reef: Benthic Water Temperature, ongoing since 2005. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.6073/PASTA/02E0FA99C6FCA29A1BDD26C46013F0F7\u003c/span\u003e\u003cspan address=\"10.6073/PASTA/02E0FA99C6FCA29A1BDD26C46013F0F7\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMorais J, Morais R, Tebbett SB, Bellwood DR (2022) On the fate of dead coral colonies. Funct Ecol 36:3148\u0026ndash;3160. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/1365-2435.14182\u003c/span\u003e\u003cspan address=\"10.1111/1365-2435.14182\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMorais J, Santos BA (2018) Limited potential of deep reefs to serve as refuges for tropical Southwestern Atlantic corals. Ecosphere 9:e02281. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/ecs2.2281\u003c/span\u003e\u003cspan address=\"10.1002/ecs2.2281\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMoritz C, Brandl S, Rouz\u0026eacute; H, Vii J, P\u0026eacute;rez-Rosales G, Bosserelle P, Chancerelle Y, Galzin R, Liao V, Siu G, Taiarui M, Nugues M, H\u0026eacute;douin L (2021) Long-term monitoring of benthic communities reveals spatial determinants of disturbance and recovery dynamics on coral reefs. Mar Ecol Prog Ser 672:141\u0026ndash;152. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3354/meps13807\u003c/span\u003e\u003cspan address=\"10.3354/meps13807\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMuir PR, Done T, Aguirre JD, Keith S (2021) High regional and intrageneric variation in susceptibility to mass bleaching in Indo-Pacific coral species. Glob Ecol Biogeogr 30:1889\u0026ndash;1898. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/geb.13353\u003c/span\u003e\u003cspan address=\"10.1111/geb.13353\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMuir PR, Marshall PA, Abdulla A, Aguirre JD (2017a) Species identity and depth predict bleaching severity in reef-building corals: shall the deep inherit the reef? Proc R Soc B Biol Sci 284:20171551. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1098/rspb.2017.1551\u003c/span\u003e\u003cspan address=\"10.1098/rspb.2017.1551\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMuir PR, Marshall PA, Abdulla A, Aguirre JD (2017b) Species identity and depth predict bleaching severity in reef-building corals: shall the deep inherit the reef? Proc R Soc B Biol Sci 284:20171551. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1098/rspb.2017.1551\u003c/span\u003e\u003cspan address=\"10.1098/rspb.2017.1551\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMumby P, Chisholm J, Edwards A, Clark C, Roark E, Andrefouet S, Jaubert J (2001) Unprecedented bleaching-induced mortality in Porites spp. at Rangiroa Atoll, French Polynesia. Mar Biol 139:183\u0026ndash;189.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMu\u0026ntilde;iz-Castillo AI, Rivera-Sosa A, McField M, Chollett I, Eakin CM, Enr\u0026iacute;quez S, Gir\u0026oacute; A, Drysdale I, Rueda M, Soto M, Craig N, Arias-Gonz\u0026aacute;lez JE (2024) Underlying drivers of coral reef vulnerability to bleaching in the Mesoamerican Reef. Commun Biol 7:1452. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s42003-024-07128-y\u003c/span\u003e\u003cspan address=\"10.1038/s42003-024-07128-y\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNeal BP, Condit C, Liu G, Dos Santos S, Kahru M, Mitchell BG, Kline DI (2014) When depth is no refuge: cumulative thermal stress increases with depth in Bocas del Toro, Panama. Coral Reefs 33:193\u0026ndash;205. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00338-013-1081-6\u003c/span\u003e\u003cspan address=\"10.1007/s00338-013-1081-6\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOksanen J, Simpson GL, Blanchet FG, Kindt R, Legendre P, Minchin PR, O\u0026rsquo;Hara RB, Solymos P, Stevens MHH, Szoecs E, Wagner H, Barbour M, Bedward M, Bolker B, Borcard D, Borman T, Carvalho G, Chirico M, Caceres MD, Durand S, Evangelista HBA, FitzJohn R, Friendly M, Furneaux B, Hannigan G, Hill MO, Lahti L, McGlinn D, Ouellette M-H, Cunha ER, Smith T, Stier A, Braak CJFT, Weedon J (2025) vegan: Community Ecology Package.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOliver TA, Palumbi SR (2011) Do fluctuating temperature environments elevate coral thermal tolerance? Coral Reefs 30:429\u0026ndash;440. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00338-011-0721-y\u003c/span\u003e\u003cspan address=\"10.1007/s00338-011-0721-y\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePage CA, Giuliano C, Bay LK, Randall CJ (2023) High survival following bleaching underscores the resilience of a frequently disturbed region of the Great Barrier Reef. Ecosphere 14:e4280. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/ecs2.4280\u003c/span\u003e\u003cspan address=\"10.1002/ecs2.4280\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePenin L, Adjeroud M, Schrimm M, Lenihan HS (2007) High spatial variability in coral bleaching around Moorea (French Polynesia): patterns across locations and water depths. C R Biol 330:171\u0026ndash;181. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.crvi.2006.12.003\u003c/span\u003e\u003cspan address=\"10.1016/j.crvi.2006.12.003\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePereira PHC, Lima GV, Pontes AVF, C\u0026ocirc;rtes LGF, Gomes E, Sampaio CLS, Pinto TK, Miranda RJ, Cardoso ATC, Araujo JC, Seoane JCS (2022) Unprecedented Coral Mortality on Southwestern Atlantic Coral Reefs Following Major Thermal Stress. Front Mar Sci 9:725778. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3389/fmars.2022.725778\u003c/span\u003e\u003cspan address=\"10.3389/fmars.2022.725778\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eP\u0026eacute;rez-Rosales G, Pichon M, Rouz\u0026eacute; H, Vill\u0026eacute;ger S, Torda G, Bongaerts P, Carlot J, Under The Pole Consortium, Parravicini V, H\u0026eacute;douin L (2022) Mesophotic coral ecosystems of French Polynesia are hotspots of alpha and beta generic diversity for scleractinian assemblages. Divers Distrib 28:1391\u0026ndash;1403. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/ddi.13549\u003c/span\u003e\u003cspan address=\"10.1111/ddi.13549\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eP\u0026eacute;rez-Rosales G, Rouz\u0026eacute; H, Torda G, Bongaerts P, Pichon M, Under The Pole Consortium, Parravicini V, H\u0026eacute;douin L (2021) Mesophotic coral communities escape thermal coral bleaching in French Polynesia. R Soc Open Sci 8:210139. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1098/rsos.210139\u003c/span\u003e\u003cspan address=\"10.1098/rsos.210139\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePisapia C, Burn D, Pratchett MS (2019) Changes in the population and community structure of corals during recent disturbances (February 2016-October 2017) on Maldivian coral reefs. Sci Rep 9:8402. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41598-019-44809-9\u003c/span\u003e\u003cspan address=\"10.1038/s41598-019-44809-9\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePratchett MS, McCowan D, Maynard JA, Heron SF (2013) Changes in Bleaching Susceptibility among Corals Subject to Ocean Warming and Recurrent Bleaching in Moorea, French Polynesia. PLoS ONE 8:e70443. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1371/journal.pone.0070443\u003c/span\u003e\u003cspan address=\"10.1371/journal.pone.0070443\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eR Core Team (2024) R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRiegl B, Glynn PW, Banks S, Keith I, Rivera F, Vera-Zambrano M, D\u0026rsquo;Angelo C, Wiedenmann J (2019) Heat attenuation and nutrient delivery by localized upwelling avoided coral bleaching mortality in northern Galapagos during 2015/2016 ENSO. Coral Reefs 38:773\u0026ndash;785. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00338-019-01787-8\u003c/span\u003e\u003cspan address=\"10.1007/s00338-019-01787-8\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRiegl B, Purkis S (2015) Coral population dynamics across consecutive mass mortality events. Glob Change Biol 21:3995\u0026ndash;4005. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/gcb.13014\u003c/span\u003e\u003cspan address=\"10.1111/gcb.13014\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRoberts TE, Moloney JM, Sweatman HPA, Bridge TCL (2015) Benthic community composition on submerged reefs in the central Great Barrier Reef. Coral Reefs 34:569\u0026ndash;580. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00338-015-1261-7\u003c/span\u003e\u003cspan address=\"10.1007/s00338-015-1261-7\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRobinson JPW, Wilson SK, Jennings S, Graham NAJ (2019) Thermal stress induces persistently altered coral reef fish assemblages. Glob Change Biol 25:2739\u0026ndash;2750. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/gcb.14704\u003c/span\u003e\u003cspan address=\"10.1111/gcb.14704\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRuss GR, Rizzari JR, Abesamis RA, Alcala AC (2021) Coral cover a stronger driver of reef fish trophic biomass than fishing. Ecol Appl 31:e02224. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/eap.2224\u003c/span\u003e\u003cspan address=\"10.1002/eap.2224\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSafaie A, Silbiger NJ, McClanahan TR, Pawlak G, Barshis DJ, Hench JL, Rogers JS, Williams GJ, Davis KA (2018) High frequency temperature variability reduces the risk of coral bleaching. Nat Commun 9:1671. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41467-018-04074-2\u003c/span\u003e\u003cspan address=\"10.1038/s41467-018-04074-2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSakai K, Singh T, Iguchi A (2019) Bleaching and post-bleaching mortality of \u003cem\u003eAcropora\u003c/em\u003e corals on a heat-susceptible reef in 2016. PeerJ 7:e8138. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.7717/peerj.8138\u003c/span\u003e\u003cspan address=\"10.7717/peerj.8138\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSampayo EM, Ridgway T, Bongaerts P, Hoegh-Guldberg O (2008) Bleaching susceptibility and mortality of corals are determined by fine-scale differences in symbiont type. Proc Natl Acad Sci 105:10444\u0026ndash;10449. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1073/pnas.0708049105\u003c/span\u003e\u003cspan address=\"10.1073/pnas.0708049105\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchramek TA, Colin PL, Merrifield MA, Terrill EJ (2018) Depth-Dependent Thermal Stress Around Corals in the Tropical Pacific Ocean. Geophys Res Lett 45:9739\u0026ndash;9747. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1029/2018GL078782\u003c/span\u003e\u003cspan address=\"10.1029/2018GL078782\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSeabrook S, Correa AM, Conetta D, Grupstra CG, Howe-Kerr LI, McConnell K, Schmeltzer ER, Vega Thurber R, Thurber AR (2023) Coral Bleaching Induced Mortality Transforms Local and Global Carbon Cycles: An Unrecognized Feedback Loop That May Accelerate Reef Decline. EGUsphere 2023:1\u0026ndash;24.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSmith TB, Glynn PW, Mat\u0026eacute; JL, Toth LT, Gyory J (2014) A depth refugium from catastrophic coral bleaching prevents regional extinction. Ecology 95:1663\u0026ndash;1673. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1890/13-0468.1\u003c/span\u003e\u003cspan address=\"10.1890/13-0468.1\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSmith TB, Gyory J, Brandt ME, Miller WJ, Jossart J, Nemeth RS (2016) Caribbean mesophotic coral ecosystems are unlikely climate change refugia. Glob Change Biol 22:2756\u0026ndash;2765. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/gcb.13175\u003c/span\u003e\u003cspan address=\"10.1111/gcb.13175\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSouter D, Planes S, Wicquart J, Logan M, Obura D, Staub F (2020) Status of Coral Reefs of the World: 2020. 20.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSpeare KE, Adam TC, Winslow EM, Lenihan HS, Burkepile DE (2022) Size-dependent mortality of corals during marine heatwave erodes recovery capacity of a coral reef. Glob Change Biol 28:1342\u0026ndash;1358. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/gcb.16000\u003c/span\u003e\u003cspan address=\"10.1111/gcb.16000\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSuggett DJ, Smith DJ (2020) Coral bleaching patterns are the outcome of complex biological and environmental networking. Glob Change Biol 26:68\u0026ndash;79. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/gcb.14871\u003c/span\u003e\u003cspan address=\"10.1111/gcb.14871\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSully S, Burkepile DE, Donovan MK, Hodgson G, van Woesik R (2019) A global analysis of coral bleaching over the past two decades. Nat Commun 10:1264. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41467-019-09238-2\u003c/span\u003e\u003cspan address=\"10.1038/s41467-019-09238-2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSzereday S, Voolstra CR, Amri AY (2024) Back-to-back bleaching events in Peninsular Malaysia (2019\u0026ndash;2020) selectively affect hard coral taxa across- and within-reef scales. Mar Biol 171:183. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00227-024-04495-2\u003c/span\u003e\u003cspan address=\"10.1007/s00227-024-04495-2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTamir R, Eyal G, Kramer N, Laverick JH, Loya Y (2019) Light environment drives the shallow-to‐mesophotic coral community transition. Ecosphere 10:e02839. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/ecs2.2839\u003c/span\u003e\u003cspan address=\"10.1002/ecs2.2839\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTavakoli-Kolour P, Sinniger F, Morita M, Hazraty-Kari S, Nakamura T, Harii S (2024) Shallow corals acclimate to mesophotic depths while maintaining their heat tolerance against ongoing climate change. Mar Pollut Bull 209:117277. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.marpolbul.2024.117277\u003c/span\u003e\u003cspan address=\"10.1016/j.marpolbul.2024.117277\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTrygonis V, Sini M (2012) photoQuad: a dedicated seabed image processing software, and a comparative error analysis of four photoquadrat methods. J Exp Mar Biol Ecol 424:99\u0026ndash;108. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jembe.2012.04.018\u003c/span\u003e\u003cspan address=\"10.1016/j.jembe.2012.04.018\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVan Rein H, Schoeman DS, Brown C, Quinn R, Breen J (2011) Development of benthic monitoring methods using photoquadrats and scuba on heterogeneous hard-substrata: a boulder-slope community case study. Aquat Conserv Mar Freshw Ecosyst 21:676\u0026ndash;689. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/aqc.1224\u003c/span\u003e\u003cspan address=\"10.1002/aqc.1224\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVan Woesik R, Sakai K, Ganase A, Loya Y (2011) Revisiting the winners and the losers a decade after coral bleaching. Mar Ecol Prog Ser 434:67\u0026ndash;76. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3354/meps09203\u003c/span\u003e\u003cspan address=\"10.3354/meps09203\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVanderWeele T, Vansteelandt S (2014) Mediation Analysis with Multiple Mediators. Epidemiol Methods 2: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1515/em-2012-0010\u003c/span\u003e\u003cspan address=\"10.1515/em-2012-0010\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWall CB, Kaluhiokalani M, Popp BN, Donahue MJ, Gates RD (2020) Divergent symbiont communities determine the physiology and nutrition of a reef coral across a light-availability gradient. ISME J 14:945\u0026ndash;958. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41396-019-0570-1\u003c/span\u003e\u003cspan address=\"10.1038/s41396-019-0570-1\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWatt-Pringle R, Smith DJ, Ambo-Rappe R, Lamont TAC, Jompa J (2022) Suppressed recovery of functionally important branching Acropora drives coral community composition changes following mass bleaching in Indonesia. Coral Reefs 41:1337\u0026ndash;1350. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00338-022-02275-2\u003c/span\u003e\u003cspan address=\"10.1007/s00338-022-02275-2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWhitaker H, DeCarlo T (2024) Re(de)fining degree-heating week: coral bleaching variability necessitates regional and temporal optimization of global forecast model stress metrics. Coral Reefs 43:969\u0026ndash;984. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00338-024-02512-w\u003c/span\u003e\u003cspan address=\"10.1007/s00338-024-02512-w\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWinslow EM, Speare KE, Adam TC, Burkepile DE, Hench JL, Lenihan HS (2024) Corals survive severe bleaching event in refuges related to taxa, colony size, and water depth. Sci Rep 14:9006. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41598-024-58980-1\u003c/span\u003e\u003cspan address=\"10.1038/s41598-024-58980-1\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWinston M, Oliver T, Couch C, Donovan MK, Asner GP, Conklin E, Fuller K, Grady BW, Huntington B, Kageyama K, Kindinger TL, Kozar K, Kramer L, Martinez T, McCutcheon A, McKenna S, Rodgers K, Shayler CK, Vargas-Angel B, Zgliczynski B (2022) Coral taxonomy and local stressors drive bleaching prevalence across the Hawaiian Archipelago in 2019. PLOS ONE 17:e0269068. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1371/journal.pone.0269068\u003c/span\u003e\u003cspan address=\"10.1371/journal.pone.0269068\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWyatt ASJ, Leichter JJ, Toth LT, Miyajima T, Aronson RB, Nagata T (2020) Heat accumulation on coral reefs mitigated by internal waves. Nat Geosci 13:28\u0026ndash;34. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41561-019-0486-4\u003c/span\u003e\u003cspan address=\"10.1038/s41561-019-0486-4\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWyatt ASJ, Leichter JJ, Washburn L, Kui L, Edmunds PJ, Burgess SC (2023) Hidden heatwaves and severe coral bleaching linked to mesoscale eddies and thermocline dynamics. Nat Commun 14:25. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41467-022-35550-5\u003c/span\u003e\u003cspan address=\"10.1038/s41467-022-35550-5\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYu Q, Li B (2017) mma: an R package for mediation analysis with multiple mediators. J Open Res Softw 5:11\u0026ndash;11.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"coral-reefs","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"core","sideBox":"Learn more about [Coral Reefs](http://link.springer.com/journal/338)","snPcode":"338","submissionUrl":"https://submission.nature.com/new-submission/338/3","title":"Coral Reefs","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Coral bleaching, Depth, Coral mortality, Reef resilience, Marine heatwave, Community composition","lastPublishedDoi":"10.21203/rs.3.rs-8725130/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8725130/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eMarine heatwaves (MHWs) are intensifying and increasingly threatening coral reef persistence by inducing widespread bleaching and mortality. Although depth is often proposed as a natural refuge, its protective role, especially in reducing post-bleaching coral loss, remains debated. This study quantified how depth (5, 12 and 20m) shaped bleaching and post-bleaching mortality and evaluated drivers of depth-mediated protection during the unprecedented severe 2019 MHW in Mo\u0026rsquo;orea (French Polynesia), across 6 sites.\u003c/p\u003e \u003cp\u003eDespite thermal stress levels considered moderate according to global standards (3\u0026deg;C-weeks), the event caused unexpectedly severe impacts, with widespread bleaching resulting in the loss of half the coral cover. Bleaching was only weakly mitigated by depth with heightened protection (\u0026lt;\u0026thinsp;30% reduction) from 12 to 20m in sensitive taxa such as \u003cem\u003eAcropora\u003c/em\u003e, \u003cem\u003eMontipora\u003c/em\u003e and \u003cem\u003ePocillopora\u003c/em\u003e, not present at all in others, indicating refuge from bleaching was not universal. Mortality was concentrated in branching \u003cem\u003eAcropora\u003c/em\u003e and \u003cem\u003ePocillopora\u003c/em\u003e, while other heavily bleached taxa showed no mortality. Relative total coral cover loss fell 3.5-fold from 45% at 5m to 13% at 20m, driving 3-fold weaker community shifts, revealing that depth more strongly limited post-bleaching mortality. This effect was partly mediated by differences in pre-bleaching community composition and coral cover, yet ~\u0026thinsp;74% of the protective effect remained unexplained, suggesting that additional depth-related environmental or physiological mechanisms, such as light attenuation or colony size, were at play. These findings position deeper reefs as potential natural buffers, even in the absence of effective thermal stratification, and priorities for proactive conservation under accelerating climate change.\u003c/p\u003e","manuscriptTitle":"When depth fails to prevent bleaching but limits coral death: insights from the 2019 heatwave in Mo’orea","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-02-09 08:34:46","doi":"10.21203/rs.3.rs-8725130/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-02-07T23:36:01+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-02-05T09:55:54+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"330354017668394915475308938398267767581","date":"2026-02-05T04:14:19+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-02-05T03:21:47+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-02-05T03:19:31+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-02-03T13:37:59+00:00","index":"","fulltext":""},{"type":"submitted","content":"Coral Reefs","date":"2026-01-28T20:57:36+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"coral-reefs","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"core","sideBox":"Learn more about [Coral Reefs](http://link.springer.com/journal/338)","snPcode":"338","submissionUrl":"https://submission.nature.com/new-submission/338/3","title":"Coral Reefs","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"7159ab63-fbdb-4187-9c42-f839cb553b4f","owner":[],"postedDate":"February 9th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2026-03-09T16:11:04+00:00","versionOfRecord":{"articleIdentity":"rs-8725130","link":"https://doi.org/10.1007/s00338-026-02839-6","journal":{"identity":"coral-reefs","isVorOnly":false,"title":"Coral Reefs"},"publishedOn":"2026-03-08 15:58:25","publishedOnDateReadable":"March 8th, 2026"},"versionCreatedAt":"2026-02-09 08:34:46","video":"","vorDoi":"10.1007/s00338-026-02839-6","vorDoiUrl":"https://doi.org/10.1007/s00338-026-02839-6","workflowStages":[]},"version":"v1","identity":"rs-8725130","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8725130","identity":"rs-8725130","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}