Co-occurrence patterns of photosymbiont genera in reef-building corals: a global review and meta-analysis

Co-occurrence patterns of photosymbiont genera in reef-building corals: a global review and meta-analysis | Authorea try { document.documentElement.classList.add('js'); } catch (e) { } var _gaq = _gaq || []; _gaq.push(['_setAccount', 'G-8VDV14Y67G']); _gaq.push(['_trackPageview']); (function() { var ga = document.createElement('script'); ga.type = 'text/javascript'; ga.async = true; ga.src = ('https:' == document.location.protocol ? 'https://ssl' : 'http://www') + '.google-analytics.com/ga.js'; var s = document.getElementsByTagName('script')[0]; s.parentNode.insertBefore(ga, s); })(); Skip to main content Preprints Collections Wiley Open Research IET Open Research Ecological Society of Japan All Collections About About Authorea FAQs Contact Us Quick Search anywhere Search for preprint articles, keywords, etc. Search Search ADVANCED SEARCH SCROLL This is a preprint and has not been peer reviewed. Data may be preliminary. 1 October 2025 V1 Latest version Share on Co-occurrence patterns of photosymbiont genera in reef-building corals: a global review and meta-analysis Authors : Corinne Allen 0009-0004-5406-1914 [email protected] , Eve Hinchliffe 0009-0004-5289-3733 , David Suggett , Michael Kühl 0000-0002-1792-4790 , Teina Rongo , Wing Chan , Madeleine van Oppen , and Matthew Nitschke Authors Info & Affiliations https://doi.org/10.22541/au.175934402.21612098/v1 485 views 314 downloads Contents Abstract Supplementary Material Information & Authors Metrics & Citations View Options References Figures Tables Media Share Abstract Coral reefs are experiencing unprecedented global declines driven by intensifying marine heatwaves, deoxygenation, and other consequences of climate change. One trait that may enhance coral resilience is their ability to harbour diverse communities of microalgal symbionts (Symbiodiniaceae). Cladocopium and Durusdinium are two common Symbiodiniaceae genera harbouring functionally different traits; however, the extent to which they co-occur within coral hosts has not yet been explored. Here, we conducted a systematic literature review and meta-analysis to assess global patterns of Cladocopium-Durusdinium co-occurrence and co-phylogeny, synthesising data from ~36,000 coral colonies across 378 studies, 98 coral genera, 77 countries and territories, 11 genetic markers, and 13 genetic analysis techniques. Co-occurrence prevalence across these studies was 9.6%, with estimates varying across methodological, biological, spatial, and study variables. Incorporation of more recent high-sensitivity genotyping and intra-colony sampling increased this estimate by up to 7.1-fold, suggesting that Cladocopium-Durusdinium co-occurrence may be more common than previously recognised. We also identified a co-phylogenetic signal between putative Cladocopium and Durusdinium taxa, suggesting that in some cases, these symbionts have co-evolved. Together, these findings contribute to the understanding of eco-evolutionary relationships between symbionts within coral hosts, providing important insight into the adaptive capacity of corals in a changing climate. Introduction Coral reefs are among the most productive and ecologically important ecosystems in the world. Reef persistence is contingent on the mutualistic relationship formed between reef-building corals (Scleractinia) and their microalgal symbionts (family Symbiodiniaceae) (Trench 1979) that provide both O 2 and organic carbon to the host in exchange for protection, inorganic carbon, and nutrients. However, stability of this symbiosis is sensitive to changes in environmental conditions. Deviations in water quality, light, or temperature can disrupt the delicate partnership, resulting in a phenomenon known as coral bleaching (i.e., the loss of Symbiodiniaceae cells) (Helgoe et al. 2024; Suggett & Smith 2020). As anthropogenic stressors intensify, particularly in the form of marine heatwaves, understanding coral adaptation and the role of Symbiodiniaceae in driving adaptative capacity is becoming increasingly important. Advances in molecular techniques have transformed our understanding of Symbiodiniaceae genetic diversity (Davies et al. 2023). Once considered a monotypic genus consisting of the species Symbiodinium microadriaticum (Freudenthal 1962), symbionts were later classified into distinct ‘clades’, which have now been formally revised as eleven genera (plus several unclassified clades) within the family Symbiodiniaceae (LaJeunesse et al. 2018). Of these genera, four ( Symbiodinium, Breviolum, Cladocopium , and Durusdinium ) are commonly found in symbiosis with corals. Cladocopium and Durusdinium are particularly widespread, found in many coral species throughout the Atlantic, Pacific, and Indian Ocean basins (Nitschke et al . 2022). Significant differences occur in the functional ecology of Cladocopium and Durusdinium species, reflected in their capabilities for photosynthetic efficiency (Starzak et al. 2014), oxidative stress resistance (McGinty et al. 2012), nutrient sharing (Wall et al. 2020), and high and low temperature tolerances (Berkelmans & van Oppen 2006; Silverstein et al. 2017). For instance, Durusdinium may enhance heat tolerance by up to 1-1.5°C compared to Cladocopium (Berkelmans & van Oppen 2006; Silverstein et al. 2015); however, Durusdinium often exhibts reduced photosynthate translocation to the host that can impact fitness through slower growth rates and lower reproductive output (Cantin et al. 2009; Jones & Berkelmans 2010, 2011). Thus, the physiological performance of the coral host can differ based on the physiology (and thus taxonomic identity) of the associated symbionts (Cooper et al. 2011; Mieog et al. 2009). Furthermore, when corals harbour both Cladocopium and Durusdinium symbionts simultaneously, the functional interactions between these taxa remain poorly understood, but are hypothesised to enhance the host’s physiological flexibility and ecological niche breadth (Silverstein et al. 2012). Symbiont community assembly within corals is driven by the interaction of host characteristics (e.g., species, morphology, or life stage) and the external environment (e.g., light, temperature, latitude, depth, and local availability of symbionts) (Dubé et al. 2021; Trivedi et al. 2020). This process is dynamic and may change across space and time within a single coral colony due to the presence of intra-colony resource gradients (Kühl et al. 1995; Nitschke et al. 2022; Wangpraseurt et al. 2012) and ontogenetic shifts as corals mature (Chan et al. 2019; Gómez-Cabrera et al. 2008; Little et al. 2004). While some corals and symbionts associate with a broad range of partners (i.e., generalists), others are more specific in their associations (i.e., specialists) (Baker 2003; LaJeunesse et al. 2008; Saad et al. 2022). However, specificity does not imply that symbiont communities are fixed, as environmental changes may trigger shifts in symbiont community structure to a composition better adapted to the prevailing conditions (Buddemeier & Fautin 1993). Changes can occur via symbiont shuffling (changes in the relative abundance of symbionts that were already present) (Baker et al. 2004) or symbiont switching (where corals acquire novel symbionts from their surrounding environment) (Boulotte et al. 2016). Such dynamic processes are typically observed during summer months, when corals associate with more heat tolerant symbionts such as Durusdinium . However, these changes can be transient, with corals reverting to their original community compositions after the stressful condition(s) subside (LaJeunesse et al. 2009; Thornhill et al. 2006). The ability for corals to alter their symbiont communities to the prevailing conditions is thought to be one of the main ways by which they can adapt to climate change (Baker 2004). However, many symbiont sampling methods only capture snapshots of the community composition at specific moments in space and time (Davies et al. 2023). As such, it is plausible that what is observed as symbiont co-occurrence (spatial or temporal) may, in some cases, represent a more persistent state of coexistence (stable over space and time). If the presence of both Cladocopium and Durusdinium is indeed common and stable in corals, then it is likely coral adaptive capacity remains underestimated. Furthermore, when multiple symbiont lineages stably co-occur within hosts over evolutionary timescales, functional integration and symbiont-symbiont co-speciation may occur in the forms of: a) vicariance, where parallel branching in the symbiont phylogenies emerges because the evolutionary history of both organisms tracks similar abiotic histories, b) phylogenetic tracking of one another due to asymmetry in the interaction (i.e., one symbiont genus facilitates the other) or c) co-evolution where the interaction among symbionts bi-directionally influences fitness (Russo et al. 2018). Recognising that symbiont co-evolutionary history may constrain adaptive potential reframes coral resilience as a function of long-term integration among hosts and multiple symbiont lineages. Here, we systematically review the literature to explore patterns of Cladocopium and Durusdinium co-occurrence across space, time, and coral genera. We specifically use relevant studies to conduct a meta-analysis estimating the prevalence of co-occurrence and assess how study-specific methodological approaches, environmental conditions, and biological factors (including the cnidarian phylogeny) shape these estimates. Additionally, we perform a co-phylogenetic analysis to determine evolutionary relationships between co-occurring Cladocopium and Durusdinium . Finally, we identify research gaps and biases in the current knowledge of symbiont co-occurrence, framing next steps for building a more robust understanding of co-occurrence dynamics and how they may influence adaptive potential. Literature review In November 2022, a web-based search (Google Scholar) was conducted to identify studies that reported the presence of Cladocopium and/or Durusdinium in scleractinian corals. Due to a recent taxonomic revision (LaJeunesse et al. 2018), two search terms were used: a) coral AND s ymbiodinium AND “ clade C” OR “ clade D” (to capture relevant articles from 1990-2018) and b) coral AND symbiodiniaceae AND cladocopium OR durusdinium (to capture relevant articles from 2018-present). All articles were assigned a unique index and imported into PaperPile where a duplicate detection and cleanup was completed. A list of all articles following the automated duplicate detection search can be found in Table S1. No articles were included following this search (i.e., only through October 2022). Title and abstract screen Following duplicate deletion, remaining results were screened for relevance based on their title and abstract. Studies only passed the screening stage if they met the following eligibility criteria: a) the study was determined to be primary literature from a peer-reviewed scientific journal (e.g. no reviews, theses, book chapters, preprints, etc.) with b) scleractinian corals as the focal study subject. Two independent reviewers (CEA and MRN) performed all aspects of the search strategy and abstract screening, with any uncertainties resolved through discussion. Full text review, inclusion criteria, and data extraction Following the title/abstract screen, PDF files of all records were retrieved and screened against a library of 92 terms (Table S2) to sort articles by relevance (see Supplementary Methods). The full text of each article was then reviewed to determine if it met the following criteria for inclusion in the review and meta-analysis: (i) the study was conducted on a scleractinian coral host; (ii) the study genotyped the symbionts and reported presence of Cladocopium and/or Durusdinium within a coral at some point in space or time (binary or quantitative); (iii) there was a reasonable opportunity for detection of Cladocopium and Durusdinium (e.g., a Cladocopium- only culture inoculation of coral larvae under laboratory conditions would constitute a ‘biased inoculation’ and would be excluded; the use of genus-specific genotyping approaches, where only a single genus was targeted in the study, would also be excluded); and (iv) the study clearly reported methodology and the sample sizes of corals within the study. If inclusion criteria were met, data were extracted from the article and recorded in a database (Table S3). This database was structured using 44 fields falling within 7 attributes, which are presented in Table 1. Details and rationales for each attribute and field are included in the Supplementary Methods. If the study reported data from multiple coral genera or species, localities, depths, genetic markers, techniques, etc., a new row was added under the same index. Similarly, if two colonies had different symbiont communities these colonies were also reported in separate rows. If a field was not applicable, NA was recorded. If data or information for a specific field was unclear, insufficient, or could not be found in the article, ND (no data) was recorded. The only exception for this was when latitude and longitude were not reported, in which coordinates were estimated using Google Earth based on the locality provided in the article and recorded as “generalised” in our database. Manual data extraction was primarily completed by the lead author (CEA), some of which was assisted and quality checked by two other authors (ECH and MRN). All uncertainties were resolved through discussion by the authors. Details on included and excluded studies can be found in Table S1. SymPortal data extraction If the article utilised SymPortal (symportal.org; (Hume et al. 2019) for analysis of symbiont community composition, data was extracted semi-automatically. Data files and code for extraction are available as supplementary material. Briefly, SymPortal results accompanying 41 published studies were downloaded from the Data Explorer at symportal.org. Three SymPortal output files served as input for extraction: ‘####.seqs.absolute.abund_and_meta.txt’ and ‘####.seqs.fasta’ from the post_med_seqs directory, and ‘####.profiles.absolute.abund_and_meta.txt’ from the its2_type_profiles directory (#### representing the study-specific prefix). Associated SymPortal sample metadata were quality checked and subset to published observations. Basic quality control criteria were applied, including a minimum sequencing depth per sample of 1000 reads, and the removal of any non-coral samples (e.g., PCR controls, Symbiodiniaceae cultures, environmental samples, etc.) according to SymPortal metadata. Since use of PCR controls varied among studies, we used a data-driven approach to sequence filtering (detailed in the Supplementary Methods) to establish a filtering threshold of 1.03%. This threshold was applied to each sample of the complete dataset, after which samples were categorised according to the presence of Cladocopium and/or Durusdinium sequences as above. Articles were then manually reviewed, also as per the methods above, to capture information in fields that were not reported as part of the SymPortal metadata (e.g., intra-colony sampling, temporal sampling, etc.). Database cleaning, quality control, and post hoc additions The resulting database was first cleaned to standardise key variables due to the numerous approaches of reporting across studies. For example, if two genotyping techniques or markers were used, the results were compared, but ultimately only one was used for downstream analyses (staging steps detailed in the Supplementary Methods and Table S4). For spatial information, all coordinates were converted into decimal degrees. If a range of coordinates was reported, the midpoint was calculated, added to the database, and used for subsequent qualitative mapping. However, only studies that reported exact coordinates and depths were used for quantitative analyses of depth and distance from shore. Biogeographic regions were assigned as defined by the Global Coral Reef Monitoring Network (GCRMN) and included Australia, Brazil, the Caribbean, the East Asian Seas (EAS), the Eastern Tropical Pacific (ETP), the Pacific, the Red Sea and Gulf of Aden region (PERSGA), the Regional Organization for the Protection of the Marine Environment (ROPME), South Asia, and the Western Indian Ocean (WIO) (Souter et al. 2021). Host information, including genus and species names, were quality checked for spelling errors. To account for taxonomic updates, relevant species, genus, and family lists were retrieved from the World List of Scleractinia (World Register of Marine Species [WoRMS]; https://www.marinespecies.org/ accessed on February 14, 2025) and aligned with our extracted data. Additionally, the symbiont transmission mode, reproductive mode, and coral morphology data from the Coral Trait database (https://www.coraltraits.org/ accessed on March 14, 2025) (Madin et al. 2016) were incorporated to supplement information not commonly reported within articles. Statistical analyses for prevalence of co-occurrence All statistical analyses were conducted using R version 4.1.1 (R Core Team 2021). Meta-analytic and meta-regression models were implemented using the package metafor (Viechtbauer 2010). We used a meta-analysis of prevalence approach with a random-effects model to determine prevalence of co-occurrence in sampled coral colonies. Proportions of colonies exhibiting co-occurrence were calculated for each study. Given that many studies resulted in observed proportions close to 0 or 1, we used the escalc function to transform the observed proportions using the logit (log odds) framework (“PLO”). The meta-analysis estimated the average proportion of all studies, weighting each by the inverse of its sampling variance. Consequently, studies with larger sample sizes were often weighed more than studies with smaller sample sizes. To examine sources of heterogeneity, we conducted random-effects subgroup analyses and meta-regressions using rma.mv based on study-level characteristics. Data were analysed across eleven potential explanatory variables including genotyping technique, intra-colony sampling, life stage, reproductive mode, symbiont transmission mode, region, depth, distance from shore, study type, total sample size, and year coral was sampled. For each subgroup analysis, the variable was run as a fixed effect, while index was included as a random effect to account for heterogeneity between studies. I² and marginal R² values were calculated to estimate the residual heterogeneity and determine the proportion of variance explained by each fixed effect, and z-tests were used to test whether estimates from each level of the fixed effect were significantly different from the overall pooled estimate. While publication bias is typically evaluated in meta-analyses, the standard tests (i.e., funnel plots or Egger’s regression) are developed in the context of comparative data and assume that studies with ‘positive’ results are more likely to be published over studies with ‘negative’ results (Barker et al. 2021). As the proportional data in our study does not align with these assumptions, we did not account for publication bias here. Scleractinia phylogenetic linear mixed model We obtained information on the phylogenetic history of coral genera based on a published phylogeny in Quek et al. (2023) and fitted a phylogenetic linear mixed model using the pglmm function in the phyr package (Li et al . 2020). The response variable followed a binomial distribution , with presence of co-occurrence within a genus and region denoted as 1 and absence denoted as 0 for each study. Biogeographic region was included as a fixed effect, with coral genus, phylogeny, and the interaction between region and phylogeny modeled as random effects to account for evolutionary history and regional variation. Contribution of each random effect was interpreted using its estimated variance, with larger variance associated with a stronger role in shaping the response variable. Statistical significance of the effects was then assessed using likelihood ratio tests. Due to data sparsity, only regions with five or more studies were included. Furthermore, only genera present in both the phylogeny and the database were included in the model (n=67 genera). Co-occurrence estimates were also calculated for each genus as described above. The host phylogeny and associated data were visualised using the ggtree package (Yu et al. 2017). Detecting co-phylogeny between Cladocopium and Durusdinium We used procrustean superimposition ( paco package ; Balbuena et al. 2013) of genetic and co-occurrence data from 41 SymPortal analyses (see above) to detect signatures of co-phylogeny. Briefly, a co-occurrence matrix among ‘profiles’ (i.e., a hypothetical taxon according to SymPortal analysis), and inter-profile genetic distances were extracted from Cladocopium and Durusdinium SymPortal outputs. These were input to the PACo function (Balbuena et al. 2013) using a symmetric model, which treats both symbiont genera as equally important (n=5000 permutations and all other parameters set to default). A significant PACo p-test indicates non-randomness in the superimposition. Small values of squared residuals in the model indicate increasing strength of co-phylogeny. To examine overall goodness of fit, a Procrustes squared correlation value was computed as R²=1 - residual sum of squares / total sum of squares of the Cladocopium configuration. Code for the preparation of the SymPortal outputs for PACo and a complete description of the steps are available as supplementary material. Results Our literature search identified 3,293 studies that reported data on Cladocopium and/or Durusdinium presence in scleractinian corals. Following duplicate removal (n=565 removed) and an initial abstract screen (n=1,614 removed), the resulting 1,114 articles underwent full-text review. A total of 736 records were excluded at this stage, with the most common reasons being that symbionts were not genotyped (n=284), there was insufficient information regarding sample size (n=75), or the study involved a biased inoculation (e.g., a Cladocopium- only culture inoculation of coral larvae) (n=44). After the exclusion process, 378 articles were included in the database and resulting analyses (Fig. 1), from which data was extracted on 36,008 coral colonies, spanning 410 coral species, 98 coral genera, 77 countries and territories, 11 genetic markers, and 13 genotyping techniques. Characteristics of included studies Studies identified through our initial search were published between 1997 and 2022, with exponential increases in all publications over time (R 2= 0.85) and publications ultimately meeting our inclusion criteria (R 2= 0.83) (Fig. 2A). While articles were conducted in 77 countries and territories worldwide, many studies were based in Australia, the United States, Japan, and China. Spatial distribution of colonies sampled followed a similar trend, although higher numbers of colonies were sampled in Taiwan and several countries in Africa and the Middle East despite fewer studies being conducted there (Fig. 2B). The most common study factor was coral host (i.e., genus or species; n=165 articles), followed by site (n=77), and temperature (i.e., thermal or cold stress; n=63). The most common response variables were symbiont community composition (n=270), photophysiology (n=42), and bacterial community composition (n=39). Sample sizes from each study varied widely, ranging from a single colony to ~2000 colonies as a result of study-specific context; notably, larger sample sizes were typically associated with observational studies (63% of articles), and smaller sample sizes more common in controlled laboratory or field experiments (21% and 7% of articles, respectively). The other 14% of articles were methodology studies (3%) or a combination of study types. Intra-colony sampling was employed in 32% of studies (n=122), with n=53 occurring through space, n=42 occurring through time, and n=27 occurring through space and time. The internal transcribed spacer-2 (ITS2) region was the most common genetic marker (58% of articles, 67% of colonies), with next generation sequencing (NGS) (33% of articles, 25% of colonies) and denaturing gradient gel electrophoresis (DGGE) (20% of articles, 32% of colonies) the most common genotyping techniques. More details on techniques and markers can be found in the Supplementary Results. Of all coral colonies included in the database (n=36,008), 71.1% (n=25,618) were reported to harbour Cladocopium only, 14.3% (n=5,148) Durusdinium only, and 14.6% (n=5,242) both Cladocopium and Durusdinium. Meta-analysis of prevalence While the raw proportion of colonies showing co-occurrence of Cladocopium and Durusdinium was 0.146 (14.6%), the random effects meta-analysis of all combined studies (n=378) estimated a slightly lower proportion of 0.096 (9.6%; 95% CI: [0.08-0.12]), accounting for between-study variability. Considerable heterogeneity was evident among studies ( I 2 : 93.2%; Tau 2 : 4.4; Q: 5547.3) (Fig. 4A), which we explored by testing the influence of different methodological, biological, spatial, and study-level moderators (Table 2). Nearly all variables were significantly different from the null hypothesis of zero (p<0.01), indicating that co-occurrence was observed across all environmental and biological conditions. Given these findings, we conducted subgroup analyses to determine differences from the global pooled estimate. Methodological variables Approximately one third of articles (n=122) used intra-colony sampling over space and/or time, yielding a co-occurrence estimate of 0.15 [0.11-0.19] – 1.5x higher than the overall estimate and 1.7x higher than in articles that did not use intra-colony sampling (0.09 [0.07-0.11]) (Fig. 4B). However, intra-colony sampling contributed to a relatively small amount of variance (R 2 =1.83%). In contrast, genotyping technique explained the largest amount of variance (R 2 =37.65%). Approximately 40% of studies in the database used high sensitivity techniques, such as qPCR or NGS, to genotype symbionts from coral samples. These two techniques yielded the highest symbiont co-occurrence estimates, at 0.52 [0.35-0.68] and 0.27 [0.21-0.34], respectively, which were higher than the overall pooled estimate (both p<0.0001) (Fig. 4C). This was followed by 454 pyrosequencing (0.05 [0.01-0.19]) and cloning and sequencing (0.05 [0.03-0.1]), although these two techniques did not differ from the pooled estimate. DGGE, which was one of the most common techniques (used in 20% of the articles and 31% of coral colonies), had the lowest estimate of co-occurrence at 0.03 [0.02-0.04], along with Sanger sequencing (0.03 [0.02-0.06]), single-strand conformation polymorphism (SSCP) (0.04 [0.02-0.07]), and restriction fragment length polymorphism (RFLP) (0.05 [0.03-0.09]. These latter techniques were all lower than the pooled estimate (all p<0.05). Upon running a meta-regression model of the effects sampling and genotyping techniques on the estimate of co-occurrence, intra-colony sampling always increased the estimate irrespective of the technique used. Furthermore, using both intra-colony sampling and a high sensitivity technique led to an observed 46% of colonies harbouring Cladocopium and Durusdinium and a co-occurrence estimate of 0.30 [0.23-0.39] (NGS) and 0.68 [0.48-0.83] (qPCR) (Fig. S3). These co-occurrence estimates were up to 7.1-fold higher than the global estimate and up to 34-fold higher than colonies with no intra-colony sampling genotyped using DGGE (0.020 [0.012-0.033]). Biological variables As most of the coral samples in the database came from adult colonies (93%), this subgroup level did not differ from the overall pooled estimate (0.09 [0.08-0.12]). Early life stages showed higher co-occurrence estimates (gametes: 0.12 [0.07-0.19], larvae: 0.27 [0.15-0.43], recruits or juveniles: 0.17 [0.10-0.26]); however, only larvae (p<0.001) and recruits/juveniles (p<0.05) differed from the pooled estimate (Fig. 4D). However, these observations should be interpreted cautiously, given that the number of effect sizes was small (k=3, 11, and 21, respectively), largely since many studies on early life stages were excluded from our database due to use of biased inoculations. Information on reproductive mode was obtained for 228 coral species and 71 coral genera within our database, with 57 genera (80%) being broadcast spawners. Spawners and brooders exhibited similar co-occurrence estimates (spawners: 0.11 [0.09-0.14]), brooders: 0.13 [0.09-0.17]), with neither differing from the pooled estimate (Fig. 4E). Overall, reproductive mode had a negligible contribution to the explained heterogeneity (R 2 <0.01%). In contrast, symbiont transmission mode explained more variance (R 2 =6.14%), although information on transmission mode was only obtained from 84 coral species and 42 genera within the database. From this subset of corals, horizontal transmitters (0.17 [0.13-0.21]) had a higher estimate (p<0.0001) than the pooled mean, and 2.4x higher than vertical transmitters (0.07 [0.05-0.10]) (Fig. 4F). Morphological information was obtained for 368 coral species and explained 2.16% of variance. Massive was the most common coral morphology in the database (32%; n=9,498 colonies), followed by branching (closed) (22%; n=6,636), and corymbose (12%, n=3,679). Co-occurrence estimates varied across morphologies, ranging from 0.09 [0.07-0.12] in closed branching corals to 0.24 [0.09-0.50] in hispidose corals (Fig. 4G). While morphologies that prioritise vertical extension (e.g., massive, submassive, columnar, and closed branching) did not differ from the pooled estimate, higher estimates of co-occurrence were observed for several morphologies that prioritise horizontal or radial extension, including digitate (0.14 [0.10-0.19]), tables and plates (0.14 [0.1-0.21]), encrusting (0.15 [0.11-0.19]), and corymbose (0.20 [0.15-0.27]) morphologies (all p<0.05). Spatial variables Biogeographic region explained a comparatively high amount of variance within the model (R 2 =6.32%), with several regions deviating from the pooled estimate. Notably, the ETP had the highest estimate at 0.24 [0.16-0.35] (Fig. 4H). High estimates were also evident for the EAS and Pacific regions (both 0.14 [0.10-0.19]), both of which contributed to a substantial portion of the dataset (k=100 and k=79, respectively). More moderate estimates were observed in the WIO, Australia, ROPME, and the Caribbean, with none of these regions differing from the pooled estimate. Co-occurrence estimates were lower for the Red Sea (0.04 [0.02-0.08]) (p<0.05) and South Asia (0.03 [0-0.24]). Brazil also had a low estimate, although the effect size for this region was the smallest (k=2). Sampling depth in the database ranged from 0 to 172 m; however, many studies either reported depth as a range (43%) or did not provide depths (26%). Filtering studies that reported specific depths for colonies sampled (29% of studies) enabled a meta-regression on 8,129 coral colonies. While this dataset (k=677) yielded no effect of depth on co-occurrence estimates (p=0.43, R 2 =0.29%) (Fig. S4), many of the effect sizes were at depths less than 25 m (k=523). Once filtered to this shallower subset, depth became a significant predictor (p<0.05), with co-occurrence estimates declining with increasing depth (R 2 =1.47%) (Fig. 4J). Distance from shore was calculated for 20,231 colonies across 67% of studies reporting coordinates. Distances varied widely, ranging from inshore locations (~4 m from shore) to remote locations (~500 km from shore). When considering all effect sizes (k=972), there was a significant relationship between distance from shore and co-occurrence estimates (p<0.05) (Fig. S4). However, as with depth, there was a limited number of effect sizes at greater offshore distances. Upon restricting the analysis to colonies located within 100 km of shore (k=926), distance from shore remained a significant predictor (p<0.001), with co-occurrence estimates increasing with closer proximity to the coast (Fig. 4K). Under both analyses, the variance explained remained small (full dataset: R 2 =0.29%; restricted dataset: R 2 =0.78%). Study characteristics Study type contributed to 4.52% of explained variance in co-occurrence estimates, largely driven by laboratory-based experiments that yielded a higher co-occurrence estimate (0.16 [0.10-0.25]) (p<0.05). Other study types, such as field-based experimental studies, methodology studies, and combinations of these approaches also tended to show higher estimates (up to 0.19) (Fig. 4I) but were not significant due to their smaller sample sizes and larger variability. Observational studies, which comprised most of the dataset (k=237), had a co-occurrence estimate of 0.08 [0.06-0.1] and did not differ from the overall pooled estimate. Other study-level variables contributed to small or negligible amounts of explained variance (year sampled: R 2 =<0.01%; sample size: R 2= 1.61%) (Fig. 4L, 4M). Studies with smaller sample sizes yielded slightly higher co-occurrence estimates than ones with larger sample sizes, although this effect was minor. Scleractinia phylogenetic linear mixed model Colonies in our database originated from 98 coral genera and 23 coral families (Fig. 5; Fig. S5). However, colonies were heavily concentrated to three genera, Acropora, Porites, and Pocillopora , accounting for 48% of the database. While co-occurrence was observed in nearly all families (87%), it was only observed in 59 genera (60%), with discrepancies in the co-occurrence estimates between genera. For instance, of genera with effect sizes of k > 10, Isopora and Galaxea experienced the highest estimates (0.49 [0.37-0.61] and 0.32 [0.25-0.40]), while Porites experienced one of the lowest (0.07 [0.05-0.09]) (Fig. 5). To explore whether evolutionary relationships influence patterns of co-occurrence, we implemented a phylogenetic linear mixed model testing for phylogenetic signal, while accounting for biogeographic variation. Random effects analysis confirmed an effect of host genus on symbiont co-occurrence (variance=0.553, LR=9.03, p<0.0001), but the variance attributed to phylogenetic relatedness alone was minimal (0.0001) and non-significant. While this outcome would suggest weak phylogenetic signal, the addition of region as an interacting factor with phylogeny yielded a variance of 0.620, with likelihood ratio tests revealing the interaction as significant (LR=123.5; p<0.0001). Consequently, the influence of evolutionary history on co-occurrence appears context-dependent, shaped by biogeographic distributions of both coral hosts and symbionts. Co-phylogeny analysis For observations extracted semi-automatically from SymPortal analyses, 1024 samples (15%) originating from 23 studies exhibited Cladocopium-Durusdinium co-occurrence. After filtering out single observations, 140 unique co-occurrences of Cladocopium ITS2 profiles with Durusdinium ITS2 profiles were observed (Fig. 6). In the procrustean superimposition of phylogenies, the test was significant (p<0.001) indicating the phylogenies are significantly more similar than expected by chance. However, a low R 2 (0.25) suggests weak to moderate co-phylogenetic congruence. Notable examples exhibiting the lowest squared residuals, and thus strongest signals of co-phylogeny, include profiles containing majority ITS2 sequences C42a/C42.2 with D1-D1as-D1 observed in Pocillopora meandrina and C31/C17d with D4/D1-D6-D1ab in Montipora capitata . At the opposite high end of the squared residual distribution, a notable set of observations is C27/C3/C3u-C115 that co-occurs with multiple distinct Durusdinium profiles, such as D1-D4-D4c-D1h-D1c in Pachyseris speciosa and D1/D4/D17f-D9a in Podabacia crustacea. Other examples with notably high squared residuals include Cladocopium lineages C3f and C15 that form associations with diverse Durusdinium , as above, with examples including C15-C15y-C15a-C15ab-C15w with D1-D4-D1b-D4c-D4f and D1-D4-D6-D4c-D10, both observed in Porites. Discussion Investigating eco-evolutionary relationships between coral symbionts and their hosts is critical in understanding how corals may respond to climate change (Swain et al. 2017). Our global meta-analysis uniquely quantifies patterns in co-occurrence and co-phylogeny between Cladocopium and Durusdinium symbionts in corals, providing new implications for coral adaptive capacity and recommendations to the field for future research. Although our searches returned 3,293 eligible records, two-thirds were excluded. The most common reason was that symbiont identification was not conducted within the scope of the study. Such oversight has been similarly reported in a review by McLachlan et al. (2020) where only 22% of coral heat-stress experiments incorporated Symbiodiniaceae identification. Considering the functional importance of symbionts in coral performance, the widespread absence of symbiont identification presents a common operational research gap that fundamentally limits capacity to resolve mechanisms underlying coral resilience and adaptive capacity. Despite exclusions, we still extracted data on articles (n=378). From these, the global pooled prevalence of Cladocopium-Durusdinium co-occurrence in scleractinian corals was 9.6%, supporting growing evidence that corals can associate with more than one symbiont (Mieog et al. 2007; Silverstein et al. 2012). Most moderators from our subgroup analyses were significantly different from zero (except for unfiltered depth and year sampled), indicating that symbiont co-occurrence is a feature of scleractinian corals across biological, spatial, methodological, and study variables. Biological variables Coral life stage is well known to play a role in symbiont community composition and flexibility (Chan et al. 2019; Little et al. 2004). Early life stages are often aposymbiotic and may have underdeveloped immune systems and symbiont recognition machinery (Puill-Stephan et al. 2012), which may enable the acquisition of diverse symbionts locally available to them from the water column, sediments, or neighboring corals (Coffroth et al. 2006; Nitschke et al. 2015; Williamson et al. 2021). Indeed, many studies have demonstrated the co-occurrence of both Cladocopium and Durusdinium in early life stages (Cumbo et al. 2013; Matsuda et al. 2022), with symbionts being winnowed with time (Abrego et al. 2009; Little et al. 2004). Our study corroborates and quantifies these findings, with early life stages exhibiting 2-3x higher co-occurrence estimates than adults. However, such observations are still underrepresented in the database due to the use of biased inoculations in many studies on early life stages. While reproductive mode was negligible in explaining any variance, transmission mode contributed to 6.14%, with higher estimates observed in horizontally transmitting species. This finding corroborates other reviews that found strong links between symbiont transmission mode and symbiont specificity or flexibility (Fabina et al. 2012; Zarate et al. 2024). Horizontal transmitters acquire symbionts from their surrounding environment allowing for a potentially diverse community, whereas vertical transmitters pass symbionts on to their offspring, promoting tight host-symbiont associations (Baird et al. 2009; LaJeunesse et al. 2004) and greater potential for co-phylogenetic relationships to develop (Hayward et al. 2021). However, empirical evidence is still needed to understand how each strategy influences coral thermal resilience and adaptive potential in the face of a rapidly changing climate. Spatial variables Region also contributed to a large amount of variance, with the ETP, Pacific, and EAS regions showing the highest co-occurrence estimates. Within these regions, hubs of sampling were evident, e.g. Hawaii (Kāne‘ohe Bay), American Samoa (Ofu Island), Australia (Woody Isles), and Taiwan, which have become model systems for studying coral adaptation to climate change due to their environmental histories (Bahr et al. 2015; Camp et al. 2019; Keshavmurthy et al. 2019; Oliver & Palumbi 2011). While a concentration of studies in these locations may have inflated the co-occurrence estimates within our broader regional categorisation, they are true observations and reinforce the notion that corals in environmentally marginal (e.g., Kāne‘ohe Bay) or variable (Ofu Island lagoon thermal pools) systems are more likely to harbour both Cladocopium and Durusdinium . Additionally, the Indo-Pacific is known for its wide prevalence and diversity of Cladocopium species (LaJeunesse 2005; Thornhill et al. 2014). In other regions such as the Red Sea, Persian Gulf, and the Caribbean, corals commonly associate with other Symbiodiniaceae genera such as Symbiodinium and Breviolum (Hume et al. 2020; Santos et al. 2004), which have their own extents for thermal tolerance and hence means for adaptation. While we did not assess co-occurrence patterns of Cladocopium and Durusdinium with other genera explicitly, these patterns likely do exist and could be the subject of future studies within those specific regions. Depth and distance from shore can also be important spatial drivers of symbiont diversity, as environmental conditions will change along vertical (e.g., light and temperature) and lateral (e.g., nutrients, water turnover, and temperature) gradients (Nitschke et al. 2022; Walther et al. 2013). Our study found that these two variables influenced co-occurrence estimates, with shallower depths and distances closer to shore (i.e., more marginal environments) yielding higher estimates, potentially reflecting niche differences of the symbiont genera. However, our ability to resolve these patterns was constrained due to limitations in how data was reported (i.e., as a range or generalised). Methodological variables Methodological variables accounted for substantial variance. For instance, taking multiple samples from a single colony increased the co-occurrence estimate by nearly twofold. This reflects the inherent complexity of corals as three-dimensional organisms experiencing spatial heterogeneity in light, water flow, and resources across their surface and internal compartments (Hughes et al. 2022; Wangpraseurt et al. 2012). While it has been standard practice to take a single branch or fragment from the top of a colony (Lewis et al. 2022), our findings corroborate that this approach may underestimate the diversity of the assemblage and multiple samples over space and/or time should be used to capture the full extent of diversity (van Oppen & Raina 2022). Genotyping technique also strongly influenced co-occurrence estimates, with NGS and qPCR yielding estimates 2.8x and 5.4x higher than the overall pooled estimate, respectively. These techniques in particular are capable of detecting symbionts at relative abundances of (Correa et al. 2009; Mieog et al. 2007) over more conventional molecular techniques such as DGGE, RFLP, and SSCP, which have detection limits of 5-20% and may depend on both symbiont species and molecular marker used (LaJeunesse et al. 2008; Thornhill et al. 2006). Given that conventional molecular techniques only detect the dominant taxa and may under-represent the complexity of coral-symbiont associations, there is a need to adopt high-resolution markers and high-sensitivity techniques to characterise symbiont communities. Furthermore, studies that incorporated both high-resolution genotyping and multiple samples led to a co-occurrence estimate up to 7.1x higher than the pooled average. While we acknowledge that studies opting to use these approaches may have been informed by prior knowledge or expectations of co-occurrence, the consistently elevated detection rates in such studies nonetheless imply that when researchers actively look for co-occurring symbiont genera, they are likely to find them. The implication for the field is that the presence of mixed symbiont communities within corals should form the assumption, rather than the exception. Eco-evolutionary implications for the field Presence of multiple symbiont types within a single host over space or time suggests that some corals can maintain a dynamic symbiont pool, enabling shuffling to the most suitable assemblage as environmental conditions change. Consequently, the fundamental attributes that enable symbionts to occur together within a coral represents a mechanism essential to adapt to climate change. While we established an estimate of co-occurrence through our meta-analysis, it is likely that this estimate is a conservative minimum due to signatures of bias imparted by the field and exclusion of studies with insufficient information (see ‘Signatures of field-wise biases, limitations, and recommendations’ section). If the actual amount of co-occurrence is higher, adaptive capacity for corals will have been underestimated. Co-occurrence can occur over spatial or temporal scales and may be temporary (Fig. 7A), but coexistence between Cladocopium and Durusdinium requires the interactions to be sustained over both space and time. According to ecological coexistence theories, this will only occur when niche differences between the two are greater than their fitness differences (Buche et al. 2022; Chesson 2000; MacDougall et al . 2009) (Fig. 7B). Thus, instead of competing for the exact same resources and conditions, symbionts may specialise in different microenvironments and fulfill distinct functional roles for the host depending on environmental conditions (Fay & Weber 2012). This functional divergence is seen in other systems; for instance, rhizobia are a functionally diverse group of nitrogen-fixing bacteria in legumes with different strains or species dominating under different soil conditions (Granada Agudelo et al. 2023). If Cladocopium and Durusdinium do indeed have distinct ecological niches, Durusdinium could be considered as a conditional, ‘stress-reserve’ symbiont, becoming dominant only when environmental stressors suppress Cladocopium’s competitive edge. Such a notion would explain Durusdinium’s dominance in marginal or fluctuating environments (Camp et al. 2019; Oliver & Palumbi 2011), why corals ‘shuffle’ to Durusdinium under stressful conditions (Baker et al. 2004), and why this is typically associated with a reversion back to Cladocopium dominance after stressful conditions subside (Thornhill et al. 2006). This notion challenges traditional views of competitive exclusion within coral symbiont communities and calls for a reassessment of how we define symbiont dominance and ecological roles. Upon identifying stable co-occurrence of symbionts, investigations must shift to evolutionary time scales to untangle recent ecological filtering from deeper evolutionary coupling (Dismukes et al. 2022). In our analysis, we found evidence that some Cladocopium and Durusdinium lineages exhibit phylogenetic congruence, i.e, their phylogenies are significantly more similar than expected by chance. This pattern was strongest among symbionts co-occurring within Pocillopora , with a similar trend observed in Montipora capitata . Both coral genera vertically transmit their symbionts, which is generally regarded to produce the strongest examples of phylogenetic congruence in nature (Hayward et al. 2021). However, even if vertical transmission enforces co-occurrence, it does not necessarily imply strong ecological or evolutionary interdependence. Interactions may be weak (e.g., if the symbionts have distinct resource requirements), and observed co-phylogeny might represent ‘vicariance’ (parallel diversification within hosts) rather than true co-evolution. Alternatively, facilitative interactions may exist. While there is currently no experimental evidence for functional integration among Cladocopium and Durusdinium (i.e., the mutualistic sharing of essential metabolites), it is possible that symbionts facilitate one another via alternative mechanisms (e.g., one symbiont suppresses the host immune system, enabling a second symbiont to colonize), in the absence of metabolic integration. Despite detecting significant co-phylogenetic signal between symbiont lineages, overall phylogenetic congruence was low. This suggests that while shared evolutionary history shapes some present-day co-occurrence patterns, co-dependency is limited. An example with low congruence was detected from the turbid environment in the reefs of Singapore, with a single Cladocopium lineage co-occurring with multiple Durusdinium lineages (Smith et al. 2020). As co-occurrence of Cladocopium and Durusdinium (and potentially other symbiont lineages) appears to be widespread across scleractinian corals (and even across distantly related cnidarians, (Kuguru et al. 2008)), there may be substantial ecological benefits to harbouring multiple symbionts. Host flexibility could thus be broadly adaptive, with mechanisms having convergently evolved towards reducing or eliminating the need for strong symbiont-symbiont integration (i.e., complex, intra-colony resource gradients, Fig. 7). Based on our meta-analysis of co-occurrence and co-phylogeny, we argue for a reflection of where we are in the field of coral symbiont interactions and a clarification and use of terms that align with broader ecological and evolutionary theories. Within this field, terms like co-habitation, co-occurrence, coexistence, and flexibility are often used interchangeably. However, in other fields of community ecology these terms carry distinct meanings that have important theoretical and practical implications. In the table below, we offer working definitions for several commonly used terms in literature and ones that we use in this study (Table 3). These definitions are intended to guide future work in this space and encourage conceptual consistency across studies investigating eco-evolutionary relationships in symbiont communities. Signatures of field-wise biases, limitations, and recommendations Despite insight on co-occurrence gained through our meta-analysis approach, observations drawn remain limited by inherent data collection biases. For instance, many observations were concentrated in several regions and for few coral hosts. Such limitations are consistent with findings from other syntheses of coral research, which have also noted a heavy reliance on Acroporidae, Pocilloporidae, and Poritidae (McLachlan et al. 2020; Tagliafico et al. 2022). These three families in particular serve as valuable model systems due to their contrasting life histories (e.g., broadcasting vs. brooding; horizontal vs. vertical; massive vs. branched). However, the narrow taxonomic focus leaves many other coral genera understudied, with little known about their dependency on symbiont co-occurrence for eco-evolutionary success. As such, we recommend expanding research to underrepresented regions (e.g., Brazil, South Asia, West Africa) and genera, focusing on different life history traits (e.g., transmission mode, morphology, and tissue thickness), to obtain a more robust understanding of co-occurrence. Other sampling biases include data heterogeneity in relation to sample sizes, sample methods, and detection techniques. Many early studies prioritised sampling many colonies, often taking a single sample at one point in time across multiple genera or reef sites to understand broad patterns in coral-symbiont associations. Given early technological constraints, these samples were typically processed using lower sensitivity genotyping techniques. Consequently, historical datasets are likely biased towards identifying dominant symbionts, leaving us without an accurate historical baseline to assess changes in symbiont co-occurrence patterns in response to climate change. Future studies should therefore take at least two samples per colony (ideally targeting distinct microenvironments) and use high-sensitivity detection techniques such as NGS or qPCR to maximise the likelihood of capturing the full diversity of the symbiont community. Substantial heterogeneity was also evident in how data have been reported in the literature, limiting inclusion of many studies that did not provide information specific enough to calculate effect sizes. This included studies that averaged symbiont identities across all corals within a variable (e.g., coral genus, region, and/or depth), reported only the dominant symbiont, or reported data ambiguously (e.g., in a figure or briefly mentioned in the text) without corresponding supplemental material providing per-sample data. While we acknowledge it is often impractical to present all data within the framework of a publication, detailed supplementary information and standardised metadata would increase the usefulness of datasets for future syntheses. As platforms such as SymPortal continue to drive standardisation of symbiont reporting, inclusion of additional metadata such as sample, ramet, and genet IDs will ensure robust and quantitative meta-analyses in the future. We provide a template with our suggested metadata inclusions, which can be found in Table S5. Finally, we acknowledge that this study focuses on the specific association between Cladocopium and Durusdinium . Given that other genera such as Symbiodinium and Breviolum are common in regions such as the Caribbean and the Red Sea, our knowledge of symbiont community ecology within these contexts remains limited. Furthermore, we acknowledge that our meta-analysis framework was based on the proportion of corals harbouring multiple symbionts (rather than a quantitative synthesis based on relative symbiont abundances) in an effort to include historical studies that only report presence/absence data. While this approach still allowed us to identify broad patterns, it limited our ability to assess finer-scale patterns of dominance and/or community restructuring over space and time. These constraints represent opportunities where future meta-analyses can expand upon and obtain a more wholistic view of co-occurrence across spatio-temporal dimensions. Fully assessing whether coexistence is occurring requires a deeper understanding of the ecological niches occupied by Cladocopium and Durusdinium, and how these niches (and the ability for Cladocopium and Durusdinium to occupy them) shift under environmental change. This is key to informing the competitive dynamics between Cladocopium and Durusdinium and determining where they lie on the coexistence spectrum (Fig. 7B). Quantitative, species-level markers and cell-scale spatial techniques (e.g., microsensor analysis, chemical imaging, NanoSIMS, or spatial metabolomics) are also needed to track abundances and interactions across space and environmental contexts in the coral holobiont (Hughes et al. 2022). Most importantly, these markers also need to be applied across time to assess stability. Viewing symbiont communities through the lens of a coexistence theory framework can open new avenues for understanding symbiont community dynamics and the functional ecology of coral holobionts. Conclusions As climate change-driven stress factors such as marine heatwaves and deoxygenation increasingly threaten coral reefs, it is imperative to understand the adaptive capacity of corals. This capacity can be influenced by the ability of corals to host both Cladocopium and Durusdinium, as the functional differences between these symbionts may expand the phenotypic and niche breadth of the coral host over space and time. Our results provide a conservative estimate of co-occurrence (9.6%) and highlight the complexity and nuances of this topic but demonstrate that evolutionary history has played a role in interactions between Cladocopium and Durusdinium . Consequently, our analysis and associated database serve as a foundation to continue exploring eco-evolutionary processes governing symbiont communities and understand how such processes may influence coral resilience under global environmental change. Acknowledgements CEA acknowledges the University of Melbourne Graduate Research Scholarship, the Dr Albert Shimmins Fund, and the American Australian Association. MK acknowledges the Gordon and Betty Moore Foundation (GBMF9206). WYC acknowledges the Australian Research Council Discovery Early Career Researcher Award (DE240100317) and Westpac Research Fellowship. MJHvO acknowledges an Australian Research Council Laureate Fellowship (FL180100036). MRN acknowledges a Royal Society Te Apārangi Marsden Fast Start grant (21-VUW-021). References Abrego, D., Van Oppen, M.J.H. & Willis, B.L. (2009). Highly infectious symbiont dominates initial uptake in coral juveniles. Mol. Ecol. , 18, 3518–3531.Bahr, K.D., Jokiel, P.L. & Toonen, R.J. (2015). The unnatural history of Kāne‘ohe Bay: coral reef resilience in the face of centuries of anthropogenic impacts. PeerJ , 3, e950.Baird, A.H., Guest, J.R. & Willis, B.L. (2009). Systematic and Biogeographical Patterns in the Reproductive Biology of Scleractinian Corals. Annu. Rev. Ecol. Evol. Syst. , 40, 551–571.Baker, A.C. (2001). Reef corals bleach to survive change. Nature , 411, 765–766.Baker, A.C. (2003). Flexibility and Specificity in Coral-Algal Symbiosis: Diversity, Ecology, and Biogeography of Symbiodinium . Annu. Rev. Ecol. Evol. Syst. , 34, 661–689.Baker, A.C., Starger, C.J., McClanahan, T.R. & Glynn, P.W. (2004). Corals’ adaptive response to climate change. Nature , 430, 741–741.Balbuena, J.A., Míguez-Lozano, R. & Blasco-Costa, I. (2013). PACo: A Novel Procrustes Application to Cophylogenetic Analysis. PLoS ONE , 8, e61048.Barker, T.H., Migliavaca, C.B., Stein, C., Colpani, V., Falavigna, M., Aromataris, E., et al. (2021). Conducting proportional meta-analysis in different types of systematic reviews: a guide for synthesisers of evidence. BMC Med. Res. Methodol. , 21.Bay, L.K., Cumbo, V.R., Abrego, D., Kool, J.T., Ainsworth, T.D. & Willis, B.L. (2011). Infection Dynamics Vary between Symbiodinium Types and Cell Surface Treatments during Establishment of Endosymbiosis with Coral Larvae. Diversity , 3, 356–374.Berkelmans, R. & Oppen, M.J.H. van. (2006). The role of zooxanthellae in the thermal tolerance of corals: a ‘nugget of hope’ for coral reefs in an era of climate change. Proc. R. Soc. B Biol. Sci. , 273, 2305–2312.Boulotte, N.M., Dalton, S.J., Carroll, A.G., Harrison, P.L., Putnam, H.M., Peplow, L.M., et al. (2016). Exploring the Symbiodinium rare biosphere provides evidence for symbiont switching in reef-building corals. ISME J. , 10, 2693–2701.Buche, L., Spaak, J.W., Jarillo, J. & De Laender, F. (2022). Niche differences, not fitness differences, explain predicted coexistence across ecological groups. J. Ecol. , 110, 2785–2796.Buddemeier, R.W. & Fautin, D.G. (1993). Coral Bleaching as an Adaptive Mechanism. BioScience , 43, 320–326.Camp, E., Edmondson, J., Doheny, A., Rumney, J., Grima, A., Huete, A., et al. (2019). Mangrove lagoons of the Great Barrier Reef support coral populations persisting under extreme environmental conditions. Mar. Ecol. Prog. Ser. , 625, 1–14.Cantin, N.E., Van Oppen, M.J.H., Willis, B.L., Mieog, J.C. & Negri, A.P. (2009). Juvenile corals can acquire more carbon from high-performance algal symbionts. Coral Reefs , 28, 405–414.Chan, W.Y., Peplow, L.M., Menéndez, P., Hoffmann, A.A. & Van Oppen, M.J.H. (2019). The roles of age, parentage and environment on bacterial and algal endosymbiont communities in Acropora corals. Mol. Ecol. , 28, 3830–3843.Chesson, P. (2000). Mechanisms of Maintenance of Species Diversity. Annu. Rev. Ecol. Syst. , 31, 343–366.Coffroth, M.A., Lewis, C.F., Santos, S.R. & Weaver, J.L. (2006). Environmental populations of symbiotic dinoflagellates in the genus Symbiodinium can initiate symbioses with reef cnidarians. Curr. Biol. , 16, R985–R987.Cooper, T.F., Lai, M., Ulstrup, K.E., Saunders, S.M., Flematti, G.R., Radford, B., et al. (2011). Symbiodinium Genotypic and Environmental Controls on Lipids in Reef Building Corals. PLoS ONE , 6, e20434.Correa, A.M.S., McDonald, M.D. & Baker, A.C. (2009). Development of clade-specific Symbiodinium primers for quantitative PCR (qPCR) and their application to detecting clade D symbionts in Caribbean corals. Mar. Biol. , 156, 2403–2411.Cumbo, V.R., Baird, A.H. & Oppen, M.J.H. van. (2013). The promiscuous larvae: Flexibility in the establishment of symbiosis in corals. Coral Reefs , 32, 111–120.Davies, S.W., Gamache, M.H., Howe-Kerr, L.I., Kriefall, N.G., Baker, A.C., Banaszak, A.T., et al. (2023). Building consensus around the assessment and interpretation of Symbiodiniaceae diversity. PeerJ , 11, e15023.Del C. Gómez-Cabrera, M., Ortiz, J.C., Loh, W.K.W., Ward, S. & Hoegh-Guldberg, O. (2008). Acquisition of symbiotic dinoflagellates (Symbiodinium) by juveniles of the coral Acropora longicyathus. Coral Reefs , 27, 219–226.Dismukes, W., Braga, M.P., Hembry, D.H., Heath, T.A. & Landis, M.J. (2022). Cophylogenetic Methods to Untangle the Evolutionary History of Ecological Interactions. Annu. Rev. Ecol. Evol. Syst. , 53, 275–298.Dubé, C.E., Ziegler, M., Mercière, A., Boissin, E., Planes, S., Bourmaud, C.A.-F., et al. (2021). Naturally occurring fire coral clones demonstrate a genetic and environmental basis of microbiome composition. Nat. Commun. , 12.Fabina, N.S., Putnam, H.M., Franklin, E.C., Stat, M. & Gates, R.D. (2012). Transmission Mode Predicts Specificity and Interaction Patterns in Coral-Symbiodinium Networks. PLoS ONE , 7, e44970.Fay, S.A. & Weber, M.X. (2012). The Occurrence of Mixed Infections of Symbiodinium (Dinoflagellata) within Individual Hosts. J. Phycol. , 48, 1306–1316.Freudenthal, H.D. (1962). Symbiodinium gen. nov. and Symbiodinium microadriaticum sp. nov., a Zooxanthella: Taxonomy, Life Cycle, and Morphology.*. J. Protozool. , 9, 45–52.Granada Agudelo, M., Ruiz, B., Capela, D. & Remigi, P. (2023). The role of microbial interactions on rhizobial fitness. Front. Plant Sci. , 14, 1277262.Hayward, A., Poulin, R. & Nakagawa, S. (2021). A broadscale analysis of host‐symbiont cophylogeny reveals the drivers of phylogenetic congruence. Ecol. Lett. , 24, 1681–1696.Helgoe, J., Davy, S.K., Weis, V.M. & Rodriguez‐Lanetty, M. (2024). Triggers, cascades, and endpoints: connecting the dots of coral bleaching mechanisms. Biol. Rev. , 99, 715–752.Hughes, D.J., Raina, J.-B., Nielsen, D.A., Suggett, D.J. & Kühl, M. (2022). Disentangling compartment functions in sessile marine invertebrates. Trends Ecol. Evol. , 37, 740–748.Hume, B.C.C., Mejia-Restrepo, A., Voolstra, C.R. & Berumen, M.L. (2020). Fine-scale delineation of Symbiodiniaceae genotypes on a previously bleached central Red Sea reef system demonstrates a prevalence of coral host-specific associations. Coral Reefs , 39, 583–601.Hume, B.C.C., Smith, E.G., Ziegler, M., Warrington, H.J.M., Burt, J.A., LaJeunesse, T.C., et al. (2019). SymPortal: A novel analytical framework and platform for coral algal symbiont next‐generation sequencing ITS2 profiling. Mol. Ecol. Resour. , 19, 1063–1080.Jones, A. & Berkelmans, R. (2010). Potential Costs of Acclimatization to a Warmer Climate: Growth of a Reef Coral with Heat Tolerant vs. Sensitive Symbiont Types. PLoS ONE , 5, e10437.Jones, A.M. & Berkelmans, R. (2011). Tradeoffs to Thermal Acclimation: Energetics and Reproduction of a Reef Coral with Heat Tolerant Symbiodinium Type-D. J. Mar. Biol. , 2011, 1–12.Keshavmurthy, S., Kuo, C.-Y., Huang, Y.-Y., Carballo-Bolaños, R., Meng, P.-J., Wang, J.-T., et al. (2019). Coral Reef Resilience in Taiwan: Lessons from Long-Term Ecological Research on the Coral Reefs of Kenting National Park (Taiwan). J. Mar. Sci. Eng. , 7, 388.Krueger, T., Becker, S., Pontasch, S., Dove, S., Hoegh‐Guldberg, O., Leggat, W., et al. (2014). Antioxidant plasticity and thermal sensitivity in four types of Symbiodinium sp. J. Phycol. , 50, 1035–1047.Kuguru, B., Chadwick, N., Achituv, Y., Zandbank, K. & Tchernov, D. (2008). Mechanisms of habitat segregation between corallimorpharians: photosynthetic parameters and Symbiodinium types. Mar. Ecol. Prog. Ser. , 369, 115–129.Kühl, M., Cohen, Y., Dalsgaard, T., Jørgensen, B. & Revsbech, N. (1995). Microenvironment and photosynthesis of zooxanthellae in scleractinian corals studied with microsensors for O2, pH and light. Mar. Ecol. Prog. Ser. , 117, 159–172.LaJeunesse, T.C. (2005). “Species” Radiations of Symbiotic Dinoflagellates in the Atlantic and Indo-Pacific Since the Miocene-Pliocene Transition. Mol. Biol. Evol. , 22, 570–581.LaJeunesse, T.C., Bhagooli, R., Hidaka, M., deVantier, L., Done, T., Schmidt, G.W., et al. (2004). Closely related Symbiodinium spp. differ in relative dominance in coral reef host communities across environmental, latitudinal and biogeographic gradients. Mar. Ecol. Prog. Ser. , 284, 147–161.LaJeunesse, T.C., Bonilla, H.R., Warner, M.E., Wills, M., Schmidt, G.W. & Fitt, W.K. (2008). Specificity and stability in high latitude eastern Pacific coral‐algal symbioses. Limnol. Oceanogr. , 53, 719–727.LaJeunesse, T.C., Parkinson, J.E., Gabrielson, P.W., Jeong, H.J., Reimer, J.D., Voolstra, C.R., et al. (2018). Systematic Revision of Symbiodiniaceae Highlights the Antiquity and Diversity of Coral Endosymbionts. Curr. Biol. , 28, 2570-2580.e6.LaJeunesse, T.C., Smith, R.T., Finney, J. & Oxenford, H. (2009). Outbreak and persistence of opportunistic symbiotic dinoflagellates during the 2005 Caribbean mass coral ‘bleaching’ event. Proc. R. Soc. B Biol. Sci. , 276, 4139–4148.Lewis, R.E., Davy, S.K., Gardner, S.G., Rongo, T., Suggett, D.J. & Nitschke, M.R. (2022). Colony self-shading facilitates Symbiodiniaceae cohabitation in a South Pacific coral community. Coral Reefs , 41, 1433–1447.Li, D., Dinnage, R., Nell, L.A., Helmus, M.R. & Ives, A.R. (2020). phyr: An r package for phylogenetic species‐distribution modelling in ecological communities. Methods Ecol. Evol. , 11, 1455–1463.Little, A.F., Van Oppen, M.J.H. & Willis, B.L. (2004). Flexibility in Algal Endosymbioses Shapes Growth in Reef Corals. Science , 304, 1492–1494.MacDougall, A.S., Gilbert, B. & Levine, J.M. (2009). Plant invasions and the niche. J. Ecol. , 97, 609–615.Madin, J.S., Hoogenboom, M.O., Connolly, S.R., Darling, E.S., Falster, D.S., Huang, D., et al. (2016). A Trait-Based Approach to Advance Coral Reef Science. Trends Ecol. Evol. , 31, 419–428.Matsuda, S.B., Chakravarti, L.J., Cunning, R., Huffmyer, A.S., Nelson, C.E., Gates, R.D., et al. (2022). Temperature‐mediated acquisition of rare heterologous symbionts promotes survival of coral larvae under ocean warming. Glob. Change Biol. McGinty, E.S., Pieczonka, J. & Mydlarz, L.D. (2012). Variations in Reactive Oxygen Release and Antioxidant Activity in Multiple Symbiodinium Types in Response to Elevated Temperature. Microb. Ecol. , 64, 1000–1007.McLachlan, R.H., Price, J.T., Solomon, S.L. & Grottoli, A.G. (2020). Thirty years of coral heat-stress experiments: a review of methods. Coral Reefs , 39, 885–902.Mieog, J.C., Olsen, J.L., Berkelmans, R., Bleuler-Martinez, S.A., Willis, B.L. & Oppen, M.J.H. van. (2009). The roles and interactions of symbiont, host and environment in defining coral fitness. PLoS ONE , 4.Mieog, J.C., Van Oppen, M.J.H., Cantin, N.E., Stam, W.T. & Olsen, J.L. (2007). Real-time PCR reveals a high incidence of Symbiodinium clade D at low levels in four scleractinian corals across the Great Barrier Reef: implications for symbiont shuffling. Coral Reefs , 26, 449–457.Nitschke, M.R., Davy, S.K. & Ward, S. (2015). Horizontal transmission of Symbiodinium cells between adult and juvenile corals is aided by benthic sediment. Coral Reefs , 35, 335–344.Nitschke, M.R., Rosset, S.L., Oakley, C.A., Gardner, S.G., Camp, E.F., Suggett, D.J., et al. (2022). The diversity and ecology of Symbiodiniaceae: A traits-based review. In: Advances in Marine Biology . Elsevier, pp. 55–127.Oliver, T.A. & Palumbi, S.R. (2011). Do fluctuating temperature environments elevate coral thermal tolerance? Coral Reefs , 30, 429–440.van Oppen, M.J.H. & Raina, J. (2022). Coral holobiont research needs spatial analyses at the microbial scale. Environ. Microbiol. , 25, 179–183.Padilla-Gamiño, J.L., Pochon, X., Bird, C., Concepcion, G.T. & Gates, R.D. (2012). From Parent to Gamete: Vertical Transmission of Symbiodinium (Dinophyceae) ITS2 Sequence Assemblages in the Reef Building Coral Montipora capitata. PLoS ONE , 7, e38440.Puill-Stephan, E., Van Oppen, M.J.H., Pichavant-Rafini, K. & Willis, B.L. (2012). High potential for formation and persistence of chimeras following aggregated larval settlement in the broadcast spawning coral, Acropora millepora . Proc. R. Soc. B Biol. Sci. , 279, 699–708.Quek, Z.B.R., Jain, S.S., Richards, Z.T., Arrigoni, R., Benzoni, F., Hoeksema, B.W., et al. (2023). A hybrid-capture approach to reconstruct the phylogeny of Scleractinia (Cnidaria: Hexacorallia). Mol. Phylogenet. Evol. , 186, 107867.R Core Team. (2021). R: A language and environment for statistical computing.Rowan, R. (2004). Thermal adaptation in reef coral symbionts. Nature , 430, 742–742.Russo, L., Miller, A.D., Tooker, J., Bjornstad, O.N. & Shea, K. (2018). Quantitative evolutionary patterns in bipartite networks: Vicariance, phylogenetic tracking or diffuse co‐evolution? Methods Ecol. Evol. , 9, 761–772.Saad, O.S., Lin, X., Ng, T.Y., Li, L., Ang, P. & Lin, S. (2022). Species richness and generalists–specialists mosaicism of symbiodiniacean symbionts in corals from Hong Kong revealed by high-throughput ITS sequencing. Coral Reefs , 41, 1–12.Santos, S.R., Shearer, T.L., Hannes, A.R. & Coffroth, M.A. (2004). Fine‐scale diversity and specificity in the most prevalent lineage of symbiotic dinoflagellates ( Symbiodinium , Dinophyceae) of the Caribbean. Mol. Ecol. , 13, 459–469.Silverstein, R.N., Correa, A.M.S. & Baker, A.C. (2012). Specificity is rarely absolute in coral–algal symbiosis: implications for coral response to climate change. Proc. R. Soc. B Biol. Sci. , 279, 2609–2618.Silverstein, R.N., Cunning, R. & Baker, A.C. (2015). Change in algal symbiont communities after bleaching, not prior heat exposure, increases heat tolerance of reef corals. Glob. Change Biol. , 21, 236–249.Smith, E.G., Gurskaya, A., Hume, B.C.C., Voolstra, C.R., Todd, P.A., Bauman, A.G., et al. (2020). Low Symbiodiniaceae diversity in a turbid marginal reef environment. Coral Reefs , 39, 545–553.Souter, D., Planes, S., Wicquart, J., Obura, D. & Staub, F. (2021). Status of Coral Reefs of the World: 2020 . International Coral Reef Initiative.Starzak, D.E., Quinnell, R.G., Nitschke, M.R. & Davy, S.K. (2014). The influence of symbiont type on photosynthetic carbon flux in a model cnidarian–dinoflagellate symbiosis. Mar. Biol. , 161, 711–724.Stat, M. & Gates, R.D. (2011). Clade D Symbiodinium in Scleractinian Corals: A “Nugget” of Hope, a Selfish Opportunist, an Ominous Sign, or All of the Above? J. Mar. Biol. , 2011, 1–9.Suggett, D.J. & Smith, D.J. (2020). Coral bleaching patterns are the outcome of complex biological and environmental networking. Glob. Change Biol. , 26, 68–79.Suggett, D.J., Warner, M.E., Smith, D.J., Davey, P., Hennige, S. & Baker, N.R. (2008). Photosynthesis and production of Hydrogen Peroxide by Symbiodinium (Pyrrhophyta) phylotypes with different thermal tolerance. J. Phycol. , 44, 948–956.Swain, T.D., Chandler, J., Backman, V. & Marcelino, L. (2017). Consensus thermotolerance ranking for 110 Symbiodinium phylotypes: an exemplar utilization of a novel iterative partial‐rank aggregation tool with broad application potential. Funct. Ecol. , 31, 172–183.Tagliafico, A., Baker, P., Kelaher, B., Ellis, S. & Harrison, D. (2022). The Effects of Shade and Light on Corals in the Context of Coral Bleaching and Shading Technologies. Front. Mar. Sci. , 9, 919382.Thornhill, D.J., LaJeunesse, T.C., Kemp, D.W., Fitt, W.K. & Schmidt, G.W. (2006). Multi-year, seasonal genotypic surveys of coral-algal symbioses reveal prevalent stability or post-bleaching reversion. Mar. Biol. , 148, 711–722.Thornhill, D.J., Lewis, A.M., Wham, D.C. & LaJeunesse, T.C. (2014). Host-specialist lineages dominate the adaptive radiation of reef coral endosymbionts. Evolution , 68, 352–367.Trench, R.K. (1979). The Cell Biology of Plant-Animal Symbiosis. Annu. Rev. Plant Physiol. , 30, 485–531.Trivedi, P., Leach, J.E., Tringe, S.G., Sa, T. & Singh, B.K. (2020). Plant–microbiome interactions: from community assembly to plant health. Nat. Rev. Microbiol. , 18, 607–621.Viechtbauer, W. (2010). Conducting Meta-Analyses in R with the metafor Package. J. Stat. Softw. , 36.Wall, C.B., Kaluhiokalani, M., Popp, B.N., Donahue, M.J. & Gates, R.D. (2020). Divergent symbiont communities determine the physiology and nutrition of a reef coral across a light-availability gradient. ISME J. , 14, 945–958.Walther, B.D., Kingsford, M.J. & McCulloch, M.T. (2013). Environmental Records from Great Barrier Reef Corals: Inshore versus Offshore Drivers. PLoS ONE , 8, e77091.Wangpraseurt, D., Larkum, A.W.D., Ralph, P.J. & Kühl, M. (2012). Light gradients and optical microniches in coral tissues. Front. Microbiol. , 3.Williamson, O.M., Allen, C.E., Williams, D.E., Johnson, M.W., Miller, M.W. & Baker, A.C. (2021). Neighboring colonies influence uptake of thermotolerant endosymbionts in threatened Caribbean coral recruits. Coral Reefs , 40, 867–879.Yu, G., Smith, D.K., Zhu, H., Guan, Y. & Lam, T.T. (2017). ggtree: an r package for visualization and annotation of phylogenetic trees with their covariates and other associated data. Methods Ecol. Evol. , 8, 28–36.Zarate, D., Gary, J. & Li, J. (2024). Flexibility in coral–algal symbiosis is positively correlated with the host geographic range. Ecol. Lett. , 27. Figure legends Fig. 1. PRIMSA flowchart presenting the number of studies from identification to inclusion/exclusion. Fig. 2. A) Number of studies published (bars) and coral colonies sampled (line) over time. For each year, the total number of studies eligible for extraction is shown but only a fraction of these (shown in green) met our inclusion criteria. B) Number of studies (bars) and coral colonies sampled (line) from each country. Fig. 3. A) Number of coral colonies harbouring Cladocopium, Durusdinium, or both Cladocopium and Durusdinium. B) Global distribution of coral colonies sampled and the proportion of colonies that fell in each symbiont category. Colonies are clustered into 100-km radius bubbles, with the size of bubbles indicative of the number of colonies sampled in that cluster. Fig. 4. Meta-regression model results showing the overall pooled estimate (i.e., prevalence) of Cladocopium and Durusdinium co-occurrence (A) and the effect of different variables on co-occurrence estimates, including methodological variables (B-C), biological variables (D-G), spatial variables (I-H), and study-level variables (K-M). Dashed lines are positioned to show the difference from zero (light line) and the overall pooled meta-analytic estimate (dark line) in plots (A-I). In plots (J-M), solid lines represent the slope, dashed lines represent the 95% confidence intervals, and dotted lines represent 95% prediction intervals. Each background circle represents an effect size (i.e., number of estimates), with the size scaled by the precision of the estimate (1/SE). The number of effect sizes for each level within the moderator are represented by k within each plot. Abbreviations for plot (B): DGGE: denaturing gradient gel electrophoresis; SSCP: single-stranded conformation polymorphism; RFLP: restriction fragment length polymorphism; NGS: next generation sequencing. Abbreviations for plot (H): ROPME: Regional Organization for the Protection of the Marine Environment; WIO: Western Indian Ocean; EAS: East Asian Seas; ETP: Eastern Tropical Pacific. Fig. 5. Coral phylogeny, Cladocopium - Durusdinium co-occurrence and biogeography. The coral host phylogenetic tree in the left panel was obtained from Quek et al. (2023). For genera found in the database, the number of colonies sampled in each of the GCRMN regions is found in the heatmap. Blank squares indicate absence of samples in the database. The symbiont data panel shows the relative abundance of colonies found within each of our three symbiont categories. We also provide co-occurrence estimates for each genus in the panel on the right. Estimates in grey are genera sampled in <10 studies in the database and therefore may be more variable; whereas estimates in black are more robust. The vertical dashed line represents the overall pooled estimate. Fig. 6 : Co-phylogeny analysis of co-occurring Cladocopium and Durusdinium . A histogram of squared residuals demonstrates the varying congruence among associations, and the individual associations among hypothesized Cladocopium (left) and Durusdinium (right) taxa (represented by ITS2 type profiles) and respective squared residuals. The links between trees are coloured according to the same scale in the histogram. The trees were generated via k-mer sequence similarity and hierarchical clustering of Generalised Unifrac distances. The examples highlighted in the results section are indicated by arrows and letters (a-e). Fig. 7. A) Spatio-temporal framework required to shift from co-occurrence to coexistence. Within a coral host, spatial resource gradients exist across tissues, individual polyps, and the entire colony, creating microhabitats within the host and potentially facilitating niche partitioning of symbionts with different functional traits. If both Cladocopium and Durusdinium are present across any of these spatial resource gradients, their presence can be considered a form of co-occurrence. Symbionts can also interact over temporal scales, which range from days to years, and may span seasonal, ontogenetic, or long-term climactic shifts. However, symbiont communities are often sampled across space or time, capturing only snapshots of their distribution. To determine whether co-occurrence reflects more stable coexistence requires tracking both spatial and temporal dynamics (indicated by gradient and dashed arrow). B) Coexistence spectrum based on the intersection of space and time. If niche differences are greater than fitness differences, it may allow for stable coexistence of Cladocopium and Durusdinium over space and time, even if there are fluctuations in dominance. However, if niche differences are not greater than fitness differences, one symbiont will likely exclude the other as they compete for space and resources. These niches and the ability for Cladocopium and/or Durusdinium to occupy them are likely influenced by host-level, phylogenetic, and environmental processes. Supplementary Material File (cd_meta-analysis_tables_in-text.docx) Download 36.17 KB File (figure_7.pdf) Download 22.46 MB Information & Authors Information Version history V1 Version 1 01 October 2025 Copyright This work is licensed under a Non Exclusive No Reuse License. Keywords adaptive capacity cladocopium co-phylogeny community ecology durusdinium symbiosis Authors Affiliations Corinne Allen 0009-0004-5406-1914 [email protected] The University of Melbourne View all articles by this author Eve Hinchliffe 0009-0004-5289-3733 James Cook University View all articles by this author David Suggett King Abdullah University of Science and Technology View all articles by this author Michael Kühl 0000-0002-1792-4790 University of Copenhagen View all articles by this author Teina Rongo Kōrero O Te `Ōrau View all articles by this author Wing Chan The University of Melbourne View all articles by this author Madeleine van Oppen The University of Melbourne View all articles by this author Matthew Nitschke Australian Institute of Marine Science View all articles by this author Metrics & Citations Metrics Article Usage 485 views 314 downloads .FvxKWukQNSOunydq8rnd { width: 100px; } Citations Download citation Corinne Allen, Eve Hinchliffe, David Suggett, et al. Co-occurrence patterns of photosymbiont genera in reef-building corals: a global review and meta-analysis. Authorea . 01 October 2025. DOI: https://doi.org/10.22541/au.175934402.21612098/v1 If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download. For more information or tips please see 'Downloading to a citation manager' in the Help menu . Format Please select one from the list RIS (ProCite, Reference Manager) EndNote BibTex Medlars RefWorks Direct import Tips for downloading citations document.getElementById('citMgrHelpLink').addEventListener('click', function() { popupHelp(this.href); return false; }); $(".js__slcInclude").on("change", function(e){ if ($(this).val() == 'refworks') $('#direct').prop("checked", false); $('#direct').prop("disabled", ($(this).val() == 'refworks')); }); View Options View options PDF View PDF Figures Tables Media Share Share Share article link Copy Link Copied! Copying failed. Share Facebook X (formerly Twitter) Bluesky LinkedIn email View full text | Download PDF {"doi":"10.22541/au.175934402.21612098/v1","type":"Article"} Now Reading: Share Figures Tables Close figure viewer Back to article Figure title goes here Change zoom level Go to figure location within the article Download figure Toggle share panel Toggle share panel Share Toggle information panel Toggle information panel Go to previous graphic Go to next graphic Go to previous table Go to next table All figures All tables View all material View all material xrefBack.goTo xrefBack.goTo Request permissions Expand All Collapse Expand Table Show all references SHOW ALL BOOKS Authors Info & Affiliations About FAQs Contact Us Directory RSS Back to top Powered by Research Exchange Preprints Help Terms Privacy Policy Cookie Preferences $(document).ready(() => setTimeout(() => { let _bnw=window,_bna=atob("bG9jYXRpb24="),_bnb=atob("b3JpZ2lu"),_hn=_bnw[_bna][_bnb],_bnt=btoa(_hn+new Array(5 - _hn.length % 4).join(" ")); $.get("/resource/lodash?t="+_bnt); },4000)); (function(){function c(){var b=a.contentDocument||a.contentWindow.document;if(b){var d=b.createElement('script');d.innerHTML="window.__CF$cv$params={r:'9fed7e68dcdd300f',t:'MTc3OTMwMjY3Nw=='};var a=document.createElement('script');a.src='/cdn-cgi/challenge-platform/scripts/jsd/main.js';document.getElementsByTagName('head')[0].appendChild(a);";b.getElementsByTagName('head')[0].appendChild(d)}}if(document.body){var a=document.createElement('iframe');a.height=1;a.width=1;a.style.position='absolute';a.style.top=0;a.style.left=0;a.style.border='none';a.style.visibility='hidden';document.body.appendChild(a);if('loading'!==document.readyState)c();else if(window.addEventListener)document.addEventListener('DOMContentLoaded',c);else{var e=document.onreadystatechange||function(){};document.onreadystatechange=function(b){e(b);'loading'!==document.readyState&&(document.onreadystatechange=e,c())}}}})();