ARC Centre of Excellence for Coral Reef Studies
Emeritus Professor Terry Hughes of James Cook University, who served as the ARC Centre of Excellence for Coral Reef Studies Director from 2005 to 2020, topped the list. In 2016, Nature named Hughes one of the “10 people who mattered this year” for addressing the widespread coral bleaching event brought on by climate change. Hughes’s research has led to practical solutions to improve marine environmental management [ 57 ]. His work on the effects of climate change on coral reefs has been widely cited, especially his paper on the resistance of some coral reefs to climate change and anthropogenic factors [ 58 ]. Second on the list was Professor Ove Hoegh-Guldberg from the University of Queensland (UQ), Australia. He serves as Director of the Global Change Institute at UQ and also as a Chief Investigator at the ARC Centre of Excellence for Coral Reef Studies [ 59 ]. Dr. Katharina Fabricus is a coral reef ecologist and a Senior Principal Research Scientist at the Australian Institute of Marine Science [ 60 ]. Many of her highly cited publications are on topics related to ocean acidification [ 61 , 62 , 63 , 64 ], the impacts of water quality on coral reefs [ 65 , 66 ] and understanding the effects of terrestrial run-off on coral reefs [ 67 , 68 ].
Dr. Peter W. Glynn, from the National Center for Coral Reef Research, University of Miami, was among the pioneers in analyzing and reporting the impacts of the 1982–1983 El Niño warming event on Eastern Pacific coral reefs [ 69 ]. This was followed by Tim McClanahan, a senior conservation zoologist at the Wildlife Conservation Society and also an associate at the ARC Centre of Excellence for Coral Reef Studies [ 70 ]. His global study of more than 2500 reefs produced a Bayesian hierarchical model to predict how reef fish biomass is related to 18 socioeconomic drivers and environmental conditions [ 71 ]. Dr. Peter Mumby is a coral reef biologist from the University of Queensland and also a Chief Investigator at the ARC Centre of Excellence for Coral Reef Studies [ 72 ]. He collaborated with Professor Ove Hoegh-Guldberg to publish “Coral Reefs under Rapid Climate Change and Ocean Acidification”, which is one of the most cited papers in the field (see Table 4 ) [ 22 ]. Another study published in Nature reported the resilience of Caribbean coral reefs against moderate hurricanes [ 73 ]. Dr. David Bellwood, an Australian Laureate Fellow and Distinguished Professor at James Cook University [ 74 ], has reported on the effects of climate change on coral reef ecosystems, even though his primary research interests are in biology and the evolution of reef fish [ 58 , 75 , 76 ].
The most highly cited references about coral reefs and climate change.
Reference | Citations | Burst Index * | Burst Period | Journal | Centrality ** | Cluster |
---|---|---|---|---|---|---|
Hughes et al. [ ] | 546 | 157.67 | 2018–2021 | Nature | 0.7 | #0, #4 |
Hughes et al. [ ] | 359 | 125.52 | 2018–2021 | Science | 0 | #4 |
Hoegh-Guldberg et al. [ ] | 351 | 166.14 | 2008–2012 | Science | 0.3 | #0, #2 |
Hughes et al. [ ] | 252 | 87.69 | 2018–2021 | Nature | 0.4 | #0, #7 |
Hughes et al. [ , ] | 246 | 94.42 | 2019–2021 | Nature | 0.01 | #8 |
De’Ath et al. [ ] | 204 | 79.65 | 2013–2017 | PNAS | 0.6 | #7, #8 |
Pandolfi et al. [ ] | 175 | 68.19 | 2012–2016 | Science | 0.4 | #0, #2 |
Fabricius et al. [ ] | 162 | 63.09 | 2012–2016 | Nature Climate Change | 0.05 | #2 |
LaJeunesse et al. [ ] | 134 | 68.39 | 2013–2017 | Current Biology | 0 | #6 |
Hughes et al. [ ] | 122 | 48.71 | 2011–2015 | Trends in Ecology & Evolution | 0.01 | #0 |
Heron et al. [ ] | 118 | 33.85 | 2018–2021 | Scientific Reports | 0.6 | #4 |
Ainsworth et al. [ ] | 117 | 28.75 | 2017–2021 | Science | 0.4 | #0, #4 |
Palumbi et al. [ ] | 116 | 39.87 | 2015–2019 | Science | 0.05 | #0 |
Baker et al. [ ] | 108 | 50.7 | 2010–2013 | CETP | 0.04 | #5, #6 |
Anthony et al. [ ] | 105 | 45.6 | 2009–2013 | Global Change Biology | 0.4 | #10 |
Kroeker et al. [ ] | 104 | 39.22 | 2014–2018 | Global Change Biology | 0.4 | #2, #10 |
Zeebe and Wolf-Gladrow [ ] | 97 | 39.97 | 2010–2014 | Gulf Professional Publishing | 0.6 | #4, #10 |
Jackson et al. [ ] | 96 | 34.08 | 2016–2019 | Global Coral Reef Monitoring Network | 0.4 | #7, #18 |
Bruno and Selig [ ] | 95 | 44.45 | 2008–2012 | PLoS one | 0.05 | #0, #7 |
* Burst index: the value generated from the CiteSpace indicates the level of importance of each article in the field. ** Centrality: the main focused article between cited references in the publications.
Dr. John Pandolfi from the University of Queensland is a paleoecologist and a Chief investigator at ARC. His research integrates long-term ecological and environmental time series data to discover past and future influences of natural variability, human impact, and climate change on coral reef resilience. Among his highly cited works is a projection of the future of coral reefs under global warming and ocean acidification [ 20 ]. Dr. Kenneth RN Anthony, an associate scientist at the Australian Institute of Marine Science and director of Environmental Strategies ES5, has published widely on ocean acidification [ 77 , 78 , 79 ]. Dr. John Bruno, from the University of North Carolina, is a marine ecologist focusing on the impacts of climate change on marine ecosystems, particularly coral reef ecology. His publication with Dr. Ove Hoegh-Guldberg, on the effects of climate change on global marine ecosystems is one of his most cited works [ 80 ].
Next on the list is Emeritus Professor Barbara E. Brown from Newcastle University, who conducted extensive research on coral bleaching, specifically on the role of zooxanthellae [ 81 ]. Next on the list is Professor Andrew C. Baker, a marine biologist at the University of Miami, who studies coral reefs and climate change. He leads the Coral Reef Futures Lab and focuses on developing and testing methods to increase coral reef resilience [ 82 ].
Glenn De’ath and Ray Berkelmans, both from The Australian Institute of Marine Science, are also highly cited for their research on coral reefs. De’ath’s work involves statistics and ecology, specifically on the Great Barrier Reef coral cover decline [ 83 ], while Berkelmans’ research focuses on thermal stress, adaptation to climate warming, the resilience of reef communities, and upwelling [ 84 ].
Joan Kleypas, a Senior Scientist from the National Center for Atmospheric Research, is also on the list, and her highly cited works revolve around the impact of ocean acidification on coral reefs [ 85 , 86 ]. Next is Professor Nick Graham from Lancaster University, who assesses the impacts of climate-induced coral bleaching on coral reef fish assemblages, fisheries, and ecosystem stability [ 87 ].
Emeritus Professor Michael Lesser from the University of New Hampshire is also highly cited for his work on climate change-related stressors’ biochemical and physiological impacts on coral reefs [ 88 ]. Professor Joseph Loya from Tel Aviv University quantifies changes in biodiversity and assesses reef health [ 89 , 90 ], while Professor Peter Edmunds focuses on the physiological ecology of tropical coral reefs [ 91 ].
Lastly, Toby Gardner is a Senior Research Fellow from Stockholm Environment Institute, known for his extensive work on Caribbean corals. He co-leads SEI’s Initiative on producer-to-consumer sustainability and the transparency for sustainable economies platform. His long-term observations revealed that the coral cover of the Caribbean basin declined by 80% in just thirty years [ 92 ].
CiteSpace’s “Category” node type was used to generate a visual map showing research disciplinary categories represented by papers addressing issues related to climate change’s impact on coral reefs. The centrality of a network (i.e., the center of collaborative activities) comprising 135 nodes and 336 links was computed after the data were simplified and merged (i.e., automatically generated from the CiteSpace algorithm and programming) ( Figure 6 ). The five disciplines with the most publications in descending order were marine and freshwater biology, environmental sciences, ecology, oceanography, and geosciences. The study of coral reefs is a multifaceted research topic that includes many fields of study, as demonstrated by the distribution map. Disciplines in related subjects such as biodiversity conservation, geography, physical sciences, biology, evolutionary biology, geology, paleontology, and water resources, show strong connections, represented by the sizes of the nodes. The number of published papers is comparably low in some research disciplines, such as toxicology, biotechnology and applied microbiology, green and sustainable science and technology, and biochemistry and molecular biology. However, the relatively high betweenness centrality values of these fields suggest their significant contribution to interdisciplinary research, signifying their pivotal position in the scientific network. This centrality may also hint at their potential for future development and advancement in the field.
Network of linked research disciplines. The sizes of the modes are proportional to the frequency of the subject category cooccurrence. The thickness of the lines between the two nodes is proportional to the strength of the linkages between the two research disciplines.
Cluster analysis is a popular method of statistical data analysis and knowledge discovery because of its ability to uncover latent semantic themes in textual data [ 93 , 94 ]. Cluster analysis can divide a large body of research data into various units based on the relative degree of term correlation, making it easier to identify the research themes, trends, and connections within a given field of study [ 94 , 95 ]. A cluster’s homogeneity can be quantified using an index called the mean silhouette, with values ranging from −1 to 1. The average silhouette value for each cluster was determined using CiteSpace. The higher the value, the more similar the cluster’s members are to one another [ 96 ]. The network showed 24 clusters in the context of the scientometric analysis mapping the link between climate change and coral reefs ( Figure 7 ).
The reference co-citation research cluster network. Based on a one-year interval, a 24-cluster network of document co-citation with burst detection from 1976 to 2021. Node sizes are proportional to the frequency of the publications’ co-citations.
The largest cluster (#0) has 291 members (i.e., number of publications) and a silhouette value of 0.863 and is labeled as “coral reef.” The most cited article of this cluster is by Gilmour et al. [ 97 ]. They monitored and assessed the impacts of the 2016 heat stress event on Western Australian coral reefs. They found that mass bleaching in 2016 reduced coral cover by 70% at Scott Reef and caused widespread mortality (>30%) at Christmas Island, Ashmore Reef, and inshore reefs in southern Kimberley. A coral phase shift is characterized by a rapid decline in coral abundance or cover and an accompanying rise in non-reef-building organisms, like algae and soft corals [ 98 , 99 ]. The second largest cluster (#1) has 247 members and a silhouette value of 0.876, labeled as “phase shift.” This publication by Brodie et al. [ 100 ], entitled “Terrestrial pollutant run-off to the Great Barrier Reef: An update of issues, priorities and management responses”, is the most cited article in this cluster. They addressed findings from studies of problems caused by surface run-off of pollutants like nitrates from fertilizers, herbicides from crops, etc. Within the cluster of the study, there are three different types of management generated automatically and have mentioned (i) Reef Plant 2009, (ii) Reef Rescue, and the (iii) Reef Protection Package in the analysis. These topics are just some of the initiatives set up to continuously monitor and report on levels of discharges into the Great Barrier Reef. Multiple observations of specific facets of the topic have been published; Hughes [ 101 ]; McManus and Polsenberg [ 102 ]; Idjadi et al. [ 103 ]; Norström et al. [ 104 ]; Graham et al. [ 105 ]; Crisp et al. [ 106 ]. Many anthropogenic stressors have been linked to this phenomenon [ 106 , 107 , 108 ]. Nutrients play a pivotal role in conceptual models of how coral reef communities form. These studies show that corals have a competitive advantage over macroalgae in low nutrient conditions but that the advantage shifts to macroalgae in higher nutrient conditions [ 102 , 109 ]. Siltation, resulting in mud-bacterial complexes, collectively known as “marine snow,” is another factor that hinders coral growth. In addition, excess nutrients resulting in plankton blooms reduce light, thereby inhibiting coral growth [ 110 ].
The fourth largest cluster (#3), “ocean acidification”, has 233 members and a silhouette value of 0.959. The most cited article of this cluster is Bates [ 111 ], which reported twenty years (1996 to 2016) of marine carbon cycle observations at Devils Hole, Bermuda. Her findings shed light on the dynamic nature of biogeochemical processes like primary production, respiration, calcification, and CaCO 3 deposition in the Bermuda reef system. During this period, neither warming nor cooling of any significance was observed. However, increases in inorganic carbon in onshore waters were primarily due to increased salinity (45%), uptake of anthropogenic CO 2 (25%), and changes in Bermuda reef biogeochemical processes (30%). Increases in atmospheric carbon dioxide concentrations result in the absorption of more carbon dioxide by oceans, which in turn causes a decrease in pH [ 112 , 113 , 114 , 115 ]. The majority of research on ocean acidification has focused on the impact of changes in ocean chemistry towards suboptimal states of aragonite and calcite saturation on the calcification processes of pelagic and benthic organisms [ 77 , 116 , 117 , 118 , 119 ]. However, it is likely that ocean acidification also has an effect on other physiological processes, such as growth and reproduction in significant reef-building species [ 77 ].
The Indo-Pacific region’s Coral Triangle, the world’s epicenter of marine biodiversity [ 120 , 121 ], is predicted to become a “marginal” coral habitat between 2020 and 2050 unless CO 2 emissions are reduced [ 122 , 123 ]. In addition to reducing coral diversity, acidification also results in a decline in shellfish and fish species due to the loss of reef structure, which provides habitat for these other species and reduces the reefs’ capacity to mitigate the effects of storm waves and erosion [ 122 , 124 ]. Ocean acidification has a devastating impact on the economies of ocean-dependent sectors of the global economy. Previous studies have provided estimates of the economic impact of ocean acidification on marine mollusk and shellfish production, as well as the bioeconomic costs associated with coral reef damage [ 125 ]. These studies have shed light on the detrimental effects of ocean acidification on marine ecosystems, which in turn, can have severe economic implications. Estimating these costs can aid in developing policies aimed at reducing the negative effects of ocean acidification and promoting the sustainable use of marine resources. For instance, according to a study by Narita et al. [ 126 ], the global annual loss of mollusk production due to the fact of ocean acidification could amount to between USD 6 billion and USD 100 billion. Commercially valuable finfish populations will suffer as a result of global ocean changes that reduce coral reef coverage, resulting in a loss of habitat, reduced availability of prey, and increased predation [ 125 , 127 , 128 ]. The scientometric analysis has identified four prominent clusters, also referred to as topics, which represent distinct research areas based on their geographic location. These clusters include the “central red sea” (#3), the “eastern pacific” (#5), the “great barrier reef” (#8), and the “Dominican Republic” (#18). These geographic regions are frequently cited in scientific research as they represent the study location of many relevant studies.
The most cited article of the “Red Sea” cluster is by Osman et al. [ 129 ], which mapped coral microbiome composition along the northern Red Sea. The Red Sea is a distinctive body of water that is an evaporative basin with a high salinity above 38 ppt [ 130 , 131 ]. It is home to some of the world’s most thriving and productive coral reef ecosystems [ 132 ]. Osman et al. [ 129 ] research offered a fresh understanding of the coral microbiome’s exclusive and endemic characteristics along the northern Red Sea refugia. They looked into the surface mucus layer (SML) for bacterial communities from six dominant coral species and discovered five novel algal endosymbionts. Over the past four decades, the average annual sea surface temperature in most of the world’s tropics and subtropics has risen between 0.4 °C and 1 °C. However, in the central Red Sea, where reef growth and scleractinian coral diversity are abundant, warming is more extensive than the observed mean tropical temperature increase [ 133 ]. The 2010 “Thuwal bleaching” in the central Red Sea was caused by a temperature rise of 10–11 °C, the largest coral bleaching event ever recorded. Furby et al. [ 131 ] conducted a survey and found that the “Thuwal bleaching” event caused more severe bleaching of inshore reefs (74% of hard corals were bleached) than offshore reefs (14% of hard corals were bleached). One mechanism that can lead to higher tolerance is repeated exposure to thermal stress [ 134 , 135 ]. Based on current knowledge, it is hypothesized that the reefs in the Red Sea will be relatively resistant to bleaching as sea temperatures rise, as noted in a study by Grimsditch and Salm [ 136 ]. However, reports indicate that bleaching is beginning to occur in the Red Sea, as documented by Kleinhaus et al. [ 137 ]. For instance, Rich et al. [ 138 ] reported a winter bleaching event in the central Red Sea in January 2020 due to sea surface temperatures (SSTs) falling below 18 °C. Additionally, inshore bleaching events in the central Arabian Red Sea were observed during the “3rd global coral bleaching event” in 2015, as reported by Monroe et al. [ 139 ].
The Eastern Tropical Pacific (ETP) comprises the ocean basin extending from the Gulf of California in México to Peru and includes areas of the continental shelf and offshore islands (Coco Island, the Galápagos Islands, the Revillagigedo Archipelago and Clipperton Atoll). The most cited article of the cluster “Eastern Pacific” is Spencer [ 140 ], which discussed potentialities, uncertainties and complexities in the response of coral reefs to future sea-level rise of reef islands in the Pacific Ocean and the Caribbean Sea. Throughout the Holocene, sea levels rose without being stabilized, and reefs in the Caribbean grew in tandem with these elevation changes [ 141 ]. The once structurally complex coral reefs in the Caribbean have suffered a dramatic decline since the 1970s, with only a minority of reefs maintaining a mean live coral cover of 10% or more [ 142 ]. A strong hurricane season brought on by unusually warm waters in the tropical Atlantic, and the Caribbean in 2005 caused the worst bleaching event ever observed in the basin [ 143 ]. There was a 60% decline in coral cover on reefs in the US Virgin Islands due to the fact of a severe disease outbreak brought on by the 2005 bleaching events in the Caribbean region, as reported by Miller et al. [ 144 ].
The Great Barrier Reef is the largest coral reef ecosystem, with over 348,000 km 2 of coverage consisting of 2900 individual reefs and 900 islands stretching over 2300 kilometers [ 145 ]. Three major coral bleaching events within a span of five years (2016, 2017, and 2020) along with the effects of severe tropical cyclones, poor water quality from catchment run-off, population growth and urbanization, overexploitation of marine resources, and habitat loss have all been the factors towards the degradation of coral reefs in the Great Barrier Reefs [ 56 , 146 ]. Cluster #4, which is the fifth largest cluster, contains 176 publications and has a silhouette value of 0.9. This cluster is strongly associated with cluster #6, “symbiotic dinoflagellate,” and cluster #12, “coral disease”. The high silhouette value of 1.0 indicates that there is a focused field of study in the context of coral reefs and climate change. The most cited article in cluster #4 is by Reaser et al. [ 147 ] on scientific findings and policy recommendations for coral bleaching and global climate change. Coral bleaching, which is the ability of animals with a symbiotic relationship with Symbiodinium to turn white, is an important issue associated with climate change-based literature. According to Douglas [ 148 ], all animals that have a symbiotic association with the dinoflagellate algae of the Symbiodinium genus, which are also referred to as zooxanthellae, have the ability to undergo bleaching. Symbiodinium have been reported to form extracellular symbioses with giant clams and intracellular symbioses with various organisms, including corals, anemones, jellyfish, nudibranchs, ciliophora, foraminifera, zoanthids, and sponges [ 149 , 150 ].
Fujise et al. [ 151 ] reported that the expulsion mechanisms of Symbiodinium were temperature-dependent; however, under non-thermal stress conditions, the expulsions of this algae were part of a regulatory mechanism to maintain a constant Symbiodinium density. In response to moderate thermal stress, Symbiodinium becomes damaged, and corals either selectively digest or expel the damaged cells. During extended periods of thermal stress, damaged Symbiodinium may accumulate in coral tissues, resulting in coral bleaching. Multiple factors have been shown to cause bleaching, including high oxidative stress [ 152 ], intense light [ 153 ], high temperature [ 154 ], low salinity [ 155 ], sedimentation [ 156 ], pollutants [ 157 ], decreased seawater temperature [ 158 ], diseases [ 159 ], bacterial infection [ 160 , 161 ], and ENSO-related marine heatwave events [ 162 , 163 ]. Degree heating weeks (DHW), defined as 1°C above the long-term climate level for the warmest month at a given locality, have become a common global predictor of bleaching [ 164 ]. Severe bleaching is typical at 8 DHW and above [ 165 , 166 ]. A global analysis report of coral bleaching from 1998 to 2017 [ 166 ] found that coral bleaching was most prevalent in regions with high-intensity and high-frequency thermal-stress anomalies. In areas where sea-surface temperature (SST) anomalies varied greatly, such as the Gulf of Aqaba region [ 167 ], the Caribbean Sea [ 168 ], and the Indo-Pacific [ 169 ], coral communities were significantly less susceptible to coral bleaching [ 166 ].
Globally, coral reefs have been threatened by coral disease, which is now recognized as one of the biggest threats to these ecosystems [ 170 ]. Similar to bleaching, coral disease was not considered a severe threat to coral reefs until recently [ 170 ], despite its first documentation in 1965 [ 171 ]. Since their initial descriptions, both the variety of coral diseases and the number of reported cases have skyrocketed [ 172 , 173 ]. Approximately 76% of all coral diseases described worldwide are found within this relatively small basin, leading experts to label the Caribbean a “hot spot” for disease [ 174 ]. For example, two dominant Acropora species in the Caribbean have been replaced by low-encrusting Agaricia due to the fact of coral disease [ 175 , 176 ]. Common coral diseases include Black band disease, which is caused by increased seawater temperature and anthropogenic factors, ciliates cause the Brown band disease, Cyanobacteria cause the Red band disease, and the White plague is caused by a bacterial infection [ 177 ].
The timeline for the document co-citation analysis is an important indicator to explain the period when the study got the attention of the researcher worldwide ( Figure 8 ). From 2010 to 2021, there have been bursts in citations for research clusters on (#0) “coral reefs”, (#2) “ocean acidification”, (#3) “central sea”, (#11) “sea level rise”, and (#5) “eastern pacific”. When taken as a whole, these studies shed light on the growing interest in studying the effects of ocean acidification and sea level rise on coral reefs, with particular attention paid to the plight of these ecosystems within the eastern pacific area, such as in the central Red Sea and the Dominican Republic.
A timeline co-citation analysis. Nodes represent references, whereas lines represent connections between those references. Larger nodes indicate higher frequencies of citations. References with strong citation bursts are shown as red circles, whereas references with high centrality are shown as yellow circles. Longer line segments indicate longer time spans.
CiteSpace’s visualized analysis of 7743 publications yielded a co-citation network (frequency of two different documents are cited together in other documents) with 2525 cited documents (nodes) and 5440 links or connections indicating co-citations between nodes [ 178 ]. The larger the node, the more often a document is cited, demonstrating its impact on coral reef and climate change research. Document co-citation analysis locates essential literature. The given references (in the article) were the most cited among 7743 Web of Science references. Table 4 presents the twenty most cited references based on co-citation analysis along with their frequency, burstiness, and centrality indices. An increase in citations reflects increased interest in that topic. “Citation bursts” demonstrate correlations between publications and sudden increases in citations. When comparing clusters, the centrality index indicates how well they are connected (i.e., coral reef and climate change). An elevated centrality score indicates that the publication is located between two or more sizable subclusters [ 179 ]. Dr. Terry Hughes’s research was widely cited, with three of his Nature and one of his Science publications ranking among the top five most-cited references. The publication entitled Global warming and recurrent mass bleaching of corals topped the list with a frequency index of 546, a burst index of 157 and a centrality index of 0.7. The burst in citation period of this publication was from 2018 to 2021. The findings were based on the third global-scale pan-tropical coral bleaching episode that occurred between 2015 and 2016. The reef ecosystem of eastern and western Australia was studied using aerial and underwater surveys along with sea surface temperatures obtained from satellites. According to their findings, the devastating bleaching event in 2016 was only slightly impacted by water quality and fishing pressure, indicating that local reef protection offers little to no protection against extreme heat.
The second paper on the list, with a frequency index of 359 and burst index of 125, was a global study analyzing the bleaching records of 100 globally distributed reefs from 1980 to 2016 [ 28 ]. According to their findings, mass coral bleaching events happen every year regardless of the presence or absence of El Nino. They forecast that the intervals between recurrent events will eventually become too short to permit a complete recovery of mature coral assemblages, typically taking 10 to 15 years to reach the fastest-growing species. They warned that if temperatures rise by 1.5 or 2 degrees Celsius above preindustrial levels, it will exacerbate the already severe decline of coral reefs around the world. Similar findings were found in another study of his that also appeared on the list of the most-cited research. Research into the effects of climate change on coral reef ecosystems, with a special emphasis on the Great Barrier Reef, ranked fifth [ 28 ]. They found that the Great Barrier Reef’s 2016 record-breaking heatwave had caused widespread loss of functionally diverse corals across the reef’s most remote and pristine regions. Ranked third on the list was the study by Hoegh-Guldberg et al. [ 22 ], which investigated the effects of climate change and ocean acidification on coral reefs. This study was closely linked to the 3rd (#2) research cluster, also known as “ocean acidification”. The research review presented future scenarios for coral reefs, which suggested increasingly detrimental impacts on various sectors, including tourism, coastal protection, and the fisheries industry. These predictions were based on the assumption that global temperatures would rise by at least 2 °C between 2050 and 2100, coupled with atmospheric carbon dioxide concentrations exceeding 500 ppm. The findings of this study emphasize the urgent need for effective measures to mitigate climate change and ocean acidification to ensure the long-term survival and sustainability of coral reefs and the associated ecosystems. The article by LaJeunesse et al. [ 180 ] on coral endosymbionts has garnered significant attention, with a citation frequency of 134, a burst index of 68, and a burst period spanning from 2013 to 2017. This publication is associated with cluster (3), also known as “symbiotic dinoflagellate”. The article describes Symbiodinium clades and proposes that the divergent evolutionary Symbiodinium “clades” correspond to genera within the Symbiodiniaceae family. The study affirms that the long evolutionary history of the Symbiodiniaceae family is appropriately acknowledged within the suggested framework. The findings of this study provide valuable insights into the evolutionary relationships and ecological functions of these endosymbionts, highlighting the critical role they play in the health and survival of coral reefs.
The list of 11 to 20 top-cited articles on climate change on coral reefs cover a wide range of topics, including ocean acidification, declining coral cover, Symbiodinium diversity, and coral reef resilience. Several studies indicate that warming trends and bleaching stress are increasing, and coral bleaching protection mechanisms are becoming less effective, ultimately leading to significant declines in coral populations. Research on the impacts of ocean acidification and warming on marine organisms, as well as the interactions between these factors, has also shed light on the mechanisms underlying the sensitivity of coral reefs to climate change [ 79 , 181 , 182 ]. Studies on the diversity, distribution and stability of Symbiodinium [ 183 ] have provided insights into the potential for coral resilience, while research on the decline of coral cover in the Indo-Pacific region has highlighted the extent and timing of this phenomenon [ 183 , 184 ]. These studies demonstrate that collaborative research efforts are essential to understanding the impacts of climate change on coral reefs and developing more efficient conservation and management strategies.
Using the co-cited keyword analysis performed in CiteSpace, 963 unique keywords were generated. In order to better understand the connections between these terms, a clustering tool was used to categorize them into groups ( Figure 9 ). This generated seven major clusters consisting of “sea surface temperature”, “Symbiodinium”, “coral reef fish”, “marine protected area”, “water quality”, “ocean acidification”, and “hydrocorals”. Each of these clusters can be analyzed independently to determine which descriptors are most applicable. The major keywords used to discuss “sea surface temperature”—record, Indian ocean, and reef; “Symbiodinium”—scleractinian coral, diversity, zooxanthellae, population, nutrient enrichment, elevated pressure, and oxidative stress; “coral reef fish”—phase shift, ecosystem, disturbance, fish, dynamics, community, recruitment, abundance, thermal tolerance, Stylophora pistillata , and ecology; “marine protected area”—management, assemblage, biodiversity, susceptibility, degradation, adaptation, response, and recovery; “water quality”—sea level, Great barrier reef, rate, transport, coral bleaching, French Polynesia, and fringing reef; “ocean acidification”—climate, El Niño, impact temperature, coral reef, calcification, and carbon; and “hydrocorals”—seawater and carbon dioxide.
Distribution of co-cited keywords in the field of coral reef and climate change.
Table 5 displays the top 10 keywords with the strongest citation burst. With the exception of “ocean warming,” all the most frequently cited keywords emerged in the early 1990s and experienced a citation burst that extended until the late 2000s. The keywords identified were “French Polynesia,” “record,” “El Niño,” “Australia,” “Indian Ocean,” “Continental,” “Shelf,” “Sea level,” “Sea surface,” “temperature,” and “Island.” Notably, the keyword “ocean warming” only gained popularity in 2017, with citations peaking from 2018 to 2021. This demonstrates the significance of research on climate-related temperatures in the field of coral reefs and climate change. The keywords “Australia” and “Continental shelf” demonstrated citation bursts lasting over 15 years. In contrast, “French Polynesia” had the highest frequency of citations during a relatively shorter period, commencing in 1992 and concluding in 2006. French Polynesia, situated in the westernmost region of the South Pacific, comprises 118 islands and atolls, classified into five main clusters: the Marquesas, Society, Tuamotu, Gambier, and Austral islands [ 191 ]. These regions exhibit a north–south gradient for variables such as sea surface temperature (SST), solar insolation, evaporation, and humidity. The Millennium Coral Reef Mapping Project (MCRMP) has successfully mapped the Austral, Gambier, Society, and Tuamotu islands and atolls; however, significant research remains to be undertaken in this extensive region, which accounts for the enduring citation burst for this keyword. The findings of this study highlight the scientific interest and importance of French Polynesia as a unique and diverse region for further research and conservation efforts.
Top 10 keywords with the strongest citation bursts.
Keyword | Year | Strength | Begin | End | 1976–2021 |
---|---|---|---|---|---|
French Polynesia | 1992 | 24.51 | 1992 | 2006 | ▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▃▃▃▃▃▃▃▃▃▃▃▃▃▃▃▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂ |
Record | 1992 | 24.36 | 1992 | 2005 | ▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▃▃▃▃▃▃▃▃▃▃▃▃▃▃▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂ |
El Niño | 1995 | 22.08 | 1995 | 2006 | ▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▃▃▃▃▃▃▃▃▃▃▃▃▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂ |
Australia | 1992 | 21.33 | 1992 | 2007 | ▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▃▃▃▃▃▃▃▃▃▃▃▃▃▃▃▃▂▂▂▂▂▂▂▂▂▂▂▂▂▂ |
Indian Ocean | 1990 | 17.72 | 1998 | 2010 | ▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▃▃▃▃▃▃▃▃▃▃▃▃▃▂▂▂▂▂▂▂▂▂▂▂ |
Continental shelf | 1993 | 17.14 | 1993 | 2010 | ▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▃▃▃▃▃▃▃▃▃▃▃▃▃▃▃▃▃▃▂▂▂▂▂▂▂▂▂▂▂ |
Sea level | 1991 | 16.43 | 1991 | 2005 | ▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▃▃▃▃▃▃▃▃▃▃▃▃▃▃▃▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂ |
Sea surface temperature | 1996 | 15.91 | 1996 | 2006 | ▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▃▃▃▃▃▃▃▃▃▃▃▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂ |
Ocean warming | 2017 | 15.09 | 2018 | 2021 | ▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▃▃▃▃ |
Island | 1991 | 14.65 | 1991 | 2005 | ▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▃▃▃▃▃▃▃▃▃▃▃▃▃▃▃▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂ |
Figure 10 illustrates the dual-map overlay of the number of articles pertaining to the type or focus of the journal. The map labels represent the research subjects covered by the journals, with the citing journals displayed on the left side and the cited journals on the right. The trajectory of the citation links provides valuable insights into inter-specialty relationships. A shift in trajectory from one region to another would indicate the influence of articles from another discipline on a specific field. In the domain of coral reef and climate change interaction, the dominant fields were found to be “ecology, earth, and marine”. The most influential discipline was “plant, ecology, and zoology”, with a z-score of 7.66, followed by “earth, geology, and geophysics”, with a z-score of 4.99 and, lastly, “molecular, biology, and genetics”, with a z-score of 2.80. These findings provide a valuable understanding of the interdisciplinary relationships within the field of coral reef and climate change research, highlighting the influence of various disciplines in shaping the current research landscape.
Dual-map overlay on the impact of climate change on coral reefs research.
Coral reefs are a vital marine ecosystem service, providing high biodiversity and supporting the livelihoods of coastal communities. However, ocean warming and temperature are the largest threats to corals from anthropogenic climate change [ 192 ]. Between 1997 and 2018, the global average percentage of coral cover was approximately 32%, but by 2100, RCP 8.5 predicts a global decline in coral cover of 5 and 15%, equating to a relative global decline of more than 40% [ 115 , 193 ]. This decline is due to the fact that sea surface temperatures (SSTs) are projected to increase by more than 3 °C by the turn of the century [ 194 ]. These declines could have significant ecological and socioeconomic impacts, particularly in coastal communities that rely on coral reefs for food, tourism, and other ecosystem services.
For example, The Republic of Palau, a small Micronesian nation, has already experienced significant losses in coral reef cover [ 195 ]. Over 87% of Palau’s households are linked to coral reef-associated activities, which are critical to the country’s economic and social well-being. While tourism, particularly ecotourism, is a significant contributor to GDP growth, tax revenue, and employment, climate change-related stressors have caused a steady decline in coral reef cover. This decline has indirectly caused a major decline in tourism, threatening the country’s economic sustainability [ 196 ]. According to Barnett [ 197 ], climate change is a significant threat to food security for people in Pacific SIDS, primarily due to the decline in fisheries output resulting from the impact of climate change on total coral cover.
Apart from impacting the socioecological structure, the impact of climate change can have cascading effects on the entire reef ecosystem, affecting the abundance and diversity of other marine species that depend on corals for food and shelter. Up to 14% of species may be in imminent danger of extinction at a warming of 1.5 °C and up to 29% at a warming of 3 °C. This rise in ocean temperature will probably force coral to colonize higher latitudes that currently lack reefs [ 198 , 199 , 200 ]. However, various factors, including the need for a suitable substrate [ 201 ], connectivity to other reefs [ 202 ], ocean acidification [ 203 ], and light intensity [ 204 ], may outweigh the advantages of reefs as they expand to high latitudes [ 193 ].
Through the timeline co-citation analysis, we have observed a significant increase in research interest in the topic of climate change impacts on coral reefs between 2010 and 2022. The analysis identified several research clusters that gained traction in the scientific community, including those related to “coral reefs,” “ocean acidification,” “central red sea”, “Great Barrier Reef”, and “sea level rise”. These clusters have evolved to become research hotspots under the overarching topic of climate change impacts on coral reefs. For example, the research clusters related to the central red sea and the Great Barrier Reef have emerged as prominent research areas, given their unique characteristics and ecological importance. Similarly, the impact of ocean acidification and sea level rise on coral reefs has gained significant research interest, given their severe consequences on the health and survival of coral reefs. To better understand the impact of these research clusters, our overall discussions have been designed to incorporate the subtopics “climate change threats to coral reefs” and “adaptive strategies for coral resistance and resilience”.
Climate-induced changes in temperature are a major threat to coral reef ecosystems, and extensive research has highlighted several key areas for investigation [ 53 , 205 ], with marine heatwaves, solar radiation, heat tolerance, and thermal thresholds representing the most promising areas for future research. Marine heat waves have become increasingly prevalent and intense as a result of climate change. These extreme events, characterized by prolonged periods of elevated water temperatures, significantly impact coral reef ecosystems. For instance, the mass global coral bleaching event of 2016–2017 was the most extensive and long lasting on record, as documented by Eakin et al. [ 206 ]. The event, which was associated with the El Niño Southern Oscillation (ENSO), had varying impacts on coral reefs worldwide [ 207 ], with some regions experiencing more severe bleaching than others, as reported by Kim et al. [ 208 ].
Corals are thermophilic, but their thermal tolerance is narrowly defined [ 169 , 209 ]. For instance, the rate of calcification increases with temperature up to a threshold level, beyond which it declines [ 210 , 211 , 212 ]. Tropical corals live close to their upper thermal limits and are, therefore, highly sensitive to periods of elevated sea surface temperatures and ocean warming [ 187 , 213 ]. Coral reefs in the Persian Gulf have been observed to have the highest upper-temperature thresholds of approximately 35–36 °C [ 214 ]. However, it has also been noted that these corals remain highly vulnerable to thermal stress when temperatures surpass their local maximum summer temperatures [ 215 ]. The escalating frequency and gravity of thermally induced mass bleaching events have sparked worldwide attention to the elevated temperature impacts on corals [ 28 ]. As a result, research endeavors have focused on establishing maximum thermal tolerance thresholds and variations in diverse coral species and regions and exploring potential coral refugia to brace for future ocean warming [ 216 ].
Corals rely on their symbiotic relationship with unicellular algae of the genus Symbiodinium for photosynthesis, and over 90% of their energy budget is needed for essential functions, such as calcification, tissue growth, and reproduction [ 212 ]. This critical association is threatened when corals experience thermal stress, such as elevated sea surface temperatures (SST), resulting in coral bleaching, where the algal endosymbionts are expelled. The resulting impairment and expulsion of the algal symbionts are linked to reactive oxygen species (ROS) generation from the host, the algal symbiont, or both, triggering a host immune response [ 217 ].
Protracted coral bleaching can lead to extensive coral mortality, severely affecting the ecosystem and associated reef fauna. Based on the timeline cocitation analysis, it was evident that the Red Sea (Cluster #3) and Great Barrier Reef (GBR) (cluster #8) are major research hotspots in terms of geographic regions. Although the Persian Gulf is a hot sea that supports coral reef ecosystems, the Red Sea harbors corals with greater thermal stress tolerance, with some coral genotypes capable of surviving temperatures over 5 °C above their summer maxima [ 216 , 218 ]. Corals in the southern end of the Red Sea are more heat resistant, surviving prolonged high temperatures, while the northern Red Sea benefits from heat-resistant genotypes that have migrated from the south [ 219 ]. The importance of broad latitudinal temperature gradients in promoting adaptation to high temperatures and exchanging heat-resistant genotypes across latitudes for genetic rescue in coral reefs is exemplified in the evolutionary history of coral reefs in the northern Red Sea [ 9 , 216 ]. On the other hand, the GBR, known as the world’s largest coral ecosystem, was severely impacted by the 2015–2016 climate change-amplified strong El Niño event that triggered the warmest temperatures on record. This resulted in a massive bleaching event affecting nearly 90% of reefs along the northern region, leading to a loss of approximately 30% of live coral cover in the following six months [ 28 , 220 , 221 ]. Research has increasingly linked climate change to a rise in coral diseases. Bruno et al. [ 222 ] used a high-resolution satellite dataset to investigate the relationship between temperature anomalies and coral disease on a large spatial scale of 1500 km in Australia’s Great Barrier Reef. Their findings showed a significant positive correlation between warm temperature anomalies and the incidence of the white syndrome, an emergent disease in Pacific reef-building corals. In a similar vein, Tignat-Perrier et al. [ 223 ] noted a decline in populations of two gorgonian species ( Paramuricea clavata and Eunicella cavolini ) found in the Mediterranean Sea due to the fact of microbial diseases during thermal stress events. These studies illustrate the growing concern that climate change is contributing to the increased incidence and severity of coral diseases, which could ultimately lead to a decline in the health of marine ecosystems.
In the past, studies on the impact of climate change on coral reefs primarily centered on the thermal tolerance of corals and the consequences of massive, abrupt coral loss on organisms associated with reefs [ 224 ]. However, research has recently shifted towards investigating the distinct and synergistic effects of ocean warming and ocean acidification resulting from increased atmospheric CO 2 levels. The timeline co-citation analysis reveals that these emerging research fields are highly significant with recent citation bursts, as evidenced by their identification as Cluster #2 (Ocean acidification) and Cluster #10 (Elevated CO 2 ), respectively.
The escalation of atmospheric carbon dioxide (CO 2 ) concentrations has resulted in ocean acidification, which is among the foremost threats to coral reef ecosystems. Forecasts for 2100 anticipate a rise in CO 2 concentrations to between 540 and 970 ppm, leading to a global decrease in seawater pH by 0.14 to 0.35 units [ 31 , 68 , 116 , 225 ]. As demonstrated by Fabricius et al. [ 68 ], ecological traits of coral reefs will gradually transform as seawater pH decreases to 7.8, and a decline below this level (at 750 ppm pCO 2 ) would be catastrophic for these ecosystems. Ocean acidification reduces the availability of carbonate ions that corals require to form their calcium carbonate skeletons, ultimately leading to a decrease in coral calcification rates [ 33 ]. Ocean acidification has also been shown to decrease the ability of coral larvae to settle and survive [ 226 ] and increase their susceptibility to disease [ 227 ]. Research has shown that even modest increases in ocean acidity can impact the physiological processes of corals. For example, exposure to high levels of CO 2 reduces coral growth and calcification rates [ 68 , 226 ]. In addition to the direct effects on coral physiology, ocean acidification can have cascading impacts on the entire coral reef ecosystem. For instance, reduced calcification by corals can reduce the complexity of the coral reef structure, potentially leading to the loss of important habitats for fish and other marine organisms [ 228 ]. Furthermore, ocean acidification can impact the symbiotic relationship between corals and their algal symbionts, potentially leading to a decline in the productivity of the reef ecosystem as a whole [ 229 ]. The combination of ocean warming and acidification is particularly concerning, as they act synergistically to exacerbate the negative impacts on coral reef ecosystems [ 22 ]. With continuing increases in atmospheric CO 2 levels, the effects of ocean acidification on coral reefs are expected to become even more pronounced, highlighting the need for urgent action to reduce greenhouse gas emissions and protect these valuable and vulnerable ecosystems.
The rate of atmospheric CO 2 increase continues to accelerate, with emission scenarios predicting CO 2 concentrations of 540–970 ppm and a decline in seawater pH by 0.14–0.35 units globally for 2100 [ 68 , 225 ]. Fabricius et al. [ 68 ] demonstrated that many ecological properties in coral reefs will gradually change as pH declines to 7.8 and that it would be catastrophic for coral reefs if seawater pH dropped below 7.8 (at 750 ppm pCO 2 ).
Coral resistance and resilience are scientific constructs that pertain to the capacity of coral reefs to withstand and recuperate from various stressors. Coral resistance is defined as the ability of corals to endure or tolerate perturbations and stressors, such as variations in water temperature, ocean acidification, pollution, and physical injury. Corals that possess a greater resistance to these stressors exhibit a greater ability to sustain their structure and function despite disturbances and are less prone to suffering from coral bleaching, disease, or mortality [ 229 , 230 ]. A myriad of studies has reported on the bleaching thresholds of corals inhabiting the Persian Gulf, despite conditions at least 2 °C higher than other coral reef ecosystems worldwide [ 231 ]. Additionally, corals from the Indo-Pacific and Caribbean regions have been found to maintain calcification rates even in low aragonite saturation states, present in naturally acidified locales [ 68 , 232 ]. The eastern Pacific region of Palau has revealed the thriving of reefs in waters with natural acidification, resulting from biological processes and reef system circulation patterns [ 232 , 233 ]. However, it is noteworthy that coral communities in Palau’s relatively acidic reef zones developed over thousands of years, fostering an inherent resistance that differs from coral communities in regions affected by higher anthropogenic interventions.
Coral resilience, in contrast, refers to the ability of coral reefs to recover from disturbances and stressors. Corals that exhibit higher resilience can reproduce, regenerate, and rebuild their structural complexity after experiencing bleaching [ 234 ]. These mechanisms are attributable to genetic diversity within coral populations and their symbiotic association with Symbiodinium algae, which are critical to their health and survival [ 235 , 236 ]. Genetic adaptation in corals is mediated through various factors, including the activation of heat-shock proteins, oxidoreductase enzymes, and microsporine-like amino acids. The coral surface micro-layer that absorbs UV radiation has also been identified as a significant mechanism for adaptation [ 180 , 237 , 238 ]. In-depth research on corals that thrive in the warm waters of the Persian Gulf has demonstrated their capacity for resilience, attributable to metabolic trade-offs, unique physiological characteristics, and specific genetic signatures, including a heat-specialist algal endosymbiont, Symbiodinium thermophilum [ 236 , 239 ]. S. thermophilum can thrive in high-temperature and high-salinity environments, allowing the coral to develop a temperature-stress-resistant phenotype [ 239 ].
Symbiodinium, a diverse group of dinoflagellates, is classified into nine clades (A–I) based on their phylogenetic characteristics [ 240 ]. Among these clades, Symbiodinium clade D has garnered attention for its exceptional thermal resilience ability, despite its relatively low representation (less than 10%) in the endosymbiotic community of coral hosts [ 241 ]. Various coral species, including fast-growing branching types, such as Acropora, Stylophora, and Pocillopora, as well as slow-growing massive, encrusting, and solitary corals, have been associated with Symbiodinium clade D [ 242 ]. The prevalence of clade D Symbiodinium in corals from the Persian Gulf has been linked to their higher thermal tolerance, particularly in comparison to corals associated with clade C, which is the dominant lineage in corals from the Great Barrier Reef and other Pacific coral reef ecosystems [ 243 ], and clade B in corals from the Atlantic [ 244 ]. These findings highlight the significance of Symbiodinium diversity in understanding the thermal resilience of coral reefs and the potential mechanisms underlying their adaptation to changing environmental conditions.
McCulloch et al. [ 234 ] explored the ability of coral species to withstand the adverse impacts of ocean acidification and global warming on coral reefs. Their study revealed that some coral species (i.e., Stylophora pistillata and Porites spp.) exhibit the capacity to increase pH levels within their calcifying fluid, crucial for the deposition of calcium carbonate and maintenance of the coral structure, even in the face of declining seawater pH levels. The study demonstrated the significance of acid-base regulation mechanisms for corals’ resilience to the effects of ocean acidification, allowing them to maintain or increase their calcification rates despite rising ocean acidification. Moreover, the study indicated that corals could acclimate to extended acidification, which enables them to maintain or increase their calcification rates by upregulating their internal pH levels, thus providing insight into potential strategies for mitigating the effects of climate change on coral reefs. A similar adaptation resilience strategy against ocean acidification was observed in cold-water scleractinian corals (i.e., Caryophyllia smithii , Desmophyllum dianthus , Enallopsammia rostrata , Lophelia pertusa , and Madrepora oculate ) [ 245 ].
Oceanographic processes, such as upwelling and tidal currents, also play a significant role in helping corals avoid bleaching. In areas where upwelling events mix deeper, cooler water with shallow warmer water, thermal stress is reduced [ 246 , 247 ]; for example, in northern Galapagos during the 2015/16 ENSO [ 248 ] and Nanwan Bay, southern Taiwan, during summer [ 249 ]. Similarly, a coral reef’s ability to resist bleaching is bolstered by the elimination of potentially damaging oxygen radicals due to the swift water flow associated with tidal currents [ 230 , 250 , 251 ].
Therefore, in summary, the scientific community has identified various adaptive strategies that could enhance the resilience and resistance of coral reefs to these challenges. Going forward, it is crucial to continue ongoing research efforts to better understand the mechanisms underlying coral resilience and resistance, identify research gaps, and develop new management strategies for protecting these vital ecosystems. This can be achieved through a multidisciplinary approach that combines laboratory-based experimentation, field research, and community engagement. In addition, collaborations between the scientific community and policymakers can facilitate the implementation of evidence-based management practices that promote the resilience and resistance of coral reefs to climate change and other stressors.
The scientometric analysis that is presented in this article demonstrates that research on coral reefs in relation to climate change has emerged as one of the potential fields, with interest in this topic has grown steadily since the 2000s. The increasing global temperatures are posing a significant threat to coral reefs, leading to widespread coral bleaching and mortality. Moreover, changes in ocean chemistry brought on by an increase in carbon dioxide levels lead to ocean acidification, which can worsen the effects of rising temperatures on corals [ 22 ]. In addition, sea level rise and coastal development are transforming the physical structure of coral reef ecosystems, exacerbating the negative effects of the other stressors [ 252 ]. These changes are harmful not only to the coral reefs but also to the plethora of species that rely on them for survival and the communities that rely on them for livelihoods and for protecting the coast. Future challenges for developing countries like those within the coral reef triangle initiative (i.e., Indonesia, Malaysia, the Philippines, Papua New Guinea, Timor Leste, and the Solomon Islands) will center on access to funding for conservation-restoration efforts and continued monitoring studies. There are several ways that ongoing research and coordinated action can help coral reefs cope with the effects of climate change:
The present study was supported by the Department of Higher Education, Ministry of Higher Education Malaysia, under the LRGS program (LRGS/1/2020/UMT/01/1; LRGS UMT Vot No. 56040) entitled “Ocean Climate Change: Potential Risk, Impact and Adaptation Towards Marine and Coastal Ecosystem Services in Malaysia’. The work was also supported by PASIFIC program GeoReco project funding from the European Union’s Horizon 2020 Research and Innovation programme, under the Marie Sklodowska-Curie grant agreement No. 847639, and from the Ministry of Education and Science.
Conceptualization, C.S.T. and M.N.A.; methodology, M.N.A.; software, C.S.T.; validation, J.B., Z.V.-G. and V.R.; formal analysis, F.L.; investigation, G.S.; resources, I.G.; data curation, J.B.; writing—original draft preparation, C.S.T.; writing—review and editing, M.N.A.; visualization, F.L., G.S., I.G., V.R. and Z.V.-G.; supervision, M.N.A.; project administration, G.S.; funding acquisition, J.B., Z.V.-G., V.R. and I.G. All authors have read and agreed to the published version of the manuscript.
Not applicable.
Data availability statement, conflicts of interest.
The authors declare no conflict of interest.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
Click through the PLOS taxonomy to find articles in your field.
For more information about PLOS Subject Areas, click here .
Loading metrics
Open Access
Peer-reviewed
Research Article
Roles Conceptualization, Formal analysis, Investigation, Methodology, Validation, Visualization, Writing – original draft, Writing – review & editing
Affiliation Department of Biology, Dalhousie University, Halifax, Nova Scotia, Canada
Roles Conceptualization, Funding acquisition, Methodology, Project administration, Resources, Supervision, Validation, Writing – review & editing
* E-mail: [email protected]
Affiliation Department of Oceanography, Dalhousie University, Halifax, Nova Scotia, Canada
Artificial reefs (ARs) have been used on coral reefs for ecological research, conservation, and socio-cultural purposes since the 1980s. We examined spatio-temporal patterns in AR deployment in tropical and subtropical coral reefs (up to 35° latitude) and evaluated their efficacy in meeting conservation objectives, using a systematic review of the scientific literature. Most deployments (136 studies) were in the North Atlantic and Central Indo-Pacific in 1980s – 2000s, with a pronounced shift to the Western Indo-Pacific in 2010s. Use of ARs in reef restoration or stressor mitigation increased markedly in response to accelerating coral decline over the last 2 decades. Studies that evaluated success in meeting conservation objectives (n = 51) commonly reported increasing fish abundance (55%), enhancing habitat quantity (31%) or coral cover (27%), and conserving target species (24%). Other objectives included stressor mitigation (22%), provision of coral nursery habitat (14%) or source populations (2%) and addressing socio-cultural and economic values (16%). Fish (55% of studies) and coral (53%) were the most commonly monitored taxa. Success in achieving conservation objectives was reported in 33 studies. Success rates were highest for provision of nursery habitat and increasing coral cover (each 71%). Increasing fish abundance or habitat quantity, mitigating environmental impacts, and attaining socio-cultural objectives were moderately successful (60–64%); conservation of target species was the least successful (42%). Failure in achieving objectives commonly was attributed to poor AR design or disruption by large-scale bleaching events. The scale of ARs generally was too small (m 2 –10s m 2 ) to address regional losses in coral cover, and study duration too short (< 5 years) to adequately assess ecologically relevant trends in coral cover and community composition. ARs are mostly likely to aid in reef conservation and restoration by providing nursery habitat for target species or recruitment substrate for corals and other organisms. Promoting local socio-cultural values also has potential for regional or global impact by increasing awareness of coral reef decline, if prioritized and properly monitored.
Citation: Higgins E, Metaxas A, Scheibling RE (2022) A systematic review of artificial reefs as platforms for coral reef research and conservation. PLoS ONE 17(1): e0261964. https://doi.org/10.1371/journal.pone.0261964
Editor: Maura (Gee) Geraldine Chapman, University of Sydney, AUSTRALIA
Received: September 10, 2021; Accepted: December 14, 2021; Published: January 21, 2022
Copyright: © 2022 Higgins et al. This is an open access article distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the paper and its Supporting Information files.
Funding: Funding for this study was provided by the Natural Sciences and Engineering Research Council of Canada Discovery Grants program (RGPIN-2016-04878 to AM and RGPIN-2016-04878 to RES). EH was partially supported by scholarships from the Natural Sciences and Engineering Research Council of Canada CGS Masters, Nova Scotia Research and Innovation Scholarship, and the Faculty of Graduate Studies at Dalhousie. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
The global cover of scleractinian corals has declined dramatically since 1985 due to synergistic effects of increased ocean temperatures and acidification, predation, biological invasions, mechanical damage, and disease [ 1 , 2 ]. The increasing frequency and intensity of natural and anthropogenic stressors has altered coral reefs, contributing to large-scale phase shifts, in some regions, to alternative stable communities dominated by fleshy macroalgae [ 3 , 4 ], soft corals, corallimorpharia, or sponges [ 5 ]. It is estimated that more than 800 million people worldwide depend on coral reefs for food, coastal protection, and tourism [ 6 – 8 ], and that persistence of alternative stable states will cause a significant reduction in these ecosystem services [ 9 ].
Traditional conservation measures (e.g. no take-zones, reserves, and marine protected areas) have been used on coral reefs for decades [ 10 – 12 ], but attention has progressively shifted toward active restoration methods as a consequence of accelerating coral decline [ 13 , 14 ]. Ecological restoration is the process which assists the recovery of a degraded, damaged or destroyed ecosystem [ 15 ]. Since it may not be possible to remove the threat responsible for degradation or damage, the trajectory of recovery may allow adaptation to local and global changes [ 16 ]. The United Nations General Assembly recognized the pressing need to restore damaged ecosystems and proclaimed 2021–2030 to be the United Nations Decade on Ecosystem Restoration, with the primary goal being to prevent, halt and reverse the degradation of ecosystems worldwide. The United Nations Environment Assembly adopted a resolution that requested UNEP to specifically better define best practices for coral restoration [ 17 ]. Since the main threat to coral reefs is climate change [ 18 ], their restoration is likely most effective as a complementary tool in a larger management portfolio or as a temporary measure to minimize loss while global solutions are sought [ 17 , 19 ]. However, restoration of coral reefs has lagged behind and the spatial extent of restoration is the smallest compared to other major marine coastal ecosystems [ 20 ]. Thus, our knowledge on best practices for coral reef restoration is limited.
Motivations for coral reef restoration have ranged from ecological to cultural to legislative reasons, but experimental reasons appear to dominate [ 21 , 22 ]. Experimental approaches to active restoration include direct transplantation of corals, coral gardening, larval propagation, substrate manipulation, and substrate addition through the deployment of artificial reefs [ 17 , 19 ]. All approaches of active restoration have had certain shortcomings, such as short monitoring periods (average = 18 months) and small scales (< 100 m 2 ) and have often lacked objectives [ 19 ]. Of these, artificial reefs (ARs), although popular for fish enhancement, have not been used as extensively for coral restoration [ 19 ], possibly because of the logistics of deployment and, on average, an order of magnitude greater cost than other approaches [ 20 ]. ARs have been deployed in coral ecosystems globally to address various conservation objectives, including enhancing fish and invertebrate biomass [ 23 ], increasing habitat quantity and structural complexity of denuded reefs [ 24 , 25 ], conservation of target species [ 26 , 27 ], and as nursery habitat for transplantation initiatives [ 28 ]. Examining the objectives of artificial coral reefs, success in meeting these objectives, and assessing their potential benefits as a restoration strategy can inform management decisions in different regions and under projected climate scenarios. However, for management decisions to be effective, the benefits of AR must be quantified and the efficacy of the methodologies (e.g. AR type, size, distribution, deployment location and period) evaluated.
ARs deployed in different temperate and tropical ecosystems can provide benefits to both benthic and pelagic communities [ 29 ] by supplying additional hard substrate for settlement [ 30 ], reducing fishing and tourism pressure on natural reefs [ 31 ], increasing heterogeneity of natural substrata [ 32 , 33 ], and providing shelter from predators and human disturbances [ 34 , 35 ]. As with other active restoration approaches, clearly defined objectives for the deployment of ARs are not always provided [ 29 ], presenting challenges with monitoring their effectiveness. There is also concern that the scale of ARs is too small to have long-term impacts on conservation or restoration of target species and their functional relationships [ 36 ]. It has been argued that ARs can introduce alien materials onto reefs that may harm the recipient community by leaking toxic compounds [ 37 ] or by scouring natural reefs if detached during coastal storms [ 38 ]. Additionally, there is debate as to whether ARs act as a source or sink for fish and invertebrate populations [ 39 – 42 ].
To assess the functional importance of ARs, an understanding of the dynamics of established benthic communities and their relationship with demersal and pelagic species is imperative [ 35 ]. Deploying ARs for restoration of coral ecosystems specifically is a relatively new strategy, and most research to date has been largely descriptive [ 43 ], with few replicated comparisons to natural reefs [ 44 ]. For example, there is increasing evidence that fish and invertebrate assemblages on ARs deployed in coral ecosystems do not mimic those on natural reefs [ 45 – 47 ]; the role of ARs in colonization by reef invertebrates is unknown [ 35 ]. Long-term data on species’ residence time, growth and survival, and production patterns on adjacent natural coral reefs rarely are collected during studies of ARs [ 34 , 40 ].
Planning AR deployments in coral ecosystems with specific goals and objectives coupled with long-term monitoring plans can allow the assessment of conservation outcomes from these interventions [ 17 , 19 , 29 ]. Here, we present results of a systematic review of the scientific literature focussed specifically on the use of ARs as an active restoration strategy for coral ecosystems. In particular, we examine stated objectives of ARs over the past 100 years and across 8 marine realms, along with records of the spatial scale, monitored taxa, and study duration. For studies that recorded progress toward meeting conservation objectives, we evaluate and discuss the reported success, and identify factors that may limit the attainment of objectives. Based on our findings, we propose that among all prospective conservation objectives for artificial coral reefs, the provision of nursery habitats and additional hard substrate for colonization, and the promotion of local socio-cultural values are those most likely to achieve conservation success. However, given the limited evidence of setting conservation objectives specific to deployment, the large variation in size, spacing and monitoring effort, and the potential cost, much more research is needed to assess the use of ARs as a coral restoration strategy globally.
We conducted searches in ISI Web of Science Core Collection (1900–2020), Scopus ( https://www.scopus.com ), and Google Scholar ( https://scholar.google.ca ) for peer-reviewed publications that measured or monitored ecological and socio-cultural variables on ARs deployed in tropical and subtropical coral reef ecosystems (up to 35° latitude). In each database, we adapted the following general search terms to account for syntax differences: (TITLE-ABS-KEY ((artificial* OR “man-made” OR construct*) W/2 (coral* OR reef* OR habitat* OR nursery*)) AND TITLE-ABS-KEY (coral* OR tropic* OR subtropic*)). The first two sets of search terms were optimized to return studies that incorporate AR structures that both were designed purposefully and became de facto ARs. The last set narrowed the scope of the search to articles pertaining to ARs deployed in coral ecosystems. Studies on both vertebrate and invertebrate groups were included. Searches in all databases were completed on 31 December 2020. To ensure scientific rigour in the assessment of conservation objectives, we did not include the 1000s of studies from the grey literature, the validity of which had not been evaluated through peer review.
Over all databases, the search terms returned 4088 articles after duplicates were removed. All article citations and abstracts were imported into the web-based software review program Covidence ( https://www.covidence.org ), their titles and abstracts were screened, and 802 studies were extracted that included research on AR structures in coral reef ecosystems ( Fig 1 ). A full text review was conducted for 530 articles, and data were extracted from 136 that met one or more of the following secondary inclusion criteria: 1) included a date, precise location, and depth of deployment, 2) included the precise dimensions and number of ARs in the study, and 3) stated an objective of AR deployment.
https://doi.org/10.1371/journal.pone.0261964.g001
Articles were divided into two categories: 1) those that directly measured the success of meeting the objective(s) of ARs, and 2) those that were deployed for the purposes of scientific experiments or as de facto submergences (e.g. accidental ship groundings, dumping vehicles or building materials as waste). All 136 studies from both categories were surveyed for 1) duration of study, 2) clear description of AR dimensions, 3) targeted taxonomic groups, and 4) socio-cultural and ecological response variables used to assess whether the conservation objective(s) of the AR was being met. Latitude and longitude were extracted for each AR and then categorized into marine realms as defined in [ 48 ]. All 136 studies were used to examine spatio-temporal patterns of AR deployments as presented in the scientific literature. For the analyses of global AR abundance over time and to ensure representation of definitions used in the studies, we included all structures clearly defined as ARs by study authors and with a minimum area of ≥ 0.25 m 2 .
To ensure ecological relevance of conclusions about conservation success, studies reporting on progress towards attaining conservation objectives of deployed ARs fulfilled all secondary inclusion criteria listed above as well as two additional ones: (1) the monitored ARs were ≥ 1 m 2 to allow for comparison with natural reef formations and knolls; and (2) for studies reporting on multiple ARs, individual AR structures were defined as such if they were at least 2 m from the nearest adjacent AR. This spacing reflects what is considered an AR by study authors and the methods used to ensure connectivity of motile organisms and larvae between ARs. It has been shown that ARs > 2 m apart can form distinct benthic communities [ 49 ]. A total of 53 studies fulfilled all inclusion criteria and were used in this analysis of conservation objectives.
Studies monitoring the success of an AR towards achieving one or more conservation objectives were further sub-classified into 8 categories of objectives: increase fish abundance, increase coral cover, conservation of target species (i.e. reef species of significant ecological or socio-cultural importance), socio-cultural value (e.g. economic evaluation, attractiveness to divers or tourists), serving as a source population for recruitment to the surrounding ecosystems, nursery or coral garden, increase habitat quantity, and stressor mitigation (i.e. deployment following catastrophic events, such as bleaching, severe tropical storms, and dredging). The ecological response variables used to assess success in meeting the conservation objective(s) of ARs were categorized according to the measurements (abundance, diversity, cover, recruitment, biomass, size distributions, survival/mortality, growth and reproduction rates, species turnover, connectivity/space use, and structural complexity) and by broad taxonomic groups (fish, coral, other invertebrates, and algae).
Definitions of ar.
There is little standardization or agreement about the definition of AR in the scientific literature. Definitions within the studies examined in this review were disparate or absent. Authors reported on a vast array of structures, from de facto or accidental deployments to purposefully designed and deployed ARs. De facto or accidentally deployed ARs are wide ranging. Most are wrecks (or pieces of wrecks) of various numbers (15 in one case [ 50 ]), sizes and types of vessels; retired oil rigs [ 51 ], breakwaters and coastal jetties [ 52 ], and ropes in a tuna farm [ 53 ] were also considered ARs. Purposefully deployed ARs ranged from piles of rocks [ 54 ] or tires on the seafloor [ 55 ] to specifically engineered structures optimized for recruitment of target species for conservation, such as casitas or Autonomous Reef Monitoring Structures [ 28 , 56 , 57 ]. This is a similar range in structures and materials as for all ARs and is not particular to tropical reefs [ 58 ]. We used a broad AR definition when examining spatio-temporal patterns of AR deployment to accurately characterize the wide variety of structures that are currently being categorized as ARs in the peer-reviewed literature.
There is also little consistency in AR area within the peer-reviewed literature. Deployments of ARs for conservation purposes were conducted on a larger scale than ARs deployed for scientific experimentation. Most ARs used in experimental studies (70%) were 1–5 m 2 ( Table 1 ), while more than a half (60%) of ARs with conservation objectives were > 150 m 2 ( Table 2 ). The small size in experimental studies likely reflects logistical constraints of monitoring large reef structures in scientific experiments or of experimentally controlling and disentangling confounding abiotic effects of reef development on larger ARs [ 36 ]. Spacing between individual ARs is not well reported in studies examining structures with conservation objectives, which often neglect to distinguish between ARs and AR modules. Nearly all studies that monitored communities on de facto reefs reported that the structures were > 150 m 2 ; only two studies monitored response variables on ARs of smaller area ( Table 1 ).
https://doi.org/10.1371/journal.pone.0261964.t001
https://doi.org/10.1371/journal.pone.0261964.t002
There were only 4 reports of AR deployments in the scientific literature until the mid-twentieth century. More than 2200 ARs were deployed in the 1960s, most in Hawaii ( Fig 2 ). Comparatively few deployments were recorded from the 1970s to the 1990s, with a greater than 2-fold increase from the 1960s in the 2000s, followed by a similar increase between the 2000s and the 2010s ( Fig 2 ). The increase in the 2000s corresponds to the increased focus on effects of climate change on coral reefs in the late 1990s following the first major global bleaching event in 1998 [ 59 ]. In the 2010s, > 10000 deployments were reported during a single study on the Indian shelf [ 60 ], resulting in the highest recorded number of ARs in coral ecosystems globally. These temporal patterns parallel those for ARs in other coastal ecosystems, reflecting a general global transition in AR research [ 61 ].
Inset shows marine realms defined by [ 48 ].
https://doi.org/10.1371/journal.pone.0261964.g002
Following the large number of AR deployments in the Western Indo-Pacific, the Tropical Atlantic region has the next greatest number of AR deployments to date, with a coral restoration effort in Antigua contributing substantially to the region’s deployments (~ 3500 ARs deployed in 2004). However, the high abundance of ARs from the Tropical Atlantic is biased by a high intensity of study and frequency of publication from the southern United States (particularly Florida) from the 1960s onward [ 61 , 62 ]. Florida has a long history of AR deployment, with reefs often made from cheap waste materials (tires, metal construction materials, automotive parts) or de facto structures (sunken vessels, planes) [ 63 ].
The Central and Eastern Indo-Pacific exhibit similar numbers of AR deployments ( Fig 2 ). AR deployments in the Eastern Indo-Pacific are attributed mostly to a single location (Hawaii), while in the Central Indo-Pacific they are distributed across several countries (e.g. Australia, Indonesia, Malaysia, Taiwan, Thailand, Vietnam) but largely concentrated in Indonesia. The regional interest of ARs in the Central Indo-Pacific may be a consequence of increasing exploitation of marine habitats [ 64 ] and the reliance of Southeast Asian countries on the economic value of ecosystem services associated with coral reefs (e.g. fisheries, tourism, shoreline protection) [ 65 ].
Studies reporting on ARs that did not have a direct conservation-oriented objective were classified as either scientific experimentation or de facto submergences ( Table 1 ) and were not included in our exploration of AR conservation objectives ( Table 2 ). Over one third of the studies examined in this review (53 of 136) reported on scientific experiments conducted on ARs and 42% of these were conducted in the Tropical Atlantic realm. Overall, studies addressing only scientific objectives were marginally shorter than conservation-oriented projects, with mean durations of 1.7 y ( Table 1 ) and 2.0 y ( Table 2 ), respectively.
ARs recorded in peer-reviewed literature and deployed in the 1920s – 1950s were unplanned ship groundings that later were observed to have an AR effect by attracting fish and invertebrate colonizers [ 66 ]. Research efforts on de facto reefs (22 of 136 studies) reflect largely opportunistic monitoring, with data most often collected through digital imagery and a few manipulative experiments ( Table 1 ). De facto ARs are the most variable in terms of study duration, ranging from 4 months to 11 years.
The Tropical Atlantic and Central Indo-Pacific realms have the highest number of AR deployments for scientific objectives or de facto deployments, with 414 and 402, respectively. In the past two decades, more than half (57%) of all such AR deployments are from the Central Indo-Pacific, a region which has experienced significant coral mortality since 1998 [ 4 , 67 ].
Conservation objectives of ars..
The three most-commonly cited conservation objectives of ARs were increasing fish abundance (55%), increasing habitat quantity (31%), and increasing coral cover (27%) ( Table 2 ). These conservation objectives were most common in the Western and Central Indo-Pacific, Tropical Atlantic and Temperate Australasia ( Fig 3 ). Many of these ARs are in countries with substantial government funding for research and conservation, notably the USA (Florida) and Israel. In the Central Indo-Pacific, ARs with conservation objectives were predominantly deployed in countries with well-established national programs for AR development (Thailand, Malaysia), as well as those which received international funding in response to reef decimation by the Indian Ocean tsunami of 2004 (Thailand, Indonesia) [ 68 , 69 ].
See Fig 2 for map of bioregions.
https://doi.org/10.1371/journal.pone.0261964.g003
The most frequently cited conservation objectives reflect a concentration on enhancing the quantity of hermatypic coral habitat and its most economically valuable inhabitants, including commercial fish ( Fig 3 ). Fewer studies reported on ARs deployed for objectives related to the mitigation of natural and anthropogenic impacts on reef communities, such as conservation of target species (24%), mitigation of environmental stressors (22%), and provision of coral nurseries (14%), a relatively new restoration goal [ 70 ]. These AR conservation objectives are particularly common in the Central and Western Indo-Pacific ( Fig 3 ). Studies addressing socio-cultural value and economic analyses on ARs (16%) were most frequently conducted in the Western Indo-Pacific. More specifically, 8 out of 51 studies were from the Middle East, where sea surface temperatures (SST) have increased more than 3 times the global average since 1985 [ 1 ]. This region is a global hotspot for AR research, leading the publication output in many categories of conservation objectives ( Fig 3 ). Two studies (both from Malaysia) stated their conservation objective was to deter fishing trawlers and were not included in Table 2 .
Globally, fish and coral (29 and 26 studies, respectively) were the most frequently monitored taxonomic groups in the 51 studies assessing progress towards achieving conservation objectives of an AR. Most studies on corals (79%) were conducted in the Central and Western Indo-Pacific, while studies addressing fish populations were more evenly distributed across realms ( Fig 4 ). Publications on corals were most frequent in the Central and Western Indo-Pacific, indicating biases in the AR conservation literature.
https://doi.org/10.1371/journal.pone.0261964.g004
ARs have been deployed on coral reefs to assess and increase abundance of fish populations since the 1980s, and fish taxa were monitored in 55% of studies evaluating the conservation success of ARs ( Fig 4 ). This is largely in response to declining fisheries on coral reefs due to overfishing and harmful fishing practices that have had catastrophic effects on coral reef fish since the 1980s, such as cyanide and dynamite fishing [ 4 , 71 ]. Many studies have focused on the population dynamics and behaviour of commercially or recreationally desirable fish species on and near ARs [ 72 , 73 ]. In the 1980s and 1990s, publications focused on protecting and increasing target fish species on reefs [ 74 – 76 ]. From the 1990s to 2010s, research effort on fish taxa has continually increased ( Fig 5 ). In the late 2010s and 2020, a few conservation-oriented publications from Temperate Australasia and the Eastern Indo-Pacific focused on space use or connectivity of fish populations on ARs and adjacent natural reefs, likely reflecting an increased focus on the importance of connectivity in the persistence of reef fish assemblages [ 77 – 79 ].
https://doi.org/10.1371/journal.pone.0261964.g005
Scleractinian corals were the other most frequently monitored (53%) taxonomic group on ARs in the peer reviewed literature. Similarly to fish population metrics, the number of studies monitoring coral communities increased every decade from the 1990s to 2010s ( Fig 5 ), reflecting the increasing scale and severity of anthropogenic impacts on coral reefs [ 80 , 81 ]. Due to the alarming decline in coral cover and associated biodiversity worldwide, objectives of ARs that focus on coral conservation (e.g. coral nurseries or transplantation initiatives) [ 10 , 82 ] will likely continue to increase into the 2020s and beyond. More studies on coral conservation were published in the first five months of 2020 than during an entire decade in the 1980s and 1990s ( Fig 5 ).
Benthic algae and invertebrates other than corals were the least monitored taxonomic groups on ARs ( Fig 4 ). Understanding the successional patterns of these organisms on different AR structures is important because they can attract or deter target species [ 35 ]. Monitoring frequency of these underrepresented groups has increased since the 1990s, but they were still only measured in 0.05% (algae) and 20% (other invertebrates) of conservation studies published in the 2010s ( Fig 5 ). However, despite increasing awareness of the importance of these groups for attaining conservation objectives of ARs, monitoring is still lacking in many regions [ 35 , 83 ]. Non-coral invertebrate groups were monitored in studies from the Indo-Pacific realms (16%), the Tropical Atlantic (17%), and Temperate Australasia (33%) ( Fig 4 ). Only 1% of conservation studies measured benthic algae, all in the Indo-Pacific and the Tropical Atlantic. Fouling invertebrates and macroalgae growing on ARs can attract fish and motile invertebrate grazers [ 84 – 86 ]. Structures designed to support the growth of these organisms on coral reefs can enhance reef complexity and the abundance of local consumer populations [ 87 , 88 ]. Alternatively, excessive fouling by toxic invertebrates (e.g. ascidians and sponges) and some species of macroalgae deter coral larvae from settling and increase post-recruitment mortality rates [ 89 – 91 ]. Therefore, it is unclear whether ARs designed to promote fouling communities for the attraction of target fish species are conducive to coral recruitment.
Reported success of achieving ar conservation objectives.
Deployment of ARs with specific conservation objectives has varied over time ( Fig 6 ) and geographic locations ( Fig 3 ). Of the 51 studies, 65% reported success or progress towards achieving the conservation objective of AR deployment. Objectives with the highest reported rates of success were provision of nursery habitat and increasing coral cover (each 71%), followed by increasing fish abundance and mitigating effects of environmental impacts (each 64%), and increasing habitat quantity and attaining socio-cultural objectives (each 63%) and ( Table 2 ). Conservation of target species was reported as successful in only 42% of studies. The most-commonly cited reasons for not achieving conservation objectives were poor AR design for target species and extensive bleaching during the study period ( Table 2 ). Effective AR design considerations can be integrated into management strategies and deployment plans; however, reducing the level of extensive bleaching on artificial and natural reefs will require global cooperation for reducing carbon emissions [ 92 ].
Numbers above bars indicate number of studies.
https://doi.org/10.1371/journal.pone.0261964.g006
Many studies reported multiple conservation objectives for each AR ( Table 2 ), and 35% did not draw conclusions on all stated objectives. For example, if an AR was deployed for both increasing fish abundance and mitigating an environmental stressor, researchers may have recorded progress towards attaining only one of the two objectives due to constraints of logistics or expertise. Deploying ARs with multiple conservation objectives may reduce the likelihood of evaluating success or measuring ecological function of the AR. Structural design, site, and monitoring should be tailored for specific conservation objectives to limit ambiguous conclusions about success.
While ARs have been deployed to increase fish abundance since the 1980s, many studies monitoring their success did not measure appropriate ecological response variables for detecting increased fish production on the reef ( Fig 7 ). For example, few studies examining the success of ARs in increasing fish abundance effectively monitored fish recruitment and movement between natural reefs and ARs. Therefore, authors were not able to distinguish whether ARs are attracting fish from adjacent habitats or enhancing abundance of resident populations. The three-dimensional structure and physical relief of the AR plays a significant role in attracting adult and juvenile fish from the water column [ 34 , 93 , 94 ]. Factors that contribute to the species composition of the colonizing fish community on ARs include distance from suitable substrate, distance from source populations, access by predators, access to food, and shelter for protection and egg-laying [ 34 , 63 ]. Disentangling whether ARs actually enhance production of fish or simply redistribute them within the ecosystem would enable researchers to evaluate whether ARs can be used to increase absolute fish abundance on coral reefs. This knowledge gap is well cited within the AR literature [ 34 , 40 , 42 ] and new approaches, such as modelling of biomass flux, may prove useful [ 95 ]. However, our results indicate that the gap remains poorly addressed in coral reef ecosystems specifically.
https://doi.org/10.1371/journal.pone.0261964.g007
Increasing coral cover has been a relatively successful AR conservation strategy ( Table 2 ). Overall, peer-reviewed studies used appropriate monitoring strategies for determining the success of this objective; however, there was regional variation in the measured response variables. Studies done in marine realms that encompassed ocean warming hotspots (Western and Central Indo-Pacific) concentrated on response variables pertaining to specific coral life history events (e.g. recruitment, survival/mortality, reproduction, and growth) ( Fig 7 ). However, the scale of ARs has been too small to address regional losses in coral cover and the study duration has been too short to adequately assess a sustained increase in coral cover ( Table 2 ), which can take decades to detect [ 96 , 97 ]. Small-scale rehabilitation projects using ARs to increase coral cover in denuded areas might be successful if proper design considerations and environmental stressors are taken into account [ 17 ]. For example, suspended ARs could be deployed on shallow water reefs and moved to deeper or cooler water during periods of peak SST to avoid bleaching [ 47 ].
Protecting select ecologically and socio-culturally important species was addressed through the objective of conserving target species. Authors reported limited success for this objective, with many studies citing inappropriate design for target species as the reason ( Table 2 ). One study reported that colonization of target fish species was interrupted by the presence of invasive lionfish [ 98 ]. Structural design and site selection must be considered using species-specific requirements to increase the overall success of this conservation objective [ 17 ]. ARs deployed for the purpose of restoring, rehabilitating, or mitigating reef degradation for conservation of selected species need to be specifically engineered to enhance settlement and survival of targeted species [ 94 ].
Stressor mitigation has been increasingly used as a conservation objective for ARs over the past two decades ( Fig 6 ). This is most likely a response to the increasing frequency and severity of coral bleaching events and concurrent climate change perturbations since the 1990s [ 1 , 80 , 99 ]. While this objective can be met on small spatial scales (e.g. preventing impacts of wave action and sedimentation) [ 60 ], our results suggest limited success when ARs are deployed to address ecosystem-wide stressors because ARs operate on a much smaller scale (m– 100s m) than natural reefs (10s – 100s km). Both scientific and conservation projects on ARs can be interrupted by large-scale bleaching events during the study period, making it difficult or impossible to assess the efficacy of ARs in mitigating stressors [ 26 , 68 ]. ARs do not directly alleviate underlying environmental stressors and may only be effective at remediating damages once the original perturbation has been substantially reduced or removed [ 17 , 100 ]. Coral reef restoration (including through the deployment of ARs) is most effective as an integrated component of wider management frameworks that include stressor mitigation [ 17 ]. The mismatch between the increasing spatial scale of stressors and the small scale of management interventions, such as ARs, reinforce the urgency for developing comprehensive management frameworks [ 101 ].
Arguably the most successful application of ARs is as nursery habitat for coral transplantation or source populations for which specific and appropriate ecological response variables (i.e. coral growth, reproduction, and survival) were used to determine success ( Table 2 ). As long as coral colonies or fragments of colonies experience low mortality, increased larval production and a high yield of functional adult colonies with low environmental impact are possible [ 102 , 103 ]. Native species predicted to respond well to anticipated climatic changes can be selectively bred as a biological bank to re-populate natural reefs after disturbances [ 104 ]. If ARs are suspended or designed to detach from the seafloor, they also can be moved horizontally or vertically to avoid unfavorable growing conditions [ 70 ]. While nurseries operate on a relatively small scale compared to natural reefs, the likelihood of an AR functioning as a small source population in the region can be maximized by seeding it with high densities of coral species [ 28 ]. As with many studies published on active coral restoration strategies, publications examining the success of ARs as coral nurseries were exclusively from the Western and Central Indo-Pacific ( Fig 3 ).
ARs deployed to increase habitat have been largely successful, likely because they add hard substrate to the benthic environment, making this a relatively attainable objective [ 39 , 105 , 106 ]. Measured response variables focused on benthic community development and fish presence at the AR ( Table 2 ). Study durations for this objective were too short (0.01–3.5 y) to characterize success beyond initial recruitment and colonization phases for fish and invertebrates [ 63 ]. However, increasing hard substrate is not considered a high priority in reef conservation compared to addressing large-scale tissue loss of scleractinian corals caused by ocean acidification and warming [ 36 ].
In studies where deployment of ARs for socio-cultural purposes was the primary goal, the ARs were monitored appropriately and can be considered successful. However, in studies that combined socio-cultural and ecological objectives, conclusions were only drawn about the latter. Studies that monitored AR success using socio-cultural objectives employed a variety of socio-cultural variables, which can be separated into those monitoring human behaviour and emotions relative to ARs and those concerned with economic valuation ( Table 2 ). In the Western Indo-Pacific, researchers surveyed the attractiveness of ARs to divers and diver behaviour on ARs [ 107 ]. Some studies examining the economic value of ARs lacked secondary inclusion criteria for this review but conducted a cost-benefit analysis [ 108 ] or estimated gross revenue generated from commercial fisheries as a consequence of ARs [ 109 ].
Overall, our results indicate that ARs have limited success in meeting regional-scale conservation objectives, such as increasing abundance of coral and fish species or stressor mitigation. Nonetheless, these objectives are being increasingly cited in studies examining AR success, likely because of the acceleration of coral decline globally and the increasing call for remediating losses with active restoration strategies [ 14 ]. Because ARs mostly operate on a much smaller scale than natural reefs (except possibly small patch reefs), their success in addressing large-scale objectives must be assessed. Reference or control sites can provide context for the observed outcomes on ARs [ 29 ]. For example, a meta-analysis of 39 studies documented no difference in fish community metrics between natural and artificial reefs [ 41 ]. While it has been suggested that larger ARs (> 150 m 2 ) support higher fish abundances [ 23 ], the extent to which ARs function as a source of fish production remains poorly understood [ 34 , 40 , 42 ]. Further, larger ARs are logistically difficult to fund, deploy, and monitor. The introduction of networks of ARs to regions with minimal environmental stressors may increase the success of abundance-oriented conservation objectives (i.e. increasing fish abundance and coral cover) by increasing colonizable reef area while fostering connectivity of fish and invertebrate species between degraded natural reefs [ 100 ]. Overall, small-scale objectives of ARs (e.g. increasing public education, selective coral breeding programs, training scientific and recreational divers) are far more achievable because they do not require additional intensive, long-term studies to determine their contribution to reef conservation and are generally successful when well defined and monitored.
Among all studies considered in this review, more than 73% spanned 3 years or less, which is too short a period for elucidating or predicting long-term shifts in coral reef populations. Studies that examined the success of ARs in meeting conservation objectives spanned 1 week to 5 years. This period matches the average for monitoring studies of several different coral restoration approaches [ 19 , 10 ] and may be adequate for addressing short-term goals of restoration at local scales [ 17 ]. For example, observation periods of months to years can allow monitoring colonization patterns in many short-lived organisms, such as reef-associated invertebrates (e.g. ascidians, bryozoans, and some sponges), that can settle, reproduce, and die on a substrate within months [ 47 , 110 , 111 ]. These durations also may be effective for monitoring fish populations on ARs, as many fish species have a life expectancy of under 5 years due to their inherent longevity or high rates of juvenile mortality [ 112 – 114 ]. Changes in coral community composition and dynamics, however, take much longer to detect [ 115 , 116 ]. For example, scleractinian coral communities require multidecadal monitoring to properly assess ecologically relevant trends in coral cover and species composition [ 96 , 97 ]. Longer monitoring periods also may be needed to capture effects of aperiodic or stochastic events, such as heatwaves or storms. Future studies examining the success of ARs in achieving coral-oriented conservation objectives must adjust study duration according to the relevant time scales of biotic and abiotic factors that govern the underlying ecological processes.
ARs can have some potentially negative impacts on the surrounding ecosystems. Often the materials used in ARs, such as rubber or plastics, are not biodegradable or may even leach toxic substances into the surrounding ecosystems [ 63 ]. Concrete, which is used for many ARs because of ease of production and low cost, in addition to leaching metals has a high alkalinity that may inhibit colonization [ 117 , 118 ]. To increase the ecological value of artificial structures, new materials using aggregate concrete with different chemistries are being developed [ 117 , 118 ]. ARs can also facilitate the introduction and spread of invasive species [ 119 , 120 ]; modification of the physical and chemical properties of the ARs and pre-seeding by native species may minimize colonization by non-native species [ 119 ]. Engineering solutions can provide potential mitigation strategies for the negative impacts of ARs.
Based on their reported success as active restoration tools for tropical coral reefs, ARs are most likely to achieve their conservation objectives by providing nursery habitat for rearing target reef species or by supplying additional hard substrate for settlement and recruitment of corals and other marine organisms. We suggest that promoting local socio-cultural values also has potential for success if it is prioritized as an objective and properly monitored. This objective can be effective also globally, by increasing awareness of coral reef decline among tourists who mostly originate from countries without corals. While the effectiveness of ARs per se in achieving regional-scale conservation objectives may be limited, their integration into a larger restoration program could prove beneficial if used conjunction with other conservation strategies. However, given their relatively high cost, the implementation of ARs into larger restoration programs would require the development of better practices in identifying objectives, selecting the appropriate designs, and monitoring the relevant ecological responses.
S1 checklist. prisma 2009 checklist..
https://doi.org/10.1371/journal.pone.0261964.s001
https://doi.org/10.1371/journal.pone.0261964.s002
We thank Claire Attridge for her help with the literature review and the preparation of the manuscript.
Discover the world's research
share this!
August 7, 2024
This article has been reviewed according to Science X's editorial process and policies . Editors have highlighted the following attributes while ensuring the content's credibility:
fact-checked
trusted source
by Australian Institute of Marine Science
Most of the underwater surveys contributing to these findings , published today, were conducted before and during the recent mass bleaching event, one of the most extensive and serious on record, and have not yet captured how many corals survived or died following the bleaching.
Surveys in the Central region were also completed before the passage of tropical Cyclone Jasper in December 2023.
AIMS' Long-Term Monitoring Program (LTMP) leader Dr. Mike Emslie said coral cover increases were a positive sign but did not reflect the potentially destructive consequences of the 2024 mass bleaching event.
"We saw evidence of early onset mortality, particularly in the Southern region, but the full picture of mortality was not yet apparent during this year's surveys," he said.
"While bleached corals are very stressed, they are still alive and are recorded as live coral on our surveys.
"Some types of corals can remain bleached for months, remaining on a knife edge between survival and death. This is why returning and repeating surveys of the reefs in this vast, complex and dynamic system is so important. This year's results serve as a very important reference against which to measure the impacts of the summer's events."
The next (LTMP) survey season recommences in September and will capture impacts on coral cover from this summer's mass bleaching event and the cyclones, with a full assessment complete by mid-2025.
"Climate change remains the greatest threat to the Reef because it drives these mass bleaching events. This most recent one was the fifth such event since 2016. These more frequent and extensive marine heat waves will lead to shortened 'windows' for coral recovery. Recent gains, while encouraging, can be lost in a short amount of time," Dr. Emslie said.
Surveys were conducted at 94 Reefs spread through the Northern, Central and Southern Great Barrier Reef between August 2023 and June 2024.
The Report recorded the following average hard coral coverage:
The AIMS report finds that small rises in coral cover this year brought the Northern and Central regions to their highest levels in 38 years of monitoring.
The surveys also found that crown-of-thorns starfish outbreaks have persisted on some reefs in the Southern region.
The long term monitoring team surveyed reefs off Townsville after the passage of tropical Cyclone Kirrily in late January, finding evidence of storm damage and declines in hard coral cover ranging from 6% to 10% at Kelso, John Brewer, Helix and Chicken Reefs. Other reefs appear to have escaped with little impact.
AIMS Research Program Director Dr. David Wachenfeld said the regional increases in coral cover are encouraging, showing the Reef's capacity for recovery after reaching their lowest levels within the last 15 years. However, climate change and other disturbances mean this recovery is fragile and Reef resilience is not limitless.
"In many ways the Reef has had some lucky escapes in recent years. The 2020 and 2022 mass bleaching events had levels of heat stress that were not as intense as the 2016 and 2017 events or the 2024 event. Coupled with very few other events causing widespread coral death, that has led to the levels of coral cover increase we have seen," he said.
"But the frequency and intensity of bleaching events is unprecedented, and that is only forecast to escalate under climate change, alongside the persistent threat of crown-of-thorns starfish outbreaks and tropical cyclones."
Aerial surveys undertaken by AIMS and the Great Barrier Reef Marine Park Authority in February and March found bleached corals in the shallows of 73% of reefs surveyed across all three regions.
In recent weeks, AIMS scientists in separate monitoring programs observed substantial mortality in reefs that were particularly hard hit by the 2024 event.
"We are only one large scale disturbance event away from a reversal of the recent recovery. The 2024 bleaching event could be that event—almost half of the 3000 or so reefs that make up the marine park experienced more heat stress than ever recorded," Dr. Wachenfeld said.
"We still don't know how much mortality this event has caused. Our monitoring over the next 12 months will help us to understand how this bleaching event stacks up against the others in the last decade."
AIMS CEO Professor Selina Stead said AIMS was prioritizing research to develop scientific solutions to boost reef resilience under a warming climate.
"Climate change is increasing pressure on reef systems around the world," she said. "The 2024 bleaching event was part of the fourth global bleaching event, announced in April.
"These vitally important ecosystems that millions rely upon need strong greenhouse gas emissions reduction, science-based management of local pressures, and input from multiple fields of research if they are to endure.
"At AIMS we are developing a toolbox of interventions to help reefs adapt to and recover from the effects of climate change ."
Provided by Australian Institute of Marine Science
Explore further
Feedback to editors
2 hours ago
Relevant physicsforums posts, strategies and tips for first responders interacting with autism spectrum disorder patients.
21 minutes ago
29 minutes ago
Using capsaicin to get really high, therapeutic interfering particle.
Aug 14, 2024
Aug 8, 2024
More from Biology and Medical
Aug 8, 2023
Aug 3, 2022
Mar 26, 2024
Apr 5, 2024
Mar 8, 2024
Jun 26, 2024
12 hours ago
Aug 15, 2024
Let us know if there is a problem with our content.
Use this form if you have come across a typo, inaccuracy or would like to send an edit request for the content on this page. For general inquiries, please use our contact form . For general feedback, use the public comments section below (please adhere to guidelines ).
Please select the most appropriate category to facilitate processing of your request
Thank you for taking time to provide your feedback to the editors.
Your feedback is important to us. However, we do not guarantee individual replies due to the high volume of messages.
Your email address is used only to let the recipient know who sent the email. Neither your address nor the recipient's address will be used for any other purpose. The information you enter will appear in your e-mail message and is not retained by Phys.org in any form.
Get weekly and/or daily updates delivered to your inbox. You can unsubscribe at any time and we'll never share your details to third parties.
More information Privacy policy
We keep our content available to everyone. Consider supporting Science X's mission by getting a premium account.
Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.
Nature volume 632 , pages 320–326 ( 2024 ) Cite this article
17k Accesses
1 Citations
2993 Altmetric
Metrics details
Mass coral bleaching on the Great Barrier Reef (GBR) in Australia between 2016 and 2024 was driven by high sea surface temperatures (SST) 1 . The likelihood of temperature-induced bleaching is a key determinant for the future threat status of the GBR 2 , but the long-term context of recent temperatures in the region is unclear. Here we show that the January–March Coral Sea heat extremes in 2024, 2017 and 2020 (in order of descending mean SST anomalies) were the warmest in 400 years, exceeding the 95th-percentile uncertainty limit of our reconstructed pre-1900 maximum. The 2016, 2004 and 2022 events were the next warmest, exceeding the 90th-percentile limit. Climate model analysis confirms that human influence on the climate system is responsible for the rapid warming in recent decades. This attribution, together with the recent ocean temperature extremes, post-1900 warming trend and observed mass coral bleaching, shows that the existential threat to the GBR ecosystem from anthropogenic climate change is now realized. Without urgent intervention, the iconic GBR is at risk of experiencing temperatures conducive to near-annual coral bleaching 3 , with negative consequences for biodiversity and ecosystems services. A continuation on the current trajectory would further threaten the ecological function 4 and outstanding universal value 5 of one of Earth’s greatest natural wonders.
Like many coral reefs globally, the World Heritage-listed GBR in Australia is under threat 4 , 6 . Mass coral bleaching, declining calcification rates 5 , 7 , outbreaks of crown-of-thorns starfish ( Acanthaster spp.) 8 , severe tropical cyclones 9 and overfishing 10 have placed compounding detrimental pressures on the reef ecosystem. Coral bleaching typically occurs when heat stress triggers the breakdown of the symbiosis between corals and their symbiotic dinoflagellates 11 . Although coral bleaching can occur locally as a result of low salinity, cold waters or pollution, regional and global mass bleaching events, in which the majority of corals in one or more regions bleach at once, are strongly associated with increasing SST linked to global warming 2 .
The first modern observations of mass coral bleaching on the GBR occurred in the 1980s, but these events were less widespread and generally less severe 3 than the bleaching events in the twenty-first century 4 . Stress bands in coral skeletal cores have provided potential evidence for pre-1980s bleaching in the GBR and Coral Sea, such as during the 1877–78 El Niño 12 . However, stress bands are evident in relatively few cores before 1980 (ref. 12 ), suggesting that severe mass bleaching did not occur in the 1800s and most of the 1900s.
As the oceans have warmed, however, mass coral bleaching events have become increasingly lethal to corals 4 . Coral bleaching on the GBR 1 in 1998 coincided with a strong eastern-Pacific El Niño, and in 2002 with a weak El Niño. El Niño events can induce lower cloud cover and increased solar irradiance over the GBR 13 , increasing the risk of thermal stress and mass bleaching events 14 . In 2004, water temperatures were anomalously warm, and although bleaching occurred in the Coral Sea 15 , it was not widespread in the GBR, probably because there was reduced upwelling and an associated reduced influence of nutrients on symbiotic dinoflagellate expulsion 16 .
However, in the nine January–March periods from 2016 to 2024 (inclusive) there were five mass coral bleaching events on the GBR. Each was associated with high SSTs and affected large sections of the reef. GBR mass bleaching occurred in both 2016 and 2017, influenced by the presence of an El Niño event in 2016, and led to the death of at least 50% of shallow-water (depths of 5–10 m) reef-building corals 4 . Major bleaching events occurred again in quick succession in 2020 and 2022, with the accumulated heat stress for large sections of the GBR reaching levels conducive to widespread bleaching but lower levels of coral mortality 1 . The bleaching event in 2022 occurred, unusually, during a La Niña event, which is typically associated with cooler summer SSTs, higher than average rainfall and higher cloud cover on the GBR 1 . At the time of writing, researchers are assessing the impacts of the 2024 mass bleaching event.
The frequency of recent mass coral bleaching and mortality on the GBR is cause for concern. In 2021, the World Heritage Committee of the United Nations Educational, Scientific and Cultural Organization (UNESCO) drafted 17 a decision to inscribe the GBR on the List of World Heritage in Danger, stating that the reef is “facing ascertained danger”, citing recent mass coral bleaching events and insufficient progress by the State Party (Australia) in countering climate change, improving water quality and land management issues. The committee’s adopted decisions 18 have not included inscription of the ‘in danger’ status, but the draft inscription highlights the seriousness of the recent mass coral bleaching events. Authorities in Australia 5 have noted that climate change and coral bleaching have deteriorated the integrity of the outstanding universal value of the GBR, a defining feature of its World Heritage status.
Although rapidly rising SSTs are attributed to human activities with virtual certainty 19 , understanding the multi-century SST history of the GBR is critical to understanding the influence of SST on mass coral bleaching and mortality in recent decades. Putting aside a problematic attempt to do this 20 , which was discredited 21 , 22 , knowledge of the long-term context for GBR SSTs comes primarily from two multi-century reconstructions based on the geochemistry of coral cores collected from the inner shelf 23 and outer shelf 24 (Flinders Reef) in the central GBR. These reconstructions showed that SSTs in the early 2000s were not unusually high relative to levels in the past three centuries, with five-year mean SSTs (and salinities) estimated to be higher in the 1700s than in the 1900s. However, these records were limited by their relatively coarse five-year sampling resolution and their most recent data point being from the early 2000s. After these studies were published, SSTs in the GBR have continued to rise. Updated analysis of coral data from Flinders Reef provides valuable improved temporal resolution 25 , but interpretations of these records remain limited spatially.
Here, we investigate the recent high SST events in the GBR region in the context of the past four centuries. We combine a network of 22 coral Sr/Ca and δ 18 O palaeothermometer series (Supplementary Tables 1 and 2 ) located in and near to the Coral Sea region to infer spatial mean SST anomalies (SSTAs) for January–March, the months when maximum SST and thermal bleaching are most likely to occur in the Coral Sea 16 , 26 , each year from 1618 to 1995 ( Methods and Supplementary Information ). Anthropogenic climate change began and proceeded entirely within the multi-century lives of some of these massive coral colonies, offering a continuous multi-century record covering the industrial era. We use this 1618–1995 reconstruction and the available 1900–2024 instrumental data to contextualize the modern trend and rank four centuries of January–March SSTAs with greater precision than was previously possible. We then assess the degree of human influence on ocean temperatures in the region using climate model simulations run both with and without anthropogenic forcing.
Mass coral bleaching on the GBR in 2016, 2017, 2020, 2022 and 2024 during January–March coincided with widespread warm SSTAs in the surrounding seas 1 , including the Coral Sea (Fig. 1a–e , using ERSSTv5 data 27 ). The Coral Sea and GBR have experienced a strong warming trend since 1900 (Fig. 1f ). January–March SSTAs averaged over the GBR are strongly correlated ( ρ = 0.84, P ≪ 0.01) with those in the broader Coral Sea (Fig. 1f ), including when the long-term warming trend is removed from both time series ( ρ = 0.69, P < 0.01; Supplementary Fig. 4 ). Based on the strength of this correlation, we associate high January–March area-averaged Coral Sea SSTAs with increased thermal bleaching risk in the GBR.
a – e , SSTAs (using ERSSTv5 data) for January–March in the Australasian region relative to the 1961–90 average for the five recent GBR mass coral bleaching years: 2016, 2017, 2020, 2022 and 2024. The black box shows the Coral Sea region (4° S–26° S, 142° E–174° E). f , Coral Sea and GBR mean SSTAs for 1900–2024 in January–March relative to the 1961–90 average. The black vertical lines indicate the five recent GBR mass coral bleaching years.
Record temperatures were set in 2016 and 2017 in the Coral Sea, and in 2020 they peaked fractionally below the record high of 2017. The January–March of 2022 was another warm event, the fifth warmest on record at the time. Recent data (ERSSTv5) indicate that 2024 set a new record by a margin of more than 0.19 °C above the previous record for the region. The January–March mean SSTs averaged over the five mass bleaching years during the period 2016–2024 are 0.77 °C higher than the 1961–90 January–March averages in both the Coral Sea and the GBR. The multidecadal warming trend, extreme years and association between GBR and Coral Sea SSTs are similar for the HadISST 28 gridded SST dataset, with some notable differences in the 1900–40 period (Supplementary Fig. 3 ). Furthermore, analysis of modern temperature-sensitive Sr/Ca series from GBR corals for 1900–2017 provides coherent independent evidence of statistically significant multi-decadal warming trends in January–March SSTs in the central and southern GBR (Supplementary Information section 4.2 ).
Reconstructing Coral Sea January–March SSTs from 1618 to 1995 extends the century-long instrumental record back in time by an additional three centuries (Fig. 2a and Methods ). The reconstruction (calibrated to ERSSTv5) shows that multi-decadal SST variability was a persistent feature in the past. At the centennial timescale, there is relative stability before 1900, with the exception that cooler temperatures prevailed in the 1600s. Warming during the industrial era has been evident since the early 1900s (Fig. 2a ). There is a warming trend for January–March of 0.09 °C per decade for 1900–2024 and 0.12 °C per decade for 1960–2024 (Fig. 1f ) using ERSSTv5 data. Calibrating our reconstruction to HadISST1.1 yields similar results, with some differences in the degree of pre-1900 variability at both multi-decadal and centennial timescales (Supplementary Information section 5.2.6 ).
a , Reconstructed and observed mean January–March SSTAs in the Coral Sea for 1618–2024 relative to 1961–90. Dark blue, highest skill (maximum coefficient of efficiency) reconstruction with the full proxy network; light blue, 5th–95th-percentile reconstruction uncertainty; black, observed (ERSSTv5) data. Red crosses indicate the five recent mass bleaching events. Dashed lines indicate the best estimate (highest skill, red) and 95th-percentile (pink) uncertainty bound for the maximum pre-1900 January–March SSTA. b , Central GBR SSTA for the inner shelf 23 in thick orange and outer shelf 25 (Flinders Reef) in thin orange lines; these series are aligned here (see Methods ) with modern observations of mean GBR SSTAs for January–March relative to 1961–90. Observed data are shown at annual (grey line) and five-year (black line with open circles, plotted at the centre of each five-year period and temporally aligned with the five-year coral series 23 ) resolution. Dashed lines indicate best-estimate pre-1900 January–March maxima for refs. 23 (red) and 25 (pink). Orange shading indicates 5th–95th-percentile uncertainty bounds. Red crosses indicate the five recent mass bleaching events. c , Evaluation metrics for the Coral Sea reconstruction (Supplementary Information section 3.1 ); RE, reduction of error; CE, coefficient of efficiency; Rsq-cal, R-squared in the calibration period; Rsq-ver, R-squared in the verification (evaluation) period. d , Coral data locations relative to source data region (orange box) and Coral Sea region (red box). Coral proxy metadata are given in Supplementary Tables 1 and 2 .
Our best-estimate (highest skill; Methods ) annual-resolution Coral Sea reconstruction (Fig. 2a ), using the full coral network calibrated to the ERSSTv5 instrumental data, indicates that the January–March mean SSTAs in 2016, 2017, 2020, 2022 and 2024 were, respectively, 1.50 °C, 1.54 °C, 1.53 °C, 1.46 °C and 1.73 °C above the 1618–1899 (hereafter ‘pre-1900’) reconstructed average. Using the same best-estimate reconstruction, Coral Sea January–March SSTs during these GBR mass bleaching years were five of the six warmest years the region has experienced in the past 400 years (Fig. 2a ).
By comparing the recent warm events to the reconstruction’s uncertainty range ( Methods ), we quantify, using likelihood terminology consistent with recent reports from the Intergovernmental Panel on Climate Change 19 , that the recent heat extremes in 2017, 2020 and 2024 are ‘extremely likely’ (>95th percentile; Fig. 2a ) to be higher than any January–March in the period 1618–1899. Furthermore, the heat extremes in 2016 and 2022 are (at least) ‘very likely’ (>90th percentile) to be above the pre-1900 maximum. We perform a series of tests that verify that our findings are not simply an artefact of the nature of the coral network itself (Supplementary Information section 5.2 ). In a network perturbation test, we generate 22 subsets of the reconstruction by adding proxy records incrementally in order from the highest to the lowest correlation with the target (Supplementary Information section 5.2.5 ). We confirm that 2017, 2020 and 2024 were ‘extremely likely’ (>95th percentile) to have been warmer than any year pre-1900 (using ERSSTv5 data) for all of these proxy subsets. Furthermore, in 20 of the 22 subsets, 2016 was also ‘extremely likely’ (>95th percentile), rather than ‘very likely’, to be warmer (2022 was ‘extremely likely’ in 14 of the 22 subsets). All our additional tests, including a reconstruction with only Sr/Ca coral data (thereby omitting the possibility of any non-temperature signal in δ 18 O coral on the reconstruction), achieve high reconstruction skill and confirm the extraordinary nature of recent extreme temperatures in the multi-century context (Supplementary Information section 5.2 ). Analyses using HadISST1.1 generally show lower correlations with the coral data and reconstructions with slightly warmer regional SSTs before 1900, along with more-muted centennial and multi-decadal variability in the pre-instrumental period. Nevertheless, the HadISST1.1-calibrated reconstructions show that the recent thermal extremes are well above the best estimate (highest skill) of the pre-1900 maximum of reconstructed January–March SSTAs (Supplementary Fig. 42 ). Furthermore, lower SSTAs (in the HadISST1.1 data) relative to the previous three centuries (as in our reconstructions calibrated to HadISST1.1), coupled with the recently observed mass coral bleaching events, could indicate that long-lived corals have a greater sensitivity to warming than is currently recognized.
Reconstructed regional GBR SSTAs based on a five-year-resolution, multi-century coral δ 18 O record from the central inshore GBR 23 (Fig. 2b ) show similarly strong warming since 1900 but more multi-decadal-to-centennial variability than the Coral Sea reconstruction. Recent five-year mean January–March GBR SSTAs narrowly exceed the best estimate of the maximum pre-1900 five-year mean since the early 1600s (Fig. 2b ). The averages for the five-year periods centred on 2018 and 2022 exceed the pre-1900 maximum by 0.11 °C and 0.06 °C, respectively. Results are similar using the five-year-resolution Flinders Reef (central outer shelf) 24 record (Supplementary Fig. 24 ), although its interpretation is limited by the lack of uncertainty estimates available for that record. Our Coral Sea reconstruction incorporates an updated (annual resolution) record from Flinders Reef 25 , which indicates similar centennial trends (thin orange line in Fig. 2b ) and shows that the recent high January–March SSTA events have approached the estimated local pre-1900 maximum SSTA. Although contiguous multi-century cores from within the GBR are limited in their spatial extent, twentieth-century warming is evident in these records.
The extraordinary nature of the recent Coral Sea January–March SSTs in the context of the past 400 years is further illustrated by comparing the ranked temperature anomalies (Fig. 3 ) for the combined reconstructed and instrumental period from 1618–2024, incorporating reconstruction uncertainty ( Methods ). The mass coral bleaching years of 2016, 2017, 2020, 2022 and 2024, and the heat event of 2004, stand out as the warmest events across the whole 407-year record. The warmest three years (2024, 2017 and 2020) exceed the upper uncertainty bound (95th percentile) of the warmest reconstructed January–March in the pre-1900 period (pink (upper) dashed line in Fig. 3 ); 2016, 2004 and 2022 exceed the 90th percentile bound (red (lower) dashed line in Fig. 3 ). The warming trend is clear in the association between the ascending rank of the temperature anomalies and the year (shown as the colour of the filled circles in Fig. 3 ). Despite high interannual variability, 78 of the warmest 100 January–March periods between 1618 and 2024 occurred after 1900, and the 23 warmest all occur after 1900. The warmest 20 January–March periods all occur after 1950, coinciding with accelerated global warming.
Ranked January–March SSTAs for 1618–2024 relative to 1961–90 (coloured circles) from the best-estimate (highest skill, full coral network) reconstruction (1618–1899) and instrumental (ERSSTv5) data (1900–2024). The year is indicated by the colour of the filled circles. The 5th–95th-percentile uncertainty bounds of the pre-1900 reconstructed SSTAs are shown by small grey dots. The year labels indicate the warmest six years on record, five of which were mass coral bleaching years on the GBR. The pink (upper) dashed line indicates the 95th-percentile uncertainty bound of the maximum pre-1900 reconstructed SSTA; the red (lower) dashed line indicates the 90th-percentile limit.
Using climate model simulations from the most recent (sixth) phase of the Coupled Model Intercomparison Project 29 (CMIP6), we assess the human influence on January–March SSTAs in the Coral Sea. The model simulations are from two experiments in the Detection and Attribution Model Intercomparison Project (DAMIP) 30 . The first set of simulations represents historical climate conditions, including both the natural and human influences on the climate system over the 1850–2014 period (‘historical’; red in Fig. 4 ). The second experiment is a counterfactual climate that spans the same period and uses the same models but includes only natural influences on the climate, omitting all human influences (‘historical-natural’; blue in Fig. 4 ). The historical experiment includes anthropogenic emissions of greenhouse gases and aerosols, stratospheric ozone changes and anthropogenic land-use changes; the historical-natural experiment does not. Variations in natural climate forcings, such as from volcanic eruptions and solar variability, are incorporated in both experiments. We include models that have a transient climate response (the global mean surface-temperature anomaly at the time of a doubling of atmospheric CO 2 concentration) in the range 1.4–2.2 °C, which is deemed ‘likely’ by the science community 31 ( Methods and Supplementary Information ).
Climate-model simulations of Coral Sea January–March SSTAs relative to the 1850–1900 average for the period 1850–2014, for models within the ‘likely’ range for their transient climate response 31 . The blue line (median) and light blue shading (5th–95th-percentile limits) are from the ‘historical-natural’ climate model simulations (no anthropogenic climate forcing); the red line and light red shading are from the ‘historical’ simulations (anthropogenic influences on the climate included) using the same set of climate models. The climate-model-derived time of emergence of anthropogenic climate change, shown by the grey and black vertical lines (1976 and 1997), is when the ratio of the climate change signal to the standard deviation of noise/variability 32 across model ensemble members first rises above 1 and 2, respectively. All models are represented equally in the model ensemble.
It is only with the incorporation of anthropogenic influences on the climate that the model simulations capture the modern-era warming of the Coral Sea January–March SSTA (Fig. 4 ). The median of the historical simulations has statistically significant warming trends of 0.05 °C, 0.10 °C and 0.15 °C per decade for the periods from 1900, 1950 and 1970 to 2014, respectively; the equivalent historical-natural trends are smaller in magnitude than ±0.01 °C per decade. To further explore the centennial-scale trends, we use a bootstrap ensemble ( Methods ) of the two sets of 165-year simulations from 1850–2014. We found that 100% of the historical bootstrap ensemble has statistically significant positive trends ( Methods ) for 1900–2014, but this value is 0% for the historical-natural ensemble. The observed (ERSSTv5) mean SSTA for 2016–2024 of 0.60 °C relative to 1961–90 is warmer than any nine-year sequence in the 7,095 simulated years in the historical-natural experiments from models with transient climate responses in the ‘likely’ range 31 .
We also use the simulations to estimate the time of emergence of the anthropogenic influence on January–March Coral Sea SSTAs above the natural background variability. The anthropogenic warming signal 32 increases from near zero in 1900 to around 0.5 standard deviations of the variability (‘noise’) in 1960. The climate change signal-to-noise ratio then increases rapidly from 1960 to 2014, exceeding 1.0 in 1976, 2.0 in 1997 and around 2.8 by 2014, the end of these simulations (Fig. 4 , Methods and Supplementary Fig. 50 ). Anthropogenic impacts on the climate are virtually certain to be the primary driver of this long-term warming in the Coral Sea.
Previously, our knowledge of the SST history of the GBR and the Coral Sea region has been highly dependent on instrumental observations, with the exception of the five-year-resolution multi-century coral Sr/Ca and U/Ca SST reconstructions from the two point locations in the central GBR 23 , 24 , an update at one of these locations 25 , seasonal resolution ‘floating’ (in time) chronologies from the GBR in the Holocene 33 , 34 and point SST estimates further back in time 35 . Thus, the context of recent warming trends in the Coral Sea and GBR and their relation to natural variability on decadal to centennial timescales is largely unknown without reconstructions such as the one we developed here.
Our coral proxy network is located mostly beyond the GBR, in the Coral Sea, and some series are located outside the Coral Sea region (Fig. 2d ). The selection of the Coral Sea as a study region allowed for a larger sample of contributing coral proxy data than exists for the GBR. However, coral bleaching on the GBR can be influenced by factors other than large-scale SST, including local oceanic and atmospheric dynamics that can modulate the occurrence and severity of thermal bleaching and mortality events 13 . Nonetheless, warming of seasonal SSTs over the larger Coral Sea region is likely to prime the background state and increase the likelihood of smaller spatio-temporal-scale heat anomalies. Furthermore, where we use only the five-year resolution series directly from the GBR to reconstruct GBR SSTAs, we draw similar conclusions about the long-term trajectory of SSTAs as for our full coral network (Fig. 2b and Supplementary Fig. 24 ). Furthermore, short modern coral series from within the GBR, analysed in this study, document a multi-decadal warming signal that is coherent with instrumental data (Supplementary Figs. 29 and 30 ). Nonetheless, additional high-resolution, multi-century, temperature-sensitive coral geochemical series from within the GBR would help unravel the local and remote ocean–atmosphere contributions to past bleaching events and reduce uncertainties.
The focus on the larger Coral Sea study region also takes advantage of the global modelling efforts of CMIP6. The large number of ensemble members available for CMIP6 means that greater climate model diversity, and therefore greater certainty in our attribution analysis, is possible compared with most single model analyses. There is also a methodological benefit in having high replication of the same experiments run with multiple climate models. However, coarse-resolution global-scale models do not accurately simulate smaller-scale processes, such as inshore currents and mesoscale eddies in the Coral Sea or the Gulf of Carpentaria, which probably affect local surface temperatures and variations in nutrient upwelling in the GBR 36 , 37 . Upwelling on the GBR is linked to the strength of the East Australian Current 16 , the southward branch of the South Pacific subtropical gyre. The CMIP-scale models we use do capture these gyre dynamics. The models show that the East Australian Current is expected to increase in strength as the climate continues to warm through this century 38 , and this may lead to more nutrient inputs that can exacerbate coral sensitivity to rising heat stress 39 , 40 . As well as focusing our model analysis on the larger Coral Sea region, we use a three-month time step. In doing so, we minimize the impact of model spatio-temporal resolution on our inferences about the role of anthropogenic greenhouse-gas emissions on the SST conditions that give rise to GBR mass bleaching.
We present analyses and interpretations that are as robust as possible given currently available data and methods. However, several sources of remaining uncertainty mean that future reconstructions of past Coral Sea and GBR SSTs could differ from those presented here. Although bias corrections are applied to observational SST datasets such as ERSST and HadISST, these datasets probably retain biases, especially for the period during and before 1945 (ref. 41 ), and these may not be fully accounted for in the uncertainty estimates 42 . Because our reconstructions are calibrated directly to these datasets, future observational-bias corrections are likely to improve proxy-based reconstructions.
Reconstructions of SST that use coral δ 18 O records may be susceptible to the influence of changes in the coral δ 18 O–SST relationship on time periods longer than the instrumental training period, along with non-SST changes in the δ 18 O of seawater, which can covary with salinity. As such, new coral records of temperature-sensitive trace-element ratios such as Sr/Ca, Li/Mg or U/Ca may prove influential in future efforts to distinguish between changes in past temperature and hydroclimate. Owing to the limited availability of multi-century coral data from within the GBR itself, the reconstructed low-frequency variability of GBR SSTs in recent centuries is likely to change as more temperature proxy data become available. It is also likely that new sub-annual resolution records would aid in removing potential signal damping or bias from our use of some annual-resolution records to reconstruct seasonal SSTAs.
With global warming of 0.8–1.1 °C above pre-industrial levels 19 there has been a marked increase in mass coral bleaching globally 43 . Even limiting global warming to the Paris Agreement’s ambitious 1.5 °C level would be likely to lead to the loss of 70–90% of corals that are on reefs today 44 . If all current international mitigation commitments are implemented, global mean surface temperature is still estimated to increase in the coming decades, with estimates varying between 1.9 °C (ref. 45 ) and 3.2 °C (ref. 46 ) above pre-industrial levels by the end of this century. Global warming above 2 °C would have disastrous consequences for coral ecosystems 19 , 44 and the hundreds of millions of people who currently depend on them.
Coral reefs of the future, if they can persist, are likely to have a different community structure to those in the recent past, probably one with much less diversity in coral species 4 . This is because mass bleaching events have a differential impact on different coral species. For example, fast-growing branching and tabulate corals are affected more than slower-growing massive species because they have different thermal tolerance 4 . The simplification of reef structures will have adverse impacts on the many thousands of species that rely on the complex three-dimensional structure of reefs 4 . Therefore, even with an ambitious long-term international mitigation goal, the ecological function 4 of the GBR is likely to deteriorate further 5 before it stabilizes.
Coral adaptation and acclimatization may be the only realistic prospect for the conservation of some parts of the GBR this century. However, although adaptation opportunities may be plausible to some extent 47 , they are no panacea because evolutionary changes to fundamental variables such as temperature take decades, if not centuries, to occur, especially in long-lived species such as reef-building corals 48 . There is currently no clear evidence of the real-time evolution of thermally tolerant corals 48 . Most rapid changes depend on a history of exposure to key genetic types and extremes, and there are limitations to genetic adaptation that prevent species-level adaptation to environments outside of their ecological and evolutionary history 19 . Model projections also indicate that rates of coral adaptation are too slow to keep pace with global warming 49 . In a rapidly warming world, the temperature conditions that give rise to mass coral bleaching events are likely to soon become commonplace. So, although we may see some resilience of coral to future marine heat events through acclimatization, thermal refugia are likely to be overwhelmed 50 . Global warming of more than 1.5 °C above pre-industrial levels will probably be catastrophic for coral reefs 44 .
Our new multi-century reconstruction illustrates the exceptional nature of ocean surface warming in the Coral Sea today and the resulting existential risk for the reef-building corals that are the backbone of the GBR. The reconstruction shows that SSTs were relatively cool and stable for hundreds of years, and that recent January–March ocean surface heat in the Coral Sea is unprecedented in at least the past 400 years. The coral colonies and reefs that have lived through the past several centuries, and that yielded the valuable Sr/Ca and δ 18 O data on which our reconstruction is based, are themselves under serious threat. Our analysis of climate-model simulations confirms that human influence is the driver of recent January–March Coral Sea surface warming. Together, the evidence presented in our study indicates that the GBR is in danger. Given this, it is conceivable that UNESCO may in the future reconsider its determination that the iconic GBR is not in danger. In the absence of rapid, coordinated and ambitious global action to combat climate change, we will likely be witness to the demise of one of Earth’s great natural wonders.
The Coral Sea and GBR area-averaged monthly SSTAs relative to 1961–90 for January–March are obtained from version 5 of the Extended Reconstructed Sea Surface Temperature dataset (ERSSTv5) 27 . We compare our results using ERSSTv5 with those generated using the Hadley Centre Sea Ice and Sea Surface Temperature dataset (HadISST1.1) 28 . We use only post-1900 instrumental SST observations here. Although gridded datasets have some coverage before 1900, ship-derived temperature data in the region for that period are too sparse to be reliable for calibrating our reconstruction (Supplementary Information section 1.2 ). The regional mean for the GBR is computed using the seven grid-cell locations used by the Australian Bureau of Meteorology (Supplementary Information section 1.1 ). We define the Coral Sea region as the ocean areas inside 4° S–26° S, 142° E–174° E.
We use a network of 22 published and publicly available sub-annual and annual resolution temperature-sensitive coral geochemical series (proxies; Fig. 2d , Supplementary Tables 1 and 2 , and Supplementary Fig. 5a–v ) from the western tropical Pacific in our source data region (4° N–27° S, 134° E–184° E) that cover at least the period from 1900 to 1995. Of these 22 series, 16 are δ 18 O, which are in per mil (‰) notation relative to Vienna PeeDee Belemnite (VPDB) 51 ; the remaining six are Sr/Ca series. The coral data are used as predictors in the reconstruction of January–March mean SSTAs in the Coral Sea region. We apply the inverse Rosenblatt transformation 52 , 53 to the coral data to ensure that our reconstruction predictors are normally distributed. Sub-annually resolved series are converted to the annual time step by averaging across the November–April window. This maximizes the detection of the summer peak values, allowing for some inaccuracy in sub-annual dating and the timing of coral skeleton deposition 54 , 55 . A small fraction (less than 0.8%) of missing data is infilled using the regularized expectation maximization (RegEM) algorithm 56 (Supplementary Information section 2.3 ), after which the proxy series are standardized such that each has a mean of zero and a standard deviation of one over their common 1900–1995 period.
To produce our Coral Sea reconstruction, we use nested principal component regression 57 (PCR), in which the principal components of the network of 22 coral proxies are used as regressors against the target-region January–March SSTA relative to the 1961–90 average. We perform the reconstructions separately for each nest of proxies, where a nest is a set of proxies that cover the same time period. The longest nest dates back to 1618, when at least two series are available. The nests allow for the use of all coral proxies over the full time period of their coverage. The 96-year portion of the instrumental period (1900–1995) that overlaps with the reconstruction period is used for calibration and evaluation (or equivalently, verification) against observations. We reconstruct regional SSTAs from the principal components of the coral network of δ 18 O and Sr/Ca data, rather than their local SST calibrations, to minimize the number of computational steps and to aid in representing the full reconstruction uncertainty.
Principal component analysis (PCA) is used to reduce the dimensionality of the proxy matrix, as follows. Let P ( t , r ) denote the palaeoclimate-data matrix during the time period t = 1,..., n at an annual time step for proxy series r = 1,..., p . PCA is undertaken on this matrix during the calibration period, P cal . We obtain the principal component coefficients matrix P coeff ( r , e ) for principal components e = 1,..., n PC and principal component scores P score ( t , e ), which are representations of the input matrix P cal in the principal component space. P score is truncated to include n PC,use principal components to form \({P}_{{\rm{score}}}^{{\prime} }\) such that the variance of the proxy network explained by the n PC,use principal components is greater than \({\sigma }_{{\rm{expl}}}^{2}\) (which we set to 95%). Reconstruction tests in which \({\sigma }_{{\rm{expl}}}^{2}\) is varied from 70% to 95% show that our results are not strongly sensitive to this choice, and tests based on lag-one autoregressive noise for \({\sigma }_{{\rm{expl}}}^{2}\) from 50% to 99% further support this choice (Supplementary Information section 3.2 ). These principal components are used as predictors against which the Coral Sea January–March instrumental SSTAs are regressed. We regress the standardized SSTA target data during the calibration period, I cal , against the retained principal components of the predictor data, \({P}_{{\rm{score}}}^{{\prime} }\) :
Thus, we obtain n PC,use estimates of the regression coefficients γ e with gaussian error term ε t ~ N (0, \({\sigma }_{N}^{2}\) ). The principal components are extended back into the pre-instrumental period by multiplying the entire proxy matrix P ( t , p ) with the truncated principal component coefficient matrix \({P}_{{\rm{coeff}}}^{{\prime} }\) ( t , e ) to obtain \({Q}_{{\rm{coeff}}}^{{\prime} }\) :
The reconstruction proceeds with the fitted regression coefficients γ e and extended coefficient matrix \({Q}_{{\rm{coeff}}}^{{\prime} }\) to obtain a reconstruction time series R m ( t ) for a given nest of proxy series
The standardized reconstruction R m ( t ) is then calibrated to the instrumental data such that the standard deviation and mean of the reconstruction and target during the calibration interval are equal. As well as obtaining reconstructions for each nest of available proxies, we compute stitched reconstructions S c ( t ) for each calibration period c , which include at each time step the reconstructed data for the proxy nest with maximum coefficient of efficiency 58 , 59 (Supplementary Information section 3.1 ). This procedure is performed for contiguous calibration intervals between 60 and 80 years duration between 1900 and 1995, with interval width and location increments of two years, reserving the remaining data in the overlapping period for independent evaluation, and for all proxy nests. The reconstruction error is modelled with a lag-one autoregressive process fitted to the residuals. We evaluate the capacity of our reconstruction method to achieve spurious skill from overfitting by performing a test in which we replace the coral data with synthetic noise (Supplementary Information section 3.2i ). We find that reconstructions based on synthetic noise achieve extremely low or zero skill and as more noise principal components are included in the regression, the evaluation metrics indicate declining skill. Our reconstruction and evaluation methods therefore guard against the potential for spurious skill.
Our reconstruction method is further evaluated by using a pseudo-proxy modelling approach based on the Community Earth System Model (CESM) Last Millennium Experiment (LME) 60 , for which there are 13 full-forcing ensemble members covering the period 850–2005. We use the pseudo-proxy reconstructions to evaluate our reconstruction method and coral network in a fully coupled climate-model environment. We form pseudo-proxies by extracting from each LME ensemble member the SST and sea surface salinity (SSS) from the 1.5° × 1.5° grid cell located nearest to our coral data. We then apply proxy system models in the form of linear regression models, basing δ 18 O on both SST and SSS, and Sr/Ca on SST only (Supplementary Information section 3.3 ). We set the spatial and temporal availability of the pseudo-coral network to match that of the coral network. We then apply our PCR reconstruction and evaluation procedure to the pseudo-proxy network, taking advantage of the availability of the modelled Coral Sea SSTA data across the multi-century period of 1618–2005, which allows for the evaluation of the pseudo-proxy reconstruction over this entire time period. We first test our method using a ‘perfect proxy’ approach (with no proxy measurement error) before superimposing synthetic noise on the pseudo-proxy time series, evaluating our methodology at two separate levels of measurement error, quantified by signal-to-noise ratios of 1.0 and 4.0. The evaluation metrics for these tests indicate that our coral network and reconstruction method obtain skilful reconstructions of Coral Sea SSTAs in the climate-model environment (Supplementary Figs. 17b , 18 , 20b , 21 , 22b and 23 ).
We use two multi-century five-year-resolution coral series from the central GBR 23 , 24 (Fig. 2b and Supplementary Fig. 24 ) and a network of sub-annual and annual resolution modern coral series (dated from 1900 onwards but not covering the full 1900–1995 period) from 44 sites in the GBR (Supplementary Information section 4.2 ) for independent evaluation of coral-derived evidence for warming in the region. We estimate five-year GBR SSTAs (Fig. 2b ) by aligning the post-1900 mean and variance of the proxy and instrumental (ERSSTv5) data.
Of the 22 available coral series, 16 are records of δ 18 O, a widely used measure of the ratio of the stable isotopes 18 O and 16 O. In the tropical Pacific Ocean, δ 18 O is significantly correlated with SST 61 , 62 , 63 , 64 . Coral δ 18 O is also sensitive to the δ 18 O of seawater 65 , which can reflect advection of different water masses and/or changes in freshwater input, such as from riverine sources or precipitation, which in turn co-vary with SSS. Thus, it is generally considered that the main non-SST contributions to coral δ 18 O are processes that co-vary with SSS 62 , 66 . Our methodology minimizes the influence of non-temperature impacts on the reconstruction by exploiting the contrast in spatial heterogeneity between SST and SSS in January–March (Supplementary Information section 5.1 ). SSS is spatially inhomogeneous in the tropical Pacific 66 , 67 , leading to low coherence in SSS signals across our coral network. By contrast, the strong and coherent SST signal across our coral network locations and the Coral Sea region leads to principal components that are strongly representative of SST variations. This produces a skilful reconstruction of SST, as determined by evaluation against independent observations, and low correlations with SSS across the Coral Sea region (Supplementary Fig. 31 ).
Although the likelihood of non-SST influences on our SST reconstruction is low, we nonetheless test the sensitivity of our reconstruction and its associated interpretations to the possibility of these influences on the coral data. The tests compute the correlations between our best-estimate SSTA reconstruction (highest coefficient of efficiency) and observations of SSS, along with a series of additional reconstructions based on subsets of our coral network. The correlations between our highest coefficient of efficiency January–March Coral Sea SSTA reconstruction and January–March SSS are mapped for the Coral Sea and its neighbouring domain using three instrumental SSS datasets (Supplementary Fig. 31 ). Correlations are not statistically significant over most of the domain. Noting differing spatial correlation patterns between the instrumental SSS datasets 68 , which also cover different time periods (Supplementary Information section 5.1 ), we undertake six sensitivity tests using subsets of the coral network (Supplementary Information section 5.2 ). We use the following combinations of coral series: (1) the full network of 22 δ 18 O and Sr/Ca series (Figs. 2a and 3 ); (2) a subset of the six available Sr/Ca series (Supplementary Figs. 32 – 33 ), to test how the reconstruction is influenced by the inclusion of coral δ 18 O records; (3) a fixed nest subset of the five longest coral series, extending back to at least 1700 (Supplementary Figs. 34 – 35 ), to test for the potential influence of combining series of differing lengths (from our splicing of portions of the best reconstructions from each nest); (4) a subset of the ten coral series that are most strongly correlated with the target (Supplementary Figs. 36 and 37 ), to test how our reconstruction is influenced by the inclusion of coral series that are less strongly correlated with our target; (5) a subset of coral series that excludes the six records that are reported to potentially include biological mediation or non-climatic effects, or have low correlation with the target (Supplementary Figs. 38 and 39 ), to test their influence on the reconstruction; and (6) a network perturbation test comprising 22 separate subsets of proxies, in which proxy records are added incrementally in order of highest to lowest correlation with the target, starting with a single coral series and increasing the number of included proxies to all 22 series in our network (Supplementary Information section 5.2.5 ), to systematically quantify the influence of gradually including more coral datasets on our reconstruction and its interpretations.
The evaluation metrics (Fig. 2c and Supplementary Figs. 32b , 34b , 36b and 38b ) indicate a skilful reconstruction back to 1618 for the reconstructions based on the Full, Sr/Ca only, Long, Best-10 and OmitBioMed networks. These reconstructions explain 82.7%, 80.6%, 77.6%, 79.8% and 80.4% (R-squared values) of the variance in January–March SSTAs, respectively, in the independent evaluation periods (using ERSSTv5b). All coral subsets in the network perturbation test produce skilful reconstructions (Supplementary Fig. 40 ). The highest-skill reconstructions for all subsets in the network perturbation test align with our key interpretations (Supplementary Figs. 41 and 42 ). Together, our sensitivity tests show that the coral network, observational data and reconstruction methodology are a sound basis for reconstructing Coral Sea January–March SSTAs in past centuries and contextualizing recent high-SST events ( Supplementary Information ).
The multi-model attribution analysis used here is based on simulations from CMIP6. We analyse simulations from the historical experiment (including natural and anthropogenic influences for 1850–2014) and the historical-natural experiment (natural-only forcings for 1850–2014). We select climate models for which monthly surface temperature is available in at least three historical and historical-natural simulations (Supplementary Table 5 ). All model simulations are interpolated to a common regular 1.5° × 1.5° latitude–longitude grid. January–March SSTAs relative to 1961–90 are calculated for each simulation. The full historical all-forcings ensemble is composed of 14 models with 268 simulations for 1850–2014. The natural-only ensemble is composed of the same 14 models with 95 individual simulations. A subset of climate models in the CMIP6 ensemble are considered by the science community to be ‘too hot’, simulating warming in response to increased atmospheric carbon dioxide concentrations that is larger than that supported by independent evidence 31 . We omit these models from our analysis by including only models with a transient climate response in the ‘likely’ range 31 of 1.4–2.2 °C. Our results are not strongly sensitive to this selection (Supplementary Information section 6.3 ). The ten remaining models yield a total of 25,410 years from 154 historical ensemble members and 7,095 years from 43 historical-natural ensemble members. We weight the models equally in our analysis using bootstrap sampling. We report linear trends based on simple linear regression models fitted with ordinary least squares. The statistical significance of linear trends is assessed using the Spearman’s rank correlation test 69 .
We assess the anthropogenic influence on SSTAs in the Coral Sea region by starting with the assumption that any anthropogenic influence on SSTAs in the Coral Sea is indistinguishable from natural variability at the commencement of the model experiments. We measure the impact of anthropogenic influence on the climate in the region using a signal-to-noise approach 32 , 70 . We calculate the anthropogenic ‘signal’ as the mean of the difference between the smoothed (using a 41-year Lowess filter) modelled historical Coral Sea SSTA and the mean smoothed modelled historical-natural SSTA. Our ‘noise’ is the standard deviation of the difference between the modelled historical SSTA and its smoothed time series (Supplementary Information section 6 ).
Methods additionally rely on Supplementary Information and refs. 71 , 72 , 73 , 74 , 75 , 76 , 77 , 78 , 79 , 80 , 81 , 82 , 83 , 84 , 85 , 86 , 87 , 88 , 89 , 90 , 91 , 92 , 93 , 94 , 95 , 96 , 97 , 98 , 99 , 100 , 101 , 102 , 103 , 104 .
The ERSSTv5 instrumental SST data are available from the US National Oceanic and Atmospheric Administration at https://psl.noaa.gov/data/gridded/data.noaa.ersst.v5.html . The HadISST1.1 data are available from the UK Met Office at https://www.metoffice.gov.uk/hadobs/hadisst/ . The original coral palaeoclimate data are available at the links provided in Supplementary Table 2 . Land areas for maps are obtained from the Mapping Toolbox v.23.2 in Matlab v.2023b and the Global Self-consistent, Hierarchical, High-resolution Geography (GSHHS) Database at https://www.soest.hawaii.edu/pwessel/gshhg/ through the m_map toolbox by R. Pawlowicz, available at https://www.eoas.ubc.ca/%7Erich/map.html . Prepared data from the coral geochemical series, reconstructions and climate models that support the findings of this study are available at: https://doi.org/10.24433/CO.4883292.v1 .
The code that supports the findings of this study is available and can be run at : https://doi.org/10.24433/CO.4883292.v1 .
Australian Institute of Marine Science. Long-Term Monitoring Program. https://www.aims.gov.au/research-topics/monitoring-and-discovery/monitoring-great-barrier-reef/long-term-monitoring-program (2024).
Hughes, T. P. et al. Global warming and recurrent mass bleaching of corals. Nature 543 , 373–377 (2017).
Article ADS CAS PubMed Google Scholar
Hoegh-Guldberg, O. Climate change, coral bleaching and the future of the world’s coral reefs. Mar. Freshw. Res. 50 , 839–866 (1999).
Google Scholar
Hughes, T. P. et al. Global warming transforms coral reef assemblages. Nature 556 , 492–496 (2018).
Great Barrier Reef Marine Park Authority. Great Barrier Reef Outlook Report 2019 (Great Barrier Reef Marine Park Authority, 2019).
Hughes, T. P. et al. Coral reefs in the Anthropocene. Nature 546 , 82–90 (2017).
Davis, K. L., Colefax, A. P., Tucker, J. P., Kelaher, B. P. & Santos, I. R. Global coral reef ecosystems exhibit declining calcification and increasing primary productivity. Commun. Earth Environ. 2 , 105 (2021).
Article ADS Google Scholar
Westcott, D. A. et al. Relative efficacy of three approaches to mitigate Crown-of-Thorns Starfish outbreaks on Australia’s Great Barrier Reef. Sci. Rep. 10 , 12594 (2020).
Article ADS CAS PubMed PubMed Central Google Scholar
Mellin, C. et al. Spatial resilience of the Great Barrier Reef under cumulative disturbance impacts. Glob. Chang. Biol. 25 , 2431–2445 (2019).
Article ADS PubMed Google Scholar
Jackson, J. B. C. et al. Historical overfishing and the recent collapse of coastal ecosystems. Science 293 , 629–637 (2001).
Article CAS PubMed Google Scholar
Hoegh-Guldberg, O. & Smith, G. J. The effect of sudden changes in temperature, light and salinity on the population density and export of zooxanthellae from the reef corals Stylophora pistillata Esper and Seriatopora hystrix Dana. J. Exp. Mar. Biol. Ecol. 129 , 279–303 (1989).
Article Google Scholar
DeCarlo, T. M. et al. Acclimatization of massive reef-building corals to consecutive heatwaves. Proc. Biol. Sci. 286 , 20190235 (2019).
PubMed PubMed Central Google Scholar
McGowan, H. & Theobald, A. ENSO weather and coral bleaching on the Great Barrier Reef, Australia. Geophys. Res. Lett. 44 , 10,601–10,607 (2017).
Zhao, W., Huang, Y., Siems, S. & Manton, M. The role of clouds in coral bleaching events over the Great Barrier Reef. Geophys. Res. Lett. 48 , e2021GL093936 (2021).
Oxley, W. G., Emslie, M., Muir, P. & Thompson, A. Marine Surveys Undertaken in the Lihou Reef National Nature Reserve (Australian Institute of Marine Science, 2004).
DeCarlo, T. M. & Harrison, H. B. An enigmatic decoupling between heat stress and coral bleaching on the Great Barrier Reef. PeerJ 7 , e7473 (2019).
Article PubMed PubMed Central Google Scholar
UNESCO World Heritage Committee. Extended 44th Session of the World Heritage Committee, Fuzhou (China) 16–31 July 2021 . Draft decision 44 COM 7B.90. https://whc.unesco.org/document/188005 (UNESCO, 2021).
UNESCO World Heritage Committee. Extended 45th Session of the World Heritage Committee, Riyadh (Saudi Arabia) 10–25 September 2023 . Decision 45 COM 7B.13. https://whc.unesco.org/document/199654 (UNESCO, 2023).
IPCC. Summary for Policymakers. In Climate Change 2021: The Physical Science Basis. Contribution of Working Group I to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change (eds Masson-Delmotte, V. et al.) (Cambridge Univ. Press, 2021).
Kamenos, N. A. & Hennige, S. J. Reconstructing four centuries of temperature-induced coral bleaching on the Great Barrier Reef. Front. Mar. Sci. 5 , 283 (2018).
Hoegh-Guldberg, O. et al. Commentary: reconstructing four centuries of temperature-induced coral bleaching on the Great Barrier Reef. Front. Mar. Sci. 6 , 86 (2019).
DeCarlo, T. M. Commentary: reconstructing four centuries of temperature-induced coral bleaching on the Great Barrier Reef. Front. Mar. Sci. 7 , 30 (2020).
Hendy, E. J. et al. Abrupt decrease in tropical Pacific sea surface salinity at end of Little Ice Age. Science 295 , 1511–1514 (2002).
Calvo, E. et al. Interdecadal climate variability in the Coral Sea since 1708 A.D. Palaeogeogr. Palaeoclimatol. Palaeoecol. 248 , 190–201 (2007).
Zinke, J. et al. North Flinders Reef (Coral Sea, Australia) Porites sp. corals as a candidate global boundary stratotype section and point for the Anthropocene series. Anthropocene Rev. 10 , 201–224 (2023).
Spady, B. L. et al. Global Coral Bleaching Database (NCEI Accession 0228498) (NOAA National Centers for Environmental Information, 2022); https://www.ncei.noaa.gov/archive/accession/0228498 .
Huang, B. et al. Extended reconstructed sea surface temperature, Version 5 (ERSSTv5): Upgrades, validations, and intercomparisons. J. Clim. 30 , 8179–8205 (2017).
Rayner, N. A. et al. Global analyses of sea surface temperature, sea ice, and night marine air temperature since the late nineteenth century. J. Geophys. Res. Atmos. 108 , 4407 (2003).
Eyring, V. et al. Overview of the Coupled Model Intercomparison Project Phase 6 (CMIP6) experimental design and organization. Geosci. Model Dev. 9 , 1937–1958 (2016).
Gillett, N. P. et al. The Detection and Attribution Model Intercomparison Project (DAMIP v1.0) contribution to CMIP6. Geosci. Model Dev. 9 , 3685–3697 (2016).
Hausfather, Z., Marvel, K., Schmidt, G. A., Nielsen-Gammon, J. W. & Zelinka, M. Climate simulations: recognize the ‘hot model’ problem. Nature 605 , 26–29 (2022).
Hawkins, E. et al. Observed emergence of the climate change signal: from the familiar to the unknown. Geophys. Res. Lett. 47 , e2019GL086259 (2020).
Gagan, M. K. et al. Temperature and surface-ocean water balance of the mid-Holocene tropical western Pacific. Science 279 , 1014–1018 (1998).
Arzey, A. K. et al. Coral skeletal proxy records database for the Great Barrier Reef, Australia. Preprint at Earth Syst. Sci. Data https://doi.org/10.5194/essd-2024-159 (2024).
Brenner, L. D. et al. Coral record of Younger Dryas Chronozone warmth on the Great Barrier Reef. Paleoceanogr. Paleoclimatol. 35 , e2020PA003962 (2020).
Furnas, M. J. & Mitchell, A. W. Nutrient inputs into the central Great Barrier Reef (Australia) from subsurface intrusions of Coral Sea waters: a two-dimensional displacement model. Cont. Shelf Res. 16 , 1127–1148 (1996).
Wolanski, E., Andutta, F., Deleersnijder, E., Li, Y. & Thomas, C. J. The Gulf of Carpentaria heated Torres Strait and the Northern Great Barrier Reef during the 2016 mass coral bleaching event. Estuar. Coast. Shelf Sci. 194 , 172–181 (2017).
Oliver, E. C. J. & Holbrook, N. J. Extending our understanding of South Pacific gyre ‘spin-up’: modeling the East Australian Current in a future climate. J. Geophys. Res. Oceans 119 , 2788–2805 (2014).
DeCarlo, T. M. et al. Nutrient-supplying ocean currents modulate coral bleaching susceptibility. Sci. Adv. 6 , eabc5493 (2020).
Article ADS PubMed PubMed Central Google Scholar
Wiedenmann, J. et al. Nutrient enrichment can increase the susceptibility of reef corals to bleaching. Nat. Clim. Chang. 3 , 160–164 (2013).
Article ADS CAS Google Scholar
Chan, D. & Huybers, P. Correcting observational biases in sea surface temperature observations removes anomalous warmth during World War II. J. Clim. 34 , 4585–4602 (2021).
Chan, D., Kent, E. C., Berry, D. I. & Huybers, P. Correcting datasets leads to more homogeneous early-twentieth-century sea surface warming. Nature 571 , 393–397 (2019).
Hughes, T. P. et al. Spatial and temporal patterns of mass bleaching of corals in the Anthropocene. Science 359 , 80–83 (2018).
Hoegh-Guldberg, O. et al. Chapter 3: Impacts of 1.5°C global warming on natural and human systems. In Global Warming of 1.5°C (eds Masson-Delmotte, V. et al.) (IPCC, 2018).
Meinshausen, M. et al. Realization of Paris Agreement pledges may limit warming just below 2 °C. Nature 604 , 304–309 (2022).
Matthews, H. D. & Wynes, S. Current global efforts are insufficient to limit warming to 1.5 °C. Science 376 , 1404–1409 (2022).
Coles, S. L. et al. Evidence of acclimatization or adaptation in Hawaiian corals to higher ocean temperatures. PeerJ 6 , e5347 (2018).
Hughes, T. P., Baird, A. H., Morrison, T. H. & Torda, G. Principles for coral reef restoration in the anthropocene. One Earth 6 , 656–665 (2023).
Logan, C. A., Dunne, J. P., Ryan, J. S., Baskett, M. L. & Donner, S. D. Quantifying global potential for coral evolutionary response to climate change. Nat. Clim. Chang. 11 , 537–542 (2021).
Dixon, A. M., Forster, P. M., Heron, S. F., Stoner, A. M. K. & Beger, M. Future loss of local-scale thermal refugia in coral reef ecosystems. PLOS Clim. 1 , e0000004 (2022).
Coplen, T. B. Discontinuance of SMOW and PDB. Nature 375 , 285 (1995).
van Albada, S. J. & Robinson, P. A. Transformation of arbitrary distributions to the normal distribution with application to EEG test-retest reliability. J. Neurosci. Methods 161 , 205–211 (2007).
Article PubMed Google Scholar
Emile-Geay, J. & Tingley, M. Inferring climate variability from nonlinear proxies: application to palaeo-ENSO studies. Clim. Past 12 , 31–50 (2016).
Barnes, D. J., Taylor, R. B. & Lough, J. M. On the inclusion of trace materials into massive coral skeletons. Part II: distortions in skeletal records of annual climate cycles due to growth processes. J. Exp. Mar. Biol. Ecol. 194 , 251–275 (1995).
Article CAS Google Scholar
Gagan, M. K., Dunbar, G. B. & Suzuki, A. The effect of skeletal mass accumulation in Porites on coral Sr/Ca and δ 18 O paleothermometry. Paleoceanogr. Paleoclimatol. 27 , PA1203 (2012).
ADS Google Scholar
Schneider, T. Analysis of incomplete climate data: estimation of mean values and covariance matrices and imputation of missing values. J. Clim. 14 , 853–871 (2001).
PAGES 2k Consortium. Consistent multidecadal variability in global temperature reconstructions and simulations over the Common Era. Nat. Geosci. 12 , 643–649 (2019).
Cook, E. R., Briffa, K. R. & Jones, P. D. Spatial regression methods in dendroclimatology: a review and comparison of two techniques. Int. J. Climatol. 14 , 379–402 (1994).
Nash, J. E. & Sutcliffe, J. V. River flow forecasting through conceptual models part I − a discussion of principles. J. Hydrol. 10 , 282–290 (1970).
Otto-Bliesner, B. L. et al. Climate variability and change since 850 CE: an ensemble approach with the Community Earth System Model. Bull. Am. Meteorol. Soc. 97 , 735–754 (2016).
Evans, M. N., Kaplan, A. & Cane, M. A. Optimal sites for coral-based reconstruction of global sea surface temperature. Paleoceanogr. Paleoclimatol. 13 , 502–516 (1998).
Russon, T., Tudhope, A. W., Hegerl, G. C., Collins, M. & Tindall, J. Inter-annual tropical Pacific climate variability in an isotope-enabled CGCM: Implications for interpreting coral stable oxygen isotope records of ENSO. Clim. Past 9 , 1543–1557 (2013).
PAGES Hydro2k Consortium. Comparing proxy and model estimates of hydroclimate variability and change over the Common Era. Clim. Past 13 , 1851–1900 (2017).
Freund, M. B. et al. Higher frequency of Central Pacific El Niño events in recent decades relative to past centuries. Nat. Geosci. 12 , 450–455 (2019).
Gagan, M. K. et al. New views of tropical paleoclimates from corals. Quat. Sci. Rev. 19 , 45–64 (2000).
Thompson, D. M., Ault, T. R., Evans, M. N., Cole, J. E. & Emile-Geay, J. Comparison of observed and simulated tropical climate trends using a forward model of coral δ 18 O. Geophys. Res. Lett. 38 , L14706 (2011).
LeGrande, A. N. & Schmidt, G. A. Global gridded data set of the oxygen isotopic composition in seawater. Geophys. Res. Lett. 33 , L12604 (2006).
Reed, E. V., Thompson, D. M. & Anchukaitis, K. J. Coral-based sea surface salinity reconstructions and the role of observational uncertainties in inferred variability and trends. Paleoceanogr. Paleoclimatol. 37 , e2021PA004371 (2022).
Khaliq, M. N., Ouarda, T. B. M. J., Gachon, P., Sushama, L. & St-Hilaire, A. Identification of hydrological trends in the presence of serial and cross correlations: A review of selected methods and their application to annual flow regimes of Canadian rivers. J. Hydrol. 368 , 117–130 (2009).
Mahlstein, I., Hegerl, G. & Solomon, S. Emerging local warming signals in observational data. Geophys. Res. Lett. 39 , L21711 (2012).
Freeman, E. et al. ICOADS Release 3.0: a major update to the historical marine climate record. Int. J. Climatol. 37 , 2211–2232 (2017).
Huang, B. et al. Uncertainty estimates for sea surface temperature and land surface air temperature in NOAAGlobalTemp version 5. J. Clim. 33 , 1351–1379 (2020).
Druffel, E. R. M. & Griffin, S. Variability of surface ocean radiocarbon and stable isotopes in the southwestern Pacific. J. Geophys. Res. 104 , 23607–23613 (1999).
DeLong, K. L., Quinn, T. M., Taylor, F. W., Lin, K. & Shen, C.-C. Sea surface temperature variability in the southwest tropical Pacific since AD 1649. Nat. Clim. Change 2 , 799–804 (2012).
Quinn, T. et al. A multicentury stable isotope record from a New Caledonia coral: Interannual and decadal SST variability in the southwest Pacific since 1657. Paleoceanography 13 , 412–426 (1998).
Quinn, T. M., Crowley, T. J. & Taylor, F. W. New stable isotope results from a 173-year coral from Espiritu Santo, Vanuatu. Geophys. Res. Lett. 23 , 3413–3416 (1996).
Alibert, C. & Kinsley, L. A 170-year Sr/Ca and Ba/Ca coral record from the western Pacific warm pool: 1. What can we learn from an unusual coral record? J. Geophys. Res. Oceans 113 , C04008 (2008).
Tudhope, A. W. et al. Variability in the El Niño-Southern Oscillation through a glacial-interglacial cycle. Science 291 , 1511–1517 (2001).
Urban, F. E., Cole, J. E. & Overpeck, J. T. Influence of mean climate change on climate variability from a 155-year tropical Pacific coral record. Nature 407 , 989–993 (2000).
Guilderson, T. P. & Schrag, D. P. Reliability of coral isotope records from the western Pacific warm pool: A comparison using age-optimized records. Paleoceanography 14 , 457–464 (1999).
Quinn, T. M., Taylor, F. W. & Crowley, T. J. Coral-based climate variability in the Western Pacific Warm Pool since 1867. J. Geophys. Res. 111 , C11006 (2006).
Gorman, M. K. et al. A coral-based reconstruction of sea surface salinity at Sabine Bank, Vanuatu from 1842 to 2007 CE. Paleoceanography 27 , PA3226 (2012).
Bagnato, S., Linsley, B. K., Howe, S. S., Wellington, G. M. & Salinger, J. Coral oxygen isotope records of interdecadal climate variations in the South Pacific Convergence Zone region. Geochem. Geophys. Geosyst. 6 , Q06001 (2005).
Linsley, B. K. et al. Tracking the extent of the South Pacific Convergence Zone since the early 1600s. Geochem. Geophys. Geosyst. 7 , Q05003 (2006).
Cole, J. E., Fairbanks, R. G. & Shen, G. T. Recent variability in the Southern Oscillation: Isotopic results from a Tarawa Atoll coral. Science 260 , 1790–1793 (1993).
Dassié, E. P. et al. A Fiji multi-coral δ 18 O composite approach to obtaining a more accurate reconstruction of the last two-centuries of the ocean-climate variability in the South Pacific Convergence Zone region. Paleoceanography 29 , 1196–1213 (2014).
Carton, J. A., Chepurin, G. A. & Chen, L. SODA3: A new ocean climate reanalysis. J. Clim. 31 , 6967–6983 (2018).
Zuo, H., Balmaseda, M. A., Tietsche, S., Mogensen, K. & Mayer, M. The ECMWF operational ensemble reanalysis-analysis system for ocean and sea ice: A description of the system and assessment. Ocean Sci. 15 , 779–808 (2019).
Cheng, L. et al. Improved estimates of changes in upper ocean salinity and the hydrological cycle. J. Clim. 33 , 10357–10381 (2020).
Thompson, D. M. et al. Identifying hydro‐sensitive coral δ18O records for improved high‐resolution temperature and salinity reconstructions. Geophys. Res. Lett. 49 , e2021GL096153 (2022).
Wu, Y., Fallon, S. J., Cantin, N. E. & Lough, J. M. Assessing multiproxy approaches (Sr/Ca, U/Ca, Li/Mg, and B/Mg) to reconstruct sea surface temperature from coral skeletons throughout the Great Barrier Reef. Sci. Total Environ. 786 , 147393 (2021).
Sadler, J., Webb, G. E., Leonard, N. D., Nothdurft, L. D. & Clark, T. R. Reef core insights into mid-Holocene water temperatures of the southern Great Barrier Reef. Paleoceanography 31 , 1395–1408 (2016).
Roche, R. C. et al. Mid-Holocene sea surface conditions and riverine influence on the inshore Great Barrier Reef. Holocene 24 , 885–897 (2014).
Reed, E. V., Cole, J. E., Lough, J. M., Thompson, D. & Cantin, N. E. Linking climate variability and growth in coral skeletal records from the Great Barrier Reef. Coral Reefs 38 , 29–43 (2019).
Razak, T. B. et al. Use of skeletal Sr/Ca ratios to determine growth patterns in a branching coral Isopora palifera. Mar. Biol. 164 , 96 (2017).
Marshall, J. F. Decadal-scale, High Resolution Records of Sea Surface Temperature in the Eastern Indian and South Western Pacific Oceans from Proxy Records of the Strontium/calcium Ratio of Massive Porites Corals PhD thesis, Australian National Univ. (2000).
Marshall, J. F. & McCulloch, M. T. An assessment of the Sr/Ca ratio in shallow water hermatypic corals as a proxy for sea surface temperature. Geochim. Cosmochim. Acta 66 , 3263–3280 (2002).
Gagan, M. K. et al. Coral oxygen isotope evidence for recent groundwater fluxes to the Australian Great Barrier Reef. Geophys. Res. Lett. 29 , 43-1–43-4 (2002).
D’Olivo, J. P., Sinclair, D. J., Rankenburg, K. & McCulloch, M. T. A universal multi-trace element calibration for reconstructing sea surface temperatures from long-lived Porites corals: Removing ‘vital-effects’. Geochim. Cosmochim. Acta 239 , 109–135 (2018).
Fallon, S. J., McCulloch, M. T. & Alibert, C. Examining water temperature proxies in Porites corals from the Great Barrier Reef: a cross-shelf comparison. Coral Reefs 22 , 389–404 (2003).
Brenner, L. D., Linsley, B. K. & Potts, D. C. A modern Sr/Ca-δ 18 O-sea surface temperature calibration for Isopora corals on the Great Barrier Reef. Paleoceanography 32 , 182–194 (2017).
Alibert, C. et al. Source of trace element variability in Great Barrier Reef corals affected by the Burdekin flood plumes. Geochim. Cosmochim. Acta 67 , 231–246 (2003).
Murty, S. A. et al. Spatial and temporal robustness of Sr/Ca-SST calibrations in Red Sea corals: Evidence for influence of mean annual temperature on calibration slopes. Paleoceanogr. Paleoclimatol. 33 , 443–456 (2018).
Sayani, H. R., Cobb, K. M., DeLong, K., Hitt, N. T. & Druffel, E. R. M. Intercolony δ 18 O and Sr/Ca variability among Porites spp. corals at Palmyra Atoll: Toward more robust coral-based estimates of climate. Geochem. Geophys. Geosyst. 20 , 5270–5284 (2019).
Otto, F. E. L. Geert Jan van Oldenborgh 1961–2021. Nat. Clim. Chang. 11 , 1017 (2021).
Download references
We acknowledge the originators of the coral data cited in Supplementary Tables 1 and 2 ; S. E. Perkins-Kirkpatrick and the deceased G. J. van Oldenborgh 105 for contributions to an earlier version of this manuscript; E. P. Dassié and J. Zinke for discussions and data; R. Neukom for advice on an earlier version of the reconstruction code; and B. Trewin and K. Braganza for advice about the Bureau of Meteorology GBR SST time series. B.J.H. and H.V.M. acknowledge support from an Australian Research Council (ARC) SRIEAS grant, Securing Antarctica’s Environmental Future (SR200100005), and ARC Discovery Project DP200100206. A.D.K. acknowledges support from an ARC DECRA (DE180100638) and the Australian government’s National Environmental Science Program. B.J.H. and A.D.K. acknowledge an affiliation with the ARC Centre of Excellence for Climate Extremes (CE170100023). H.V.M. acknowledges support from an ARC Future Fellowship (FT140100286). A.K.A. acknowledges support from an Australian government research training program scholarship and an AINSE postgraduate research award. Funding was provided to B.K.L. by the Vetlesen Foundation through a gift to the Lamont-Doherty Earth Observatory. Grants to B.K.L. enabled the generation of coral oxygen isotope and Sr/Ca data from Fiji that were used in our reconstruction (US National Science Foundation OCE-0318296 and ATM-9901649 and US National Oceanic and Atmospheric Administration NA96GP0406). We acknowledge the support of the NCI facility in Australia and the World Climate Research Programme’s working group on coupled modelling, which is responsible for CMIP. We thank the climate-modelling groups for producing and making available their model output. For CMIP, the US Department of Energy’s Program for Climate Model Diagnosis and Intercomparison provided coordinating support and led the development of software infrastructure in partnership with the Global Organisation for Earth System Science Portals.
Authors and affiliations.
Environmental Futures, School of Earth, Atmospheric and Life Sciences, University of Wollongong, Wollongong, New South Wales, Australia
Benjamin J. Henley, Helen V. McGregor & Ariella K. Arzey
Securing Antarctica’s Environmental Future, University of Wollongong, Wollongong, New South Wales, Australia
School of Agriculture, Food and Ecosystem Sciences, University of Melbourne, Parkville, Victoria, Australia
Benjamin J. Henley
School of Geography, Earth and Atmospheric Sciences, University of Melbourne, Parkville, Victoria, Australia
Andrew D. King & David J. Karoly
ARC Centre of Excellence for Climate Extremes, University of Melbourne, Parkville, Victoria, Australia
Andrew D. King
School of the Environment, The University of Queensland, Brisbane, Queensland, Australia
Ove Hoegh-Guldberg
Australian Institute of Marine Science, Townsville, Queensland, Australia
Janice M. Lough
ARC Centre of Excellence for Coral Reef Studies and School of Earth Sciences, University of Western Australia, Crawley, Western Australia, Australia
Thomas M. DeCarlo
Department of Earth and Environmental Sciences, Tulane University, New Orleans, LA, USA
Lamont-Doherty Earth Observatory of Columbia University, Palisades, NY, USA
Braddock K. Linsley
You can also search for this author in PubMed Google Scholar
B.J.H., H.V.M. and A.D.K. conceived the study and developed the methodology. B.J.H. did most of the analysis. A.K.A. contributed analysis of modern coral data (Supplementary Information section 4.2 ). T.M.D. contributed analysis of instrumental data coverage (Supplementary Information section 1.2 ). B.K.L. contributed sub-annual coral data. B.J.H. and H.V.M. led the preparation of the manuscript, with contributions from A.D.K., O.H.-G., A.K.A., D.J.K., J.M.L., T.M.D. and B.K.L. Generative artificial intelligence was not used in any aspect of this study or manuscript.
Correspondence to Benjamin J. Henley .
Competing interests.
The authors declare no competing interests.
Peer review information.
Nature thanks Simon Michel, Miriam Pfeiffer, Claudia Tebaldi and the other, anonymous, reviewer(s) for their contribution to the peer review of this work.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary information ., rights and permissions.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ .
Reprints and permissions
Cite this article.
Henley, B.J., McGregor, H.V., King, A.D. et al. Highest ocean heat in four centuries places Great Barrier Reef in danger. Nature 632 , 320–326 (2024). https://doi.org/10.1038/s41586-024-07672-x
Download citation
Received : 02 November 2022
Accepted : 04 June 2024
Published : 07 August 2024
Issue Date : 08 August 2024
DOI : https://doi.org/10.1038/s41586-024-07672-x
Anyone you share the following link with will be able to read this content:
Sorry, a shareable link is not currently available for this article.
Provided by the Springer Nature SharedIt content-sharing initiative
Great barrier reefâs temperature soars to 400-year high.
Nature (2024)
By submitting a comment you agree to abide by our Terms and Community Guidelines . If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.
Sign up for the Nature Briefing: Anthropocene newsletter — what matters in anthropocene research, free to your inbox weekly.
COMMENTS
Coral Reefs is a multidisciplinary journal focusing on diverse aspects of coral reef science and management. Official publication of the International Coral Reef Society. Publishes analytical and theoretical papers, full-length reports, reviews, and short notes. Covers broad fields such as reef structure, sedimentology, biogeochemical cycles ...
Approaches for projecting coral reef futures. A search of articles in the peer-reviewed literature found 79 studies published between 1999 and 2023 that modeled coral reef responses to future ...
Research Open Access 29 Feb 2024 Communications Biology Volume: 7, P: 246 Coral thermal stress and bleaching enrich and restructure reef microbial communities via altered organic matter exudation
Across the entire Great Barrier Reef, 34.8% of individual reefs experienced at least 4 °C-weeks, and 20.7% of reefs were exposed to 8 °C-weeks or more of accumulated heat stress in 2016 (Fig. 1b).
This paper acts as a scientific reference for the mapping framework and summary statistics for the Allen Coral Atlas. In summary, for shallow tropical reefs (−30° to 30° latitude) at 5-m-pixel resolution, we present: (1) their global extent and distribution, (2) the distribution of 11 geomorphic classes, and (3) the distribution of 6 benthic classes (Figure 1).
Coral reefs are an effective natural barrier from waves (1, 2), which has allowed human populations to settle on tropical coasts and reef islands ().The predicted future increase of sea level is often seen as the most crucial threat to these low-lying areas ().However, the interaction between higher sea levels and wave climate on coral reef flats is likely to be an important mechanism ...
For the purposes of this paper, we use the term coral reef to refer to limestone structures that are built by reef-building corals throughout ocean depths, including in shallow water (<40 m), at mesophotic depths (∼40-200 m), and in the deep sea (>200 m). While this definition is somewhat broader than many interpretations of this term, it ...
More research is needed to understand interactions between stressors, especially those which most influence other stressors, to enable more effective management prioritisation to mitigate against the most severe impacts of climate change and human activities on coral reef ecosystems (Ban et al., 2014).
Warm-water coral reefs are largely dependent on the physical and chemical changes occurring in the surface of the ocean, whereas cold-water reef systems are tied relatively more to the broad scale conditions of the bulk ocean (Freiwald et al., 2004; Eakin C. M. et al., 2010).In this respect, there are likely to be differences in terms of the rate and characteristics of the changes that are ...
The present paper describes overview of coral resources in Indian seas, their biology and taxonomy, anthropogenic stress on coral reefs, conservation and research efforts being put by various ...
Coral reefs face some of the greatest and most urgent threats from climate change. Marine heatwaves that cause mass coral bleaching are increasing in frequency and spatial distribution as a result of human-induced climatic change, jeopardizing the survival of thousands of reef-associated organisms and the critical services they provide to nature and people (1, 2).
MPAs range from ineffective "paper parks," to multiple-use areas with varying degrees of protection, to marine reserves, or no-take areas (NTAs). ... and go beyond the current emphasis on routine monitoring and mapping. Indeed, most coral reef research is parochial and short-term, and provides little insight into global or longer-term changes.
Introduction. Coral reef ecosystems have suffered an unprecedented loss of habitat-forming hard corals in the past few decades [1-5].Reefs are subject to a suite of chronic and acute anthropogenic disturbances including declining water quality, destructive fishing practices, over-harvesting of reef species, and outbreaks of coral predators and coral disease, however in the past two decades ...
Abstract. Addressing the global decline of coral reefs requires effective actions from managers, policymakers and society as a whole. Coral reef scientists are therefore challenged with the task ...
Coral reef ecosystems are biodiversity hotspots that provide a habitat for about a third of all marine species (Fisher et al, 2015)—which is why colloquially they are referred to as the "rainforests of the sea".In addition to their immense ecological importance, coral reefs offer a wealth of ecosystem services to millions of people, including the provision of food and commercial ...
The leading countries in coral reef and climate change research are shown in Figure 4. The USA and Australia produced the most publications, with 2940 and 2933 articles, respectively, more than triple the output of third-ranked England, with approximately 719 articles. The USA and Australia contributed more than 75% of all publications.
Artificial reefs (ARs) have been used on coral reefs for ecological research, conservation, and socio-cultural purposes since the 1980s. We examined spatio-temporal patterns in AR deployment in tropical and subtropical coral reefs (up to 35° latitude) and evaluated their efficacy in meeting conservation objectives, using a systematic review of the scientific literature. Most deployments (136 ...
Coral Reefs. The journal, Coral Reefs, the Journal of the International Coral Reef Society, is committed to publishing diverse and multidisciplinary papers across broad fields of coral reef science. It publishes analytical and theoretical papers focussed on understanding and managing coral reef systems and organisms.
Importance of coral reefs. • they have recreational value. • - rich source of nutrients. • -source of income for coastal area people. • -Attraction for tourists. • -helps in nitrogen ...
Coral reef ecosystems are being fundamentally restructured by local human impacts and climate-driven marine heatwaves that trigger mass coral bleaching and mortality1. ... This research was ...
June 1, 2021 — New research shows table corals can regenerate coral reef habitats on the Great Barrier Reef decades faster than any other coral type. The research suggests overall reef recovery ...
Individual coral reefs reached record levels of heat stress up to 12 weeks ahead of previously recorded peaks. Heat stress puts immense pressure on vital but fragile tropical ecosystems such as coral reefs, mangrove forests, and seagrass meadows. For example, Newfound Harbor Key, a coral reef in the Florida Keys, accumulated heat stress that ...
In an article published on August 9th in the science journal Bioscience, coral reef expert Peter Edmunds argues for further focus on the undersea creature. "If we're going to have a chance to ...
Other reefs appear to have escaped with little impact. AIMS Research Program Director Dr. David Wachenfeld said the regional increases in coral cover are encouraging, showing the Reef's capacity ...
Researchers in Australia say temperatures in and around the vast coral reef over the past decade are the highest recorded in 400 years. ... The research showed temperatures around the Great ...
High adenylyl cyclase activity and in vivo cAMP fluctuations in corals suggest central physiological role. K. L. Barott. , Y. Helman. & M. Tresguerres. Read the latest Research articles in Coral ...
Dr. Kenkel is a marine biologist who has been studying Caribbean coral reefs since 2006. Since 2018, my lab has transplanted hundreds of coral specimens to reefs in Florida to study what makes ...
Coral reefs support immense biodiversity and provide important ecosystem services to many millions of people. Yet reefs are degrading rapidly in response to numerous anthropogenic drivers. In the ...
coral reefs and related habitats, the impacts of society on coral reefs, and the impacts of coral management on communities. The survey is repeated in each jurisdiction every five to seven years in order to provide longitudinal data and information for managers to effectively conserve coral reefs for current and future generations.
Mass coral bleaching on the Great Barrier Reef (GBR) in Australia between 2016 and 2024 was driven by high sea surface temperatures (SST)1. The likelihood of temperature-induced bleaching is a key ...