Coral reef ecosystems are globally significant, yet they face increasing threats from both local (fishing, pollution, diseases, cyclones) and global (warming, acidification) stressors. The cumulative effects of these stressors raise concerns about the imminent collapse of these ecosystems. Ecosystem collapse signifies a functional extinction, where defining characteristics and functions are fundamentally transformed, leading to the permanent loss of evolutionary history, ecological functions, and ecosystem services crucial for human well-being. While live coral cover is a commonly used indicator of coral reef health, it lacks information on the broader community composition and functional aspects. This study addresses the need for a unified framework to assess the risk of coral reef collapse by applying the IUCN RLE, a standardized method for assessing ecosystem collapse risk, to the WIO. The RLE integrates multiple variables across five criteria: decline in ecosystem extent, restricted geographic distribution, abiotic disruption, biotic disruption, and quantitative model. This comprehensive approach allows for a more holistic understanding of the threats faced by WIO coral reefs and facilitates informed policy decisions.

Numerous studies have assessed live coral cover trends regionally across various locations such as the Indian Ocean, East Asian-West/Central Pacific, and Caribbean. These studies highlight drivers of decline, reef status, and management options, however methodological differences and data inconsistencies limit their synthesis and capacity to support coherent policy. The current methods lack a unifying framework to address the risk of complete ecosystem collapse. A recent study emphasized the need to bridge the gap between theory and practice in assessing ecosystem collapse risk, using the RLE framework to rigorously define collapsed states, identify drivers of change, and establish quantitative collapse thresholds. The RLE provides a standardized classification of ecosystem collapse risk, integrating multiple variables and producing an output that ranges from least concern to collapsed. This study builds on previous research by applying the RLE framework to a large dataset of WIO coral reefs, allowing for a more comprehensive and comparative assessment across ecoregions.

This study applied the IUCN RLE methodology to assess the risk of coral reef ecosystem collapse in the WIO region and its 11 nested ecoregions. The assessment utilized the Millennium Coral Reef layer, a comprehensive regional dataset incorporating hard coral, fleshy algae, and fish abundance data, along with projected sea surface temperatures. Four of the five RLE criteria were evaluated: decline in ecosystem extent (criterion A), restricted geographic distribution (criterion B), abiotic disruption (criterion C), and biotic disruption (criterion D). Criterion E (quantitative model) was not evaluated due to data limitations. For criterion A, a threshold of 10% coral cover was used to indicate the loss of reef accretion and consequently, a decline in ecosystem extent. Criterion B considered the area of occupancy and extent of occurrence of the reefs in relation to the established RLE thresholds. Criterion C, focused on abiotic disruption, used projected sea surface temperatures and degree heating weeks (DHW) to assess future thermal stress. RCP 6.0, a plausible emission pathway, was selected for analysis due to its better fit with observed bleaching events. For criterion D, assessing biotic disruption, four compartments of the ecosystem model were considered: hard coral cover, fleshy algae cover (summed across different algal types due to data inconsistencies), parrotfish abundance, and grouper abundance. Thresholds for collapse were established for each indicator based on literature review. Given the multiple compartments in the model, an algorithm was developed to assess ecological integrity and risk of collapse, considering the hierarchical interactions among the compartments. This algorithm starts with coral cover as a base, sequentially considering algae, herbivores, and piscivores, adjusting the risk level based on the relative risk of each compartment. This mitigates the potential inflation of risk due to limitations in data coverage, allowing a more nuanced assessment of ecosystem vulnerability. The overall risk of collapse for each ecoregion was determined by selecting the highest risk level among criteria A-D. The risk for the WIO region as a whole was calculated by weighting each ecoregion's score by its area of coral reefs.

The study found that WIO coral reefs are vulnerable (VU) to ecosystem collapse at the regional level. This vulnerability is primarily driven by future warming and biotic disruption, particularly the decline in piscivorous fish populations indicating fishing pressure. At a finer geographic scale, considerable variation in collapse risk was observed among the 11 ecoregions. Seven ecoregions scored the highest risk levels (four critically endangered (CR) and three endangered (EN)) due to future warming, primarily in island ecoregions across Madagascar, the Comoros, the outer Seychelles, and the Mascarene Islands. The remaining four ecoregions were assessed as vulnerable. Reefs in the large continental ecoregions (northern Tanzania-Kenya and northern Mozambique-southern Tanzania) were vulnerable due to declining populations of piscivorous fishes, while reefs in the northern Seychelles and Delagoa were vulnerable due to declines in reef areal extent and limited geographic distribution. The dominant threat was future increases in thermal stress in the island ecoregions (CR and EN), consistent with other analyses. The analysis, however, highlights the challenges in interpreting thermal stress from projected temperatures, as variance in temperature within climate model grid cells differs significantly from empirical observations. The analysis of ecological integrity considered the relative risks of coral cover, algae, herbivores, and piscivores. The decline in piscivore populations, particularly groupers, was identified as a significant factor in biotic disruption, particularly in ecoregions where fishing pressure is high. Parrotfish responses were less alarming, potentially masked by responses to coral degradation. To address this complexity and data limitations, a structured algorithm was developed to assess ecological integrity and risk of collapse, considering hierarchical interactions among ecosystem compartments. The algorithm starts with the coral risk level and incrementally increases the risk level if higher risk levels are observed in algae, herbivores, and piscivores. This method improves consistency and standardization across studies by acknowledging the variable importance of different biotic compartments and the presence of inevitable data gaps. Applying this algorithm to the WIO reefs, the final risk levels reflects the importance of piscivores in top-down control of prey populations but avoids undue inflation of overall risk if other compartments are healthy.

The findings of this study highlight the significant vulnerability of WIO coral reefs to collapse. The combined effects of climate change (thermal stress) and local stressors (fishing pressure) present a substantial threat to these ecosystems. The study's application of the IUCN RLE methodology provides a comprehensive assessment that goes beyond the traditional focus on live coral cover. The results emphasize the importance of integrating multiple indicators and accounting for the complex interactions within the coral reef ecosystem. The identification of specific ecoregions at high risk allows for targeted conservation efforts. For ecoregions less threatened by warming, prioritizing local management actions to reduce fishing pressure and promote coral recovery is vital. Conversely, for ecoregions threatened by warming, reducing local threats is crucial to maintain ecosystem function and buy time for coral adaptation. The study's emphasis on integrating climate and ecosystem-focused actions offers a robust framework for effective conservation and management strategies. The results strongly support the recommendations for separate measures of area and integrity for quantifying ecosystem health, underscoring the need for a shift beyond solely focusing on coral cover as a primary indicator. The RLE, as a composite index, is well-suited for this purpose, incorporating area and integrity into a single assessment framework.

This study provides a comprehensive assessment of the vulnerability of WIO coral reefs to collapse using the IUCN RLE methodology. The findings highlight the critical need for both climate change mitigation and adaptation strategies, coupled with effective ecosystem-based management focused on alleviating local stressors. The regionally-nested ecoregion scale provides insights for both inter- and intra-regional policy development. The methodology is replicable across other coral reef regions globally, facilitating standardization and consistency in assessing ecosystem collapse risk and supporting the implementation of global conservation and sustainability targets. Future research should prioritize improvements in data collection and analysis to enhance the accuracy and granularity of the RLE assessment, particularly at finer spatial scales.

The study acknowledges several limitations that might affect the interpretation of the results. Data gaps exist for some threatening processes, limiting the comprehensive analysis of all potential stressors. Variations in data quality and quantity across different monitoring programs and regions necessitate the aggregation of variables, potentially confounding some interpretations. The use of estimated initial values for some indicators, due to the lack of historical data extending 50 years back, introduces uncertainty. The uneven spatial coverage of monitoring sites can impact the generalizability of the findings. Despite these limitations, the comprehensive application of the RLE methodology and the consideration of multiple indicators provide a robust framework for assessing coral reef vulnerability. Continued efforts in data collection and refinement of the methodology will further improve the accuracy and robustness of future assessments.