Mesophotic coral bleaching associated with changes in thermocline depth

Coral reefs, despite their limited spatial extent, support a remarkable biodiversity, with nearly 30% of all marine fish species and potentially over a million species overall. However, these ecosystems are highly vulnerable to rising global sea surface temperatures (SSTs), leading to widespread and increasingly frequent coral bleaching events. Shallow-water coral reefs, living near their thermal limits, are particularly threatened, with projections suggesting substantial losses even with moderate temperature increases. Mesophotic coral ecosystems (MCEs), located at depths of 30–150 m, were initially considered potential refuges due to the cooler, deeper waters they inhabit. However, recent research has demonstrated that MCEs exhibit limited connectivity with shallow-water reefs, possessing unique biological and ecological characteristics and high levels of endemism. Understanding the susceptibility of these extensive and biodiverse MCEs to warming oceans is therefore crucial for effective conservation strategies. Mesophotic corals are thought to be buffered from thermal stress by the thermocline, the region of rapid temperature decrease with depth. However, vertical shifts in thermocline depth, influenced by processes like internal waves, can expose benthic communities to significant temperature changes. Monitoring these dynamic processes is significantly more challenging than tracking SST, making predictions of MCE vulnerability complex. This study focuses on the Chagos Archipelago in the central Indian Ocean to investigate mesophotic coral bleaching and the role of oceanographic factors in this phenomenon.

The literature extensively documents the impact of rising SSTs on shallow-water coral reefs, with numerous studies detailing the increasing frequency, intensity, and duration of bleaching events. The vulnerability of shallow corals is well-established, and predictive models based on SST are widely employed. Conversely, the understanding of MCEs is relatively recent, with a growing body of research emphasizing their unique biodiversity and limited connectivity with shallow-water reefs. Some early hypotheses suggested MCEs might serve as refuges for shallow-water corals, but this view is being revised based on new evidence highlighting their endemic species and distinct ecological roles. Prior studies have also investigated the role of the thermocline and internal waves in moderating thermal stress on benthic communities, but their application to MCEs and bleaching prediction remains limited. The existing literature lacks comprehensive investigations into the interplay between large-scale oceanographic processes (like the IOD) and small-scale internal wave dynamics in driving mesophotic coral bleaching.

This multidisciplinary study utilized a combination of techniques to investigate mesophotic coral bleaching at Egmont Atoll in the Chagos Archipelago. Two research cruises were conducted in November 2019 and March 2020. High-resolution digital imagery of coral colonies and their bleaching status was collected from shallow (15 m) to mesophotic depths (90 m) using a remotely operated vehicle (ROV). The ROV was equipped with cameras, lighting, and a CTD to record temperature and other oceanographic data during dives. A total of 1080 images were collected across six depth bands (15–20 m, 30–40 m, 60–70 m, 80–90 m, 110–120 m, 150–160 m), with the deepest two bands excluded from analysis due to the absence of scleractinian corals. Moored in-situ measurements of oceanographic parameters (temperature, current velocity) were obtained at two sites (Ile Des Rats and Manta Alley) using ADCPs and thermistor chains. These data were used to resolve the dynamical processes modulating thermocline depth. A high-resolution numerical model (MITgcm) was employed to simulate oceanographic conditions, incorporating bathymetric data from a multibeam survey. Publicly available numerical model outputs of Indian Ocean conditions were also analyzed to understand the regional-scale influence of the IOD. Coral bleaching prevalence (percentage of affected colonies) and severity (bleaching index, BI) were assessed from the ROV imagery using a four-category classification system (unaffected, light, moderate, severe). Statistical analysis (Kruskal-Wallis test, PERMANOVA) was used to compare bleaching levels between sites, seasons, and depth bands. NOAA Coral Reef Watch data were also consulted to assess surface thermal stress.

The study documented coral bleaching at depths up to 90 m at Egmont Atoll, the deepest recorded instance. This occurred in the absence of shallow-water bleaching, highlighting the limitations of using SST alone to predict MCE bleaching. The bleaching event coincided with an exceptionally strong positive phase of the IOD, causing thermocline deepening and bringing warmer surface waters to mesophotic depths. Analysis of NOAA CRW data and modeled local temperature fluctuations confirmed the absence of significant shallow-water thermal stress during the bleaching event. The 22 °C isotherm deepened significantly during the study period. Differences in bleaching severity between the two study sites suggested a local influence, likely due to internal wave activity. High-resolution numerical modeling, validated by in-situ data, revealed that internal waves, especially Mode 2 waves, modulated the thermocline depth. At Manta Alley, these waves enhanced warming at mesophotic depths, while at Ile Des Rats, they had a cooling effect, highlighting the site-specific nature of internal wave impacts. In-situ temperature measurements from the ROV corroborated the model predictions. Significant differences in coral community composition were observed between depth bands but less so between sites at the same depth, suggesting that community structure is not the sole driver of bleaching variation.

This study's findings challenge the assumption that MCEs universally serve as refuges from coral bleaching. The observed bleaching at mesophotic depths demonstrates the vulnerability of these ecosystems to thermal stress, emphasizing the need for their inclusion in conservation planning. The reliance on SST as an indicator of thermal stress for MCEs is inadequate. The study highlights the importance of incorporating subsurface oceanographic processes, such as IOD-induced thermocline deepening and internal wave dynamics, for accurate bleaching susceptibility predictions. The IOD, projected to increase in frequency and intensity under global warming, poses a substantial threat to MCEs. High-resolution modeling is essential for understanding the local-scale variations in thermal regimes caused by internal waves, which can significantly influence bleaching patterns over small spatial scales. The study's results underscore the need for advanced monitoring techniques and predictive models that account for both large-scale oceanographic patterns and localized processes.

This study provides compelling evidence that mesophotic coral ecosystems are vulnerable to thermal stress and bleaching, driven by both large-scale events (like the IOD) and small-scale processes (internal waves). The reliance on SST as a predictor of MCE bleaching is insufficient, necessitating the integration of subsurface oceanographic dynamics into predictive models and conservation strategies. Future research should focus on expanding monitoring efforts in MCEs, improving high-resolution modeling capabilities, and investigating the resilience and adaptation mechanisms of mesophotic corals. More detailed analysis of the interaction between large-scale climate patterns and small-scale oceanographic features will improve the accuracy of future predictions of MCE bleaching and inform conservation strategies.

The study focused on a single atoll in the Chagos Archipelago, limiting the generalizability of the findings to other geographic regions. The limited duration of the study (two cruises) might not fully capture the long-term variability of thermocline depth and internal wave activity. Technical challenges associated with ROV operations and image analysis, such as the inability to quantify image area, may have introduced some uncertainty into the bleaching assessments. The high-resolution numerical model, while validated by in-situ data, involves inherent assumptions and simplifications that could affect the accuracy of simulations.