Microbial Community Dynamics Provide Evidence for Hypoxia during a Coral Reef Mortality Event

In July 2016, a severe coral reef invertebrate mortality event occurred approximately 200 km southeast of Galveston, Texas, at the East Flower Garden Bank, wherein ∼82% of corals in a 0.06-km² area died. Based on surveys of dead corals and other invertebrates shortly after this mortality event, responders hypothesized that localized hypoxia was the most likely direct cause. However, no dissolved oxygen data were available to test this hypothesis, because oxygen is not continuously monitored within the Flower Garden Banks sanctuary. Here, we quantify microbial plankton community diversity based on four cruises over 2 years at the Flower Garden Banks, including a cruise just 5 to 8 days after the mortality event was first observed. In contrast with observations collected during nonmortality conditions, microbial plankton communities in the thermocline were differentially enriched with taxa known to be active and abundant in oxygen minimum zones or that have known adaptations to oxygen limitation shortly after the mortality event (e.g., SAR324, Thioglobaceae, Nitrosopelagicus, and Thermoplasmata MGII). Unexpectedly, these enrichments were not localized to the East Bank but were instead prevalent across the entire study area, suggesting there was a widespread depletion of dissolved oxygen concentrations in the thermocline around the time of the mortality event. Hydrographic analysis revealed the southern East Bank coral reef (where the localized mortality event occurred) was uniquely within the thermocline at this time. Our results demonstrate how temporal monitoring of microbial communities can be a useful tool to address questions related to past environmental events. IMPORTANCE In the northwestern Gulf of Mexico in July 2016, ∼82% of corals in a small area of the East Flower Garden Bank coral reef suddenly died without warning. Oxygen depletion is believed to have been the cause. However, there was considerable uncertainty, as no oxygen data were available from the time of the event. Microbes are sensitive to changes in oxygen and can be used as bioindicators of oxygen loss. In this study, we analyze microbial communities in water samples collected over several years at the Flower Garden Banks, including shortly after the mortality event. Our findings indicate that compared to normal conditions, oxygen depletion was widespread in the deep-water layer during the mortality event. Hydrographic analysis of water masses further revealed some of this low-oxygen water likely upwelled onto the coral reef.