Linking Environmental, Ecological, Anthropogenic and Host-specific Drivers to Stony Coral Condition on Florida\u27s Coral Reef (FCR) in the Southeast Florida Coral Reef Ecosystem and Conservation Area (ECA)

Populations of Orbicella faveolata are the subject of collaborative intensive disease intervention efforts. Monthly monitoring of individual colonies and successful disease intervention provide an unprecedented in situ record of the occurrence of new SCTLD infections on a set of Orbicella faveolata colonies. These data provide a measure of disease incidence and host susceptibility through time that we can relate to concurrent spatial and temporal gradients in suspected drivers. Recent spatial and temporal models by the PIs indicated various links from inland water sources to the lesion outbreaks on reef corals. Spatial models were best explained (35.1%) by septic tank density within a 21 km radius. Temporal models showed that the number of new lesions was best explained (49.7%) by the flow rate out of the inlets over the previous 7 days. This suggested a link between some aspect of the water leaving the ICAs and SCTLD incidence on the large O. faveolata corals in the ECA. These models were built off of a limited dataset which has since been expanded, therefore we propose to rebuild them with an additional 14 months of monitoring data increasing the replication of our temporal model and improving its strength/rigor and importantly capturing another full seasonal cycle of SCTLD incidence. Additionally, successful disease intervention treatments on these O. faveolata colonies have kept diseased corals alive providing a unique opportunity to test intraspecific differences between groups of corals with differing infection patterns. Some corals have lesions once, some have lesions numerous times, and some never show lesions. Colonies from all categories appear in close spatial proximity on the reef indicating that intraspecific coral differences are affecting disease dynamics. Understanding the biological reasons for these intraspecific differences will enable more robust predictions of future spatial and temporal infections. Furthermore, this ‘patient history’ provides a foundation to interpret and contextualize more probative or diagnostic analyses opening an opportunity to investigate factors that contribute to disease resistance. As part of an ongoing project by the SCTLD resistance research consortium (RRC), 90 corals are being sampled to provide a fundamental understanding of the O. faveolata holobiont at gross morphologic, genetic, biochemical and molecular scales. Analyzing samples from these groups should identify differences in endosymbionts, genotypes, metabolites, microbes, biological pathways, metabolites and antimicrobial bioactivity, immune response, and histopathological differences. We propose to collaborate with and synthesize data from the SCTLD RRC across a suite of host-specific factors and use cutting-edge machine learning to identify those factors responsible for driving host resistance or resilience to SCTLD. The SCTLD RRC models will identify the proximate factors involved in driving differences between colony infection rates which can be used to understand the underpinnings of how SCTLD affects the holobiont, why some colonies are more resistant to disease, and as bioindicators to assess a population’s vulnerability to SCTLD. This project will lead to a step-change in the understanding of the combination of factors that drive SCTLD dynamics across scales. These range from local-scale patterns of disease incidence in the ECA to broad-scale patterns of identifying why specific colonies are more resistant to SCTLD infection across the ECA and Lower Keys. Our results and outputs will directly inform local management decision making on prioritizing efforts to reduce local stressors contributing to SCTLD incidence and identify optimal times and locations to prioritize disease intervention, on how water management is linked to SCTLD disease incidence in South Florida’s ECA, and on current and future land management strategies like septic to sewar conversions to potentially reduce coral disease. Finally, we propose to compile the predictor datasets from these studies in a GIS geodatabase for inclusion in a publicly accessible decision support tool