Predicting the Distribution and Biodiversity of Structure‐Forming Deep‐Sea Corals Using Hierarchical Community Occupancy Models
Holly F. Goyert, Arliss J. Winship, Matthew Poti, Enrique J. Salgado, Rachel Bassett, Janessy Frometa, Michael Coyne, Thomas F. HouriganABSTRACT
Aim
Many deep‐sea corals form important structural habitat on the seafloor and are targets for biodiversity conservation. Structure‐forming deep‐sea corals are particularly vulnerable to human activities that disturb the seafloor, including bottom‐contact fisheries and offshore energy development. Systematic surveys of deep‐sea coral occurrence are generally not feasible, especially at regional scales. Consequently, species distribution models have become an important tool for predicting the potential distributions of deep‐sea corals from limited survey data. However, due to reliance on records from historical databases and concerns about the reliability of absence data, existing species distribution models for deep‐sea corals have often been restricted to inference of relative probability of occurrence (i.e., habitat suitability) from presence data for select species or broader groups of corals. We aimed to predict and map the distribution and biodiversity of deep‐sea corals offshore the US Gulf coast (known as the Gulf of America/Gulf of Mexico; hereafter, ‘Gulf’), USA, using absolute estimates of occurrence and richness.
Location
US Gulf of America/Mexico.
Taxon
Deep‐sea corals in the orders Scleractinia, Antipatharia, Malacalcyonacea, and Scleralcyonacea.
Methods
Deep‐sea corals are sessile, slow‐growing, and long‐lived, making them ideal candidates for occupancy models that use presence‐absence data to estimate imperfect detection and probability of occurrence. Because deep‐sea coral observations from visual surveys can be difficult to identify to species, we implemented hierarchical community occupancy models to estimate the richness of 28 deep‐sea coral genera (rather than species).
Results
The models characterised spatial patterns in richness across multiple genera. As expected, we found that common taxa had higher probabilities of detection and occurrence than rare taxa.
Main Conclusions
Hierarchical community occupancy models are a tool to account for detectability and uncertainty in estimating important conservation metrics, including the biodiversity of deep‐sea corals or other organisms with similar traits.