Deep Coral Predictive Habitat Modeling in the US Atlantic and Gulf of Mexico: Focusing on Uncharted Deep-Sea Corals
Project Status: This project began in August 2011 and was completed in September 2013
We are using statistical models that combine databases of known deep-sea coral beds with information about key habitats to predict and map suitable habitat for deep-sea corals in the US Atlantic and Gulf of Mexico. The resulting habitat suitability maps will improve the conservation planning, management, and exploration of deep-sea coral ecosystems.
Why We Care
Deep-sea coral is a diverse and valuable resource that, among many things, provides habitat for fish and invertebrates. Because of their slow growth rates and vulnerability to bottom disturbance, deep-sea coral ecosystems are particularly important to conserve. The distribution of deep-sea coral is poorly understood because of the logistical difficulty and expense of surveying the deep ocean. Predictive modeling of deep-sea coral habitats is essential for supporting conservation planning and for targeting areas for future mapping and exploration. Modeling can also lead to insights into the environmental factors driving the distribution of deep-sea corals, helping to build our knowledge base of how these unique ecosystems function.
What We’re Doing
We are using statistical modeling techniques to predict areas of the seafloor that are capable of supporting deep-sea corals. Using a sophisticated statistical machine-learning algorithm called maximum entropy (MaxEnt), we are combining databases of known deep-sea coral locations developed by the NOAA Deep-Sea Coral Research and Technology Program (DSCRTP) with environmental and oceanographic data to generate predictive models of deep-sea coral distribution. These models will be used to produce regional maps of deep-sea coral habitat.
The project is being conducted in three areas of the United States:
the Northeast/Mid-Atlantic, in partnership with the NOAA/NMFS Northeast Fisheries Science Center, Sandy Hook Laboratory, and National Systematics Laboratory;
the southeast Atlantic, in partnership with the NOAA/NMFS Southeast Fisheries Science Center, Beaufort Laboratory;
the Gulf of Mexico, in partnership with NOAA/NOS/NCCOS Center for Coastal Environmental Health and Biomolecular Research).
In these regions, deep-sea coral occurs on the continental shelves and slopes, at ocean depths of 50 to greater than 2000 meters.
What We’ve Found
Thus far we have:
Generated the highest-resolution (<500 meter grid cell size), most complete maps of potential deep-sea coral distribution in the eastern US and Gulf of Mexico regions
Determined that the amount and location of suitable habitat varies by taxonomic group (e.g. soft versus hard corals)
Identified (by predictive models) many unexplored areas that are likely to support deep-sea coral communities
Discovered that information on substrate, including hard bottom and sediment grain size, is a critical factor for predicting the distribution of many deep-coral taxa
Determined that bottom geomorphology, bottom temperature, bottom salinity, and surface productivity correlate with coral habitat suitability.
We anticipate that our work will:
Support conservation and management of deep-sea corals
Help regional fishery management councils (FMCs) plan benthic habitat management initiatives
Provide valuable information for offshore spatial planning, including planning for renewable energy installations
Lead to explorations to map previously undiscovered deep-coral communities
Lend insight into the environmental processes that shape deep-sea coral distributions.
As deep-coral databases improve, we hope to improve and validate these models with additional field data. We are assisting with the integration of these predictive maps of deep-coral habitat suitability into regional spatial planning processes.
Related Regions of Study: Atlantic Ocean, Atlantic Seaboard, Gulf of Mexico
Primary Contacts: Brian Kinlan, Chris Caldow
Science for Coastal Ecosystem Management (Seafloor Mapping, Marine Spatial Planning, Coral)
Related NCCOS Center: CCMA