Test Your Habitat IQ
Gulf Spill Restoration Website
Stay Connected

Predictive Habitat Modeling for Alaska’s Deep-Sea Coral and Sponge Resources

Project Goal

To create predictive habitat suitability models for deep-sea corals and coral and sponge “gardens” for Alaskan waters.

Geographic Location



Alaska has among the richest and most diverse deep-sea coral habitat in the nation. NOAA partnered with the Marine Conservation Institute, Bangor University (UK), and the National Fish and Wildlife Foundation on the second year of a two-year project partially funded by the Oak Foundation to predict deep-sea coral habitat in Alaskan waters. Deep-sea coral records from NOAA and the Smithsonian Institution were used to develop deep-sea coral habitat suitability models for deep-sea coral taxa in in Alaskan waters, including the U.S. Exclusive Economic Zone. Deep-sea coral habitat suitability was modeled at approximately 700 m x 700 m spatial resolution using a Maximum-Entropy (Maxent) presence-only modeling approach, in conjunction with a variety of physical, chemical and environmental variables known or thought to influence the distribution of deep-sea corals. To calculate validation metrics, the presence data was randomly partitioned to create 75% training and 25% test datasets, with test data used to calculate validation metrics. Maps were prepared of predicted habitat suitability for different orders or sub-orders of corals. Locations of dense coral and sponge “gardens” were also modeled.


Maxent models identified slope, temperature, salinity and depth as important predictors for most deep-sea coral taxa. The majority of predicted coral habitat (with the highest probabilities) occurs in the Aleutian Islands, Bering Shelf-break, Gulf of Alaska seamounts, and the Fjord region and shelf break of Southeast Alaska. The Aleutian Islands were predicted to have the highest suitability for most coral taxa. True soft corals (Suborder Alcyoniina) were predicted to have the broadest distribution, including in areas of the Bering Sea shelf that did not appear to be suitable for other corals. Large areas of highly suitable deep-sea coral habitat were predicted both within and outside of existing bottom trawl closures. Predicted habitat suitability results are not meant to identify coral areas with pin-point accuracy and probably over predict actual coral distribution due to model limitations and unincorporated variables (e.g., substrate) that are known to limit their distribution.

Importance to Management

Alaskan waters cover huge areas and include some of the nation’s most productive bottom fisheries. Research surveys will only be able to reach a fraction of these areas. Predictive habitat modeling can identify areas with the highest probability of harboring deep-sea corals in areas that have not been visited. These model outputs will help target areas for future field research and provide insights into the most important environmental and physical drivers controlling the distribution of deep sea coral and sponge habitat in Alaskan waters. It is expected that the model results along with field research conducted by NOAA’s Deep Sea Coral Research and Technology Program, will help inform the North Pacific Council’s 2nd 5-year Essential Fish Habitat Review and identify areas of potential interaction between commercial bottom fisheries and deep-sea coral and sponge habitats.


Guinotte, J.M. and A.J. Davies (2013) Predicted deep-sea coral habitat suitability for Alaskan waters. Report to NOAA-NMFS. 22 pp.

Fiscal Year 2012 Funding


Point of Contact

Thomas Hourigan, Tom.Hourigan@noaa.gov