Integrating Predictions of Deep-Sea Coral and Sponge Habitat with Ocean Mapping and Exploration Offshore the US West Coast – An Example from the EXPRESS Campaign

Information about the spatial distribution of vulnerable marine ecosystems, including deep-sea coral and sponge (DSCS) habitat, is critical for making environmentally sound decisions related to offshore activities such as commercial fishing and energy development. Spatial predictive modeling is a cost-effective tool for identifying potential DSCS habitat in broad areas where data are often sparse. In addition to providing information for managing marine resources, models can identify potential targets for ocean mapping and exploration. Data from field surveys can then be used for model validation and to create new models. Integrating predictive models with ocean mapping and exploration in this manner can provide an iterative research approach that supports conservation and management of DSCS and advances the knowledge of DSCS ecology. As part of the Expanding Pacific Research and Exploration of Submerged Systems (EXPRESS) campaign, partners at the National Oceanic and Atmospheric Administration (NOAA), the Bureau of Ocean Energy Management (BOEM), and the US Geological Survey (USGS) demonstrated this approach for the area offshore the US West Coast. Models predicting areas of suitable habitat were developed for ~50 DSCS taxa. Model outputs included maps of the predicted distributions of suitable habitat and spatially explicit depictions of prediction uncertainty. During an October 2018 cruise on NOAA Ship Bell Shimada, predictions of suitable habitat for DSCS were used to select locations for surveys with an autonomous underwater vehicle and a remotely operated vehicle at 12 sites where fishing restrictions have been modified by the Pacific Fishery Management Council to protect essential fish habitat for groundfish. In addition to providing baseline information about these sites to support management, annotations of DSCS occurrences from these surveys have been used to test the performance of the models. This data will also be included in updated models of DSCS habitat in support of ongoing and future management and exploration priorities.