"The images are amazing, and we have not seen densities like this since 2003," said Deborah Hart, a mathematical biologist at the Woods Hole Laboratory of NOAA's Northeast Fisheries Science Center (NEFSC) who also leads the agency's sea scallop stock assessment effort.
"One scallop per image is considered high density, so seeing hundreds in an image is really exciting, and potentially very good news for future harvests."
The findings were made during the annual federal sea scallop resource survey conducted on the 146-foot research vessel Hugh Sharp, operated by the University of Delaware.
The survey used NOAA’s HabCamV4, a towed platform equipped with stereo cameras and strobes, side-scan sonar, a plankton microscope, and a variety of environmental sensing instruments. It operates throughout the sea scallop fishing grounds from the Mid-Atlantic Bight to Georges Bank.
The HabCamV4, developed at WHOI, is a sophisticated Habitat Camera Mapping System. WHOI scientists, Scott Gallager and Amber York are aboard the Sharp for the survey as research collaborators.
NOAA's HabCamV4 is towed behind the Sharp at around 6 knots (about 7 miles per hour) about two meters (roughly 6 feet) above the sea floor. It takes six overlapping images every second to create a continuous mosaic of the ocean floor over many miles in great detail. Computer servers on the vessel store the images and environmental data from the sensors on the HabCam, along with information from the ship’s computer systems about its location, speed and other operating details.
One in every 50 images is annotated manually, meaning scallops, fish and other organisms and the habitat in that image are described on a computer file, while every image is classified automatically using machine vision tools, explained Mr Gallager.
"We're seeing many swimming scallops and other behaviors that are providing insights into how the animals live and interact in that environment," Ms Hart said.
"Baby scallops are seen attached to adults, and other scallops are swimming above the bottom, perhaps to diffuse to areas that are less dense and provide more room to grow."
"Swimming scallops—particularly young ones—are known to occur due to the presence of predators, but some of these scallops were over one meter (about three feet) off the bottom without any predators in the area. This means that there is a tremendous potential for transport in the water column once they leave the bottom," added Mr Gallager, a biologist at WHOI and principal developer of HabCamV4.
"The stereo imaging capability on HabCamV4 is allowing us to measure precisely the size of the scallops regardless of their orientation, their altitude off the bottom, and swimming attack angle. Knowing the altitude and attack angle, we can predict their swimming speed and how far they may be carried by the currents," Mr Gallager said.
"It may very well be that swimming is the main mechanism for post-settlement dispersal, and that is what controls where we see scallops downstream of such events."
