Benthic ROVing

During the May 2010 maintenance cruise, we also took the time to explore the seafloor surrounding the Barkley Canyon instrument platforms using the remotely operated vehicle (ROV) ROPOS (Remotely Operated Platform for Ocean Science, operated by the Canadian Scientific Submersible Facility (CSSF)). Equipped with many cameras which stream the video back to the ship, ROPOS is a most useful engine to obtain a better picture of the animals that live in the vicinity of the deployed instruments.

Explore we did, but our roving was certainly not random. At each of the main three depths, we traced eight 50m transects in a spoke-wheel pattern centered on the IP (Figure 1). Due to ROPOS' impressive navigation system, it is possible to know the coordinates, depth and altitude of the ROV for any one second. By coupling the navigation information with the time-stamped videos, it is possible to determine the position of all the larger animals which appear within the field of view of the camera.

Two camera systems were used; one high definition forward-looking and one down-looking camera, both endowed with a pair of lasers 10cm apart. The first camera is used to get a wide view of the transect in high resolution. The main issue with this system is that the camera is set at an angle; causing perspective issues and complicating the taking of measurements (see previous blog entry). The crucial measurement here being the width of the transect; needed so that densities of organisms can be estimated. To alleviate this difficulty, we use the down-looking camera. With this system, as long as the altitude of the ROV is constant, the width of the transects remains the same and can be measured using the laser beams. The first camera is used to identify the organisms we encounter, but only the ones that also appear in the second camera are included in the database (Figure 2).

Two and a half hours of video was thus collected, but it took over 10 hours to analyse and even more time to extract meaningful information from the collected data. Video analysis is a time-consuming process, but it can give us very good information regarding why certain species are only found in certain locations. In this case we were particularly interested in examining how depth might affect species presence and absence. Previous studies have suggested that more species might be present halfway down the continental slope, but this pattern is not expected to apply everywhere (Gage 1996). We are also looking at whether other environmental factors such as current or sediment properties may help explain species' distribution patterns. For example, certain species of suspension-feeders (animals which capture food particles in suspension, e.g. sea pens) might require a hard substrate to anchor and grow while others might only occur in areas with specific current velocities as these are likely to affect the quantity and catchability of drifting food particles (Wildish and Kristmanson, 1997).