Creating Benthic Habitat Maps
By using the backscatter and bathymetry data in tandem with video for ground-truthing habitat types, we hope to eventually create habitat maps that span large areas; once different backscatter returns on are ground-truthed at various depths, an algorithm can be developed that predicts the habitat type over areas not ground-truthed via C-BASS video.
Extrapolating Density to Obtain Absolute Abundances
After obtaining footage of a transect, analysis begins by counting and identifying (to lowest taxonomic level possible, dependent on visibility conditions, altitude above bottom, etc.) all individuals within the chosen camera's field-of-view (FOV). Currently, the video is analyzed in one minute clips and this is used as the sampling unit.
Upon counting and identifying fishes, the area viewed for each minute is estimated by calculating the FOV and length of transect covered. This is done via a series of trigonometrical equations and using the ship's speed over ground, respectively. The density (number per square kilometer) of each species viewed per minute can then be calculated.
Density estimates are then post-stratified according to habitat type which is done using multibeam and backscatter maps in an Geographical Information System (GIS) as all the counts/density estimates are georeferenced. Knowing the area of each stratum, or habitat type, the stratified density estimates can then be extrapolated to obtain total abundance estimates for the entire survey area.
Using Several Data Sources to Define Priority Areas
In addition to fisheries data available through NOAA (Gulf of Mexico Data Atlas), there are many other sources used to decide which areas are priority for the upcoming mapping cruises. Some of these include:
1. Loggerhead Sea Turtle sightings - this species has diet overlap with Red Grouper (Epinephalus morio) meaning it's presence could be a good indicator of the grouper's habitat
2. Previously mapped areas with multibeam and backscatter
3. Nautical Charts
4. Commercial and Recreational Fishing Locations
5. Sidescan Data
Understanding Reactive Behavior
In order to understand the sampling ability of this system, it is necessary to investigate how fish react to C-BASS during surveys. For the more conspicuous species (snappers, groupers, jacks), which are the targets of this technology, observed reactive movement is being quantified. We've found that a majority of individuals have exhibited neutral behavior which includes remaining stationary or being "in transit" as C-BASS passes by.
C-BASS was also a part of the first Untrawlable Habitat Survey Initiative (UHSI) cruise in August of 2014. The purpose of this cruise was to compare and test several survey technologies, including an ROV, AUV, and stationary cameras, to investigate what level of sampling coverage each was capable of. This initiative was funded by and a collaboration with NOAA NMFS.
Above: Photo progression showing the reaction of a Red Grouper (Epinephelus morio) from it's initial and final views on the footage. The grouper does not appear to react to C-BASS until approximately 4-seconds after being initially spotted.