Occurrence and percent cover of the colonial invasive tunicate Didemnum vexillum from near-seafloor drift transect video imagery and high-resolution digital still images in the western Bay of Fundy, Canada

Study Extent

Within the Head Harbour/West Isles/Passages EBSA in the western Bay of Fundy, Canada. This study is concentrated in the southern portion, specifically in Western Passage and Head Harbour Passage off of SW Deer Island and NW Campobello Island, respectively. All imagery was collected on September 21 and 22, 2022.

Sampling

Near-seafloor optical imagery was obtained at seven locations within the HH/WI/P EBSA on 21 and 22 September 2022 using an underwater camera system recently developed by DFO and operating off the Canadian Coast Guard Ship (CCGS) Viola M. Davidson. The system uses an optical cable slip-ring winch system (Varcon Sedna 1, with Falmat Xtreme cable) to lower the camera package to within 1 m of the seafloor, with real-time video provided to the surface to permit adjustment of the camera height above the seafloor. Surveys were conducted using a drift transect protocol, for a target 20-minute deployment time at each location. The overall distance covered over the seafloor was dependent on near-seafloor tidal current and direction during each deployment. Site selection was determined using high-resolution (1-m spatial resolution) multibeam echosounder and airborne LIDAR (Light Detection and Ranging) bathymetry data acquired from the Ocean Mapping Group (OMG) and Canadian Hydrographic Service (CHS) visualized in ArcGIS Pro v2.8. Bathymetric data was paired with previously predicted areas of high M. modiolus and Boltenia ovifera (Linnaeus, 1767) abundance. Both species generate important biogenic habitat that could be sensitive to invasive tunicate overgrowth: M. modiolus creates dense mussel reefs and B. ovifera can form widespread native tunicate fields. The camera system provided two primary digital imagery sources. A SubC 1Cam Alpha 6 video camera provided downward facing continuous video of the seafloor. High-definition (HD) video was viewable at the surface and recorded (1080 progressive/60 frames per s) on a Blackmagic Hyperdeck Studio 4K Pro deck. Illumination was provided by five Deep Sea Power and Light (DSPL) 10 000 lumen video lights. A Nikon D850 full-frame camera with a Nikon 18 – 35-mm lens, connected to a Quantum T5D-R studio flash (also oriented downward and under operator control from the surface) was used to take digital still images at periodic intervals (with a target of 30-s intervals) during each transect. Intervals between still images varied as the operator had to prioritize maneuvering the camera package around complex bottom topography. Moreover, the operator sometimes decreased intervals between still images as drift speed increased. A Valeport VA520 altimeter provided a continuous record of the height of the survey package off-bottom, and a pair of DSPL Sealaser 100 lasers (10-cm spacing) provided a means to evaluate the field of view of both the digital video and digital still imagery. For this survey, an underwater acoustic positioning system was not available, and so geographic positions were recorded from a Hemisphere R330 GNSS receiver with an A45 antenna at the surface. From prior use of similar near-seafloor camera systems within the EBSA on the CCGS Viola M. Davidson, it can be expected that spatial offsets from the survey vessel could range up to 40 m astern to 5–10 m port or starboard, depending on depth and tidal factors (pers. obs.). However, the along track line of each transect can be expected to correspond to the actual distance travelled by the camera package over the seafloor. Drift camera tracks for all transects were plotted in ArcGIS corresponding to the date and time when the camera package was in a near-seafloor survey mode. Additionally, reported depths (in m Below Chart Datum – BCD) corresponding to the location of the camera track were extracted from the 1-m multi-beam echosounder and LIDAR bathymetry data. High-definition video footage for each drift transect was divided into 20-s video segments using VLC media player. A video segment was deemed unusable and removed from the dataset if there was too much turbidity, or if the camera position was too high off-bottom (altitude > 1.5 m) to reliably image the seafloor. When present in a usable video segment, D. vexillum frequency of occurrence was classified as "Common” (continuous coverage/patches throughout the video segment), "Occasional” (individual colonies of various sizes encountered > 5 times throughout the video segment), or "Rare” (small, isolated colonies encountered ≤ 5 times throughout the video segment). Still images, targeted to be taken ~ 30 s apart along the transect, were assessed to determine the number of images from each transect with visible D. vexillum colonies. Similar to the video analysis, still images were deemed unusable and removed from the dataset if there was insufficient light levels, too much turbidity, or if the altitude exceeded 1.5 m. When D. vexillum was observed in an image, we visually categorized colony coverage as > 50%, 26–50%, 6–25%, or ≤ 5% of the image field of view. We then estimated the frequency of each coverage category occurring within each transect. Where species identification was uncertain, due to insufficient image quality, resolution, or small colony size, we did not count the colony as D. vexillum; thus, we report conservative estimates of D. vexillum presence and coverage. The actual distance traveled (m) by the camera package over the seafloor was calculated using the z-field geometry tool in ArcGIS Pro, calculated at the transect and 20-s video segment level. The distance between still images along a transect was calculated using the “Split Line at Point” tool in ArcMap (v.10.8, Advanced licencing). Field of view was estimated by measuring the length and width of a subset of still images and video frame grabs in ImageJ2, using the 10-cm laser distance for scale. Field of view was then standardized for each altitude (at a resolution of 0.1 m) and used to estimate individual digital still image area. For each 20-s video segment, an average altitude was calculated to yield a single field of view estimate from which area of the segment could be derived from the distance traveled in the segment. Segment areas were summed to estimate the overall area covered in each transect.

Method steps

1. As described above (an in detail in Teed et al. 2024), sites were selected using bathymetry data acquired from the Ocean Mapping Group (OMG (2007) Ocean Mapping Group. Southwest New Brunswick – Multibeam Mapping, Ocean Action Plan. https://omg.unb.ca/Projects/South_West_NB/html/ (accessed 28 August 2023)), as well as areas with high prediction probabilities of the horse mussel and sea potato (Mireault CA, Lawton P, Devillers R, Teed L (2023) Presence/absence and abundance of vulnerable marine ecosystem species Boltenia ovifera and Modiolus modiolus in the lower Bay of Fundy derived from high resolution still imagery. Coastal Ecosystems Science Division, Fisheries and Oceans Canada, St. Andrews, N.B. https://open.canada.ca/data/en/dataset/152ae3f1-d2b9-43d9-a7b4-d769d9e9fc41 (accessed 29 September 2023)). Distance travelled estimates for each imagery transects was calculated using ArcGIS tools (ESRI (2021) ArcGIS Pro: Release 2.8. Redlands, CA: Environmental Systems Research Institute, https://www.esri.com/en-us/home (accessed 23 September 2022)). An image's field of view was estimated using ImageJ2 (Rueden CT, Schindelin J, Hiner MC, DeZonia BE, Walter AE, Arena ET, Eliceiri KW (2017) ImageJ2: ImageJ for the next generation of scientific image data. BMC Bioinformatics 18: 1, https://doi.org/10.1186/s12859-017-1934-z). Abundance of D. vexillum was estimated using a modified SACFOR abundance scale (Hiscock K (ed) (1996) Marine Nature Conservation Review: Rationale and methods. Coasts and seas of the United Kingdom. MNCR series. Joint Nature Conservation Committee, Peterborough, 167 pp).

Identified down to species level. Distinguished from the other species present locally, Didemnum albidum, by D. vexillum's yellow-cream colour, presence of meandering, dark spicule-free lines between zooid groups and the presence of flexible, irregular, frond-like outgrowths, whereas D. albidum is bright white in colour, with a lack of bands between zooid groups and even surface with no outgrowths.

1. Didemnum vexillum
   common name: Pancake tunicate rank: species

This dataset was collected in the "Head Harbour/West Isles/Passages" Ecologically and Biologically Significant Area (EBSA) in the western Bay of Fundy, Canada, particularly in the southern portion of the EBSA.