NOAA Point Shapefile- Benthic Habitat Classifications from Phantom S2 ROV Underwater Video, US Virgin Islands, Project NF-06-03, 2006, UTM 20N WGS84

Downloadable Data

Date that the source FGDC record was last modified.

Converted from FGDC Content Standards for Digital Geospatial Metadata (version FGDC-STD-001-1998) using 'fgdc_to_inport_xml.pl' script. Contact Tyler Christensen (NOS) for details.

Partial upload of Spatial Info section only.

Replaced entire Lineage section to populate new Source Contribution field.

Partial upload of Positional Accuracy fields only.

Original underwater still photography of seafloor along transect. | Source Geospatial Form: remote-sensing image | Type of Source Media: external hard drive

Original underwater video of seafloor along transect.. | Source Geospatial Form: video | Type of Source Media: external hard drive

Data Acquisition:The NOAA ship Nancy Foster was used to collect underwater video and still photographs of benthic habitats from 3/21/06 to 4/2/06. The 36 ROV transects collected traversed over 70 km. Transect placement was designed for use in both accuracy assesment and ground truthing video (.wmv) and high-resolution digital still photographs (.jpg) were acquired using a downward pointing camera mounted on a Spectrum Phantom S2 Remotely Operated Vehicle (ROV) operated by an experienced team from NURP. Video was collected continuously and still photographs were collected every 30 seconds. The ROV followed pre-determined transect lines and was self-propelled as much as possible, but was towed by the NANCY FOSTER using a 1 inch steel cable when conditions prohibited self-propellation. The camera height off the bottom was kept consistent at approximately 2m and the ROV speed was maintained at 1 meter/second. Most transects were run during daylight hours to guarantee sufficient ambient light levels for visual interpretation, but high powered strobe lights mounted on the ROV were available to supplement ambient light and during occasional night operations. Downward pointing parallel lasers (5cm seperation) were used as scale references. Video and still photographs were transmitted via the ROV cable to shipboard hard drives as they were acquired. An ROV-mounted transducer and an acoustic receiver suspended off the side of the ship were used to determine the ROV's relative position to the ship, while the ROV's absolute geographic position was estimated using this relative position and the shipboard GPS. The positional accuracy was estimated to be within 5 m.

Benthic Habitat Interpretation:The benthic habitat was characterized by visual interpretation of both video images and still photographs by three CCMA interpreters. Interpretations were standardized among interpreters by repeatedly interpreting the same still photos and video until structure classes were the same and percent covers were within 5% of each other. After the initial standardization, interpreters were assignedrandom transects to reduce any possible regional bias.Video was used to interpret broad-scale structure and substrate rugosity, since still photos were incapable of providing this data due to their relatively small and top-down field of view. The still photos were used to estimate the percent cover of finer-scale substrate and biological cover components of the benthic habitat.Structure referred to the broad-scale habitat class within the entire field of view. Based on previous benthic habitat work in the area (Kendall et al., 2001 and the NF-04-06 NANCY FOSTER mission) the structure designation was chosen from either 1) colonized pavement, 2) colonized pavement with sand channels, 3) patch reef, 4) spur and groove, 5) scattered coral and rock in sand, or 6) unconsolidated. Substrate described the visible abiotic components of the benthic habitat. The four substrate classes, considered mutually exclusive and exhaustive were 1) consolidated material, 2) sand, 3) rubble (particles ~ 2-10 cm) and 4) rhodoliths. Substrate was measured to the nearest percent of the visible bottom. An estimate of rugosity was approximated by the vertical range of substrate in the field of view and was classified as either high (> 1m), moderate (< 1m and >0.3m), or low (<0.3m).Biological cover referred to the biotic component of the seafloor and was divided among five mutually-exclusive categories differentiated by their size and shape. These were 1) sponge (Phylum Porifera), 2) gorgonian and black coral (Subclass Octocorallia and subclass Ceriantipatharia), 3) hydrocoral and stony coral (Subclass Hexacorallia), 4) fleshy algae and 5) algae veneer/turf. The sum of all cover categories provided an estimate of total colonization. If a biological cover component could not be unquestionably identified as one of the five cover categories it was only added to the total colonization estimate.Photo stills were viewed in ESRI's ArcMap software (ArcGIS 9) with a superimposed 10 X 10 grid shapefile to aid percent cover estimations. Video was viewed simultaneously on a second monitor using windows media viewer. Interpretation results were entered into an EXCEL spreadsheet using a custom VBA script, with a single record for each sample point (i.e. every still photograph taken at 30 second intervals). These sample points and their associated benthic habitat attributes were then imported in ESRI ArcMap and joined into point shapefiles.