An experimental comparison of three towed underwater video systems using species metrics, benthic impact and performance

Abstract

1. Managing ecological systems, which operate over large spatial scales, is inherently difficult and often requires sourcing data from different countries and organizations. The assumption might be made that data collected using similar methodologies are comparable, but this is rarely tested. Here, benthic video data recorded using different towed underwater video systems (TUVSs) were experimentally compared. 2. Three technically different TUVSs were compared on different seabed types (rocky, mixed ground and sandy) in Kingmere Marine Conservation Zone, off the south coast of England. For each TUVS, species metrics (forward facing camera), seabed impact (backward facing camera) and operational performance (strengths and limitations of equipment and video footage) were compared with the aim of providing recommendations on their future use and comparability of data between different systems. 3. Statistically significant differences between species richness, density, cover and assemblage composition were detected amongst devices and were believed to be mostly due to their optical specifications. As a result of their high image definition and large field of vision both the benthic contacting heavy and benthic tending TUVS provided good quality footage and ecological measurements. However, the heaviest TUVS proved difficult to operate on irregular ground and was found to cause the most impact to the seabed. The lightest TUVS (benthic contacting light) struggled to maintain contact with the seabed. The benthic tending TUVS was able to fly over variable seabed relief and was comparably the least destructive. 4. Results from this study highlight that particular care should be given to sled and optic specifications when developing a medium-or long-term marine protected area monitoring programme. Furthermore, when using data gathered from multiple sources to test ecological questions, different equipment specifications may confound observed ecological differences. 5. A benthic tending TUVS is recommended for benthic surveys over variable habitat types, particularly in sensitive areas, such as marine protected areas.

Video Transects Reveal That Tidal Sand Waves Affect the Spatial Distribution of Benthic Organisms and Sand Ripples

“…Cameras are increasingly applied in marine habitat mapping (Guinan et al, 2009;Kendall et al, 2005;Stoner et al, 2007), and multiple guidelines exist for their onboard operation and subsequent footage analysis Hitchin et al, 2015;Sheehan et al, 2016;White et al, 2007). Video footage was analyzed by two independently working observers (Hitchin et al, 2015), and despite small absolute differences between observers, both observer-specific models and models based on the averaged data yielded similar results, showing that our methodology is robust.…”

Increasing the functionalities and accuracy of fisheries electronic monitoring systems

“…Underwater cameras could also enable monitoring of the dimensions of the purse seine during setting (e.g. CRISP, ; Sheehan et al, ; Simrad, ; Underwood, Rosen, Engas, & Eriksen, ). For example, underwater cameras are being developed to enable real‐time monitoring of catch in trawl codends (Simrad, ) and as catch pass through trawls (CRISP, ; Underwood et al, ).…”

Laser chasing behaviour of wild fishes exploited as a tool to compare space use between size, sex and species

“…(see Sheehan, Stevens, Gall, Cousens, & Attrill, 2013). 36 of these sites are within the statu- Video surveys are conducted using a towed underwater video system (TUVS) for 20 min transects of target distance 200 m, The towed camera system floats above the seabed to reduce impactsee Sheehan et al (2010) and Sheehan et al (2016) for detailed information on how the TUVS is configured and operated. In order to measure the variable field of view and the size of benthic organisms, two green 532 nm lasers (Beam of light technologies, Inc: Scuba-1 Underwater dive laser) were mounted either side of the camera, pointing forward at 30 cm apart.…”