BS ISO 23731:2021 pdf download – Marine technology — Marine environment impact assessment (MEIA) — Performance specification for in situ image-based surveys in deep seafloor environments.
4 Principle The suggested protocols are image-based as these methods are non-destructive to the monitored environments. They incorporate a temporal survey element to allow intrinsic variability within systems to be differentiated from anthropogenically-caused disturbances. The observation platform stays without motion at the observation site to obtain fixed point data. Video (or > 15 fps stills) is employed to a) increase the accuracy of mobile organism identification due to the availability of images from multiple observation angles, b) allow estimates of organism activity, c) allow calculation of current speed and direction through the tracking of suspended particles between frames/images. These in situ image-based surveys are done with monitoring by video camera systems where the field of view and depth of focus are fixed and known or recorded. The duration of the monitoring is determined prior to the set-up of the monitoring system, and monitoring data can be acquired autonomously. Examples of seafloor observatories are shown in Annex A. 5 Performance requirements and recommendations 5.1 Angle and scale of the monitoring The angle of the camera vs. the seafloor and the altitude above the seafloor should be known or recorded. Scale information and minimum resolvable dimensions should be extractable from the image/images. This may be done via calibrated stereo cameras, a laser scaling system, imaging of a physical object/scale of known dimensions or the like. 5.2 Illumination Illumination should be by white light of known and/or recorded intensity and wavelength characteristics. Red light can be used in addition. 5.3 Bait Bait of a size where the flesh is considered able to be fully consumed within 7 days should be standardized between deployments and attached to the imaging platform, or deployed where it can be imaged by an AUV or other moving platform, somewhere in the field of view to obtain information on mobile predators/scavenger types and abundances.
5.4 Image capturing schedule/timing Images shall be captured either as video or as stills at 15 frames per second (fps) or faster, with total duration per imaging event of 5 s or greater. Imaging events should be recorded at 6- to 8-h intervals or less over a continuous period (e.g. 372 days; see NOTE). For moving survey platforms, the position accuracy for repeated observations should be enough to image 50 % or more of the previously imaged area/volume [ 14 , 15 ] . If a combination (the preferred scenario) of a stationary long-term observation platform and a moving platform, able to image large horizontal distances is used, the moving platform should image at least 50 % of the area/volume imaged by the stationary platform. Video/images should be timestamped, either on the image itself or as associated data, and a CTD (conductivity temperature depth profiler), or other instrument for measuring temperature and other physicochemical environmental parameters and recording data at least for the time period that is imaged, should be deployed on the imaging platform to allow the determination of correlations between environmental parameters and organisms, particle abundances/types, bottom current speed/direction, etc.