Methodology:

The in-situ underwater still images and short videos were gathered with BAS' Shallow Underwater Camera System (SUCS) during the expedition JR17003a of RRS James Clark Ross to the eastern Antarctic Peninsula in March 2018.

The SUCS for JR17003a comprised three units:

1. The UIC unit consisting of (i) the PC with monitor, (ii) the cable metering sheave indicator and (iii) the deck box.

2. The deck unit consisting of (i) the winch, (ii) UW-cable, (ii) the deck monitor and (iii) the metering sheave on the mid-ships gantry.

3. The UW-unit of the tripod consisting of (i) the UW-housing including the camera, booster and power distribution board, (ii) the UW-light, and (iii) the USBL.

The SUCS includes 1000 m of fibre-optic cable, allowing operation to approximately 900m depth. The LabView interface together with the fibre-optic upgrade enables high- resolution photo stills (2448 x 2050) and video footage (720 x 480) to be taken simultaneously. The field of view when the SUCS is standing on the seafloor is covering 0.51 m2 of seafloor.

The SUCS and Agassiz gears, when both deployed at the same site, increase the value of the data obtained. This is because specimens trawled in the latter and identified by detailed morphological inspection or using molecular methods improve the likelihood and confidence of correct identifications of individuals seen in the SUCS images.

The SUCS was deployed at 12 stations (ranging in depth from 200 m to 800 m), according to weather and accessibility due to ice conditions. The SUCS was used extensively during the time when heavier gear could not be operated due to problems with the 30t winch, allowing coverage of a wider range of depth stations.

On one occasion SUCS deployment was delayed because of very cold temperatures (-15°C) affecting the backlight of the winch monitor screen. This was resolved by re-routing the image output to an alternative monitor, and using a portable heater for subsequent deployments.

Normal protocol involved three consecutive photo transects, the direction of which was determined by the bridge according to wind direction (to allow the ship to sit comfortably in dp), each 100 m apart, with each complete transect consisting of 10 photos, themselves each 10 m apart. Duplicates of each photo were taken with different light levels, to allow distinguishing of different features. At two locations the three transects were not fully completed because of icebergs in close vicinity to the ship or problems with gear. Additional photos were sometimes taken during the 10 m relocation if the camera showed a good view, however these photos will not be included in subsequent analysis. Short videos using the SUCS camera were taken at some transect points or during relocations.

Data collection:

The Shallow Underwater Camera System with Labview SUCS software and Prosilica camera system developed by the British Antarctic Survey, http://vocab.nerc.ac.uk/collection/B76/current/B7600022/.

High- resolution photo stills (2448 x 2050) covering 0.51 m2 of seafloor and video footage resolution (720 x 480).

Data quality:

Standard protocols for SUCS deployment and image capture were followed. Data entry of SUCS' image geolocation data was linked to vessel's USBL system. Data entry of image number was double checked by independent person.