Flying a tow-camera is kind of like trying to backseat fly an upside down, underwater kite (anchor?) attached to a 2 km long piece of string.
Monitoring the position of the towcamera relative to the ship
There are two altimeters mounted to the tow-camera that gives real-time measurements of how far away the seafloor is relative to (1) straight below the tow-camera (Imagenex_Altitude_Depth), and (2) straight in front of the HD-camera FOV (Tritech_Altitude Depth).
These numbers are constantly changing because of sea-surface conditions , speed of the ship, and how well the passive heave-compensation is working on the ship. A lot of mental math is needed to calculate a 'safe-distance' based on how much and how fast the altimeter readings are changing. Basically, when the altimeters drop < 1 m, I radio to the winch operator to go 'UP UP UPPPPPPPPP". Most of the time, I tried to keep the altimeter between 2-4 m which was only possible during optimal sea-state conditions.
Speed (in knots) is also important because it determines how fast the camera moves with each winch pull and whether the video is moving too fast to be 'useable' for research. The trick is to (1) monitor how fast the ship is moving relative to how fast the tow-camera is moving along the bottom, (2) keep the camera FOV as close to the bottom as possible, and (3) anticipate any sudden changes in seafloor slope based on the distance from bottom and ship speed.
One laptop ran the software monitoring groundfaults (which are bad for ROV ops.) and controlled the on/off switch for most of the powered systems on the tow-camera. The pan and tilt angles of the main HD-camera were also controlled through this software. Because we were often towed alongside walls and up slopes, having the ability to control the camera pan and was quite useful. We could point the camera towards the side-facing wall (data) instead of a fixed, straight ahead FOV which would have captured a black void of nothingness (no data).
How to get good quality imagery data from a towed camera system.
For this cruise, we also experimented with a 7th data-display. This one gave the pilot real-time visualization of both altimeter readings. For me, this ended up being quite handy as it offloaded a bit of the mental math. The 'game' was to monitor the amplitude of the wave-forms (surface swell) while keeping the running average 'horizontal / flat' through verbal commands to the winch operator.
I once had 4 touchdowns in one cruise
So, I did plant the tow-camera into mud this cruise (n=4). Twice occurred within the first 20 mins of my first ever practice dive. The third happened when both of my walkie-talkies died. The fourth happened when I didn't realize I needed to compensate for a decrease in the tow-camera altitude after asking the ship to slow down.
Good news is that there was no damage to the tow-camera! DFO and Highland Technologies Inc. built one heck of a science tool. There's a new sheriff in town with towed camera platforms.
Jackson W.F. Chu
Jackson is a marine biologist, photographer, and dog-owner. This blog will try to be heavy on the photos and light on the text.