- Exploring Migrating Deep-sea Scattering Layers
- Expedition Features
- Finding “Dory”
Releasing tens of thousands of dollars of instrumentation into the ocean is tricky — you may not get it back once it’s in the water. Because of this, a large part of ocean engineering consists of rigorous testing and incorporating system redundancies before deployment in the open ocean.
The National Geographic Society (NGS) Driftcams were tested during development at the Neutral Buoyancy Lab at the University of Maryland, and designed to overcome obstacles in the field, including communication errors, lost controls, and becoming physically lost. However, the true test lies at sea, where conditions may not always be favorable, predictable, or possible to simulate in lab conditions.
This expedition is the first time these new model Driftcams are being tested at sea. First, each Driftcam’s communication and buoyancy engine are tested in the ship’s dry lab. Then they are set in the water while tethered to a small boat, where fine adjustments to the ballast are made to accommodate for the local water density. After a Driftcam clears those stages, it is deployed untethered to a relatively shallow depth (50 meters or about 164 feet) to confirm that all the things tested before are working while the device is fully autonomous.
One of the Driftcams passed all previous tests and was released at dusk. Despite all previous testing, acoustic communications from the ship to the Driftcam did not work. The system was submerged and the team had no way to know its location or depth. Meanwhile, sunset was fast approaching. This set off around 15 hours of unknowns as the team worked to relocate the Driftcam, henceforth known as “Dory.”
Below is a non-exhaustive list of problems the team faced, and the system redundancies in place that allowed them to overcome the issues.
- The team was unable to communicate with the Driftcam via the acoustic modems.
The team was unable to locate the Driftcam after its intended surfacing time.
- The ship-board acoustic modem was functioning properly, so the problem was likely at the Driftcam. The Driftcam was programmed to surface after one hour at 50 meters depth. So IF the device was otherwise functioning properly, we could estimate the direction and speed the current might take the Driftcam and have the ship nearby when it surfaced.
The Driftcam was either sitting too low on the surface or had not surfaced at all.
- Once on the surface, there are three potential ways to relocate the device:
- 0-1 mile away — A bright, reflective orange flag sticks two feet above water when the Driftcam is at the surface. The video lights also enhance Driftcam visibility at night. However, in the fading light at the intended surfacing time, the researchers and crew had no luck spotting the flag.
- 1-5 miles away — A VHF (very high frequency) beacon in the Driftcam emits a steady pulse that can be picked up with a receiver and directional antenna. The team brought out the VHF tracking and scanned in all directions over the next couple of hours but were not able to detect the VHF beacon.
- Worldwide — The Driftcam includes an Iridium satellite beacon, which sends a GPS location up to satellites and is relayed via email and a website. If the device has wandered beyond the range of visual and VHF scanning, the satellite uplink could provide a location. At three hours after the intended surfacing time, no messages had come through the satellite.
- Options B and C for relocating a Driftcam depend on it being at the surface because radio signals will not pass through saltwater. These communications links are also affected by how low the Driftcam is floating and the roughness of waves. Neither the VHF nor satellite signals can get out if the Driftcam is being over-washed by frequent waves.
- There is a plan for this situation, however. There is a 15 pound drop-weight on each Driftcam. The systems are programmed to activate a “burn wire” that will drop this weight if it has not been recovered within a couple of hours. With this weight removed, the system will float higher and be able to more reliably get the VHF and satellite antennas clear of the water.
After more than three hours with no VHF signals or satellite messages, the team had to assume that the automatic burn wire release of the drop-weight had not occurred.
BUT there is a plan for that too. In case the software, batteries, or connection wires fail and cannot release this weight, there is also a GTR (galvanic timed release) in the links holding this drop-weight. This is made of magnesium, which naturally dissolves at a roughly predictable rate when in seawater, and eventually weakens to the point where the weight is dropped (over 15-36 hours, depending on the thickness of the selected link).
Because this mechanism is not precisely timed, the team could not know exactly when it would release. They took turns waking up regularly through the night, checking the satellite service website for messages.
Map of the ship’s location in the Gulf of Mexico. The yellow points are labeled with locations of interest, including location pings from Driftcam “Dory” at 4 a.m. and 6 a.m., and its projected location, based on the two previous points, at 9:45 a.m. (yellow point in the center of the screen). Image courtesy of April Cook, Nova Southeastern University. Download largest version (jpg, 5.5 MB).
Finally at about 3 a.m. the next morning, messages from Dory began appearing, providing its location. The team was able to calculate Dory’s speed and trajectory based on her location pings.
National Geographic and Second Star Robotics team members communicating with Driftcam “Dory” near the bow of the R/V Point Sur. Image courtesy of April Cook, Nova Southeastern University. Download largest version (jpg, 5.4 MB).
At around 10:30 a.m., our ship’s first mate spotted the flag affixed to Dory as the R/V Point Sur followed in the general direction of Dory’s trajectory. System redundancies in design, and the team’s flexibility and ability to troubleshoot under pressure allowed us to eventually find Dory. Of course, having an excellent crew and a capable ship was also invaluable to this effort.
Our National Geographic and Second Star Robotics team members on the small boat to retrieve Driftcam “Dory.” Dory can be seen in the distance, affixed with a bright orange and reflective flag. Image courtesy of April Cook, Nova Southeastern University. Download largest version (jpg, 6.1 MB).
Published August 2, 2021