ABOARD THE R/V THOMAS G. THOMPSON—In the early morning hours of Saturday, May 10, we were on the seafloor in the deepest part of the Kermadec Trench when all of the video screens in the Nereus control room went dark. That wasn't unusual, but it did signal a premature end to our 7-hour dive to 10,000 meters and the start of a programmed ascent sequence. What was different this time was that we also lost our acoustic communications link with the remotely operated sub that tells us roughly where it is in the water.

But we knew where the vehicle was when we lost communication and we also knew it would begin rising 30 minutes later at a known rate. So the ship stood off and we waited for the 5-hour ascent time to pass. When that time came and went, we posted look-outs around the ship to scan the water in case the malfunction that took down our communications link was also interfering with the radio beacon that allows us to track the vehicle on the surface.

After a few hours of this, it became obvious something else was wrong with the vehicle, but there were an additional set of automatic procedures designed to initiate ascent at set times. Before we could make it to the next point in that timeline, however, the bridge reported seeing white objects in the water. These turned out to be small pieces from inside the sub’s hulls, which confirmed the growing fear that some of us harbored as the afternoon wore on: Nereus was gone.

Pilot Casey Machado summed things up best in a remembrance she wrote over the weekend:

Nereus was lost doing what she was designed to do, exploring the deepest reaches of the ocean with a basket full of samples and a control room filled with scientists interacting actively for the first time with the Hadal seafloor. I am still stunned in disbelief by the whole experience and I feel as if I have lost a child. I'd greet Nereus each morning on deck and wave goodbye to her as she was released to her long commute down to work each dive. She will be missed greatly.

There is a saying among people who do this for a living: “You don't put anything in the ocean you can't afford to lose.” Although we could ill-afford to lose Nereus, everyone accepted or at least acknowledged the inherent risk in what we were doing. What is difficult to accept is that with one structural failure, the research community has lost the only active submersible that can reach the greatest depths of the oceans. We can still send down landers and fish traps, but they only tell us part of the story of the deep ocean. There is so much more to learn.

Even before the news about our loss went out, people here were taking stock of what we had on the ship and making plans for ways to carry on with our research. By evening, we launched the ship's CTD rosette, a workhorse ocean science instrument that measures conductivity (which gives us salinity), temperature and depth to gather water samples down to 5,000 meters and record chlorophyll levels in the upper water column. These will give us an idea of how much primary production (photosynthesis) is occurring at the surface and details about the microbial life in the water column as a result.

On Sunday we set about set about our return to science in earnest. This morning, small teams disassembled and inspected the glass flotation spheres that bring instruments back to the surface. (We do this after every other deployment, but today it felt extra-critical.) Another group assembled a new fish trap from spare parts. This will allow us to make more deployments up the side of the trench to explore the biogeographic transition zone over which fish suddenly disappear from the seafloor. We are now working our way up the western side of the Kermadec Trench, so fish we collect here will be compared to those we collected down the axis of the trench to address the question of whether both communities are genetically linked. We know the fish don't move much deeper, but how much do they migrate along the depth to which they seem to be confined?

Because we needed Nereus to gather specimens for the respirometer, we took most of the equipment off of the hadal elevator and converted it to a hadal water sampler. If everything goes well, this will soon have a new core sampler installed. These two instruments—the water bottles and coring device—plus the CTD will allow us to gather additional data for the microbiologists and macrobiologists who want to study the microbes, plankton and the food supply falling from the surface to hadal depths. You'll learn more about that in a coming post, but briefly, they are interested in some of the most basic questions about life in the trenches, including how much of what kinds of food makes it to hadal depths, the abundance and diversity of microbial life in deep-ocean sediments that arise as a result and the adaptations they have evolved to thrive in the trenches.

By dinner, we had both fish traps and the elevator in the water and headed to the seafloor at depths of around 7,000 meters. We also had a lucky break today and collected some pumice floating past the ship. These turned out to be tiny rafts of life carrying several different organisms, including crabs, barnacles, anemones, hydroids, and amphipods. As we discovered on our 8,000-meter dive, these rafts can eventually sink, providing yet another pathway for food to reach the deep ocean. Out here, as they are near most trenches, underwater volcanoes are common. This means pumice formed during eruptions could constitute yet another way for food to circulate to the deep ocean. But how much, no one knows. This is new territory.

The list of what we don't know about the deep ocean is long and, because we've been exploiting the ocean for a long time, a little alarming. Even though we separate it into distinct zones, the ocean is an interconnected whole and we've seen time and again that what happens in one part of the ocean can have an impact elsewhere in surprising ways.

We received a very kind message of sympathy from James Cameron, who called the loss of Nereus, “A dark day for many reasons,” one of which is the fact that the loss of this single vehicle denies us access to nearly half of the ocean's total depth range. When you consider how much of the deep ocean remains entirely unexplored and how fundamental our questions are about the hadal ecosystem, it becomes clear that we don't need a vehicle to replace Nereus – we need a fleet of them.