Hard Rock Geology above the Challenger Deep

Geologists are only now starting to understand subduction initiation because it does not happen often. When it does, it largely occurs deep underwater since the majority of subduction zones occur when a dense, old oceanic plate sinks beneath another oceanic plate. Thus, the goal of our research cruise was to use a remotely operated vehicle (ROV) called Jason to collect rock samples from the inner wall of the Mariana Trench near the Challenger Deep. This region of the trench is ideal because of the strong and geologically recent extension that occurred, allowing exposure of the Moho (the crust-mantle boundary) as well as very young arc volcanic rocks available for sampling. In addition to these goals, the seafloor near the Challenger Deep is host to unusual deep-sea vents situated on ultramafic mantle rock, which we also aimed to study using Jason.

My journey onto the University of Washington’s research vessel Thomas G. Thompson started after all members of the science party arrived in Guam. We hailed from institutions around the world, including the United States (California, Texas, Tennessee), Canada, France and Japan. After getting onto Naval Base Guam, we boarded R/V Thompson on Nov. 15. While the science team made preparations in the main lab, the Jason engineering team also had their own pre-dive tasks. Jason, along with its support vehicle Medea, are launched from the deck of the research vessel. Tethered to the ship by a 10-kilometer (6-mile)-long fiber optic cable, which delivers electrical power and communications from the on-ship control room, Jason can dive to depths of 6,500 meters (21,325 feet) and sample rocks, sediments, hydrothermal vent fluids, marine organisms, all while taking ultra 4K video of the deep sea. For our primarily geology and geochemistry-focused team, we would be using Jason’s manipulator arms to collect hard rock samples from the seafloor.

But before we could do this, the engineers needed to ensure the two-body system was ready for the target depths. Jason had not dived this deep since 2003, at this very same location in the Challenger Deep. Two things needed to occur before we could begin our scientific dives. The first was the unspooling of the entire length of the cable in order to relax it and ensure proper operation prior to the scheduled science dives. The second would be engineering test dives to the target depth.  

While the Jason engineers worked on the ROV, the science team also had much to work on. Doing fieldwork at sea, especially if it involves collecting hard rock samples, is very different from collecting rock samples from land outcrops. Generally, on land, how much rock one takes from an area isn’t particularly limited as long as there are ample bodies with backpacks and strong arms, as well as access to field vehicles to transport rocks back to the lab. However, at sea, our sampling would be severely limited by the ROV’s payload – which, on Jason, was an array of milk crates that could be configured to hold roughly 30 rock samples, depending on size. Furthermore, unlike working on land, where geologists can use drills, hammers and chisels, and even tile saws to cut rock from outcrops, none of this is possible in deep water. The ROV’s arms are limited to collecting samples that are already on the seafloor (what land geologists call “float”) or loosely held rocks in outcrops. 

In addition to the limited number of samples the ROV can collect on any given dive, we also have to work around the physical limitations of diving to 6,500 meters water depth and back to the surface. Descending takes about four hours; the same amount of time is needed for ascent. Thus, some eight hours of the day is spent just getting Jason to the seafloor and back. Once on the bottom, Jason spends anywhere from four to eight hours exploring and taking samples. One scientific dive therefore takes 12 hours on average. At sea, the science team split into two teams on 12-hour work shifts; the Jason engineering team works on four-hour shifts.

Once we left port and the Jason engineers conducted successful test dives, our first target was a vent site called the Shinkai Seep Field (SSF). It is located approximately 80 kilometers (50 miles) northeast of the Challenger Deep at water depths of 5,551 to 5,861 meters (18,211 to 19,229 feet). An important organism of the SSF is the vesicomyid clam of genus Abyssogena, which form colonies around the site. During Jason dives, we were always on the lookout for the bright white shells of these clams which we hoped would lead to new seeps.  The most spectacular aspect of the SSF is the eerie white brucite-carbonate chimneys growing straight up from the serpentinite seafloor. The chimneys looked like white cathedrals in the deep darkness of the hadal zone, and they were truly a spectacular and uncanny sight to behold.

After diving the Shinkai Seep, we moved on to the westernmost targets along our transect. At this point, we had experienced some considerable delays due to very strong trade winds which prevented Jason from being safely launched. Original plans to dive for rocks near the SSF were scrapped in favor of moving to the far west of our transect to escape the bad weather. Our next dive at the westernmost site was successful. Our objective at this site was to systematically sample rocks every 50 to 100 meters along an upslope transect along the inner wall of the Mariana Trench, starting close to the seafloor at 6,500 meters and ending at 6,066 meters water depth.  

Previous dives by our Japanese colleagues using manned submersible Shinkai 6500 revealed that the Moho is exposed well along areas of the inner trench wall. One prediction we were hoping to test was to see whether mantle and crustal rocks in the west were noticeably different in terms of composition, water content, and degree of deformation, compared to rocks in the east – the latter region experienced a longer history of subduction (starting roughly eight million years ago) whereas the former only started subduction three million years ago. At the westernmost J14 site, we recovered 35 rocks, including gabbro (a rock made up of crystals of the minerals feldspar and clinopyroxene and formed by intrusion of hot magma into the deep crust) and peridotite (mantle rock). We also recovered abundant basalts as would be expected at an active volcanic arc, most of which had very thick seawater alteration rinds and manganese crust. While collecting them off the seafloor, the basalts looked quite pale in color, which misled us since basalts on land tend to be quite dark due to their mafic (high iron and magnesium) content. However, prolonged exposure to seawater results in the breakdown of iron and magnesium minerals to lighter-colored secondary minerals. Once we cut open some of the basalts, we found only a small unaltered core representing the original basalt. The cut basalts affectionately were dubbed “century eggs” due to their resemblance to the well-known Chinese delicacy.

Once all the samples came up after the dive, the hard work of processing and characterizing them took place in the main ship lab. The first and most important step was to verify each collected sample in the payload basket against the dive log. This is critical as during each sample collection, precise location data are recorded for each rock. This is what sets ROV-based sampling apart from dredging, where a basket is simply dragged across the seafloor to collect whatever is in its path. With the spatial and video context provided by Jason, we would be able to pinpoint where each sample came from, thus being able to truly map this small part of the seafloor. After the payload is verified, every sample is then placed in its own individual labeled bin. Each sample is photographed twice (“as is” and then after being cut in half). Cutting the rocks was quite an endeavor, as sometimes this had to be done at night on a rocking ship, outside on the deck. In addition to photographs, the rock samples are weighed, measured and then described using standardized nomenclatures. All of this initial data is archived into individual dive reports for each dive. The last step in the sample processing sequence is to subsample each rock so that individual research teams from the science party have enough material to work on once they return to their home institutions.

Although our main goals of the cruise involved analyzing hard rocks of the seafloor, we also had the incredible opportunity to observe deep-sea animals in action as Jason was diving and recording video. In addition to live organisms, we observed evidence of organisms burrowing and living on the seafloor, such as “crop circles” (trails) left by what we think were acorn worms. We saw the Mariana snailfish – a beautiful, ghostly white fish with diaphanous fins, cusk eels with tiny beady eyes, and lots of sea cucumbers, both resting on the seafloor but also moving within the water column.

This cruise was an incredible opportunity for me as a primarily terrestrial geologist. I got to see first-hand how life and science at sea work. I worked with an amazing team of students, postdocs, and scientists, forging friendships and scientific collaborations. While we didn’t achieve all the intended target dives because of weather and technical challenges, we managed to dive the westernmost and easternmost endpoints along our proposed geological transect, retrieving samples from both these distinct areas. Our Japanese colleagues were also able to recover important sampling equipment they left at the Shinkai Seep eight years ago. I learned that working at sea can be tough, and things rarely go as planned, but the experience taught me much about how to approach science, as well as life.

Emily J. Chin is an associate professor in the Geosciences Research Division at Scripps Institution of Oceanography at UC San Diego.