PS101 - Weekly Report No. 5 | 09. October - 16. October 2016

Discovering the ice-covered Arctic deep-sea: Of robots, minerals and microbes

[19. October 2016] 

The fifth week of expedition PS101 has made significant progress in the study of hot vents and seamounts under the ice-covered seas. The under ice robot NUI samples the seafloor.



In mid October the Arctic has become dark, ice-cold and windy. The few remaining open water leads are freezing up, frost flowers form on top of them and the harsh winds pile up layers of snow. The ice drift gets as fast as one knot. The abyss now seems a more favorable environment, as it is warmer than the air around us, and life looks much more colorful than above. The new Underice Robot NUI maps uncharted areas of the seamounts, and brings back some unique samples of seafloor life (Fig. 1).

NUI or Nereid Under Ice is a specially designed robot prototype for ice-covered oceans that can either operate in fully autonomous mode as an Autonomous Underwater Vehicle (AUV) or as a Remotely Operated Vehicle (ROV). The NUI engineering team has been very busy developing and testing it under the real, harsh conditions of the Central Arctic throughout the cruise. The weather has constrained the number of dives available, but what NUI has achieved has been both scientifically successful and proven the full versatility of the robot.

Only during the unlucky Dive no. 13 the mission failed at its outset as the robot fell asleep under an ice-floe.  During the following dives, it mapped the summits of Karasik Seamount, collected 3D photography, and accomplished chemical sniffing for evidence of seafloor fluid flow. Then it was converted into an ROV that could conduct seafloor sampling with its manipulator arm and would display fantastic HD Video of the behavior of animals at the seafloor.  We used the opportunity to launch in the quiet center of a storm.  Diving to the summit of Karasik Seamount we were able to film the behavior of the giant sponges that dominate the seafloor and also starfish, deepwater corals and shrimp. (Abb. 2A).  The Karasik Seamount is one of  a chain of several large seamounts originating at Gakkel Ridge, and forming the Langseth Ridge (2B)

It was really exciting to watch robot pilot Casey at work – the robot television room was filled with 20 watchers of science and crew. She juggled sponges into the net, took seafloor samples with tiny cores, while we were watching close ups of shrimps feeding on bacterial mats growing on sponges (Abb. 3-4). It almost felt like we were diving ourselves, watching the large screens around us. And we were provided with samples of the giant sponges of various life cycle stages, starfish and microbal mats (Fig. 3-6). As a bonus, NUI filmed the unexpected zooplankton swarms above the mount during the down and upcast of the dive.

Such video-guided dive capabilities under the full ice-cover represent a rare, important technical innovation. Only few robots on Earth can do what NUI was able to achieve.  It is a phenomenal technical challenge to navigate under the ice, and return to the ship through a small hole in the ice, and on a cable. Just one successful dive under the ice can replace many hours of winch-operated devices, and some samples are impossible to get without such cabled robots. Yet, the deployment of robots next to the ship results in a high sensitivity to wind and weather. New ice-breaking research vessels are currently planned with a moonpool in the ship’s belly, to allow the protected launch and recovery of robots for underice diving.

Besides new types of robots and camera sledges with sonars, we also use quite classical tools for seafloor sampling: the dredge. There is one overarching question leading much of our work this week: is there a direct connection between the hydrothermally active mounts in the Gakkel Rift Valley, and the larger, apparently inactive chain of seamounts rooting in Gakkel Ridge, and forming the Langseth Ridge (Fig. 2B). Both are inhabited by very different animal communities – but are the rocks similar ? We carried out three dredge hauls with a chain bag dredge to systematically collect rocks from the flanks of the mounts. This device consists of a solid metal frame with a wide-meshed chain bag attached to it. It is towed above the seafloor for several hundred meters, scraping off and collecting larger rocks while sediments and smaller rocks pass through the mesh. This way we were able to collect about 700 kg of rock samples. And smaller rock samples were obtained by chance, via box, multiple, and gravity coring. On the basis of this sample material we were able to gain further insight into the regional geology.

While the rocks at the active Gakkel Ridge volcano exclusively contain pillow basalts, we recovered additional diverse lithologies from the seamounts forming the Langseth Ridge (Fig. 2B). These comprise fine-grained basalts with large olivine crystals, plagioclase and pyroxene bearing dolerites, and a striking brecciated rock type in which the rock fragments are as well volcanic. At the northern flank of the Langseth Ridge the dredge furthermore brought up a greenish rock type that strongly resembles serpentinised mantle rock. Especially these potential mantle as well as the brecciated rocks point to tectonic forces forming the seamount chain in addition to evidence to volcanic formation.

We furthermore sampled clearly hydrothermally influenced rocks and precipitates at both the vent mount within the Gakkel Rift valley as well as from the north flank of one of the large seamounts of Langseth Ridge (Fig. 7-9). The hydrothermal precipitates were obtained from the field right under the large plume of the vent mount. Altogether the rock samples are well suited for further detailed petrological and geochemical work that will help to better understand the magmatic-tectonic development of the Langseth Ridge.

Our four-meter long temperature probe (Fig. 10) is a useful device for comparing the heat flow distribution among the different mount types of the Gakkel Ridge. We expected to find temperature anomalies in the seabed and overlying water column associated with hydrothermalism. Our giant seafloor thermometer has seven miniaturized temperature data loggers. It is a robust high-precision device for deep-sea research in rough and uncharted terrains, however, it needs sediments to penetrate the seafloor which are rare around Gakkel Ridge.

Hence, the temperature loggers were also attached to the gravity corer, TV-MUC and OFOS, to characterize temperature anomalies in the uppermost sediment layers and bottom waters.

After 37 successful measurements, we are able to compare the temperature field around the seamounts. The heat flow was between 0.5 mW/m² up to > 130 mW/m², increasing with proximity to the plume site at the Vent Mount whereas the large seamounts of Langseth Ridge were cold.  In addition to measurements of the temperature gradient we also measured thermal conductivity as a material property, using a needle probe on recovered core material collected by the gravity corer (Fig. 10). The thermal conductivity values of the sediments in this area are quite high with values between 0.9 W/(m*K) up to 1.3 W/(m*K). This might result from the higher content of volcanic glass, sulfides and ashes trapped in the sediment.

By now our observations converge well. The hydrothermal activity in this region of the Gakkel Ridge comes from the central rift, from a seamount which hosts a small active area of chimneys and fluids sources within a 50 x 100 Meter area north of its summit. Besides many small chimneys, we see sulfide precipitates and warm fluids emanating from the steep rocky terraces. The animal communities are very different, whereas the apparently older, steeper but cold seamounts are densely populated by seafloor life, the vent mount is quite barren, apart from occasional amphipod swarms and fields of sea anemones. Only the microbes seem quite happy here, due to the venting of high concentrations of methane and hydrogen. The latter provide a potential energy sources for specific microorganisms in the otherwise energy limited deep sea. To test the microbial activity and compare it between the mounts, we filter large amounts of water (up to 500 l in 90 minutes) and retain the bacterial community on a filter with micrometer pore size. The fresh plume water is characterized by very high methane concentrations of a rather heavy carbon isotope signature of -11‰ 13C  (Abb 11). In the fresh plume, we can follow a high temperature anomaly of 0.3°C hundred above the seafloor, with methane concentrations of >200 nM.

To investigate the microbial consumption of hydrogen and methane in plume waters and bottom waters of the nearby seamounts and abyssal plains, we incubated plume water in gas-tight serum vials and incubated the samples at in situ conditions. The bacterial communities in the plume consumed hydrogen very quickly, but not methane. Outside of the plume waters, neither hydrogen nor methane were consumed. This points to a specific adaptation of the Arctic vent microbes to hydrogen as energy, but does not explain why no cold-adapted methane oxidizers can be found, which is to be investigated further in our home laboratories. It is curious to find an absence of the typical hydrothermal vent communities here in the Arctic, despite the considerable activity in fluid flow. At the larger, cold seamounts, we find remnants of what could have been a past vent community marked by mytilid shells and debris of segmented tubes. Typical for science: for every answer found, ten other questions arise.

And time is running, we have already started counting the last tasks, to fill gaps in our sampling program. Darkness, increasing ice and winds tell us it is time to go home. We were quite lucky so far and are all happy about the exciting innovations in technology and science, and our preliminary results. This cruise builds on a long-term German-US collaboration in the exploration of extreme environments, with the help of technology developped at the interface between space and deep ocean research. In Germany this approach has been pioneered by the ROBEX initiative (Robotic Exploration of Extreme Environments), and in the US the current project is sponsored by NASA’s PSTAR (Planetary Science and Technology Analogue Research) program. The ice covered Arctic is one of the least known, least explored regions of Earth, and also an analogue system to other ice-covered planets. And Earth is now recognized to be just one of up to 10 planetary bodies in our solar system that may host ice-covered liquid oceans to be explored.

All participants are in good health and are sending greetings to families, friends and colleagues.


Antje Boetius






Antje Boetius

Scientific Coordination

Rainer Knust
Rainer Knust


Sanne Bochert
Sanne Bochert