Pushing the boundaries of marine research and exploration

The ocean covers 71% of the surface of our planet and encompasses 93% of the habitable volume on Earth. However, most of it is inaccessible to humans without technical assistance. Fortunately, innovation is occurring at an exponential rate, with advances in areas such as 3D printing, artificial intelligence (AI), miniaturization and materials science rapidly increasing our understanding of the ocean and its inhabitants. The continuing evolution of marine technology promises to tackle both the vast scales needed to cover the ocean and the finer resolution needed to improve our understanding of oceanic processes and marine life. This article, with examples from the Schmidt Ocean Institute (SOI), explores some of the platforms and sensors that are pushing the boundaries of marine research, exploration and data collection.

Advancing research platforms

Traditionally, access to the ocean for scientific advancement has been from research vessels, which continue to provide a robust platform to address a range of marine research and exploration goals. The Schmidt Ocean Institute has been operating research vessels over the last 15 years and its current research platform, RV Falkor (too) (Figure 1), is a global-class state-of-the-art ocean research vessel that provides a sophisticated technical facility for scientists. In addition to eight scientific laboratories, including a seawater flow-through system designed to avoid microplastic contamination and an onboard liquid nitrogen generator for biological sample preservation, the vessel is equipped with full ocean-depth sonars (Kongsberg EM2040, EM124 and EM712), two moon pools, a 150-ton crane and dozens of oceanic and atmospheric sensors. Falkor (too) also carries a high-performance computer (HPC) onboard, which is made available to scientists to run models of the surrounding ocean and broader research area, allowing them to modify sampling strategies while at sea.

Recent improvements in low Earth orbiting satellites now provide vessels such as Falkor (too) with a more seamless ship-to-shore communication capability with a higher bandwidth connection to shoreside laboratories which, in turn, enables broader participation by those joining remotely and expands the expertise available on each expedition. Scientists also have access to SOI’s ROV SuBastian, a remotely operated vehicle capable of descending to 4,500m. In addition to data and sample collection from the deep sea, the ROV is employed as a platform to integrate new scientific equipment and testing of prototype technologies. The capabilities of a research vessel for scientific exploration and discovery are substantial – in the inaugural year of operations of Falkor (too), scientists on this vessel discovered and studied a new animal ecosystem under the seafloor, five new hydrothermal vent fields, 11 new seamounts and over 150 potential new species. 

Complementing the data gathering ability of the research vessels of today are a plethora of remote-controlled and autonomous technologies. Underwater platforms including gliders, autonomous underwater vehicles (AUVs), Argo floats and vertical sampling systems such as the Bottom Stationing Ocean Profiler, have been routinely used in oceanographic data collection for over two decades. However, the last decade has seen an exponential increase in surface and aerial platforms, including surface uncrewed vehicles that can deploy and recover subsurface devices with no humans at sea.

Autonomous surface vehicles can collect standard near-surface marine atmospheric and oceanographic data either ahead of a research vessel (Figure 2) or in areas and situations that are difficult for research vessels to operate. For example, in 2021, a Saildrone vehicle collected video, wind speed and other data on Hurricane Sam, a category-four tropical storm. Such in situ observations in the upper ocean and just above the sea surface are notoriously challenging during a cyclone but vital for improving coupled ocean-atmosphere models and tropical storm predictions.

The Ocean Discovery XPRIZE saw the dawn of a new capability for uncrewed surface vessels; the ability to deploy and recover AUVs or other underwater devices remotely by humans at mission control on land. Both the surface and subsurface components can be adaptable during a mission and configurable to work in tandem or independently while at sea to provide more flexibility and speed. One example, developed by SEA-KIT International, was used in 2022 to map the inside of the caldera following the large and violent eruption of the underwater volcano Hunga Tonga Hunga Ha’apai, which ejected its volcanic cloud 57km above the sea surface into the mesosphere. The USV Maxlimer, initially developed as part of the winning entry from The Nippon Foundation-GEBCO Alumni Team in the Shell Ocean Discovery XPRIZE, was deployed to map the caldera while it was actively venting, a scenario that was too hazardous for a crewed research vessel. The data confirmed that volcanic activity was still taking place and provided data for researchers to gain a better understanding of the impact of the eruption. In 2016, SOI’s RV Falkor mapped this undersea volcano, and a comparison with the new bathymetry showed that 9.5km³ of seafloor material was removed during the eruption but 6.3km³ was redeposited within 20km of the caldera rim, leaving 3.2km³ unaccounted for.

The coordinated use of multiple vehicles can effectively scale up marine scientific exploration. Further innovations in autonomous surface technologies that integrate AI will eventually lead to reduced human oversight and control, allowing machines to effectively gather the data needed for climate, weather and ecosystem modelling and monitoring. For example, the Mayflower Autonomous Ship (MAS) has an ‘AI Captain’ that can ingest and analyse weather data to create and follow a mission path independently, making it an appealing and low-cost platform for routine and longer-term data collection. 

Advancing sampling technologies

In addition to furthering scientific knowledge through exploration, data collection and analysis at sea, a research vessel and ROV are versatile platforms for testing prototype marine technologies, including novel sensors to study marine life. ROV SuBastian has an ultra-high-definition pan-zoom-tilt camera for video acquisition, which is frequently used to observe animal behaviour and characterize the marine ecosystem, including seafloor features. Although invaluable, visual imagery alone is inadequate for new species identification and a sample specimen needs to be captured and brought back to the lab for analysis. However, innovations in imaging technologies are beginning to provide an avenue for biological oceanographers to gather in situ morphological and taxonomic data without the need for sample collection. The Deep Particle Image Velocimetry (DeepPIV) instrument, developed by Monterey Bay Aquarium Research Institute (MBARI), was integrated on ROV SuBastian for testing. The DeepPIV consists of a laser and optics that illuminate a sheet of fluid, enabling the ROV’s science cameras to capture the movement of particles, plankton or other small creatures in the water. In addition to the rapid characterization of organisms, this allows fine-scale fluid motion to be quantified. The DeepPIV was coupled with the Eye Remote Imaging System (EyeRIS), also developed at MBARI, a real-time 3D imaging lens inspired by the multi-lens eyesight of insects. The combined use of both instruments resulted in the illumination and imaging of mid-water organisms in 3D, allowing researchers to create and manipulate a digital image to study the creature from multiple angles on a screen.

Marine biological specimens that are soft-bodied, brittle or fragile have historically been difficult to collect without damaging their frame. To address this, researchers from the City University of New York and Harvard University developed fully robotic, soft arms and fingers, called ‘squishy fingers’, which can gently interact with deep-sea animals. The soft manipulators were 3D printed onboard RV Falkor, integrated on ROV SuBastian, and used for adaptive sampling of delicate specimens (Figure 3). The instrument was tested at depths over 2,000m and successfully grabbed fragile animals such as goniasterids and holothurians.

Gelatinous creatures typically found in the mid-water column are also notoriously difficult to collect and study in situ, partly due to their fragile composition and partly because they are continuously mobile. A sphere-like encapsulation device that rapidly collects and preserves tissue samples of such delicate organisms was designed by a team led by the University of Rhode Island. The Rotary Actuated Dodecahedron (RAD) device was integrated on ROV SuBastian’s manipulator arms in 2019, then further refined (RAD2) and re-integrated again in 2021 (Figure 4). The RAD2 device allows for rapid and targeted mid-water sampling of specimens without using nets that can capture additional and superfluous specimens.

These prototype technologies, combined with advances in eDNA sampling, miniaturization of sequencing technologies, low-cost sampling and AI for visual identification, are part of a suite of tools and sensors that set a new benchmark for imaging and sampling marine life and, in the future, will result in real-time and rapid in situ taxonomic identification.

Imagining marine technologies of the future

Advances in materials science, miniaturization of sensors and the integration of intelligent technology such as robotics with haptic suits have the potential to transform oceanographic research by utilizing power sources in new ways, enabling massive data collection and altering the way in which humans interact with the ocean.

Currently, oceanographic and atmospheric sensors require batteries or tethered power sources and are heavy or bulky. New materials, including ultrathin organic solar panels, can transition traditional sensors into longer-lasting (renewable power source) and lightweight (removal of batteries), ubiquitous data collection devices. Materials scientists at King Abdullah University of Science and Technology demonstrated the creation of ultrathin solar cells by inkjet printing and their placement on surfaces as fragile as air bubbles. Massachusetts Institute of Technology is developing a robust and more scalable version that will be easier to manufacture, is transparent, flexible and conductive and can be added seamlessly onto any surface, turning it into a renewable power source. Although currently nascent technology, the small solar panels could eventually be added as a coating or film to different at-sea technologies.

The miniaturization of sensors also offers an environmentally sustainable and cheap solution for mass data collection. For example, an integrated circuit with wireless communication capabilities and a renewable power source that takes the form of a tiny microchip less than the size of a ladybird has been developed by Northwestern University. The miniature sensors were designed with natural drag mechanisms (like a maple tree propeller seed), to fall at a slow rate and in a controlled manner. As they descend, atmospheric conditions are measured and monitored, including air pollution, atmospheric pH and sun exposure as a function of altitude. Miniature, naturally descending sensors, perhaps mimicking marine snow, could in the future be used to measure and monitor the distribution of pollutants or other particles over or in the ocean and surface fluxes at the ocean-atmosphere boundary.

The ocean is a harsh environment generally inhospitable to humans. Direct human interaction is limited to the surface and scuba diving depths – barely scratching the surface of the ocean. However, integrating robotics with sophisticated haptic suits has formed real-world physical avatars on land, opening the potential for expanding human reach in the ocean in a novel way. Haptic suits and mixed reality techniques allow humans to interact remotely with objects, providing sensory input to the brain. Recently, the Italian Institute of Technology successfully demonstrated its iCub3 avatar robot connected to its iFeel haptic suit. In the demonstration, a human in Genoa, Italy, controlled a robot 300km away in Venice, Italy, and experienced what the robot sensed. Haptic technology and mixed reality techniques are already being developed for use in the ocean. For example, delicate marine specimens have been collected while scuba diving using a marine haptic-gloved human hand controlling a soft manipulator for greater precision and gentleness than a bare human hand. Adapting this technology more robustly for oceanography, where a robot avatar could be immersed in the marine environment, will allow scientists to see, feel, smell and touch organisms in remote marine locations in real time, enabling interactions as if humans were indeed in the ocean. Such capabilities would enable remote and high-risk environments to be explored in different ways, allowing scientists to work in the aquatic ecosystem while physically remaining onshore and providing an opportunity for broader engagement by those who cannot go to sea or live at great distances from the shoreline.

Conclusion

In parallel to technological development on land, the pace of advances in marine technology is happening increasingly faster, allowing scientists to collect in situ oceanographic data at larger scales, at reduced costs, more efficiently and for longer durations in every ocean environment. While vessels remain integral to conducting research at sea by providing a platform for multidisciplinary studies, testing innovative technologies and serving as a mothership for remote and autonomous vehicles, robotic platforms make remote and harsh environments more accessible, provide an avenue for routine data collection, and extend the reach and capabilities of a vessel. Ongoing innovation in sampling capabilities provides the ability to gather more diverse and robust multidisciplinary data, such as 3D imagery and in situ characterization of marine species. Scientists and engineers working across multiple disciplines are imagining and conceptualizing the possible, designing and building the technology, and interpreting and analysing the data. This cross-disciplinary innovation, combining oceanographic expertise with expertise from other fields, is exciting and the next decade will give us an incredible capacity to understand and predict the complex interactions of the ocean, land and atmosphere.

References

Burns, J. A., Becker, K. P., Casagrande, D., Daniels, J., Roberts, P., Orenstein, E., Vogt, D. M., Teoh, Z. E., Wood, R., Yin, A. H., Genot, B., Gruber, D. F., Katija, K., Wood, R. J., & Phillips, B. T. (2024). An in situ digital synthesis strategy for the discovery and description of ocean life. Sci. Adv., 10 (3), eadj4960. https://doi.org/10.1126/sciadv.adj4960

Phillips, B. T., Becker, K. P., Kurumaya, S., Galloway, K., Wittredge, G., Vogt, D., Teeple, C., Rosen, M., Pieribone, V., Gruber, D., & Wood, R. (2018). A dexterous, glove-based teleoperable low-power soft robotic arm for delicate deep-sea biological exploration. Sci. Rep. 8, 14779. https://doi.org/10.1038/s41598-018-33138-y

Seabrook, S., Mackay, K., Watson, S., Clare, M., Hunt, J., Yeo, I., Lane, E., Clark, M., Wysoczanski, R., Rowden, A., Hoffmann, L., Armstrong, E., & Williams, M. (2022). Pyroclastic density currents explain far-reaching and diverse seafloor impacts of the 2022 Hunga Tonga Hunga Ha’apai eruption. PREPRINT available at Research Square. https://doi.org/10.21203/rs.3.rs-2395332/v1

Vogt, D., Becker, K., Phillips, B., Graule, M., Rotjan, R., Shank, T., Cordes, E., Wood, R., & Gruber, D. (2018). Shipboard design and fabrication of custom 3D-printed soft robotic manipulators for the investigation of delicate deep-sea organisms. PLOS ONE 13(8), e000386. https://doi.org/10.1371/journal.pone.0200386