On 23 May 1989, a violent earthquake rumbled through the remote underwater environs near Macquarie Island, violently shaking the Australian research station on the island and causing noticeable tremors as far away as Tasmania and the South Island of New Zealand. The seismic waves it generated rippled through and around the planet, circling the surface several times before dying away.
Seismographs everywhere in the world captured the motion of these waves, and geoscientists immediately analyzed the recorded waveforms. The magnitude 8.2 strike-slip earthquake had rocked the Macquarie Ridge Complex (MRC), a sinuous underwater mountain chain extending southwest from the southern tip of New Zealand’s South Island. The earthquake’s great magnitude—it was the largest intraoceanic event of the 20th century—and its slip mechanism baffled the global seismological community: Strike-slip events of such magnitude typically occur only within thick continental crust, not thin oceanic crust.
Fast forward a few decades: For 2 weeks in late September and early October 2020, nine of us sat in small, individual rooms in a Hobart, Tasmania, hotel quarantining amid the COVID-19 pandemic and ruminating about our long-anticipated research voyage to the MRC. It was hard to imagine a more challenging place than the MRC—in terms of extreme topographic relief, heavy seas, high winds, and strong currents—to deploy ocean bottom seismometers (OBSs). But the promise of unexplored territory and the possibility of witnessing the early stages of a major tectonic process had us determined to carry out our expedition.
Where Plates Collide
Why is this location in the Southern Ocean, halfway between Tasmania and Antarctica, so special? The Macquarie archipelago, a string of tiny islands, islets, and rocks, only hints at the MRC below, which constitutes the boundary between the Australian and Pacific plates. Rising to 410 meters above sea level, Macquarie Island is the only place on Earth where a section of oceanic crust and mantle rock known as an ophiolite is exposed above the ocean basin in which it originally formed. The island, listed as a United Nations Educational, Scientific and Cultural Organization World Heritage site primarily because of its unique geology, is home to colonies of seabirds, penguins, and elephant and fur seals.
Yet beneath the island’s natural beauty lies the source of the most powerful submarine earthquakes in the world not associated with ongoing subduction, which raises questions of scientific and societal importance. Are we witnessing a new subduction zone forming at the MRC? Could future large earthquakes cause tsunamis and threaten coastal populations of nearby Australia and New Zealand as well as others around the Indian and Pacific Oceans?
Getting Underway at Last
As we set out from Hobart on our expedition, the science that awaited us helped overcome the doubts and thoughts of obstacles in our way. The work had to be done. Aside from the fundamental scientific questions and concerns for human safety that motivated the trip, it had taken a lot of effort to reach this place. After numerous grant applications, petitions, and copious paperwork, the Commonwealth Scientific and Industrial Research Organisation (CSIRO) Marine National Facility (MNF) had granted us ship time on Australia’s premier research vessel, R/V Investigator, and seven different organizations were backing us with financial and other support.
COVID-19 slowed us down, delaying the voyage by 6 months, so we were eager to embark on the 94-meter-long, 10-story-tall Investigator. The nine scientists, students, and technicians from Australian National University’s Research School of Earth Sciences were about to forget their long days in quarantine and join the voyage’s chief scientist and a student from the University of Tasmania’s Institute for Marine and Antarctic Studies (IMAS).
Together, the 11 of us formed the science party of this voyage, a team severely reduced in number by pandemic protocols that prohibited double berthing and kept all non-Australia-based scientists, students, and technicians, as well as two Australian artists, at home. The 30 other people on board with the science team were part of the regular seagoing MNF support team and the ship’s crew.
The expedition was going to be anything but smooth sailing, a fact we gathered from the expression on the captain’s face and the serious demeanor of the more experienced sailors gathered on Investigator’s deck on the morning of 8 October.
The Furious Fifties
An old sailor’s adage states, “Below 40 degrees south, there is no law, and below 50 degrees south, there is no God.”
Spending a rough first night at sea amid the “Roaring Forties,” many of us contemplated how our days would look when we reached the “Furious Fifties.” The long-feared seas at these latitudes were named centuries ago, during the Age of Sail, when the first long-distance shipping routes were established. In fact, these winds shaped those routes.
Hot air that rises high into the troposphere at the equator sinks back toward Earth’s surface at about 30°S and 30°N latitude (forming Hadley cells) and then continues traveling poleward along the surface (Ferrel cells). The air traveling between 30° and 60° latitude gradually bends into westerly winds (flowing west to east) because of Earth’s rotation. These westerly winds are mighty in the Southern Hemisphere because, unlike in the Northern Hemisphere, no large continental masses block their passage around the globe.
These unfettered westerlies help develop the largest oceanic current on the planet, the Antarctic Circumpolar Current (ACC), which circulates clockwise around Antarctica. The ACC transports a flow of roughly 141 million cubic meters of water per second at average velocities of about 1 meter per second, and it encompasses the entire water column from sea surface to seafloor.
Our destination on this expedition, where the OBSs were to be painstakingly and, we hoped, precisely deployed to the seafloor over about 25,000 square kilometers, would put us right in the thick of the ACC.
Mapping the World’s Steepest Mountain Range
Much as high-resolution maps are required to ensure the safe deployment of landers on the Moon, Mars, and elsewhere in the solar system, detailed bathymetry would be crucial for selecting instrument deployment sites on the rugged seafloor of the MRC. Because the seafloor in this part of the world had not been mapped at high resolution, we devoted considerable time to “mowing the lawn” with multibeam sonar and subbottom profiling before deploying each of our 29 carefully prepared OBSs—some also equipped with hydrophones—to the abyss.
Mapping was most efficient parallel to the north-northeast–south-southwest oriented MRC, so we experienced constant winds and waves from westerly vectors that struck Investigator on its beam. The ship rolled continuously, but thanks to its modern autostabilizing system, which transfers ballast water in giant tanks deep in the bilge to counteract wave action, we were mostly safe from extreme rolls.
Nevertheless, for nearly the entire voyage, everything had to be lashed down securely. Unsecured chairs—some of them occupied—often slid across entire rooms, offices, labs, and lounges. In the mess, it was rare that we could walk a straight path between the buffet and the tables while carrying our daily bowl of soup. Solid sleep was impossible, and the occasional extreme rolls hurtled some sailors out of their bunks onto the floor.
The seismologists among us were impatient to deploy our first OBS to the seafloor, but they quickly realized that mapping the seafloor was a crucial phase of the deployment. From lower-resolution bathymetry acquired in the 1990s, we knew that the MRC sloped steeply from Macquarie Island to depths of about 5,500 meters on its eastern flank.
We planned to search for rare sediment patches on the underwater slopes to ensure that the OBSs had a smooth, relatively flat surface on which to land. This approach differs from deploying seismometers on land, where one usually looks for solid bedrock to which instruments can be secured. We would rely on the new, near-real-time seafloor maps in selecting OBS deployment sites that were ideally not far from the locations we initially mapped out.
However, the highly detailed bathymetric maps we produced revealed extraordinarily steep and hazardous terrain (Figure 1). The MRC is nearly 6,000 meters tall but only about 40 kilometers wide—the steepest underwater topography of that vertical scale on Earth. Indeed, if the MRC were on land, it would be the most extreme terrestrial mountain range on Earth, rising like a giant wall. For comparison, Earth’s steepest mountain above sea level is Denali in the Alaska Range, which stands 5,500 meters tall from base to peak and is 150 kilometers wide, almost 4 times wider than the MRC near Macquarie Island.
A Carefully Configured Array
Seismologists can work with single instruments or with configurations of multiple devices (or elements) called arrays. Each array element can be used individually, but the elements can also act together to detect and amplify weak signals. Informed by our previous deployments of instrumentation on land, we designed the MRC array to take advantage of the known benefits of certain array configurations.
The northern part of the array is classically X shaped, which will allow us to produce depth profiles of the layered subsurface structure beneath each instrument across the ridge using state-of-the-art seismological techniques. The southern segment of the array has a spiral-arm shape, an arrangement that enables efficient amplification of weak and noisy signals, which we knew would be an issue given the high noise level of the ocean.
Our array’s unique location and carefully designed shape will supplement the current volumetric sampling of Earth’s interior by existing seismic stations, which is patchy given that stations are concentrated mostly on land. It will also enable multidisciplinary research on several fronts.
For example, in the field of neotectonics, the study of geologically recent events, detailed bathymetry and backscatter maps of the MRC are critical to marine geophysicists looking to untangle tectonic, structural, and geohazard puzzles of this little explored terrain. The most significant puzzle concerns the origin of two large underwater earthquakes that occurred nearby in 1989 and 2004. Why did they occur in intraplate regions, tens or hundreds of kilometers away from the ridge? Do they indicate deformation due to a young plate boundary within the greater Australia plate? The ability of future earthquakes and potential submarine mass wasting to generate tsunamis poses other questions: Would these hazards present threats to Australia, New Zealand, and other countries? Data from the MRC observatory will help address these important questions.
The continuous recordings from our OBSs will also illuminate phenomena occurring deep below the MRC as well as in the ocean above it. The spiral-arm array will act like a giant telescope aimed at Earth’s center, adding to the currently sparse seismic coverage of the lowermost mantle and core. It will also add to our understanding of many “blue Earth” phenomena, from ambient marine noise and oceanic storms to glacial dynamics and whale migration.
Dealing with Difficulties
The weather was often merciless during our instrument deployments. We faced gale-strength winds and commensurate waves that forced us to heave to or shelter in the lee of Macquarie Island for roughly 40% of our time in the study area. (Heaving to is a ship’s primary heavy weather defense strategy at sea; it involves steaming slowly ahead directly into wind and waves.)
Macquarie Island presents a natural wall to the westerly winds and accompanying heavy seas, a relief for both voyagers and wildlife. Sheltering along the eastern side of the island, some of the crew spotted multiple species of whales, seals, and penguins.
As we proceeded, observations from our new seafloor maps necessitated that we modify our planned configuration of the spiral arms and other parts of the MRC array. We translated and rotated the array toward the east side of the ridge, where the maps revealed more favorable sites for deployment.
However, many sites still presented relatively small target areas in the form of small terraces less than a kilometer across. Aiming for these targets was a logistical feat, considering the water depths exceeding 5,500 meters, our position amid the strongest ocean current on Earth, and unpredictable effects of eddies and jets produced as the ACC collides head-on with the MRC.
To place the OBSs accurately, we first attempted to slowly lower instruments on a wire before releasing them 50–100 meters above the seafloor. However, technical challenges with release mechanisms soon forced us to abandon this method, and we eventually deployed most instruments by letting them free-fall from the sea surface off the side of the ship. This approach presented its own logistical challenge, as we had accurate measurements of the currents in only the upper few hundred meters of the water column.
In the end, despite prevailing winds of 30–40 knots, gusts exceeding 60 knots, and current-driven drifts in all directions of 100–4,900 meters, we found sufficient windows of opportunity to successfully deploy 27 of 29 OBSs at depths from 520 to 5,517 meters. Although we ran out of time to complete mapping the shallow crest of the MRC north, west, and south of Macquarie Island, we departed the study area on 30 October 2020 with high hopes.
Earlier this year, we obtained additional support to install five seismographs on Macquarie Island itself that will complement the OBS array. Having both an onshore and offshore arrangement of instruments operating simultaneously is the best way of achieving our scientific goals. The land seismographs tend to record clearer signals, whereas the OBSs provide the spatial coverage necessary to image structure on a broader scale and more accurately locate earthquakes.
Bringing the Data Home
The OBSs are equipped with acoustic release mechanisms and buoyancy to enable their return to the surface in November 2021, when we’re scheduled to retrieve them and their year’s worth of data and to complete our mapping of the MRC crest from New Zealand’s R/V Tangaroa. In the meantime, the incommunicado OBSs will listen to and record ground motion from local, regional, and distant earthquakes and other phenomena.
With the data in hand starting late this year, we’ll throw every seismological and marine geophysical method we can at this place. The recordings will be used to image crustal, mantle, and core structure beneath Macquarie Island and the MRC and will enable better understanding of seismic wave propagation through these layers.
Closer to the seafloor, new multibeam bathymetry/backscatter, subbottom profiler, gravity, and magnetics data will advance understanding of the neotectonics of the MRC. These data will offer vastly improved views of seafloor habitats, thus contributing to better environmental protection and biodiversity conservation in the Tasmanian Macquarie Island Nature Reserve–Marine Area that surrounds Macquarie Island and the Commonwealth of Australia’s Macquarie Island Marine Park east of Macquarie Island and the MRC.
Results from this instrument deployment will also offer insights into physical mechanisms that generate large submarine earthquakes, crustal deformation, and tectonic strain partitioning at convergent and obliquely convergent plate boundaries. We will compare observed seismic waveforms with those predicted from numerical simulations to construct a more accurate image of the subsurface structure. If we discover, for example, that local smaller- or medium-sized earthquakes recorded during the experiment have significant dip-slip components (i.e., displacement is mostly vertical), it’s possible that future large earthquakes could have similar mechanisms, which increases the risk that they might generate tsunamis. This knowledge should provide more accurate assessments of earthquake and tsunami potential in the region, which we hope will benefit at-risk communities along Pacific and Indian Ocean coastlines.
Scientifically, the most exciting payoff of this project may be that it could help us add missing pieces to one of the biggest puzzles in plate tectonics: how subduction begins. Researchers have grappled with this question for decades, probing active and extinct subduction zones around the world for hints, though the picture remains murky.
Some of the strongest evidence of early-stage, or incipient, subduction comes from the Puysegur Ridge and Trench at the northern end of the MRC, where the distribution of small earthquakes at depths less than 50 kilometers and the presence of a possible subduction-related volcano (Solander Island) suggest that the Australian plate is descending beneath the Pacific plate. Incipient subduction has also been proposed near the Hjort Ridge and Trench at the southern end of the MRC. Lower angles of oblique plate convergence and a lack of trenches characterize the MRC between Puysegur and Hjort, so it is unclear whether incipient subduction is occurring along the entire MRC.
Testing this hypothesis is impossible because of a lack of adequate earthquake data. The current study, involving a large array of stations capable of detecting even extremely small seismic events, is crucial in helping to answer this fundamental question.
We thank the Australian Research Council, which awarded us a Discovery Project grant (DP2001018540). This research was supported by a grant of sea time on R/V Investigator from the CSIRO MNF. We have additional support from ANSIR Research Facilities for Earth Sounding and the U.K.’s Natural Environment Research Council (grant NE/T000082/1) and in-kind support from Australian National University, the University of Cambridge, the University of Tasmania, and the California Institute of Technology. Geoscience Australia; the Australian Antarctic Division of the Department of Agriculture, Water and the Environment; and the Tasmania Parks and Wildlife Service provided logistical support to install five seismographs on Macquarie Island commencing in April 2021. Unprocessed seismological data from this work will be accessible through the ANSIR/AuScope data management system AusPass 2 years after the planned late 2021 completion of the experimental component. Marine acoustics, gravity, and magnetics data, both raw and processed, will be deposited and stored in publicly accessible databases, including those of CSIRO MNF, the IMAS data portal, Geoscience Australia, and the NOAA National Centers for Environmental Information.
Hrvoje Tkalčić ([email protected]) and Caroline Eakin, Australian National University, Canberra; Millard F. Coffin, University of Tasmania, Hobart, Australia; Nicholas Rawlinson, University of Cambridge, U.K.; and Joann Stock, California Institute of Technology, Pasadena