Page 44 - Summer 2021

MOBILE EARTHQUAKE RECORDING
Mission Objectives
Mapping the three-dimensional structure inside our planet is key to elucidating the origin of Earth, its subse- quent evolution, and its ongoing dynamics (Romanowicz, 2008). Information on deep-Earth structure is gleaned via seismic wave speed tomography and other advanced seismic-imaging techniques (Tromp, 2020) from the transient elastic wavefield emitted by earthquakes that are large enough to be recorded worldwide.
The interior of our planet consists of roughly concentric shells named crust, mantle, liquid outer, and crystalline inner core, but such a one-dimensional subdivision is inadequate. To begin, the crust is very heterogeneous, broken up into a patchwork of tectonic plates and rang- ing in thickness from 0 km at midocean ridges to some 70 km under the Andes and the Himalayas. Further- more, the solid mantle slowly moves about, ultimately mixing but maintaining inhomogeneities of temperature, chemical composition, and crystal structure. Much like the crust, the base of the mantle, some 2,891 km down into the Earth, is also extremely heterogeneous, and the core-mantle boundary is a “mountainous” surface. Last, even the solid inner core displays strong contrasts in physical properties, related to its growth by the continued solidification of the liquid outer core. Its seismic wave speed varies from place to place, often anisotropically, that is, depending on the look angle. Taken all together, a multitude of observations shows that the interior of the Earth manifests significant lateral variations from merely depth-dependent structure, which therefore requires three-dimensional mapping.
Different rocks and minerals all have different seismic wave speeds, but, to a good approximation, seismic wave speeds in the mantle are primarily a record of its internal temperature distribution. Unlike the speed of sound in air, which increases with temperature, hot rocks transmit sound more slowly. In contrast, when rocks are colder, their wave speed increases. (For more on the acoustic properties of rocks, see TenCate and Remillieux, 2019.) Because hotter rocks are buoyant and colder rocks are denser than their surroundings, the mantle slowly con- vects, deforming internally. Tomographic images reveal zones of high seismic wave speed that outline sinking sheets of subducted material (van der Hilst et al., 1997), while isolated columnar upwellings or mantle plumes are manifest as low wave speed regions (Montelli et al.,
2006). Seismic wave speed maps provide a snapshot of the Earth’s interior temperature distribution as it slowly, but inexorably, cools down overall.
Whatever the nature of the seismological probing method, the ability to measure small variations in seismic propa- gation velocities is crucial. Land-based global networks help us map origin times and locations of earthquake sources. Seismic waves from distant earthquakes convert at the ocean bottom to acoustic pressure variations in the water column. On the receiving end, determining the location of the recording station and keeping track of time are of fundamental importance. To measure seis- mic velocities of earthquake arrivals with sensors drifting at depth, we must quickly tag the instrument’s location and time of recording. In practice, this entails surfac- ing within days, if not hours, for the Global Positioning System location determination and time acquisition to perform instrumental clock-drift corrections and to transmit the detected data via satellite.
Not all earthquakes are created equal, and to be useful for tomographic imaging of the Earth’s mantle, we must be judicious in reporting “data” and avoid false triggers. Diving, surfacing, and data transmission are energetically costly. Although future generation instruments might run on thermal-energy conversion (Jones, 2019), the instruments of today have a finite battery supply.
In the end, what we require is an autonomous robotic oceanic vehicle with a hydrophone that sensitively hears, actively listens to, and expediently reports the sounds from distant earthquakes. And, to truly conquer the oceans, we need a large number of them.
First Sound
The first prototype, MERMAID-001, fulfilled the core requirements of seismological functionality, namely, earthquake detection. Figure 3 shows the design, includ- ing the acoustic payload of an off-the-shelf hydrophone. Data were stored on a flash memory card.
Over the course of 3 recovered field tests, MERMAID-001 gathered a mere 120 hours of acoustic pressure data from a depth of around 700 m offshore from La Jolla. Several positive earthquake identifications stood out from the noise. One of these was a tremor large enough and dis- tant enough to prove the utility of the new instrument for
44 Acoustics Today • Summer 2021