University of Maryland researchers are conducting an initial simultaneous analysis of hundreds of earthquakes to identify echoes from features deep within the earth.
Geophysicists from the University of Maryland analyzed thousands of recordings of seismic waves, sound waves traveling through the earth to identify echoes from the boundary between the earth̵
Researchers are uncertain about the composition of these structures, and previous studies have provided only a limited picture of them. A better understanding of their shape and extent can help reveal the geological processes that take place deep within the earth. This knowledge can provide clues as to how plate actonics work and the development of our planet.
The new research provides the first comprehensive picture of the core mantle boundary over a wide area with such a detailed resolution. The study was published on June 12, 2020, the issue of the journal Science.
The researchers focused on echoes from seismic waves traveling under the Pacific Ocean basin. Their analysis revealed a previously unknown structure under the volcanic Marquesas Islands in the South Pacific and showed that the structure under the Hawaiian Islands is much larger than previously known.
“By looking at thousands of nuclear mantle boundaries at one time, instead of focusing on a few at a time, as usual, we’ve gained a whole new perspective,” said Doyeon Kim, a postdoctoral fellow at the Department of Geology and lead author of the paper. “This shows us that the core mantle boundary has lots of structures that can produce these echoes, and that was something we didn’t realize before because we only had a narrow view.
Earthquakes generate seismic waves beneath the earth’s surface, traveling thousands of miles. When the waves meet changes in rock density, temperature or composition, they change speed, bend or spread, providing echoes that can be detected. Echoes from nearby structures come faster, while those from larger structures are higher. By measuring the travel time and amplitude of these echoes as they arrive at seismometers at different locations, researchers can develop models of the physical properties of rocks hidden beneath the surface. This process is similar to the way bats are echolocated to map their environment.
For this study, Kim and his colleagues looked for echoes generated by a specific type of wave, called a shear wave, as it travels along the core mantle boundary. In a recording from a single earthquake, known as a seismogram, echoes from diffracted shear waves can be difficult to distinguish from random noise. But looking at many seismograms from many earthquakes at once can reveal similarities and patterns that identify echoes hidden in the data.
Using a machine learning algorithm called Sequencer, the researchers analyzed 7,000 seismograms from hundreds of earthquakes of magnitude 6.5 and higher that occurred around the Pacific from 1990 to 2018. Sequencer was developed by the new study co-author from Johns Hopkins University and Tel Aviv University. find patterns in radiation from distant stars and galaxies. When applied to earthquake seismograms, the algorithm detected a large number of shear wave echoes.
“Machine learning in earth science is growing rapidly and a method like Sequencer allows us to systematically detect seismic echoes and gain new insights into the structures at the base of the mantle, which has largely remained enigmatic,” Kim said.
The study revealed some surprises in the structure of the core mantle boundary.
“We found echoes of about 40% of all seismic wave paths, ” said Vedran Lekić, associate professor of geology at UMD and co-author of the study. “It was surprising because we expected them to be more rare, and what that means is the abnormal structures at the core mantle border are much more widespread than previously thought.”
The researchers found that the large patch of very dense, hot material at the core mantle boundary below Hawaii produced uniquely high echoes, indicating that it is even larger than previous estimates. Known as ultralow velocity zones (ULVZ), such spots are found at the roots of volcanic plumes, where hot rocks rise from the core mantle boundary to produce volcanic islands. ULVZ under Hawaii is the largest known.
This study also found a previously unknown ULVZ in the Marquesas Islands.
“We were surprised to find such a great feature in the Marquesas Islands that we didn’t even know it existed before,” Lekić said. “This is really exciting because it shows how the Sequencer algorithm can help us contextualize seismogram data around the world in a way we couldn’t before.”
# # #
Reference: “Sequencing Seismograms: A Panoptic View of Scattering in the Core-Mantle Boundary Region” by D. Kim, V. Lekić, B. Ménard, D. Baron, and M. Taghizadeh-Popp, June 12, 2020, Science.
DOI: 10.1126 / science.aba8972
This work is supported by the Packard Foundation Fellowships and the National Science Foundation (Award No. EAR1352214). The content of this article does not necessarily reflect the views of these organizations.