About Seismic Tomography 2020

Seismic shear wave speed at 100 km depth. | ETH Zurich

Now more than 40 years old, the field of seismic tomography has provided an unprecedented look at the 3D internal structures of the Earth. But as tomography enters its “middle age,” it’s time to take stock, said SSA Past-President Peter Shearer.

Shearer, a professor of geophysics at Scripps Institution of Oceanography, proposed a new SSA meeting dedicated to tomography to answer questions facing the field, such as “are our models continuing to improve or have they plateaued? Are model uncertainties well understood? And what is most important for the future: better theory, faster computers or denser data coverage?” he said.

Seismic Tomography 2020: What Comes Next?, which will be held 9-11 October 2020 in Toronto, Canada, will convene international experts to tackle these questions and others, according to the meeting’s Co-Chairs Clifford Thurber and Andreas Fichtner. 

“Parts of the field are mature,” said Thurber, a professor of seismology at the University of Wisconsin-Madison, “but there are things on the data and techniques side and the applications side that make it exciting to do this meeting right now.”

The narrow definition of seismic tomography refers to traditional efforts to measure arrival times of seismic waves as they pass once through the Earth, using those arrival times to identify different structures within the planet by how the waves are slowed or deflected. But the field has become much broader, said Fichtner and Thurber, to encompass a wide range of cutting-edge techniques used to image the internal structure of the Earth. 

Clifford Thurber and Andreas Fichtner

“There is a whole zoo of waves that bounce back and forth, reflected from internal boundaries,” noted Fichtner, an associate professor of seismology and wave physics at ETH Zürich, “and seismologists are using this whole zoo of waves.”

The co-chairs are exploring a range of topics to include in the meeting, which runs two and 1/2-days, in part guided by the types of submissions they receive. (Abstract submissions open 1 April 2020.) Research from the meeting will also be included in a collection for a future special issue of the Bulletin of the Seismological Society of America.

Along with classic tomography research, Thurber said he hopes to see some submissions on time-dependent tomography, which is used to monitor oil reservoirs and hydrological basins; studies spanning the scales from near-surface imaging to structures tens to thousands of kilometers below the surface; computationally demanding but data-rich full waveform tomography; and joint inversion techniques that include non-seismic data in their models.

Other topics might include the gaps in geographical coverage by tomography instrumentation, the development of new instruments that can be deployed in environments such as the ocean bottom or that piggyback off existing technology such as fiber optic cables, and ways to improve the computational bottleneck that tomographers often face in their modeling.

Spectral-element simulation of the 2011 Tohoku, Japan earthquake. | Salvus

“We want to have a fairly broad scope to get some cross-fertilization of ideas,” said Thurber.

Uncertainty analysis could be another important topic area to consider, especially as it is difficult to compare images made by different research groups using different methods, Fichtner said. “At the moment, we are very well able to produce colorful images of the internal structure of the Earth, but we hardly know how good they are.”

Fichtner also suggested that studies in quantitative interpretation are a big part of moving the field forward. “We produce these 3D maps of how fast waves propagate in certain regions of the earth, but what people are more interested in is the temperature, composition and water content of these regions, since these are the actual physical quantities that drive the dynamics of the Earth,” he explained.

Thurber and Fichtner say that the meeting will likely attract tomography experts, and that they plan to invite prominent scientists from adjacent fields to interact with the tomographers. 

The co-chairs may ask presenters to offer ambitious, “potentially crazy” ideas about the future of seismic tomography. “We want to hear some courageous, forward-looking, wild ideas of where we want to go,” said Fichtner, “not so standard or conservative—even if it doesn’t sound realistic.”