From Blind Spots to Blueprints: Adaptive Imaging of Earth and Planetary Interiors with Novel Seismic Data

Project Summary:

How can we probe hidden parts of a planet’s interior? This project brings together researchers from the USA and Bristol to develop strategies for applying state-of-the-art computational methods to seismic data from Earth’s oceans and terrestrial planets. Seismology reveals Earth’s inaccessible interior, illuminating mantle circulation and plate tectonics. A long-standing challenge, however, is the lack of seismic coverage in the 
oceans. To help overcome this gap, the MERMAID project records seismic signals using autonomous robots drifting in the deep ocean. Integrating these complex new observations with traditional seismic records requires advanced self-parameterizing inversions and Bristol’s Isambard-AI supercomputing facility. These approaches will address fundamental questions about mantle processes and the structure beneath oceanic hotspots. Methods developed for MERMAID data also translate naturally to planetary seismology, where data coverage is similarly sparse and waveforms complex. Our work will therefore contribute to deciphering the diverging evolutionary histories of the Moon, Mars, and Earth.

Visitor Biography:

Vedran (Ved) Lekić is a Professor and Director of Graduate Studies in Geological, Environmental, and Planetary Sciences at the University of Maryland, College Park. He received his A.B. from Harvard University (2004), his Ph.D. from the University of California, Berkeley (2009), and postdoctoral training at Brown University. His research spans scientific targets from Earth's inner core to the critical zone, and from the Martian interior to the icy moons of the outer solar system. He has published on large low-shear-velocity provinces, upper mantle tomography, lithospheric structure and receiver functions, ice shelf seismicity, near-surface characterization for landmine and unexploded ordnance detection, and interior structures of Mars, the Moon, and icy satellites. Across these targets, he develops novel inverse methods, full-waveform modelling, machine learning approaches, and multi-geophysics imaging techniques. He has received a Packard Fellowship, an NSF CAREER Award, the Seismological Society of America Charles F. Richter Award.