Making the invisible Earth visible:
Our group uses recordings of ground vibrations from seismic sensors across the globe, combined with advances in high-performance computing and machine learning, for high-resolution subsurface Earth-imaging and improved earthquake detection with sparse seismic networks. We are addressing important Earth-imaging problems, which require extracting and interpreting data obtained in challenging, data-poor settings, from marine environments and sparse seismic networks on the African continent.
Submarine Imaging of Ocean Plates
Our planet is 70% ocean-blue. We explore fundamental and applied problems in the field of marine seismology and geophysics. How sharp is the base of the oceanic plate? Can we develop techniques that will enable “Submarine Seismic Imaging of the Base of the Oceanic Plate, by Silencing the 'Singing' of Sediments?” This work will improve the utility of seismic data obtained from ocean bottom experiments (Ph.D. student: Evan Zhang)
Lithospheric Discontinuities in the US
Why is there a global discontinuity internal to the old and cold continental lithosphere? Can we use multiple geophysical observations to test models that explain the presence of these discontinuities in the oceans and continents? Through a funded NSF grant, our research group is collaborating with leading scientists and using multiple geophysical datasets provided by EarthScope to explore this question.
Geophysics in Africa
Lithospheric Layering in Africa
The African continent hosts remarkable geologic and tectonic features that make the continent not only full of natural resources but also susceptible to natural hazards. These features include the oldest cratons, the African Superwell in the South, and the divergence with the American plates in the West. Our group is using advances in computational seismology to investigate cratonic layering on the continent with implications for understanding a crucial part of the Earth's history.
Autoadaptive Earth Imaging
Bayesian Earth Imaging
The geophysical investigation of the puzzle in the cratonic lithosphere requires new methods of extracting data from noisy datasets, sparse seismic networks, and using different geophysical datasets. members of our group are training machines to learn from data using high-performance computing tools and stochastic methods, e.g., Bayesian statistical techniques to improve the resolution and fine-scale structure of the crust and upper mantle lithosphere.