We advance geophysical and multi-modal research of explosion and related phenomena to support the nation’s efforts to better quantify, understand and monitor for nuclear testing activity.
The GEM team specializes in the quantitative study of waveform processes that are sourced by explosions emplaced near Earth’s surface over multiple spatial and temporal scales. Team members lead efforts to model dynamic features of explosion sources, the propagation of their transient signals through and along medium interfaces, and the detection, analysis and inference of source parameters and medium properties from these signals. In particular, the GEM team maintains an end-to-end research capability to acquire, detect, locate and characterize observables from both natural and human-made sources of mechanical and optical energy.
Coordinate and perform large-scale physical and virtual monitoring experiments of buried and aboveground explosions
Exploit new technologies to develop next-generation capabilities that include Distributed Acoustic Sensing (DAS), machine learning and artificial intelligence plus virtual containers.
Measure and quantify impacts of climate change on stratospheric sound propagation, hydroacoustic propagation in the Arctic, and generation of methane-sourced gas emission craters in the Russian Arctic.
Expertise and sensing capability to fuse multi-domain, geophysical observations of explosion sources and their background emissions.
Provide cutting edge, quantitative uncertainty analysis of explosion source parameters via a toolbox of empirical and mathematical techniques.
Produce accurate, empirical estimates of explosion yields from scattered seismic waveform coda.
Quantify and greatly reduce the uncertainties in moment tensors (MTUQ), using seismic observations collected at regional distances from underground nuclear test sites.
Predict 3-D ray-paths from both stationary and moving infrasound sources over the whole atmospheric column (infraGA).
Leverage Bayesian methods to produce joint, seismo-acoustic locations from near surface explosion sources.
Use graph-theory methods to quantify false association and event building errors in seismic networks.
Combine multiple Rayleigh wave polarization measurement techniques to characterize explosion sources in very complex tectonic environments.
Empower big data techniques to predict high-fidelity maps of seismic attenuation over the entire globe.
Demonstrate probabilistic methods to predictively screen small earthquakes from explosions at local distances with seismic phase ratios.
Develop novel data stream fusion methods to integrate evidence of explosions from multiple waveform and optical modalities, for both signal detection and source identification.
Advance generalized hypothesis testing techniques to adaptively detect and characterize noisy threat sources in challenging signal environments.
Characterize climate-induced variability in both infrasonic and hydroacoustic signal environments to quantify propagation uncertainties.
Development of adaptive, eigen-processing techniques to detect signals and quantify performance of IMS arrays in near-real time.