Jinsong Chen Email & Phone Number

San Ramon, Ca, Us

Berkeley, Ca, Us

• Developed a machine-learning based Bayesian model for high-resolution imaging of oil and gas reservoirs; Demonstrated conditional generative adversarial networks (c-GAN) provide a power tool for deriving conditional probability distribution from complex data sets.• Developed a hierarchical Bayesian model for stochastic inversion of 3D full waveform seismic data; Demonstrated MPI and OpenMP hybrid programming with parallel I/O provides an efficient and scalable approach for solving large-scale inverse problems.• Developed a Bayesian model based on statistical rock-physics relationships and Markov random fields to invert large-scale seismic data.• Developed a hierarchical Bayesian model to estimate groundwater dynamics at a Colorado floodplain using multi-scale historical data and climate information.• Developed a hierarchical Bayesian model to estimate spatiotemporal distributions of geochemical parameters associated with bio-stimulation using spectral induced polarization data.• Developed a hidden Markov model (HMM) to identify field-scale biogeochemical transitions using geochemical and geophysical data.• Extended the open source Java code, ‘Delivery4D’, by incorporating converted-wave data into stochastic inversion of seismic data.

Berkeley, Ca, Us

• Developed a hierarchical Bayesian model to estimate reservoir parameters using marine seismic amplitude versus angle (AVA) and controlled-source electromagnetic (CSEM) data• Investigated the effects of uncertainty in rock-physics models on reservoir parameter estimation using marine seismic AVA and CSEM data• Investigated advantages and disadvantages of Gauss-Newton based least-squares estimation methods and Markov chain Monte Carlo (MCMC) based Bayesian methods for geophysical inverse problems• Developed a Bayesian model to estimate aquifer geometry using seismic refraction data with borehole depth constraints• Developed a state-space Bayesian framework to estimate biogeochemical transformations using time-lapse geophysical data• Developed a Bayesian model to estimate hydrogeological zonation in a contaminated fractured aquifer using cross-well seismic travel time and borehole flow-meter test data• Developed a Bayesian model with Markov chain Monte Carlo (MCMC) sampling for geochemical characterization using geophysical data

Berkeley, Ca, Us

• Developed a joint inversion approach to estimate porosity and gas saturation using 1D marine seismic AVO and EM data, and applied it to the data collected from the Troll site in the North Sea• Developed a joint stochastic model to invert seismic tomographic P- and S-wave travel-time data under the constraints of 2D EM data and borehole measurements using MCMC methods.

Berkeley, Ca, Us

• Analyzed field-scale physical, geophysical, hydrogeological, geochemical, and biogeochemical data, and explored petrophysical relationships among them• Developed a Bayesian model to estimate field-scale geochemical parameters and used Gibbs sampling to explore the joint posterior distribution

北京, Beijing, Cn

• Investigated the interaction between surface and subsurface water in a large irrigation system of Northern China• Developed a model to simulate a large-scale water transfer system from Southern China to Northern China• Taught undergraduate curriculums and supervised diploma projects

Doctor Of Philosophy (Ph.D.), Environmental Engineering

Master'S Degree, Statistics

Master'S Degree, Environmental Fluid Mechanics

Master'S Degree, Hydrology And Water Resources

Bachelor'S Degree, Water Resources Engineering And Applied Mathematics

Bachelor'S Degree, Applied Mathematics

Bachelor'S Degree, Applied Mathematics