Computational Geosciences

Computational geoscience is characterized by large-scale nonlinear models that couple multiple physical, chemical, and biological processes over a wide range of length and time scales. The solution of the complex interdisciplinary problems requires advanced numerical algorithms running on high performance computers. To tackle these challenges researchers at the JSG collaborate with the Institute for Computational Engineering and Sciences (ICES) and the Texas Advanced Computing Center (TACC).

The Computational Geoscience discipline is the focal point of computational and modeling activities at the Jackson School of Geosciences and serves to:

• Bring together computational researchers from across the three units of the Jackson School and across all research themes
• Foster a culture of large-scale modeling and simulation within JSG
• Energize research at the interfaces of modeling and data, and lead to wider application of inverse methods.
• Develop a unique curriculum to educate a new generation of geoscientists well-versed in computation and ready to become leaders in their field.

Climate Modeling

Jackson School climate modeling activities include integration with comprehensive global and regional climate system models developed at NCAR and contributions of process components to these models. The research threads have emphasized fundamentals of climate dynamics, assimilation, and prediction, climate over land and land processes, especially those involving canopy radiation and those coupled to the hydrological cycle. The latter include snow, frozen ground, water tables, runoff and vector based river routing. Mechanism and processes for floods and drought are of especial current interest.

Geodynamic Modeling

Mantle convection drives plate tectonics and continental drift and, in turn, controls the occurrence of earthquakes and volcanoes, mountain building, and long-term sea level change. The major challenges in modeling global mantle convection lie in resolving the wide range of space and time scales and the orders of magnitude variation in material properties. Computational geodynamics research in the Jackson School is aimed at creating advanced mathematical and computational models of mantle convection processes that overcome the above challenges through advanced discretizations, adaptive mesh refinement, and scalable parallel solvers that run on state-of-the-art supercomputers. A new thrust is to develop inverse methods that assimilate observational data into mantle flow models.

Modeling flows in porous media

Porous media are ubiquitous throughout the geosciences and the computational modeling of porous media flows is a common interest across all three units of the Jackson School. Activities range from complex large-scale simulations of specific field sites to nanoscale transport models of fundamental geological and environmental processes. Topics include the dynamics of marine methane hydrates, fluid flow in nanopores of shale strata, reactive transport during diagenesis, partial melting and melt migration in the earth’s mantle. Researchers optimize the design of enhanced oil and gas recovery and geological CO2 storage projects or the sustainable management of water resources.

Inverse modeling

One of the central challenges in computational geosciences is the systematic assimilation of observational data into large-scale simulations to address and characterize model parameters and their associated uncertainties. This is necessary to account for measurement error, the scale-dependency of those measurements, and ambiguity in relating physical earth properties to the observations. The Jackson School has been a leader in the development of inverse methods for data assimilation and their application to such areas as seismology, thermal history and climate modeling.

Geophysics

Research in theoretical and computational geophysics includes: the solutions to inverse problems to estimate complex multi-parameter earth models from large data sets; development of numerical methods to simulate wave propagation and deformation in complex materials via finite element and finite difference methods; inference from and analysis of complex systems, such as Earth’s climate variations; and development of algorithms using parallel processing architectures. Researchers relate geophysical datasets to physical properties at scales including whole-earth structure, plate tectonics, sedimentary basins, fluid reservoirs, and pore scales.

Lithospheric deformation modeling

One of the most important problems plate tectonics is to develop a model for solid deformation of the lithosphere with localization over narrow shear zones in the rigid crust and mantle, as well as viscoplastic flow in the ductile lithosphere. To validate such models, numerical simulations of lithospheric deformation must often carry on over the tens of millions years. Therefore, a realistic description and understanding of natural processes requires both the development of a mathematical model and its accurate and fast numerical solution to identify the corresponding parameter regimes. The Jackson School has been a leader for many years in integrating new numerical techniques in computational mechanics and recent geophysical constraints. This effort has allowed for the development of new geological concepts for rifting, mountain building and subduction deriving from forward models of lithospheric deformation.

3D data analysis

Full exploitation of volumetric data sets acquired by X-ray computed tomography (CT) on rocks, fossils, meteorites, and other materials to answer geologic questions requires development of innovative and specialized analysis techniques and methodologies. The Jackson School has been a leader in creating these capabilities to exploit the unique data being generated at its world-leading CT facility. Applications include measuring size, shape and spatial and contact relationships of minerals, clasts and vesicles; measuring the density and anisotropy of trabecular bone fabrics in vertebrate fossils; imaging pore networks and fluid displacement within them; and quantifying fracture roughness and aperture variation and their effect on fluid flow.