ABSTRACT
The field of earth sciences is experiencing rapid changes as a result of the growing understanding of environmental physics, along with recent advances in measurement technologies, and dramatic increases in computing power. More complex, spatially explicit computer models are now possible, allowing for a more realistic representation of systems of interest. However, these detailed models are typically computational intensive, requiring significant time to run and produce the desired output. Perhaps more importantly, spatially explicit models typically contain a wealth of parameters that cannot be derived directly and need to be estimated through calibration. In this talk, I will highlight recent advances in adaptive Markov Chain Monte Carlo sampling for computationally efficient model calibration and uncertainty estimation. I will demonstrate the advantages of this method using several synthetic benchmark problems, including a real-world study involving the estimation of permeability and dispersion parameters in a high resolution three-dimensional flow and particle tracking model using Magnetic Resonance Imaging (MRI) data of a conservative tracer moving through a flow cell. This inverse problem is solved using high performance computing.

LINKS

• Jasper Vrugt Wins Soil Science Society of America Early Career Award
• Vrugt, J. A., H. V. Gupta, L. A. Bastidas, W. Bouten, and S. Sorooshian (2003), Effective and efficient algorithm for multiobjective optimization of hydrologic models, Water Resour. Res., 39(8), 1214, doi:10.1029/2002WR001746.
• Vrugt, J. A., W. Bouten, H. V. Gupta, and S. Sorooshian (2002), Toward improved identifiability of hydrologic model parameters: The information content of experimental data, Water Resour. Res., 38(12), 1312, doi:10.1029/2001WR001118.