Development of Efficient X-Ray CT Image Segmentation Techniques for Quantitative Analysis of Phase Distributions and Flow Processes in Porous Media Markus Tuller, Pavel Iassonov and Marcel Schaap This award is funded under the American Recovery and Reinvestment Act of 2009 (Public Law 111-5). Nondestructive imaging methods such as X-Ray Computed Tomography (CT) yield high-resolution 3 D representations of pore space and fluid distribution within porous materials. Steadily increasing computational capabilities and easier access to industrial and synchrotron X Ray CT facilities have contributed to a recent surge in microporous media research with objectives ranging from theoretical aspects of fluid and interfacial dynamics at the pore-scale to practical applications such as (D)NAPL transport and dissolution. In recent years, significant efforts and resources have been devoted to improve CT technology, micro-scale analysis, and fluid dynamics simulations. However, the development of adequate image segmentation methods for conversion of grayscale CT volumes into a discrete form that permits quantitative characterization of pore space features and subsequent modeling of liquid distribution and flow processes is lacking far behind. A preliminary study revealed that most of the documented segmentation methods that were originally developed for optical character recognition, pattern analysis, or medical research yield vastly different results when applied to X-Ray CT data of porous materials. Furthermore, there is virtually no commercial or open-source software with advanced image segmentation capabilities suitable for automated processing of large CT data sets available. To overcome these shortcomings we will identify the most promising segmentation methods and develop new hybrid algorithms tailored to industrial and synchrotron X-Ray CT data of natural porous media. Lattice Boltzmann simulations within X-Ray CT-derived discretized pore space compared to independently measured hydraulic properties will guide the development process. The generated algorithms will be combined in a user-friendly, stand-alone, and well documented software package and made available to the scientific community.