Spatiotemporal image analysis of water flow in porous media for numerical transport modelling

For more than a decade a spatiotemporal visualization tool for transport process observations in dense material by means of PET (positron emission tomography) was developed [1-5]. Such quantitative GeoPET images are exceptionally sensitive to displacements of pico molar tracer quantities detected within 1 mm grids on laboratory/drill core scale.
Now we reached a strategic milestone: A custom made image analysis algorithm is capable of quantitatively extracting velocity and porosity fields from such GeoPET image time series, even if the 4D image information includes discontinuous flow patterns (due to bottle neck effect related detection limits) and localized image artifacts. We present our approach with a concrete example: From an observed flow field in a dense core material the effective porosity and velocity field is extracted and this data is used in a finite element based transport simulation.
[1] Richter, M., et al. (2000) Z.angew.Geol. 46(2): 101-109.
[2] Gründig, M., et al. (2007) App.Geochem. 22: 2334-2343.
[3] Zakhnini, A., et al. (2013) Comp.Geosci. 57 183-196.
[4] Kulenkampff, J., et al. (2008) Phys.Chem.Earth 33: 937-942.
[5] Kulenkampff, J., et al. (2015) Clay Min. accepted 2015.