Predictive Modeling of a Simple Field Matrix Diffusion Experiment Addressing Radionuclide Transport in Fractured Rock. Is It So Straightforward?

The SKB GroundWater Flow and Transport of Solutes Task Force is an international forum in the area of conceptual and numerical modeling of groundwater flow and solute transport in fractured rocks relevant for the deep geological disposal of radioactive waste. Two in situ matrix diffusion experiments in crystalline rock (gneiss) were performed at POSIVA’s ONKALO underground facility in Finland. Synthetic groundwater containing several conservative and sorbing radiotracers was injected at one end of a borehole interval and flowed along a thin annulus toward the opposite end. Several teams performed predictive modeling of the tracer breakthrough curves using “conventional” modeling approaches (constant diffusion and sorption in the rock, no or minimum rock heterogeneity). Supporting information, derived from small-scale laboratory experiments, was provided. The teams were free to implement different concepts, use different codes, and apply the transport and retention parameters that they considered to be most suited (i.e., not a benchmark exercise). The main goal was the comparison of the different sets of results and the analysis of the possible differences for this relatively simple experimental setup with a well-defined geometry. Even though the experiment was designed to study matrix diffusion, the calculated peaks of the breakthrough curves were very sensitive to the assumed magnitude of dispersion in the borehole annulus. However, given the very different timescales for advection and matrix diffusion, the tails of the curves provided information concerning diffusion and retention in the rock matrix regardless of the magnitude of dispersion. In addition, although the task was designed to be a blind modeling exercise, the model results have also been compared to the measured experimental breakthroughs. Experimental results tend to show relatively small activities, wide breakthroughs, and early first arrivals, which are somewhat similar to model results using large dispersivity values.