Modelling matrix diffusion - Results of a bench mark study

Code comparison and benchmarking studies are important for testing the correct numerical implementation, especially in cases where analytical solutions for complex problems are not available. We will present results of such a code-comparison study for the calculation of matrix diffusion processes in fractured rocks. The benchmark investigates the accuracy of a new analytical solution which was recently introduced into the code GeoSys/RockFlow (GS/RF). We compared GS/RF with the well tested code PICNIC for advective-dispersive transport in a single fracture with constant aperture, advective-dispersive transport in a single fracture and additional matrix diffusion, and advective-dispersive transport in a single fracture plus matrix diffusion and linear sorption in the porous rock matrix. For all investigated cases the results of the two codes agree very well.