We used the finite-element computer program UNSAT2 to quantify the horizontal distribution of unsaturated downward flux in porous tuff as a function of hydrogeologic heterogeneities under two different recharge rates. The distribution of downward flux is important because it influences ground-water travel time at a potential geologic repository for high-level radioactive wastes. Hydraulic properties of the Topopah Spring Member of the Paintbrush Tuff Formation at Yucca Mountain, Nevada, were selected for use in the eight simulations that were conducted. All simulations were run essentially to steady state (referred to herein as quasi-steady state). The simulations show that (1) heterogeneities in an isotropic porous rock matrix will cause downward flux to be distributed nonuniformly, (2) zones in which saturated matrix hydraulic conductivity is less than the downward flux tend to develop positive pressures that may divert flow into fractures if they exist, (3) one-dimensional analysis of vertical flow in the unsaturated zone is insufficient to ensure that fracture flow does not occur even if the true magnitude of vertical flux is known, and (4) the true spatial distribution of hydraulic conductivity above the regional water table must be determined in order to obtain the true spatial distribution of downward flux. This analysis assumes constant spatial and temporal distribution of recharge for each simulation, but the differences among simulations verify the fact that changing recharge rates also play a significant role in the distribution of downward flux owing to the influence of recharge rate on the development of positive pore pressures in the otherwise unsaturated rock.
The primary significance of this research is as follows. (1) We have introduced a new method of UNSAT2 time step control that facilitates convergence even in complex, large-scale partially saturated hydrogeologic environments. Previous studies have experienced convergence problems with UNSAT2 under complex hydrogeologic conditions. (2) Nonuniform downward flow should be anticipated in heterogeneous hydrogeologic environments at a heterogeneity scale of 50 m and a model scale of hundreds of meters, even if recharge is uniform spatially. (3) The impact of heterogeneities in the unsaturated zone on downward flux can now be quantified. (4) Hydrogeologic heterogeneities can create zones of positive pressure and consequent possible fracture flow in the unsaturated zone even when the recharge rate is less than the average saturated hydraulic conductivity, but most certainly when the recharge rate exceeds saturated hydraulic conductivity. (5) This type of study provides a basis for designing hydrogeologic property testing programs and recharge distribution testing programs that are compatible with the design of models that can characterize downward flux. (6) This work shows that field documentation of the absence of positive pressures above the water table constitutes the best evidence of exclusive matrix flow (absence of fracture flow) within Yucca Mountain. If positive pressures are encountered in the field, then fracture flow probably exists within Yucca Mountain.