Evaluation of uncertainties due to hydrogeological modeling and groundwater flow analysis: steady flow, transient flow, and thermal studies

Starting with regional geographic, geologic, surface and subsurface hydrologic, and geophysical data for the Tono area in Gifu, Japan, we develop an effective continuum model to simulate subsurface flow and transport in a 4 km by 6 km by 3 km thick fractured granite rock mass overlain by sedimentary layers. Individual fractures are not modeled explicitly. Rather, continuum permeability and porosity distributions are assigned stochastically, based on well-test data and fracture density measurements. Lithologic layering and one major fault, the Tsukiyoshi Fault, are assigned deterministically. We conduct three different studies: (1) the so-called base case, in which the model simulates the steady-state groundwater flow through the site, and then stream trace analysis is used to calculate travel times to the model boundary from specified release points; (2) simulations of transient flow during long term pump tests (LTPT) using the base-case model; and (3) thermal studies in which coupled heat flow and fluid flow are modeled, to examine the effects of the geothermal gradient on groundwater flow. The base-case study indicates that the choice of open or closed lateral boundaries has a strong influence on the regional groundwater flow patterns produced by the models, but no field data exist that can be used to determine which boundary conditions are more realistic. The LTPT study cannot be used to distinguish between the alternative boundary conditions, because the pumping rate is too small to produce an analyzable pressure response at the model boundaries. In contrast, the thermal study shows that the temperature distributions produced by the open and closed models differ greatly. Comparison with borehole temperature data may be used to eliminate the closed model from further consideration.