Modeling fracture flow with a stochastic discrete fracture network: calibration and validation: 1. The flow model

摘要: A large-scale investigation of fracture flow was recently conducted in a granite uranium mine at Fanay-Augeres, France. Its aim to develop methodology for the possible nuclear waste repository sites crystalline environments, and thus determine what measurements make models use order predict transport properties medium, i.e., their average behaviors spatial variabilities different scales. Four types data were collected: (1) geometry network; (2) local hydraulic measured by injection tests boreholes; (3) global behavior from rate piezometric head distribution 106 m3 scale; (4) tracer performed scale up 40 m. stochastic network model assuming negligible matrix permeability developed calibrated essentially on 1 2 above; this then used 3 4 an attempt validate both parameters structure model. In first part, only problem (data 1) is discussed.