Heterogeneous hydraulic conductivity and porosity fields reconstruction through steady-state flow and transient solute transport data using the continuous adjoint state

摘要: Abstract A parameter estimation methodology has been developed on the basis of model inversion using a Quasi–Newton method and adaptive parameterization. The continuous adjoint state equations for both flow transport in porous media are employed as tool calculating gradient components objective function with respect to parameters. Solving form can be implemented independently code used solve forward problem, which renders technique non-intrusive. is applied identification hydraulic conductivity porosity fields conditioned by piezometric head data associated steady-state transient solute concentrations. Synthetic numerical experiments have undertaken test cases increasing complexity, from an almost uniform sweeping modeled domain prescribed injection at inflow boundary, spatially highly variable conditions obtained through source/sink terms within local stepwise injection. results inversions analyzed criteria based comparisons between estimated concentration reference values well (and one case) fields. show that employing computed direct problem efficient option relying jointly upon data. In reported examples characterized identifiability observations, scenarios bring few added value sought solutions conductivities. spatial structure slightly improved compared reference, but overall system distribution, main paths, transit times, only inherits cosmetics.