Local Adaptive Grid Refinement (lagr) for Contaminant Transport Modeling

The success of a groundwater contaminant transport model is dependent on: 1) how well the relevant physical, chemical, and microbiological processes controlling subsurface transport are represented with mathematical equations and their parameters, and 2) how accurately and efficiently the equations are solved. For reactive, multicomponent problems in two or more dimensions, the computational burden of meeting the second goal becomes increasingly difficult as the complexity of the problem formulation required to meet the first goal increases. To address these two goals, we have implemented a system of local adaptive grid refinement (LAGR) to achieve efficient and accurate computational results for multidimensional, multicomponent transport problems. LAGR is based on automatic generation of multiple, high resolution patch grids (finite difference in this case) placed at locations in the solution domain where necessary to achieve uniform solution accuracy. We have applied LAGR to nonreactive single component problems and to reactive multicomponent problems including such chemical processes as competitive sorption. As with familiar single component benchmark problems, numerical errors such as peak clipping, oscillations, and overand undershoot plague the coarse grid computations for our multicomponent problems. In fact, these problems are exaggerated in some multicomponent problems because the fronts are sharper as a result of the interactive chemistry. Accurate solutions to these problems are provided using LAGR at a fraction of the computational cost of using a uniform fine grid.