ÃÛÌÒ´«Ã½Æƽâ°æÏÂÔØ

Strack, Otto D.L. ¹

¹ Department of Civil, Environmental, and Geo- Engi- neering, College of Science and Technology, University of Minnesota

Groundwater flow modeling has become increasingly sophisticated as modeling tools have evolved from early computer programs that required elaborate input and produced output in terms of sheets full of numbers. Nowadays, a computer model can be constructed simply by constructing a data set, using a Graphical Information System, running it through an advanced computer program, such as MODFLOW, and combine it with an equally advanced Parameter Estimation Program, e.g., PEST. The result often is an impressive set of graphs.

Although such an approach is very powerful in the hands of an expert, there exists a danger of creating a model that looks impressive, but is unrealistic. We will attempt, in this presentation, to make a case for making use of elementary solutions to groundwater problems. Sometimes, these solutions will, by themselves, answer the question posed, but often they can help in creating an advanced model, as well as ensuring that the advanced model is likely to be realistic.

We will present examples of this idea. Several of these examples answered the question by themselves, and in other cases they were valuable in setting up an advanced model. Finally, we will briefly introduce an approach, called vertically integrated flow modeling, which can help in simplifying models, as well as serving as a tool for testing the complex model.