Groundwater Modeling Using an Integrated Large-Capacity Finite-Element Model for Multiple Aquifer Groundwater Flow

Groundwater computer codes have been used as a tool in a variety of ways in water resources planning, water rights evaluations, and groundwater contamination investigations and remediations. The use of groundwater models is burdened by the complexity of preparing the data input, evaluating model output, and difficulty in representing complex site physical features.

Micro-Fem is a software package composed of several integrated, well-defined parts and integrated programs. The software offers tremendous capabilities in terms of ease of data input preparation, checking the accuracy of the data, performing operations on the different model parameters, graphics display, and presentation and analysis of modeling results.

The Micro-Fem programs are developed to create and analyze a wide range of multilayer steady- and transient-state groundwater flow models with a maximum of 16 layers, 12,500 nodes, and 25,000 elements. The computer is used to replace the laborious finite-element administration and the file editing tasks by means of graphic screen control. The presentation of model results includes water balance calculations, flowlines, and travel time.

For the Centennial Water and Sanitation District (CWSD), groundwater is a key component in their water supply plans. Therefore, CWSD has undertaken a comprehensive groundwater modeling program to manage their groundwater resources.

The three main aquifers of the Denver Basin (Denver, Arapahoe, and Laramie-Fox Hills) developed by the CWSD were modeled using both Micro-Fem and the U.S. Geological Survey (USGS) finite-difference model, MODFLOW. Results of the two models were compared as well as the time required to prepare data and analyze modeling results. The comparison included model assessment of various scenarios of groundwater development. Also, the models were used to determine well development patterns that minimize water level declines and well replacement costs. Results of the comparison indicated that (1) Micro-Fem predicted heads were close to those predicted by MODFLOW, (2) Micro-Fem predicted pumping well cones of depression were similar to those predicted by MODFLOW, and (3) the time to prepare the input data and to display results was significantly lower than that required by MODFLOW.