Organic compounds, including hydrocarbons and chlorinated solvents, have become widespread contaminants in the soil and groundwater due to inadequate storage and disposal techniques. Considerable research has been conducted over the last two decades to develop ways of removing contamination and restoring the subsurface. Recent efforts have focused on in-situ techniques, primarily in the area of bioremediation.

In-situ bioremediation has advantages over traditional remediation such as pump-and-treat and excavation and disposal. It has the potential of destroying the contaminant in the subsurface, thereby eliminating the need to treat a waste stream. It reduces the potential of transferring contaminants to the atmosphere. Also, microbes may be able to degrade non-aqueous phase contaminant that traditional pump-and-treat systems cannot effectively move.

Many numerical models have been constructed to help evaluate relevant factors involved in the design of in-situ bioreactors. However, most of these models do not include multiple species and comprehensive collections of biological processes. There is a growing need for general purpose groundwater models that include biological processes.

This thesis presents a numerical model capable of simulating many of the important subsurface processes involved in bioremediation design. It simulates two-dimensional saturated steady-state flow aquifers with advection and dispersion of multiple reactive solutes. It includes kinetic and equilibrium Langmuir and Freundlich isotherms, as well as first-order and higher reactions. Single, double, and competitive Monod kinetic reactions are used to represent biological reactions. Certain combinations of reactions in this model can be used to simulate other processes such as intermediate toxicity and cometabolic degradation.

Construction of a new model was chosen over modification of an existing program for multiple reasons. First, it is easy to introduce programming errors while modifying existing code. Due to unfamiliarity with existing code, the programmer may be unaware of the assumptions made by it. This can lead to errors which are very difficult to locate and correct. In addition, existing programs are primarily written in FORTRAN. C++ is a better programming language to use for this type of groundwater model. The object-oriented language style lends well to dynamically including different types of reactions. Finally, a general purpose program should be easy to use. The computer program described in this thesis was written to take advantage of the Microsoft Windows™ interface for inputting data and viewing results.

The organization of this thesis is as follows: Chapter 2 presents a discussion of important subsurface processes. The numerical methodology is developed in Chapter 3. Chapter 4 evaluates model performance in a general sense. An example remedial design is presented in Chapter 5. Many of the detailed calculations are included in the appendices. The User’s Manual to the model is included in the final appendix.