Groundwater models are increasingly being used in design of remediation systems for contaminated sites. A postaudit was performed on two groundwater flow models, one analytical and one numerical, used to design a well array for a groundwater capture and containment system installed along the boundary of a manufacturing facility.
The system was initially designed based on the results of an analytical well field simulation model. Due to unexpected occurrences of light nonaqueous phase liquid (LNAPL), emulsions and resultant treatment limitations, and lower well efficiencies, pumping rates and drawdowns were lower than predicted. Enhancements to the system, including additional extraction and LNAPL recovery wells and additional treatment components, were designed and constructed based on the results of a numerical model.
A postaudit was performed on the analytical model used to design the initial system. After system performance did not meet expectations, a numerical model was used to design an enhanced system. Because the enhanced system has been in place for only a short time, a detailed postaudit of the numerical model was not performed. However, a cursory review of numerical model results versus observed conditions was performed.
Results of the postaudit indicated that the following factors contributed to the observed differences between actual and predicted water levels: (1) reduced system pumping rates due to unanticipated treatment problems, (2) limitations on total available drawdown due to well construction techniques, (3) variations in well efficiencies, (4) unexpected variations in pumping rates due to system maintenance activities, and (5) aquifer heterogeneities.