Field evaluation of near source transport of contaminants in heterogeneous media

Home / Research / Field evaluation of near source transport of contaminants in heterogeneous media
Project Number:


Funding Year:


Contract Period:

7/1/1992 - 6/30/1994

Funding Source:


  • John Hoopes

This project extended the investigation “Near Source Transport of Contaminants in Heterogeneous Media” of contaminant plume and groundwater mounding behavior by applying this work to a field site.  As geologic and aquifer properties are known only at a finite number of locations, kriging is used to estimate properties at all locations.  Stochastic and deterministic models are used in the conditional simulations to predict flow field and mounding behavior.  The goals of this project were to: (1) use conditional Monte Carlo simulation to predict the regional flow field at one field side (Lake Geneva recharge facility) with existing data; (2) assess uncertainty in prediction of mound or plume behavior (using existing models) due to imperfect knowledge of hydrogeologic parameters; (3) evaluate the effects of additional field measurements of aquifer parameters (e.g. hydraulic conductivity) on the uncertainty of predictive variables (e.g. mounding height and plume penetration); and (4) establish a tool for use at other field sites and investigate the effects of design and operating alternatives and seasonal variability on mounding and plume behavior.

The methods and accomplishments from this work are presented in two parts.  In part I, aquifer properties were estimated from kriging (first using hydraulic conductivity (K) data and second using soil type (ST) data with a triangular distribution for K and each ST).  Then two groundwater flow models were run to simulate the water table with and without recharge and compared with observation.  Results with the 2-dimensional, finite element AQUIFEM model showed that additional data on heterogeneity and K and a 3-dimensional model were needed.  One simulation (realization) with the 3-dimensional MODFLOW model gave a similar pattern and comparable magnitudes to the observations.  Multiple realizations are needed to assess uncertainty, and the plume model needs to be run to test it with the data.  In part II, a numerical method to simulate two- and three-dimensional, transient groundwater mounding was developed and tested with a Hele-Shaw model.  The boundary of the recharged water was determined by a particle tracking.  The effects on mounding of the geometry, position, orientation and hydraulic conductivity of a single heterogeneity were computed.  Three-dimensional results and application to the field site are in progress.

Project Reports: