Volume Estimation of Light Nonaqueous Phase Liquids in Porous Media


  • Discussion open until July 1, 1990.

  • Anne M. Farr is a Senior Geohydrologist at Kennedy/ Jenks/Chilton Consulting Engineers and is completing her Ph.D. at Colorado State University in the area of optimization of ground-water monitoring networks utilizing stochastic ground-water models. She previously worked for Kennedy/Jenks/Chilton Consulting Engineers focusing on hazardous waste site assessments and as a hydrologist with the USGS-WRD.

  • Robert J. Houghtalen is an Associate Professor of Civil Engineering at Rose-Hulman Institute of Technology. He recently completed his Ph.D. in Water Supply and Systems Engineering Analysis at Colorado State University as a U.S.D.A. National Needs Fellow.

  • David B. McWhorter is Professor of Agricultural and Chemical Engineering at Colorado State University. He teaches courses and conducts research in ground-water hydrology and multifluid flow in porous media.

  • Editor's Note: This paper deals with the identical subject as the paper by R. J. Lenhard and J. C. Parker in this issue. The work was done by the two groups of researchers simultaneously but with no knowledge of the other group's work. The papers were submitted within two weeks of each other in the latter part of 1988. After significant review and revision, both original pieces of work were deemed appropriate for publication inasmuch as the subject is of significant importance in ground-water hydrology, and the duplicate effort clearly enhances the validity of the work as well as its future impact.


An analytic method is described for estimating the volume of mobile Light Nonaqueous Phase Liquids (LNAPL) in porous media from observed LNAPL thicknesses in monitoring wells. Static (mechanical) equilibrium of fluids in a homogeneous porous medium is the key condition on which the method is based. Both the Brooks-Corey and van Genuchten equations, with parameters derived from laboratory column experiments reported in the literature, are used to relate fluid contents to capillary pressures. The calculations show that LNAPL in the vadose zone does not distribute itself as a distinct layer floating on the top of a capillary fringe. Rather, the traditional concept of a capillary fringe is not applicable when LNAPL is present. Further, neither the LNAPL level nor the water level in monitoring wells is equal to the water-table elevation. The water table, being the surface on which the water pressure is zero gage, is located above the LNAPL-water interface in the well, and LNAPL in the porous media will reside below the water table.

It is shown that finite volumes of LNAPL theoretically can exist in materials with positive entry pressures (e.g., Brooks-Corey porous media) without revealing their presence in the form of an LNAPL layer in monitoring wells. However, LNAPL in porous media with zero entry pressure will always appear in monitoring wells, regardless of the volume of LNAPL in the porous medium. In addition, the theory shows that the ratio of the volume of LNAPL per unit area in the vadose zone to the thickness of LNAPL in monitoring wells is strongly dependent upon the capillary properties of the porous medium. Only in porous media with very uniform pore sizes is the volume of LNAPL in the vadose zone approximately proportional to the thickness of LNAPL in monitoring wells.