Keywords: LNAPL, groundwater contamination, hydrogeological conditions, contaminant transport, unsaturated soil, LNAPL migration patterns.

Groundwater aquifers are under potential threats from various contaminants including Light Non-Aqueous Phase Liquids (LNAPLs). LNAPL migrates in a geo-environment under complex interactions with hydrogeological conditions as one of the important factors affecting the retention and flow of water and LNAPL. Subsurface soils were deposited and consolidated under a number of different processes that resulted in homogenous and heterogeneous soils being formed. The degree of heterogeneity varies significantly depending upon the active processes at a particular site and therefore it is a site specific. Additionally, groundwater fluctuations and infiltration of rainwater that are natural phenomena occurring in any geoenvironment, would affect the distribution of LNAPL. Thus, an enhanced knowledge about the migration of LNAPL in subsurface soils under various possible scenarios is imperative for a successful and economic remediation scheme to be designed. This can be acquired by obtaining high quality field and/or laboratory data. Additionally, the data generated can be an invaluable source for validation of numerical codes.

Large-scale field experiments are not favourable because of the possible adverse environmental effects as well as the associated cost and time. In stead, 1 and 2-D laboratory investigations are feasible and can be conducted under controlled conditions. A number of studies have been undertaken to investigate the migration of LNAPL in 2-D homogenous samples (see for example, [1,2]). Distribution of LNAPL (kerosene) in homogenous as well as heterogeneous samples was investigated [3]. Their results indicated that LNAPL migrates in circular shapes in the unsaturated zone. The distribution becomes irregular once LNAPL reaches the capillary fringe zone. The researchers found that in general, the distribution of LNAPL in layered samples resembles to a great extent that of the finer layer. Of note, the hydraulic relations were not established, as the suction head and degree of saturation transducers were not symmetrically located.

This paper presents the results of a laboratory investigation examining the influence of hydrogeological conditions on LNAPL migration in subsurface environments. A large-scale fully equipped 2-D tank was used to facilitate the experimental investigation. The front wall was made of glass to enable tracking down the advancing LNAPL front using electronic imaging. The back wall was used to insert TDR and pressure transducers using special fittings. Pressure transducers with hydrophilic and hydrophobic porous cups were used to measure electronically the pressure head of pore water and pore LNAPL respectively. TDR probes were employed to provide continuous measurements of volumetric water content. Two different sands with noticeable difference in the average grain size were used in this investigation to produce two homogenous sand samples and a heterogeneous sand sample. Mineral oil was used as a LNAPL. The LNAPL flow patterns during the infiltration period, at equilibrium and under fluctuating water table were continuously monitored and recorded.

Results of water drainage indicated different behaviour between homogenous and heterogeneous samples. Water was trapped at the bottom of the fine layers leaving almost half of it saturated with water. This has a great impact on subsequent LNAPL flow pattern. The results obtained in a layered sample are in disagreement with previously published by [3]. LNAPL would migrate through larger area in the case of layered sample than that for homogenous fine and coarse sand samples. The results indicate that there are major implications of the medium properties on the distribution of LNAPL during the infiltration period as well as at equilibrium. A regular and uniform LNAPL front occurs in homogenous fine sand whereas an irregular front dominated by flow through fingers was observed for coarse sand. Additionally, the lateral migration was found to increase as the grain size of homogenous samples decreases. Results of degree of saturation measured at various levels explain the flow direction of LNAPL and water and are consistent with the LNAPL flow patterns. Redistribution of LNAPL is influenced by the fluctuation of water table but it depends greatly upon the initial distribution of LNAPL and the free LNAPL volume. The results generated provide invaluable data for numerical model validation.

1

Mohamed, M.H.A., "Migration of light non-aqueous phase liquids in unsaturated and saturated sand", DPhil thesis, University of Bradford, U.K, 2003.

2

Sharma, R.S. and Mohamed, M.H.A., "Patterns and mechanisms of light non-aqueous phase liquid in unsaturated sand", Geotechnique, 53(2), 225-239, 2003. doi:10.1680/geot.53.2.225.37269

3

Pantazidou, M. and Sitar, N., "Emplacement of nonaquaeous liquids in the vadose zone", Water Resources Research, 29(3), 705-72, 1993. doi:10.1029/92WR02450

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description
purchase this book (price £135 +P&P)