| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Environmental Technology Division, Pacific Northwest National Laboratory, P.O. Box 999, MS K9-33, Richland, WA 99352
b Idaho National Engineering and Environmental Laboratory, P.O. Box 1625, Idaho Falls, ID 83415-2214
* Corresponding author (mart.oostrom{at}pnl.gov)
Received 15 May 2002.
ABSTRACT
At many contaminated sites, nonaqueous phase liquids (NAPLs) persist in the vadose zone for long periods of time. This occurs because the permeability of the NAPL becomes negligible at some saturation and downward movement ceases, resulting in residual NAPL. To obtain data that can be used to study the development of a residual NAPL saturation and to test corresponding models, a detailed transient experiment was conducted in a 170 cm long by 90 cm high by 5.5 cm wide flow cell. Fluid saturation measurements were obtained with a dual-energy 
radiation system. The experimental conditions reflected those at the Hanford Site in Washington State, where an estimated 363 to 580 m3 of carbon tetrachloride (CCl4) was disposed to the subsurface. A key subsurface feature at the Hanford Site is a sloped Plio-Pleistocene caliche layer, which was reproduced in the experiment as a sloped lens in a medium-grained, uniform, sand matrix. The caliche contains considerable amounts of CaCO3 and may have fluid wettability properties other than strongly water wet. A total of 800 mL of CCl4 was injected into the experimental domain at a rate of 0.5 mL min-1 from a small source area located at the surface. After apparent static conditions were obtained with respect to CCl4 redistribution, saturation measurements indicated that all of the dense nonaqueous phase liquids (DNAPL) that had initially moved into the caliche remained in this layer. Water was subsequently applied to the surface at a constant rate over the full length of the caliche layer to study CCl4 displacement as a result of changing water saturations. Water saturation in the caliche layer rose to as high as 0.91 during water infiltration. Results show that 25% of the DNAPL present in the caliche migrated from this layer as a consequence of water infiltration, while 75% remained in the caliche layer. The experimental results could not be reproduced with numerical multifluid flow simulations based on common constitutive theory. This indicates that improvements in constitutive theory may be needed to accurately model air-DNAPL-water flow behavior.
Abbreviations: DNAPL, dense nonaqueous phase liquids • NAPL, nonaqueous phase liquids • PCE, perchloroethylene
NONAQUEOUS PHASE LIQUIDS, like fuel hydrocarbons and chlorinated solvents, have entered the subsurface via chemical spills, leaks, and direct disposal. At many contaminated sites, NAPLs persist in the vadose zone for very long periods. This occurs because the permeability of the NAPL becomes negligible at some saturation and downward movement ceases, resulting in residual NAPL.
The term residual saturation has been applied to represent several processes or phenomena. When applied to the wetting fluid, it characteristically refers to a water saturation at which the effective permeability (i.e., the permeability at a prescribed fluid saturation) of water approaches a value of zero during water drainage. Typically, water is assumed to be immobile at the residual-water saturation. In the petroleum industry, the residual-water saturation is also called the irreducible-water saturation. This terminology has also been adapted in hydrology (e.g., Lenhard, 1992; Oostrom et al., 1999). When applied to NAPL in two-phase water-NAPL systems in strongly water-wet porous media, a common definition of residual NAPL is that which becomes discontinuous (entrapped) in the pore spaces when the apparent water saturation is on an imbibition path. However, for three phase water–NAPL–air systems, the term has also been used to indicate NAPL, not trapped by water, which is negligibly mobile. To avoid confusion with the term residual NAPL, we define NAPL that are discontinuous (i.e., blobs) and occluded by water as trapped NAPL and NAPL that are negligibly mobile and not trapped by water as residual NAPL. Residual NAPL can be either continuous throughout the pore spaces or discontinuous as a result of isolation in pore wedges or formation of lenses on water surfaces. For three-phase water–NAPL–air systems in strongly water-wet porous media, NAPL can be present in free, residual, and trapped form. However, for the same conditions in two-phase water–NAPL systems, only free and trapped NAPL might exist.
Although residual NAPL is known to occur, there are no numerical simulators that account for residual NAPL as a function of saturation-path history. The behavior of residual NAPL in three-phase air–NAPL–water systems has not been studied in great detail for environmental purposes, although its presence has broad implications for groundwater contamination. Dense vapors originating from volatile organic compounds may sink and spread rapidly in the vadose zone followed by partitioning into the groundwater. In addition, recharge water may come into contact with the NAPLs and transport dissolved components downward to aquifers.
One of the most ubiquitous NAPLs at U.S. Department of Energy sites is the DNAPL CCl4. At the Hanford Site in Washington State, CCl4 mixtures were disposed to waste sites (Last and Rohay, 1993). Plutonium recovery operations at the Plutonium Finishing Plant have resulted in organic and aqueous wastes that have been disposed at several cribs, tile fields, and drains. The major disposal facilities received an estimated 13 400 000 L of liquid waste containing 363 000 to 580 000 L of CCl4. The liquid waste included aqueous waste and DNAPL. Calculations indicated that the majority of the disposed CCl4 entered the subsurface as DNAPL. Last and Rohay (1993) conducted a rough order-of-magnitude estimate of the discharge inventory, and they computed that 21% of the liquid might have been lost to the atmosphere, 12% is retained in the vadose zone (dissolved in water, sorbed to the solid phase, and as a component of the gas phase), and 2% is dissolved in the saturated zone. The remaining 65% has not been accounted for and might have been initially present as residual liquid saturation in the vadose zone. Although a considerable amount of the disposed liquid is assumed to remain in the vadose zone as a residual liquid, the physical processes describing the formation of residual DNAPL in the vadose zone are not well understood and have not been implemented into multifluid flow simulations of the Hanford Site.
The current lack of understanding of major physical processes describing movement and redistribution of CCl4 mixtures in the vadose zone at the Hanford Site prevents significant advances in the development of a conceptual model for the site. In particular, the formation of residual saturation in the vadose zone and the movement of CCl4 in porous media with relatively high CaCO3 contents (e.g., caliche materials) are poorly understood. The caliche is a carbonate-rich sequence of the Plio-Pleistocene unit which generally represents a well-developed calcic paleosol (calcrete) or a sequence of calcic soils. The caliche materials generally consist of weathered and naturally altered sandy silts to sandy gravels that are moderately to strongly cemented with secondary pedogenic CaCO3.
A number of numerical simulators (e.g., TOUGH2, UTCHEM, STOMP) have been developed to solve the system of highly nonlinear partial differential equations governing multiphase flow in groundwater and vadose zones. None of these macroscopic continuum codes addresses the formation of residual NAPLs in three-phase systems on the basis of a consistent set of constitutive relations between relative permeability, fluid saturation, and capillary pressure (k–S–P). McBride et al. (1992) stated that the ability of k–S–P constitutive relations models to accurately describe multifluid flow in unsaturated systems has become the limiting factor in predicting NAPL movement. Hofstee et al. (1997) argued that current simulators may not predict the distinct interfacial phenomena associated with both strongly spreading and nonspreading liquids properly because a spreading coefficient of zero is implicitly assumed. The spreading behavior of oils in three-fluid systems has received considerable attention in the petroleum engineering community (e.g., Blunt et al., 1994; 1995; Vizika and Lombard, 1994). The physics behind residual formation in three-phase flow systems can be found in Blunt et al. (1994)(1995).
Although the current continuum-based models have their obvious shortcomings in predicting residual NAPL formation, it is clear that there will be an important role for these models in the future. Recently, some contributions have appeared in the literature that start to address the residual saturation formation issues in a more focused manner (Wipfler and van der Zee, 2001; Van Geel and Roy; 2002). A problem affecting the development of new constitutive relations is the lack of quality experimental data.
The relations for three-phase S–P systems (Lenhard and Parker, 1987, 1988; Parker et al., 1987) are usually extended from two-phase S–P equations based on the theoretical and experimental work of Leverett and Lewis (1941). Extending two-fluid phase S–P relations through relatively simple scaling procedures to three-fluid phase systems is commonly employed because of the experimental difficulty of directly measuring S–P relations in three-fluid phase porous medium systems. The so-called Leverett concept states that in a water-wet porous medium, when fluid wettabilities follow the order water–NAPL–air, the water content is a function of the NAPL–aqueous phase capillary pressure and the total liquid content is a function of the gas–NAPL capillary pressure. Hofstee et al. (1997) showed in their experiments that the Leverett concept might not correctly describe the perchloroethylene (PCE) distribution after drainage for columns under unsaturated conditions. It was found that PCE could not be drained from unsaturated sands below a certain minimum saturation, indicated by the authors as the critical saturation, because the PCE beaded up in distinct lenses. The critical PCE saturation in the experiments by Hofstee et al. (1997), which is a function of the saturation history at a certain location, can be viewed as being residual since its relative permeability is zero. The results found by Hofstee et al. (1997) were explained by PCE's spreading tendency on air–water interfaces. The spreading coefficient is computed as (Adamson, 1982):
 | [1] |
gl (N m-1) is the air–water interfacial tension, nl the NAPL-water surface tension, and gn the air–NAPL interfacial tension, respectively. Nonaqueous phase liquids with a positive spreading coefficient will form a film on top of an air-water interface. Other organic pollutants, such as perchloroethylene, have a negative spreading coefficient and will form microlenses as a result of the larger internal cohesion. A distinction should be made between the initial spreading coefficient, based on interfacial tensions of the pure liquid, and the equilibrium spreading coefficient, measured under chemical equilibrium. Several NAPLs, like benzene, CCl4, and trichloroethylene, for example, have initial spreading coefficients that are slightly positive. The water–air interfacial tension, however, is very sensitive to organic contamination compared with nl and gn (Corey, 1994). When NAPL vapors migrate through the pores, partitioning into the aqueous phase might lead to a distinct drop in the water–air interfacial tension and yields negative equilibrium spreading coefficients for these organic contaminants. Hofstee et al. (1997) argued that this process occurred during their experiments. Schroth et al. (1995) measured considerable decreases in nl for two nonvolatile light nonaqueous phase liquids (LNAPL) in contact with water. Such aging effects result in increases of Csp with time and render the NAPL more spreading. Simmons et al. (1992) conducted infiltration experiments with various oils in boxes packed uniformly with unsaturated sand or glass beads. It was noted that mineral oil, with a negative spreading coefficient, exhibited a rather irregular behavior, including the formation of unstable fingers. Hofstee et al. (1998) conducted an unsaturated intermediate-scale flow cell experiment in which nonspreading PCE was allowed to infiltrate into a layered system. The authors noted that the PCE ceased to spread laterally shortly after the initial infiltration. This phenomenon was attributed to the nonspreading nature of PCE.
Another major assumption in Leverett's theory is that the porous media are always water wet. This assumption might not be correct for Hanford Plio-Pleistocene materials with high CaCO3 contents. The CaCO3 might cause the porous media to have a wettability other than water wet, rendering the Leverett concept invalid. The chemistry of the solid surfaces and the fluids are major factors in determining the wettability of porous media. Crude oils are complex hydrocarbon mixtures that contain wettability-altering compounds. Willhite (1986) stated that water preferentially wets calcite and silica surfaces in the presence of pure paraffin hydrocarbons. He also noted that the addition of small amounts of polar compounds, such as hydrocarbons containing N, O, or S, or film-forming compounds can change wettability from water to intermediate or oil wet. In a table listing wettability as a function of mineralogy, the vast majority of formations derived from calcite were oil wet.
Given the potential impact of residual CCl4 in the Hanford vadose zone caliche material on groundwater contamination in particular, and the lack of understanding of the processes behind the formation of residual NAPL in general, a focused research effort was initiated to improve the knowledge of CCl4 movement and redistribution at the Hanford site. To demonstrate the most important issues related to the CCl4 contamination problem, a detailed flow cell experiment was conducted first. The goals of this experiment were to:
MATERIALS AND METHODS
Porous Medium and Fluid Properties
The three porous media used in this study were fine-grained (70 mesh; 2.12 x 10-4 m) sand, medium-grained (12/20 mesh; 1.7 x 10-3 m/8.5 x 10-4 m) sand (Unimin Corporation, Le Sueur, MN) and caliche material. The caliche material was obtained from the White Bluffs area at the Hanford Site. After removal of the CaCO3 (4.5% by weight), a particle-size analysis indicated that the caliche contains 6% clay, 71% silt, and 23% sand and can therefore be classified as a silt loam according to the USDA classification scheme.
Measured hydraulic properties of the three porous media are shown in Table 1. The saturated hydraulic conductivities were obtained using a constant head method (Klute and Dirksen, 1986) applied to a 1-m-long column with a 0.05-m diameter. The porous media in these columns were initially packed under saturated conditions. The Brooks and Corey parameters were obtained from water saturations measured in the subsequently drained column. The water table was lowered from the top to the bottom of the column at a rate of 10 cm h-1. Water saturations were obtained at 2-cm intervals using a dual-energy radiation system (Oostrom and Dane, 1990; Oostrom et al., 1998) until apparent static equilibrium had been obtained. Assuming that for these conditions the elevation above the water table is equal to the air–water capillary head, the RETC program (van Genuchten, 1985) was used to best-fit the Brooks–Corey saturation–capillary head relation to the experimental data.
|
Intermediate-Scale Experiment
The intermediate-scale experiment was conducted in a flow cell with a 170 cm long by 85 cm high by 5.8 cm wide porous medium chamber with a glass front and a kynar back panel. The glass front allows for visual observations, while the plastic kynar is easily machinable for installation of sampling ports, for example. Both materials are chlorinated-solvent resistant. In the flow cell, a sloped 90 cm long by 25 cm high caliche zone was placed in a medium-grained sand matrix. The elevation difference between the left and right boundary of the caliche layer was 5 cm, corresponding to a slope of 3.17°. The experimental design is shown in Fig. 1. Two head chambers, one on each side of the flow cell, control the water table elevation. The flow cell was packed under saturated conditions until the lower interface between the caliche and the medium-grained material was reached. Thereafter, the water table was lowered to z = 15 cm. Two days later, the remainder of the flow cell was packed. The caliche was premixed with an amount of water approximately yielding the irreducible water saturation of 0.20, which value was obtained from the one-dimensional column drainage experiment (see Table 1). The remainder of the medium sand contained an irreducible water saturation of 0.03. The reason this packing sequence was employed was to ensure that the sloped caliche layer would be unsaturated. Given the relatively high air-entry head of the caliche material, the caliche material would remain nearly saturated during the experiment if it was packed under water-saturated conditions. A picture of the packed flow cell is shown in Fig. 2.
|
|
measurements at a total of 1182 locations (Fig. 1) using procedures outlined by Oostrom and Dane (1990) and Oostrom et al. (1998). The majority of these locations (980) are in the caliche material. After completion of the packing procedure, the system was used to determine dry bulk density and water saturation values. Average porosity values were computed from the dry bulk density values assuming a particle density of 2.65 g cm-3. This value is considered to be reasonable for caliche material because only the grain coatings are calcite. Average values and associated standard deviations of the porosity values are listed in Table 1. The relatively low standard deviation indicated a uniform packing of the porous media. Figures of the measured bulk density and initial saturation distributions of the caliche layer are shown in Fig. 3a and 3b, respectively. The plots show that the packing of the sloped layer resulted in sharp interfaces between the medium-grained sand and the caliche.
|
scans of the three vertical transects were obtained. Toward the end of the CCl4 redistribution and water injection stages, scans of the whole caliche layer were taken. A full scan, with counting times of 60 s per location, lasted approximately 18 h.
Numerical Simulation
The various stages of the flow cell experiment were simulated with the nonhysteretic water–oil mode (passive air) of the fully implicit, integrated finite difference code STOMP (Subsurface Transport Over Multiple Phases) code (White and Oostrom, 2000a, b). The two-dimensional computational domain consisted of 80 (x-direction) x 50 (z-direction) = 4000 cells. The grid was refined near the source and the interfaces between the medium sand and the caliche. At the bottom of the domain, a constant water pressure was prescribed, consistent with a water table position at z = 15 cm. At the top of the domain, zero aqueous flux boundary conditions were specified during Stages I and II. During Stage III, a Neumann boundary condition was prescribed for the domain boundary located directly above the caliche layer. Hydraulic gradient boundaries were imposed on both the left- and right-hand sides of the computational domain. Zero flux boundaries were specified for CCl4 on all boundaries except for Stage I when a Neumann boundary condition was employed. For each simulation, upwind interfacial averaging was used for fluid densities and relative permeabilities. A time-step increment factor of 1.25 was applied after convergence. The maximum number of Newton–Rhapson iterations allowed was eight, with a convergence factor of 10-6, before any adjustments in the time step were imposed. Parameter values listed in Table 1 were used in the simulations. Brooks and Corey (1964) parameters were used for the S–P relations because the porous materials have distinct nonwetting fluid entry pressures, as shown in Table 1. Burdine (1953) equations were used for the k–S relations. Formation of discontinuous residual DNAPL in the vadose zone and DNAPL entrapment by imbibing water are not included in the k–S–P constitutive relations. In addition, STOMP assumes the porous media to be water wet and fluids are distributed using a spreading coefficient of zero so that when continuous DNAPL is present in a three-fluid system, it will always be located between the aqueous and the gas phases.
RESULTS AND DISCUSSION
The injected CCl4 moved uniformly through the medium-grained sand. After 22 min, the CCl4 arrived at the sand–caliche interface and subsequently migrated into the caliche. Within the caliche, the CCl4 movement was substantially slower compared with the sand, which is primarily a result of the difference in permeability between the porous media. Although the permeability of the caliche was smaller, the flux rate was such that no ponding of CCl4 was observed. In addition, lateral horizontal movement in the caliche, as a result of capillary action, was more pronounced than in the sand. Pictures of the lateral extent of the CCl4 in both materials are shown in Fig. 4a and 4b after injection of 400 and 800 mL, respectively. Figure 4b also represents the point in time when injection of CCl4 ceased.
|
data, this reduction would yield about 303 mL of water. The actual amount of water drained during the experiment (342 mL) corresponds reasonably well to this estimate based on the surface tension reduction. After cessation of the CCl4 injection, the DNAPL slowly moved laterally with time. Figure 5a and 5b show the extent of the CCl4 after 2 and 21 d of redistributing, respectively. The numbers next to the black lines indicate the number of days of redistribution. The lines with the number 0 denote the lateral extent of the DNAPL at the end of Stage I. After 21 d, no visual changes in the distribution could be observed. During CCl4 injection and subsequent redistribution, no CCl4 appeared to have moved into the sand below the lower caliche–sand interface. Apparently, the saturations were not sufficiently high to allow the CCl4 to penetrate into the sand.
|
|
|
The last part (Stage III) of the experiment involved injection of water for 4 d at a rate of 6 mL min-1 from a 90-cm-long source located on top of the sand (Fig. 1). Visual inspection, based on color changes, indicated that initially the majority of the water moved into the unsaturated caliche layer. Only after approximately 11 h did water start to drain from the flow cell. After 14 h, the drainage rate equaled the injection rate, suggesting that a steady-state situation had developed. Plots of the drainage rate and the associated CCl4 dissolved concentrations are shown in Fig. 8. The CCl4 concentration in the drained water was initially more than 400 mg L-1 and increased rapidly to about 750 mg L-1. The latter value is close to the saturated solubility of 812 mg L-1, as measured in our analytical laboratory, and indicates that most of the infiltrating water makes contact with the liquid CCl4 in the caliche layer. The relatively high initial concentration demonstrates the importance of CCl4 vapor movement followed by partitioning into the water in the saturated regions during the first two stages of the experiment. The total integrated amount of CCl4 that left the flow cell in dissolved form with the drained water was 22 g, corresponding to 13.8 mL.
|
scans in Fig. 9 after 55 and 92 h are almost identical. Water saturation measurements, after apparent steady-state conditions were obtained, showed a fairly constant total liquid saturation of approximately 0.9. Taking air entrapment into account, the apparent water saturations in the caliche layer were approaching 1.0. The value of the total liquid saturation suggests that considerable amounts of the remaining CCl4 in the caliche might have moved from being residual in a three-phase unsaturated system to entrapped under apparent saturated conditions. Some of the DNAPL left the caliche through five distinct fingers: three visible on the glass and two on the kynar side. The fingers developed between 12 and 15 h after the start of the water injection. The fingers did not appear to extend all the way to the fine-grained sand at the bottom of the flow cell and disappeared gradually during the last 2 d of the experiment. The fingers were narrow and never wider than 0.5 cm. Full scans were obtained of the caliche layer during the third and fourth day of the water injection. Both scans were quite similar, showing that CCl4 movement in the caliche had become minimal. A plot with the final saturation distribution is shown in Fig. 10. The figure shows a large zone with saturations between 0.12 and 0.18, most likely in entrapped form. It is possible, however, that some of the CCl4 in the lower part of the flow cell was still mobile, thus causing gradual movement down the slope of the caliche–sand interface. The total integrated liquid CCl4 amount from Fig. 10 is 570 mL. Taking into account the 13.8 mL that had moved out of the flow cell in dissolved form, it can be computed that 190 mL (25%) moved out of the caliche layer. At the end of the experiment, only a small amount (saturations less than 0.04) appeared at just two locations in the depression on top of the fine-grained sand interface.
|
|
|
Nonaqueous phase liquids residual saturation formation in the vadose zone has been largely ignored in constitutive k–S–P models and, consequently, also in multifluid flow simulators. Residual saturation forms because the permeability of a NAPL becomes negligible at some saturation and movement ceases. To obtain data that can be used to study the development of a residual NAPL saturation and to test corresponding models, a detailed transient experiment was conducted in an intermediate-scale flow cell. The experimental conditions reflected those at the Hanford Site in Washington State, where large amounts of CCl4 were disposed. A key subsurface feature at the Hanford Site is a sloped Plio-Pleistocene caliche layer, which was reproduced in the experiment as a sloped lens in a medium-grained, uniform, sand matrix. The experiment consisted of three stages: injection of 800 mL CCl4 at a rate of 0.5 mL min-1, CCl4 redistribution, and water injection at a rate of 6 mL min-1. During the CCl4 injection stage, the CCL4 apparently behaved as a wetting fluid and moved uniformly into the caliche. During Stage II, the DNAPL moved slowly in the lateral direction. After approximately 21 d, apparent static equilibrium was obtained with respect to the distribution of CCl4 in the caliche. Saturation measurements indicate that all of the CCl4 that initially moved into the caliche remained in this layer. A considerable amount of the CCl4 probably becomes residual in the unsaturated caliche as a result of its nonspreading behavior under equilibrium conditions. It can be shown that the suspected nonspreading behavior is primarily a result of the decrease in air–water surface tension through contamination of the water with dissolved CCl4.
Subsequent application of water to the surface at a constant rate over the full length of the caliche layer was conducted to study CCl4 displacement as a result of changing water saturations. Results show that as a result of the water injection, 25% of the CCl4 was removed from the caliche. However, 75% of the CCl4 remained in the caliche, most likely entrapped by water. The experimental results could not be reproduced with numerical multifluid flow simulations based on current k–S–P nonhysteretic constitutive relations. The lack of agreement clearly indicates that improvements in three-phase constitutive theory are needed, especially to model residual NAPL.
ACKNOWLEDGMENTS
Pacific Northwest National Laboratory is operated by the Battelle Memorial Institute for the Department of Energy under Contract DE-AC06-76RLO 1830. The laboratory experiment was supported by the Laboratory Directed Research & Development Program of the Idaho National Engineering & Environmental Laboratory (INEEL) under the Subsurface Science Initiative. The analysis component was supported by the Ground water/Vadose Zone Integration Project funded through the U.S. Department of Energy's Richland Operations Office.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  G. W. Gee, M. Oostrom, M. D. Freshley, M. L. Rockhold, and J. M. Zachara Hanford Site Vadose Zone Studies: An Overview Vadose Zone J., November 20, 2007; 6(4): 899 - 905. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Oostrom, M. L. Rockhold, P. D. Thorne, M. J. Truex, G. V. Last, and V. J. Rohay Carbon Tetrachloride Flow and Transport in the Subsurface of the 216-Z-9 Trench at the Hanford Site Vadose Zone J., November 20, 2007; 6(4): 971 - 984. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Oostrom, J. H. Dane, and T. W. Wietsma A Review of Multidimensional, Multifluid, Intermediate-Scale Experiments: Flow Behavior, Saturation Imaging, and Tracer Detection and Quantification Vadose Zone J., August 23, 2007; 6(3): 610 - 637. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Oostrom, M. D. White, R. J. Lenhard, P. J. Van Geel, and T. W. Wietsma A Comparison of Models Describing Residual NAPL Formation in the Vadose Zone Vadose Zone J., February 1, 2005; 4(1): 163 - 174. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Yoon, H. Yoon, A. J. Valocchi, and C. J. Werth Modeling the Influence of Water Content on Soil Vapor Extraction Vadose Zone J., August 1, 2003; 2(3): 368 - 381. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Crop Science | |||
| Journal of Natural Resources and Life Sciences Education |
Soil Science Society of America Journal | ||||
| Journal of Plant Registrations | Journal of Environmental Quality |
The Plant Genome | |||
