Roger B. Wallace

Effects of degree of water saturation on dispersivity and immobile water in sandy soil columns

Abstract Three natural nonaggregated soil samples, with similar grain-size distributions, have been used to determine the dispersive behavior of porous media under steady, saturated and unsaturated flow conditions. Tritium was used as a tracer and was found to have no sorption on the solid matrix. Generated breakthrough curves (BTCs) for the unsaturated experiments were symmetrical with no evidence of tailing. The unsaturated experiments for two of the soils were adequately described by considering all the water in the pore volume as mobile. However, about 10% of the pore water, independent of the degree of saturation, was found to be immobile in the case of the third soil during unsaturated flow. For this soil, there was no mass transfer between the two water regions, indicating that the immobile water is essentially isolated from the flowing water fraction. For all three soils, dispersivity under unsaturated conditions was found to be higher, independent of the degree of water saturation, than the value determined for the saturated experiments. This is inconsistent with what would be expected from the simple bundle-of-capillary-tubes model and does not agree well with a more sophisticated conceptualization of the porous medium. The data, however, clearly indicate a wider range in pore-water velocities when these soils are desaturated.

Effects of residence time and degree of water saturation on sorption nonequilibrium parameters

This study reports the impact of the degree of water saturation on sorption nonequilibrium parameters. Two nonionic organic compounds (benzene and dimethylphthalate) and three nonaggregated sandy soils were utilized. Local equilibrium assumptions were found to be invalid for describing the transport of these compounds even at pore-water velocities as low as 0.7 cm/h. Sorption nonequilibrium appeared to be of a diffusive nature rather than due to a slow chemical reaction. Sorption mass-transfer coefficients varied proportionally with pore-water velocity. A strong correlation between the mass-transfer coefficient and residence time was found utilizing present and previously reported laboratory data. A similar relationship was also found for the mass-transfer coefficient between mobile and immobile water regions. Field data indicate that the sorption mass-transfer coefficient may continue to decrease in a consistent way even at residence times as large as 5×103 h. Variations in the degree of water saturation had no impact on the value of the sorption mass-transfer coefficient other than what would be expected due to changes in the residence time. This suggested that movement into the solid grains of the large emptied pores through diffusion from the water-filled pores into stagnant water covering these grains was relatively fast compared to the sorption rate.

A laboratory method for studying the aqueous phase transport of dissolved constituents from residually held NAPL in unsaturated soil columns

Abstract The potential for soils containing residual nonaqueous-phase liquid (NAPL) to contaminate infiltrating water was investigated using laboratory experiments. An experimental procedure was developed to leach soil-filled laboratory columns containing residual levels of gasoline. The water flow in the columns was maintained under steady state, unsaturated flow conditions. The 30 cm columns were constructed of glass, stainless steel and teflon components. A syringe collection technique was developed to collect effluent samples in a vacuum chamber in order to minimize losses of volatile compounds. Tensiometers were used to monitor unsaturated flow hydraulics. Four replicate column experiments were conducted using a sandy loam soil with gasoline as the residually held hydrocarbon source. Residual levels of gasoline were achieved using ceramic pressure plates. Maintaining unsaturated flow conditions, levels of benzene, toluene, ethylbenzene, m- and p-xylene and o-xylene (BTEX) were monitored over a period of one month. Initial effluent levels of BTEX agreed well with equilibrium values determined from batch solubility tests indicating that the system was operating under local equilibrium conditions. Minimum time for complete depletion of the five compounds from the soil was calculated based on the effluent concentration levels. Results from the four replicate columns were similar.

Impact of Surfactant on Configuration of Petroleum Hydrocarbon Lens

A laboratory-scale physical model was constructed for visual observation of the basic 2-D flow characteristics of a gasoline spill through an unconfined aquifer and the subsequent treatment with a surfactant. The model consists of a parallel-plate glass tank (1 m×1 m×5 cm) packed with Ottawa sand. Gasoline was released from a point source in the vadose zone. As the specific gravity of gasoline is less than one (LNAPL), it pooled above the water saturated pores of the tension saturated region of water. Beyond the lens of gasoline, the height of the capillary fringe was reduced due to capillary pollution. The gasoline lens was then treated with an aqueous phase surfactant solution of 2% dodecyl benzene sulfonate (anionic) and 2% polyethoxylate nonyl phenol (nonionic). This surfactant solution reduced the interfacial tension between the gasoline and the aqueous phase by an order of magnitude. The surfactant solution was released from the same point source in the vadose zone as the gasoline. As a result, the ...

Residual gasoline saturation in unsaturated soil with and without organic matter

Abstract The goal of this study was to investigate the influence of one variable, natural organic matter, on residual gasoline saturation in sandy soils. Capillary pressure-saturation ( P c − S ) relationships (air-gasoline) were determined for three physically-similar sandy soils, with different organic carbon contents (0.086%, 0.89% and 1.65%) and residual gasoline saturations were compared. Two initial moisture conditions, residual water saturation and air-dry, were evaluated. One soil type was packed to two different bulk densities. Visualization of the soils using cryo-scanning electron microscopy was performed to aid in better understanding the role of the organic matter in the soil. The results showed that soils with higher organic contents had higher residual gasoline saturations when starting with an initially air-dry soil. Increasing the bulk density of the same air-dried soil resulted in an increase in residual gasoline saturation. In the presence of a residual water saturation, however, residual gasoline saturations were virtually identical for the three soils and independent of bulk density; approximately 5–10 times lower than in soil that was initially air-dry. The presence of the residual water effectively coated the surface of the soil thereby reducing or eliminating gasoline/soil interactions. Some residual water may also be occupying very small pore spaces, making these locations inaccessible to the gasoline.

Reduction of gasoline component leaching potential by soil venting

Abstract Laboratory column experiments were conducted to investigate the reduction of gasoline component leaching through the use of soil venting. Soil columns and procedures employed were aimed at simulating conditions in the vadose zone. Gasoline-contaminated soil columns that had been subjected to varying periods of soil venting were leached with water under both saturated and unsaturated flow conditions. Experiments were conducted testing for non-equilibrium transport and demonstrating reductions of gasoline constituent leaching. Glass columns 10 cm long and 5.4 cm in diameter were packed with a clean glacial outwash sand and brought to low saturations of water and gasoline by controlling fluid pressures across capillary barriers. The columns were then subjected to steady air flow, each for a different length of time, so that each represented a different stage in the soil venting process. Water was percolated through these columns and effluent was sampled and analyzed for benzene, toluene, ethylbenzene, m / p -xylenes, o -xylene and naphthalene. Results of experiments testing for non-equilibrium transport using changes in water flow rate, indicated that chemical equilibrium between phases was reached in effluent water. Results from columns vented for different lengths of time showed soil venting to be effective at reducing constituent concentrations in leachate ultimately to μ g L −1 . The results were compared to an equilibrium based soil venting model.

Influence of Initial Water Saturation on the Residual Saturation of an Organic Liquid in the Vadose Zone

The influence of initial water saturation on residual organic saturation was studied by determining pressure-saturation relationships with a pressure cell apparatus. Initial water saturations varying from 0.02 to 0.40 were investigated by allowing gasoline to imbibe into the water wet soil, displacing the mobile air. Subsequently, air was forced back into the soil in a series of incremental steps that resulted in gasoline drainage and a measurement of residual gasoline saturation. The data indicate that water saturation must be considered in order to understand retention and movement of liquid organics in the vadose zone. It is concluded that residual organic saturation decreases by the amount that initial water saturation increases, as Leveretts assumption implies, for initial water saturations less than a critical value. For initial water saturations greater than the critical values the residual oil saturation remained constant, the total liquid saturation increased and Leveretts assumption was no longer valid.

Prediction of leachate concentrations in petroleum‐contaminated soils

Abstract The efficacy of cleanup methods in reducing gasoline contamination at spill sites is typically determined by measuring benzene, toluene, xylene (BTX), and total petroleum hydrocarbon (TPH) concentrations in soil samples. Although these values may provide a direct measurement of soil contamination, they may not be indicative of what is transferred to percolating water. This study addresses this issue by measuring TPH, toluene, m‐ and p‐xylene, and naphthalene levels in gasoline‐contaminated soil columns before and after forced‐air venting and relating these values to the aqueous‐phase concentrations measured when water is percolated through the same columns. Sandy soils with and without organic matter were packed into glass columns. The soils were brought to residual water and residual gasoline saturations by applying a vacuum to a ceramic pressure plate at the column bottom. Venting was performed by passing clean, moist air through the columns. The columns were subsequently leached under unsatura...

Aqueous phase permeability in the region of discontinuous NAPL

In saturated soil, a nonaqueous phase liquid (NAPL) can become a discontinuous immobile phase when capillary forces are stronger than the opposing viscous and buoyancy forces. In regions where this occurs, the aqueous phase permeability is less than the water saturated permeability of the surrounding medium, which may cause treatment to bypass the targeted contaminated region. As an in situ treatment is applied, the saturation of residual NAPL (Srn) is reduced, while the aqueous phase saturation (Sa ) and the relative permeability of the aqueous phase (kra) increases. Hence, in situ treatment of NAPL must account for various levels of heterogeneity. In this study, column experiments measure Kra‐Sa during the application of a surfactant that enhances the dissolution of NAPL Into the aqueous phase. The results indicated that the surfactant initially decreases kra followed by a steady increase as Srn is reduced. In multiphase models, it is common practice to use empirical coefficients measured from pressure‐...

WAVE FORCES ON SUBMERGED PIPE LINES

The data for the spectra of wind-generated waves measured in a laboratory tank and in a bay are analyzed using the similarity theory of Kitaigorodski, and the one-dimensional spectra of fetch-limited wind waves are determined from the data. The combined field and laboratory data cover such a wide range of dimensionless fetch F (= gF/u ) as F : 10 ~ 10 . The fetch relations for the growthes of spectral peak frequency u)m and of total energy E of the spectrum are derived from the proposed spectra, which are consistent with those derived directly from the measured spectra.A solution of finite amplitude long waves on constant sloping beaches is obtained by solving the equations of the shallow water theory of the lowest order. Non-linearity of this theory is taken into account, using the perturbation method. Bessel functions involved in the solution are approximated with trigonometric functions. The applicable range of this theory is determined from the two limit conditions caused by the hydrostatic pressure assumption and the trigonometric function approximation of Bessel functions. The shoaling of this finite amplitude long waves on constant sloping beaches is discussed. Especially, the effects of the beach slope on the wave height change and the asymmetric wave profile near the breaking point are examined, which can not be explained by the concept of constancy of wave energy flux based on the theory of progressive waves in uniform depth. These theoretical results are presented graphically, and compared with curves of wave shoaling based on finite amplitude wave theories. On the other hand, the experiments are conducted with respect to the transformation of waves progressing on beaches of three kinds of slopes ( 1/30, 1/2.0 and 1/10 ) . The experimental results are compared with the theoretical curves to confirm the validity of the theory.Measurements of drift were made in a wind and wave facility at different elevations below the mean water level. The drift profiles were obtained for reference wind speeds, Ur = 3.1, 5.7 and 9.6 m/sec. The measurement technique involved tracing the movement of small paper discs which were soaked in water to become neutrally buoyant at the elevation of release. A logarithmic drift profile is proposed. The water shear velocity, U*w, predicts a surface stress, TS = pw U*S, in agreement with that obtained from the wind shear velocity, s = Pa U*li where pa and pw refer to air and water densities, respectively.The paper describes a procedure for obtaining field data on the mean concentration of sediments in combination of waves and currents outside the breaker zone, as well as some results of such measurements. It is assumed that the current turbulence alone is responsible for the maintenance of the concentration profile above a thin layer close to the bottom, in which pick-up of sediments due to wave agitation takes place. This assumption gives a good agreement between field data and calculated concentration profiles.A section of beach on the south coast of England has been under surveillance for five years, from March 1966 until March 1971. During this period, two permeable groynes of the Makepeace Wood type were constructed. Beach cross sectional areas and rates of accretion were compared before and after groyne construction. The groynes caused a buildup in beach levels updrift.The results of model tests, carried out to evaluate the stability of submarine slopes under wave action are presented. A Bentonite clay was sedimented in a glass walled tank 6 feet long by 0.5 feet wide by 2.5 feet deep. The sedimentation and consolidation processes were studied and sediment densities were measured at various depths in the profile. Vane shear strength profiles were also measured afvarious average degrees of consolidation. Plastic markers were placed in the sediment adjacent to a glass wall so that the soil movements under both gravity and wave induced slides could be documented by photography. Dimensional similitude is discussed and the model test data are presented in a dimensionless form. All instabilities were observed to be of the infinite slope type. Analysis of the data shows that wave action is instrumental in initiating downslope mass movements in gently to steeply sloping off-shore sediments. General lack of agreement between the model test results and published theoretical analyses was found but there was close similarity in the depths and form of failure under wave action and under gravity stresses alone. The loss of stability under wave action is analyzed on the concept that failure is gravity controlled and the soil strength is reduced to a value commensurate with gravity sliding by the cyclic shearing stresses imposed by progressive waves. A method of evaluating the stability of prototype slopes using a model test correlation and field vane strength measurements is proposed. INTRODUCTION Instabilities in submarine slopes have been observed or have been inferred over a wide range of slope angles from less than half a degree up to about 30°. These subaqueous landslides are believed to have caused rupture of submarine cables and to have generated many of the geomorphological features on the ocean bottom. There are numerous records describing these landslides but very few publications discuss the application of the principles of soil mechanics to the analysis of the stability of submarine slopes. Associate Professor of Civil Engineering, Queens University at Kingston, Canada 2 Soils Engineer, Geocon Ltd., Toronto, Canada 3 Associate Professor of Civil Engineering, Cornell University, Ithaca, N.Y.Several mathematical models have been lately presented which describe the tidal wave propagation within an estuary. The existing models derived from the method for damped co-oscillating tides are based on sinusoidal wave profile. Meanwhile a tidal wave which moves upstream, generally exhibits a progressive deformation which tends to unbalance the length of time between flood and ebb tides. The actual profile is therefore no longer sinusoidal. Our investigation uses the potential method, and takes into account the wave amplitude which is usually neglected compared with the water depth. Finally, the velocity potential is obtained explicitely, using a double iterative method. Tidal elevation, particle velocities and trajectories are given by the same computer programmed algorithm. Our study shows that l) the phenomenon can be clearly visualized on the theoretical curves and 2) the magnitude of this deformation is inversely proportional to the water depth, becoming significant when the ratio f|/h reaches the critical value of 1/10. Damping and geometrical effects are also considered and the theory was applied to the St.Lawrence Estuary. A partial positive reflection of the incoming tidal wave is assumed at the narrow section near Quebec, whereas a complete negative reflection is assumed at the entrance to Lake St.Peter. The calculated and observed wave profiles, velocity distributions, and phase shifts are in good agreement.A numerical model is presented to describe the hydromechanics of lagoons connected to the ocean by relatively narrow inlets. Because special attention is given to the flushing, all second order terms in the hydrodynamic equations are retained. The study is restricted to lagoons with a onedimensional flow pattern and water of uniform density. In designing a numerical solution to the equations, the inlet equations are regarded as implicit boundary conditions to the equations describing the flow in the lagoon proper. The advantages of this approach are: (1) the size of the computational grid in the lagoon can be chosen independently of the relatively small dimensions of the inlets and (2) the flow at branching inlets (an inlet connecting a lagoon to the ocean such that branching of the inlet flow can occur) still can be described by a one-dimensional tidal model. The predictive capability of the numerical model is confirmed by favorable comparison between measured and computed particle paths and net transport for a series of laboratory experiments. In the experiments a canal of uniform width and depth is freely connected to a tidal basin at one end and at the other end is connected to the same basin by a submerged weir.