J.-Yves Parlange

Mechanism for finger persistence in homogeneous, unsaturated porous media: Theory and verification

Wetting-front instability, or gravity-driven fingering, can occur during vertical infiltration. Previous studies found that fingers persist over long periods of constant infiltration and in subsequent infiltration cycles. We present a physically based theory to explain finger persistence. The theory is tested using a new technique in which moisture contents are visualized by the transmission of light through unsaturated sand, recorded with a video camera, and then analyzed by video digitizing computer hardware. Theory and experiments show that hysteresis in the moisture characteristic relationship explains the persistence of fingers in time.

Formation and persistence of fingered flow fields in coarse grained soils under different moisture contents

Abstract Fingers in homogeneous coarse-grained soils form from an instability in the wetting front and result in bypass flow which shortens the residence time in the vadose zone. Once formed, these fingers persist over a long period. In this paper, the formation of fingers and their persistence in initially uniform dry or wet soils are examined. Using fast-responding tensiometers in conjunction with rapid moisture measurements, it is shown that the matric potential at the finger tip and the position of the wetting curve relative to the drying curve are key factors in the persistence of the fingers. In accordance with the earlier observation, the tip and the fringes of the finger are on a wetting curve, while the remaining core of the finger is on a drying branch. This explains why large differences in moisture content can co-exist. This study also shows that the initial water content in and around the finger is crucially important in relation to which particular scanning curve is followed and, thus, to the water contents during infiltration.

Estimating basin-wide hydraulic parameters of a semi-arid mountainous watershed by recession-flow analysis

Insufficient sub-surface hydraulic data from watersheds often hinders design of water resources structures. This is particularly true in developing countries and in watersheds with low population densities because well-drilling to obtain the hydraulic data is expensive. The objective of this study was to evaluate the applicability of ‘Brutsaert’ recession flow analysis to steeper and more arid watersheds than those that were previously used. Using daily streamflow data (1962 ‐ 1992), a modified version of the analysis was used to estimate the subsurface hydraulic parameters of four semi-arid, mountainous watersheds (204 ‐ 764 km 2 ) near Oaxaca, Mexico. In this analysis, a dimensionless recession curve (DRC) was translated to best-fit observed recession flow (Q) data. The basin-wide hydraulic parameters were directly related to the magnitude of the translation used to fit the DRC to the data. One unique aspect of this study was too few high flow data to confidently fit the DRC to the data via previously published protocols. We, thus, proposed using three different approaches for translating the DRC in order to establish a range for the hydraulic parameters. Our analyses predicted a narrow range of basin-wide hydraulic parameters that were near regionally measured values and consistent within commonly published values for similar geology, suggesting that the Brutsaert method is applicable to arid, mountainous basins like those used here. This method potentially provides a costeffective alternative to traditional geohydrological field methods for determining groundwater parameters. q 2003 Elsevier B.V. All rights reserved.

Visualization by light transmission of oil and water contents in transient two-phase flow fields

Abstract The difficulty of determining transient fluid contents in a soil–oil–water system is hampering an understanding of the systems flow characteristics. In this paper, we describe a light transmission method (LTM) which can rapidly obtain oil and water contents throughout a large two-dimensional flow field of silica sand. By appropriately coloring the water with 0.005% FD&C blue #1, the hue of the transmitted light is found to be directly related to the water content within the porous media. The hue provides a high resolution measurement of the water and oil contents in transient flow fields (such as unstable flow). Evaluation of the reliability of LTM was assessed by checking the mass balance for a known water injection and its utility in visualizing a whole flow field was exemplified for unstable fingered flow by comparing fluid contents to those obtained with synchrotron X-ray radiation.

Lateral expansion of preferential flow paths in sands

The stability and persistence of preferential flow paths in sands can determine the flow paths of subsequent infiltration events. We have measured the evolution of preferential flow paths in a slab of sandy soil using an array of tensiometers and light transmission. The pressure and water content measurements show that the nonuniform moisture content exists even when the potentials are equalized horizontally and that then are the result of hysteresis in the soils pressure-saturation relationship. The equalization of potential takes place over several days, if at all, and is consistent, initially, with estimates of vapor transport out of the finger cores. Once the soil is wet enough, the remainder of water movement takes place in liquid films. Hysteresis produces another interesting situation when the pack is drained. We find that the wetter portions of the soil can be at a lower potential than the drier portions, resulting in a horizontal driving force for a flow of water from the drier to the wetter soil.

Preferential Flow and Transport of Cryptosporidium parvum Oocysts through the Vadose Zone: Experiments and Modeling

in the form of 4- to 6-m-long ovoid-shaped oocysts, with a double wall that is resistant to most oxidation As a result of Cryptosporidium parvum in drinking water, several processes such as ozonation and chlorination (Current, outbreaks of cryptosporidiosis have occurred in the last 10 yr. Al1986; Atwill et al., 1997). though it is generally believed that movement of pathogens through the soil is minimal, recent research has shown that appreciable num- During the past two decades, the presence of C. parbers of C. parvum oocysts may be transported via preferential or vum in surface- and groundwaters in the United States fingered flow to groundwater. The objective of the present research and Great Britain (Galbraith et al., 1987; Rose et al., was to further investigate and model the transport of oocysts through 1991; Craun et al., 1998) has been associated with several preferential flow paths in the vadose zone under a “worst-case” sce- major outbreaks of cryptosporidiosis (Hayes et al., 1989; nario. This was studied by adding calves feces containing C. parvum MacKenzie et al., 1994). Among the different pathways oocysts with a Cl tracer to undisturbed silt loam columns and disfor the transport of oocysts to drinking water sources, turbed sand columns during a simulated steady-state rain. The sand columns exhibited preferential flow in the form of fingers whereas downward percolation is usually considered to be insigmacropore flow occurred in the undisturbed cores. In the columns nificant, because soils are generally assumed to be an with fingered flow, oocysts and Cl were transported rapidly with the effective filter for a wide range of pathogens. Studies same velocity through the columns. Although only 14 to 86% of the of packed columns with saturated flow by Brush et al. amount applied, the number of oocysts transported across the columns (1999) and Harter et al. (2000) and undisturbed columns was several orders of magnitude above an infective dose. The macwith unsaturated flow (Mawdsley et al., 1996), however, ropore columns had only a very limited breakthrough of oocysts, showed that C. parvum oocysts could be transported which appeared several pore volumes after the Cl broke through initially. A simulation model for the transport of oocysts via preferen- rapidly downward through the soil. Although transport tial flow was developed on the basis of an existing preferential flow of C. parvum oocysts in saturated flow has been studied model for nonadsorbing solutes, with addition of a first-order sink experimentally and described mathematically (Brush et term for adsorbance of the C. parvum to the air–water–solid (AWS) al., 1999; Harter et al., 2000), detailed observations of interfaces, and with velocity and dispersivity parameters derived from the transport and persistence of C. parvum oocysts in Cl transport. The breakthrough of C. parvum oocysts could be de- unsaturated soils with preferential flow are still lacking, scribed realistically for the sand columns. However, the model could particularly in the presence of preferential flow pronot describe oocyst transport in the columns with macropores. cesses.

Closed‐form solution for finger width in sandy soils at different water contents

Wetting front instability and gravity-driven fingers can occur during infiltration in sandy soil. These soils overlay many major aquifers. Finger width and velocity are important parameters in predicting the risk of groundwater pollution in these soils. In previous studies, closed-form solutions were developed for air dry soil. In uniformly wet soils, where the fingers are much wider, predictive finger width equations have failed so far. In this paper we simplify further the Parlange-Hill closed-form solution for dry soil and show that, by taking hysteresis into account, it applies to fingers emanating from a point source in a sandy test soil at two widely different moisture contents.

Preferential transport of Cryptosporidium parvum oocysts in variably saturated subsurface environments

When oocysts of the protozoan Cryptosporidium parvum contaminate drinking water supplies, they can cause outbreaks of Cryptosporidiosis, a common waterborne disease. Of the different pathways by which oocysts can wind up in drinking water, one has received little attention to date; that is, because soils are often considered to be perfect filters, the transport of oocysts through the subsoil to groundwater is generally ignored. To evaluate the significance of this pathway, a series of laboratory experiments investigated subsurface transport of oocysts. Experiment 1 was carried out in a vertical 18-cm-long column filled either with glass beads or silica sand, under conditions known to foster fingered flow. Experiment 2 involved undisturbed, macroporous soil columns subjected to macropore flow. Experiment 3 aimed to study the lateral flow on an undisturbed soil block. The columns and soil samples were subjected to artificial rainfall and were allowed to reach steady state. At that point, feces of contaminated calves were applied at the surface along with a known amount of potassium chloride to serve as a tracer, and rainfall was continued at the same rate. The breakthrough of oocysts and chloride, monitored in the effluent, demonstrate the importance of preferential flow on the transport of oocysts. Compared with chloride, peak oocyst concentrations were not appreciably delayed and, in some cases, occurred even before the chloride peak. Recovery rates for oocysts were low, ranging from 0.1 to 10.4% of the oocysts originally applied on the columns. However, the numbers of oocysts present in the effluents were still orders of magnitude higher than 10 oocysts, the infectious dose considered by the U.S. Food and Drug Administration, Center for Food Safety and Applied Nutrition, to be sufficient to cause Cryptosporidiosis in healthy adults. These results suggest that the transport of oocysts in the subsurface via preferential flow may create a significant risk of groundwater contamination in some situations.

Hydraulic conductivity of gravel and sand as substrates in rock-reed filters

Abstract Long-term use of a constructed wetland to treat landfill leachate requires that the saturated hydraulic conductivity be maintained and clogging avoided to prevent overland flow, which bypasses the treatment process. This paper describes the application of an equation developed for prediction of cumulative drainage volume from hillslopes to measure the saturated hydraulic conductivity ( K s ) of substrates used in rock-reed filters. Outflow was measured at five intervals during the first 26 months of operation. The values of K s ) obtained by the drainage equation compared favorably with values calculated from a more dificult method based on Darcys law. Results indicate that the finest substrate (a sand-and-gravel mixture) became almost completely clogged, and that the presence of reeds ( Phragmites australis ) did not maintain or increase the conductivity. Hydraulic conductivity of pea-gravel (0.5-cm diameter) and coarse-gravel (3-cm diameter) substrates with reeds did decrease in the 26-month period.

A simplified hillslope erosion model with vegetation elements for practical applications

Soil and water conservation practices are increasingly being considered for curbing non-point source pollution from agricultural land. Several studies have demonstrated that stream power is a simple and good predictor of soil detachment and transport and can be used to predict the effect of soil and water conservation practices on soil loss. Our objective was, therefore, to develop a simple water erosion simulation model that is physically based on stream power, handles vegetation in terms of contact cover, and considers the settling velocity characteristics of the eroding sediment. The model assumes that rill flow can occur on hillslope segments with net erosion, but on segments with net deposition sheet flow is assumed. Input parameters include the depositability of the soil, rill shape, rill density, net precipitation, and an empirical power function describing the decrease of sediment concentration with vegetative cover increase. The model was evaluated by comparison of predicted and observed relationships between sediment concentration, slope, and vegetative residue cover in two experimental studies using simulated rainfall: one that involved erosion plots with various uniform slopes and levels of vegetative cover, and another that involved the observation of soil movement on mechanically shaped concave, uniform, and convex slopes with negligible vegetation. Without calibration, the model appeared to represent soil erosion relationships observed in these studies and is simple enough to be included in grid-based variable source hydrology models.