Unstable Flow during Redistribution: Controlling Factors and Practical Implications

Abstract

into a hydrophobic medium (van Ommen et al., 1988; Hendrickx et al., 1993; Ritsema et al., 1993; Ngu-of drainage in the vadose zone by inducing finger-like flow paths in soils with or without macropores. Recent studies have identified the yen et al., 1999;Carrillo et al., 2000;Wang et al., 2000b), major factors governing fingered flow to be the combined effects of soil air compression during infiltration (White et al., capillary hysteresis, the existence of a threshold water-entry value in a porous medium, and a positive matric potential gradient behind the rated infiltration under low application rates (Selker wetting front. This situation typically occurs during redistribution et al., 1992; Yao and Hendrickx, 1996; Wang et al., following high-rate infiltration, a common occurrence in hydrology. 1998b; Geiger and Durnford, 2000). In addition to these The conditions favoring instability can also develop during infiltration recognized situations, another condition predicted to into a fine-over-coarse layered soil, into hydrophobic or air-entrapped cause unstable flow in uniform porous media is redistrisoils, or even in a homogeneous coarse-textured soil if the infiltration bution following the cessation of infiltration (Raats, rate is low. An analysis of the conditions necessary for the onset of unstable flow in a uniform soil is provided in this paper. We demon-1973; Philip, 1975). Diment and Watson (1985) and Tastrate that if the matric potential gradient (dh/dz ) becomes positive mai et al. (1987) demonstrated that redistribution causes during redistribution, a perturbation at the wetting front will cause unstable flow in coarse-textured, oven-dry, uniform mafinger flow. However, if dh/dz remains negative, the perturbation will terials. However, Diment and Watson's (1985) experibe dissipated. The analysis is used to predict a critical depth of irrigaments in a small slab box showed that redistribution tion (I c ) beyond which the flow should become unstable. A series of "stabilized" when the initial water content was increased point-source and line-source infiltration experiments were conducted to only a few percent of saturation. In contrast, recent using a slab-box filled with uniform sands. The results confirmed that experiments of Wang et al. (2003a, 2003b) in the field as soon as I c is exceeded, a finger was formed at the bottom of the and with a large slab box showed that redistribution is wetting front, channeling the flow and stopping water movement in the surrounding areas. We discuss this phenomenon's implications unstable even in a very wet uniform sand. Nicholl et al. for practical irrigation and leaching designs. Unstable flow causes major uncertainties in the characterization

A review of model applications for structured soils: a) Water flow and tracer transport

“…Modelling of fingering flow (e.g., De Rooij, 2000;Wang et al, 2004), funneling flow (e.g., Ju and Kung, 1997), and reactive, colloid or particle-facilitated solute transport are excluded from this contribution. Model applications involving transport of pesticides, as representatives for reactive solutes, will be reported in a companion review article.…”

Experimental Investigation of Grouted Helical Piers for Use in Foundation Rehabilitation

“…This phenomenon was not observed in the one-g tests and thus can be attributed to the increased centrifugal field applied to the model. This type of fingering is caused by unstable flow during infiltration into unsaturated homogeneous porous media (Selker et al, 1992;Wang, et al, 2004). Note that pier rotation, deeper embedment depth and higher injection pressure all increased the lengths of the observed fingering extensions.…”

Real-time monitoring of the vadose zone as a key to groundwater protection from pollution hazard

“…Although the basic information that allows adequate characterization of flow and transport in a well-defined medium is relatively well known, the processes occurring in natural heterogeneous environments are still far from being clearly described or properly predicted. Apparently, the difficulty in characterizing water flow and contaminant transport in natural unsaturated zones at the field scale is intimately related to the instability of unsaturated flow [7][8][9][10]. In laboratory experiments where the flow characteristics through a welldefined medium may be controlled, accurate measurements and consequent close calibration of transport models may be achieved.…”