Search

Water-repellent soils occur all over the world and affect both groundwater pollution and crop yield. The finger-like wetting patterns in these soils have many similarities with unstable wetting fronts in coarse-grained sandy soils. Our objectives were to study the water movement in water-repellent sand and to examine how the theory for unstable wetting fronts applies to water-repellent sands. Infiltration experiments, in which moisture content and matric potentials were measured, were carried out in slab chambers with identical sands but with different levels of water repellence. Soil water characteristics were determined in separate experiments. Infiltration in the hydrophilic soil resulted in a uniform and horizontal front. All water-repellent sands showed a fingered flow pattern. For negative water-entry values, water infiltrated without delay. For positive values, water entered the soil only after the depth of the ponded water equaled or exceeded the water-entry pressure, which increased with increasing repellency. The finger widths predicted with the unstable flow theory agreed rather well with the observed values. In general, the research showed that the wetting patterns of water-repellent sands depended directly on the soil water characteristic curve. This implies that the type of wetting front and risk to groundwater pollution can be predicted based on laboratory-measured soil hydraulic properties.

Plant roots and soil biota require [O.sub.2] to function normally. When the soil [O.sub.2] is depleted, it must be replaced by [O.sub.2] from the atmosphere. The transport occurs primarily by gaseous diffusion. Although [O.sub.2] diffusion has been researched in homogeneous soils, and in tight clay soils with cracks, little is known about the effects of macropores on [O.sub.2] transport in typical agricultural soils. In this study we examined the impact of large pores on [O.sub.2] concentration in a sandy loam soil. Experiments were carried out with homogeneous and simulated-macropore columns with a steady-state rainfall of 3 cm [d.sup.-1] in which the [O.sub.2] concentration in the soil matrix and in the macropore was measured at 15-cm vertical intervals. The diameter of each column was 20 cm and the height was 105 cm. Groundwater depth was systematically varied between 30 and 105 cm. Soil water content did not vary with distance above the water table for the 30-, 60-, and 90-cm groundwater depths. While both the soil water for the entire column and the [O.sub.2] concentration from the surface to the 30-cm depth was the same for the homogeneous and macropore columns, the [O.sub.2] penetrated deeper in the macropore column and was 4% greater below the 30-cm depth in the macropore column than for the homogeneous column.

The often rapid arrival of pesticides at the ground water has been explained by the concept of preferential movement of water and solutes through the soil. To facilitate understanding of these transport processes, a drainage study was conducted by applying a nonadsorbed tracer to plots drained by subsurface drains. Three management practices were employed: no-till, conventional-till, and conventional-till with incorporation of the tracer. The plots were irrigated with 71-203 mm of water. Drainage line outflow and tracer concentration in the outflow were monitored for up to 52 hr. The resulting soil profiles were analyzed for tracer concentration. The main effect of plowing and incorporating the tracer was a more uniform concentration in the resulting profile. A simple mixing-layer model was used to predict the rate at which the tracer was transported out of the root zone into the layers below. The model was found to be in reasonable agreement with drainage outflow patterns, especially from the incorporated plots. These results may be helpful for future development of best-management practices for controlling the effects of agriculture on environmental pollution.

Simple methods for predicting runoff from watersheds are important in engineering practice, particularly in flood forecasting and water-balance calculation. In this paper, we illustrate that the often used SCS runoff curve-number approach in its most elementary form can be derived from assuming that only the saturated areas contribute to direct runoff. With this approach the initial abstraction or the amount of water required before runoff starts is equal to air-filled pore space per unit area for the most shallow soil in the watershed. Air-filled pore space throughout the year was calculated with aid of the simple water balance employing the Thornthwaite-Mather procedure for the most shallow soil in the watershed. By plotting the effective precipitation defined as the amount of precipitation minus the initial abstraction against the observed runoff for two watersheds in Australia and three in the northeastern United States we found that the SCS curve-number equation in its elementary form fitted the data well.

A new high-resolution quantitative method of examining quick alterations of the fluid amount present in porous media, was provided by synchrotron X-rays at the Cornell High Energy Synchrotron Source. This method is founded on controlling X-ray absorbance alterations in sand volumes crossed by an acute, highly collimated synchrotron X-ray beam. This method helps assess the physical and chemical characteristics of a rapid flow field in porous media.

Robust and noninvasive techniques for in situ measurement of soil water content are essential for monitoring and research of many hydrologic processes. Although in situ measurement of soil water content by means of a soil's electric and dielectric properties has been explored extensively, the use of ground-penetrating radar (GPR) for this purpose is not well established. We compared the soil's dielectric constant as measured by GPR and time domain reflectometry (TDR) and its relationship to water content determined independently in auger holes. We propose a calibration equation relating soil water to GPR-measured dielectric constants. The difference in calibration equations between GPR and TDR was partially attributed to the difference in frequency ranges in which TDR and GPR operate and the zero offset of the GPR.