Models for estimating capillary rise in a heavy clay soil with a saline shallow water table

Abstract

Summary. Shallow saline water tables underlie large areas of the clay soils in the Murray basin of Australia. Accurate estimation of capillary rise is important in formulating management strategies to avoid degradation of such soils. Measured capillary rise from a saline water table was compared with capillary rise estimated by three mathematical models of varying complexity and input requirement. A quasi steady state analytical model (QS-SAM), a transient state analytical model (TSAM) and a numerical model (NM) were used. An undisturbed heavy clay soil core of 0.75 m diameter and 1.4 m deep was subjected to a static saline water table at 1.2 m from the surface. A wheat crop was grown on the core and the weekly capillary rise from the water table was measured. The electrical conductivity of a I : 2 soil : water extract was determined at 0.15 m depth intervals before and 21 weeks after the introduction of the saline water table. The QSSAM did not satisfactorily estimate the initial wetting of the subsoil and the estimated capillary rise was considerably lower than the measured values. Capillary rise estimated by the TSAM was reasonably close to the measured values, but the weekly rates fluctuated considerably. The NM estimated capillary rise quite satisfactorily throughout the experiment. Except near the soil surface, the electrical conductivity values estimated by the NM were close to the measured values. For estimating total capillary rise over large areas, the TSAM is preferred over the NM because of its fewer input requirements and shorter execution time.Salinisation of soil, the increase in soluble salts in the root zone, is influenced by climate, soil type, crop, irrigation water quality and management practices, and the depth to, and salinity of the water table. The ability to estimate the rate of salinisation is vital for the management of irrigated areas in the Murray basin of Australia, where salinisation of heavy clay soils is primarily caused by capillary rise from saline shallow water tables. Offprint requests to: S. A. PrathaparA description of water transport in the unsaturated zone of a non-hysteretic, non-swelling soil is given by Richards (1931). Analytical models to estimate capillary rise from a water table could be formulated either with a steady state solution or with a transient solution for Richards' equation. Subsequently, the rate of salinisation could be estimated by combining an appropriate analytical solution for diffusive and convective solute transport with the capillary rise calculation, or by multiplying the rate of capillary rise by the concentration of salt in the water table. Numerical models to estimate capillary rise and salinisation in an unsaturated soil profile solve Richards' equation and the diffusive and convective solute transport equation simultaneously.Analytical models to estimate capillary rise are efficient and easy to use when input data are sparse and uncertain. However, the analytical models are limited to certain idealised situations such as homogeneous and is...

Observations of Salinity Patterns in Shallow Groundwater and Drainage Water From Agricultural Land in the Northern Part of the Netherlands

“…The upward movement of salts by capillary rise is well known from irrigation in arid regions (Prathapar et al, 1992;Ritzema, 1994;Zhang et al, 1999;Kelleners, 2001). It is not known to what extent capillary rise of salts affects growth and rooting depth of crops in the Netherlands.…”

Tidal impacts on riparian salinities near estuaries

“…In shallow groundwater conditions, increasing the watertable salinity may produce soil salination through capillary rise (Prathapar et al, 1992;Werner and Lockington, 2004). Where the watertable is sufficiently high, evaporative concentration of groundwater salts may also occur at the ground surface.…”

Study on capillary rise from shallow groundwater and critical water table depth of a saline-sodic soil in western Songnen plain of China

“…Ideally, such a tool should take into account well-established physical laws. Although a number of mechanistic models based on the Richards equation have been developed (Prathapar et al 1992b;Jorenush and Sepaskhah 2003;Stenitzer et al 2007), their applicability as a management tool is limited due to the model complexity and the extensive requirement of input parameters (Prathapar et al 1992a). On the other hand, many empirical models have been established (Lei et al 1988;Tang et al 1989;Prathapar et al 1992a).…”