Macroscopic approaches to root water uptake as a function of water and salinity stress

Abstract

Computer simulation models can be important tools for analyzing and managing site-specific irrigation, soil salinization, or crop production problems. For many of these problems, computer analyses require simulations of root water uptake as a function of water and salinity stress. We give an overview of macroscopic modeling approaches that are based on the Richards equation with a sink term specifying water uptake. Various parameterizations of the sink term as functions of water and salinity stress are reviewed. An example application demonstrates the simulation of drainage for a wide range of salinity and water stress conditions. The example shows that in practice it is very difficult to discriminate among the various functional forms proposed for the sink term. Future improvements to models may involve the incorporation of more dynamic root functions, such as the ability of roots to respond actively to growing conditions. As an example, we consider modeling compensated uptake, in which plants react to stress in one section of the root zone by increasing uptake in sections with more favorable conditions. Lastly, the challenge of estimating uptake reduction parameters from crop salt tolerance databases is discussed.

Introduction

Irrigation has contributed significantly to increased crop production worldwide. Unfortunately, irrigation has contributed also to increased salinization of agricultural lands, and has caused the destruction of agriculture in many areas (Van Schilfgaarde, 1980, Van Schilfgaarde, 1984, Van Schilfgaarde, 1994). About 20–30 million ha of irrigated land are seriously damaged by the build-up of salts and 0.25–0.5 million ha are estimated to be lost from production every year as a result of salt build-up (FAO, 2002). The FAO/AGL SPUSH network (http://www.fao.org/ag/agl/agll/spush/intro.htm) estimates that 19.5% of the world's 230 million ha of irrigated land is salt-affected. In the United States, 9% of cropland and pastureland suffers from reduced productivity due to salinity (Ghassemi et al., 1995). Still, despite the negative impacts, irrigation is critical to sustaining and increasing agricultural production. While about 17% of agricultural land worldwide is irrigated, this 17% accounts for about 40% of the total global food harvest (FAO, 2002). Moreover, per capita arable land has decreased over the years, from a worldwide average of 0.38 ha in 1970 to 0.28 ha in 1990, and some analysts project a further decrease to 0.15 ha in 2050 (Ghassemi et al., 1995). Hence, increased production must come from increased average yields, increases that will be possible only through high-yielding irrigated agriculture.

Effective management of salt-affected soils requires knowledge of many coupled physicochemical processes affecting soil conditions. Computer models such as SWAP (van Dam et al., 1997, Kroes and van Dam, 2003) and HYDRUS-1D (Šimunek et al., 2005) have become increasingly important tools for analyzing site-specific irrigation, soil salinization, or crop production problems. Most of these models are based on the Richards equation for variably saturated water flow and the advection–dispersion equation for solute transport. In their simplest one-dimensional forms, these equations are

$$\frac{\partial \theta }{\partial t}=\frac{\partial }{\partial z}\left[K(h)\frac{\partial h}{\partial z}-K(h)\right]-S$$

$$\frac{\partial (\theta Rc)}{\partial t}=\frac{\partial }{\partial z}\left(\theta D\frac{\partial c}{\partial z}-qc\right)-\varphi$$

respectively, where θ is the volumetric water content, h the soil water pressure head (L), t the time (T), z the depth (L), K the hydraulic conductivity (L T⁻¹), R a retardation factor accounting for sorption or exchange, c the solute concentration of the liquid phase (M L⁻³), D the solute dispersion coefficient (L² T⁻¹), q the Darcy–Buckingham volumetric water flux (LT⁻¹), and S (T⁻¹) and ϕ (M L⁻³ T⁻¹) are sinks or sources for water and solutes, respectively. In this paper, S and ϕ are associated exclusively with root uptake processes.

While models based on Eqs. (1), (2) are critical tools in irrigation and drainage studies, the equations pose challenges because of (i) the many highly nonlinear processes involved, including water and energy exchange between the biosphere and the atmosphere; (ii) issues of scale, especially for studies at larger field and watershed scales; (iii) lack of data on the many parameters required in the model. Proper parameterization of root water uptake as a function of water and salinity stress remains one of those challenges.

A large number of microscopic and macroscopic approaches to modeling water uptake have been proposed over the years. Comprehensive reviews from a mostly hydrological perspective include Molz (1981), Hopmans and Bristow (2002), Wang and Smith (2004), and Feddes and Raats (2004). The microscopic approach generally involves descriptions of radial flow to, and uptake by, individual roots (Hillel et al., 1975, Raats, in press). In contrast, modeling uptake with a sink term in the Richards equation (Eq. (1)) is a typical macroscopic approach that averages uptake over a large number of roots. The approach ignores or implicitly averages pore-scale variations in the pressure head or solute concentration in the immediate vicinity of individual roots.

This paper is limited to the macroscopic approach. Specifically, we review relatively standard macroscopic approaches to modeling root water uptake in the presence of water and salinity stress, provide an example application involving a lysimeter study, briefly raise the issue of compensatory uptake, and discuss some practical issues concerning the use of salt tolerance databases to determine uptake reduction parameters. Elsewhere in this issue, Green et al. (this issue) provide an overview of root uptake that is complimentary to our discussion, focusing on new technologies for measuring plant water use and observing compensatory uptake, and on the use of modeling to develop better irrigation policies for allocating water.