Modeling the effects of saline groundwater and irrigation water on root zone salinity and sodicity dynamics in agro-ecosystems

Recent trends and future projections suggest that the need to produce more food and fibre for the world’ s expanding population will lead to an increase in the use of marginal-quality water and land resources (<em>Bouwer, 2000; Gupta and Abrol, 2000; Wild, 2003</em>). This is particularly relevant to less-developed, arid and semi-arid countries, in which problems of soil and water quality degradation are common (<em>Qadir and Oster, 2004</em>). The aim, therefore, should be to increase yield per unit of land rather than the area cultivated. More efforts are needed to improve productivity as more lands are becoming degraded. It is estimated that about 15% of the total land area of the world has been degraded by soil erosion and physical and chemical degradation, including soil salinization (<em>Wild, 2003</em>). The main sources of soil salinity and sodicity development are groundwater and irrigation water. In discharge areas of the landscape, water exits from groundwater to the soil surface bringing the salts dissolved in it. The driving force for upward movement of water and salts is evaporation from the soil plus plant transpiration. Salt accumulation is high when the water table depth is less than a threshold. However, this threshold depth may vary depending on soil hydraulic properties and climatic conditions. Groundwater associated salinity and sodicity affects around 350 X 10⁴km²in the world (<em>Szabolcs, 1989</em>). In this thesis, the focus is to quantify and understand the salinity and sodicity dynamics, and the feedback on dynamics in groundwater dependent agro-ecosystems. First we have considered the impact of salt coming from groundwater on capillary fluxes and on the root zone water and salt dynamics. Groundwater can be a source of both water and salts in semi-arid areas, and therefore capillary pressure induced upward water flow may cause root zone salinization. To identify which conditions result in hazardous salt concentrations in the root zone, we combined the mass balance equations for salt and water, further assuming a Poisson-distributed daily rainfall and brackish groundwater quality. For the water fluxes (leaching, capillary upflow, and evapotranspiration), we account for osmotic effects of the dissolved salt mass using Van‘t Hoff’s law. Root zone salinity depends on salt transport via capillary flux and on evapotranspiration, which concentrates salt in the root zone. Both a wet climate and shallow groundwater lead to wetter root zone conditions, which in combination with periodic rainfall enhances salt removal by leaching. For wet climates, root zone salinity (concentrations) increases as groundwater is more shallow (larger groundwater influence). For dry climates, salinity increases as groundwater is deeper due to a drier root zone and less leaching. For intermediate climates, opposing effects can push the salt balance in either way. Root zone salinity increases almost linearly with groundwater salinity. With a simple analytical approximation, maximum concentrations can be related with the mean capillary flow rate, leaching rate, water saturation and groundwater salinity, for different soils, climates and groundwater depths. A Soil sodicity (quantified by <em>ESP</em>) model based on the soil salinity model (as discussed above) has been developed. For sodicity calculations, we have used the Gapon equation favored in salinity research. The simulation results show that soil salinity and sodicity development in groundwater driven agro-ecosystems play a major role in soil structure degradation. To identify which conditions can make soil sodic, we have modeled the coupled water, salt, and cation balances. The root zone salinity <em>C</em>and sodicity <em>ESP</em>gradually change to their long term average values. These long term average values are independent of the cation exchange capacity <em>CEC</em>. The rate of change depends inversely on the size of the root zone reservoir, i.e., on root zone thickness for <em>C</em>, and additionally on <em>CEC</em>, for <em>ESP.</em>Soil type can have a large effect on both the rate of approach of the long term steady state salinity and sodicity, and on the long term levels, as it affects the incoming and out-going water and chemical fluxes. Considering two possible sources of salts, i.e., groundwater and irrigation water (here represented by rainfall), the long term salt concentration <em>C</em>of the root zone corresponds well with a flux weighted average of infiltrating and upflowing salt mass divided by the average water drainage. In full analogy, the long term <em>ESP</em>can be approximated very well for different groundwater depths and climates. A more refined analytical approximation, based on the analytical solution of the water balance of <em>Vervoort and Van der Zee</em>(2008), leads to a quite good approximation of long term salinity and sodicity, for different soils, groundwater depths, and climates. Modeling is an efficient tool to investigate water and solute movement in groundwater driven agro-ecosystems. However, in most available models (<em>SWAP,</em><em>MODFLOW/MT3D</em>) continuing degradation of soil hydraulic properties as a result of rising Na⁺concentrations is ignored. Disregarding the soil hydraulic degradation due to sodicity level in some cases makes modeling water and solute movement within the soil profile questionable. We have translated the effects of soil salinity and sodicity into reduction in saturated hydraulic conductivity to quantify the feedback effects of reduction in saturated hydraulic conductivity on root zone fluxes, salinity, and sodicity under different groundwater depths and climates of Oenpelli and Tennant Creek Airport located in the North Territory of Australia. The reduction in saturated hydraulic conductivity due to salinity and sodicity <em>(Ks(C,ESP))</em>has been calculated by using the procedure developed by <em>McNeal</em>(1968). The significant feedback effects of <em>Ks(C,ESP) </em>on salt concentration and soil <em>ESP</em>depend on many important parameters like groundwater depth, leaf area index, weather seasonality and non-seasonality, and soil type. Out of these important parameters, weather seasonality is the main driver that can develop significant feedback effects of <em>Ks(C,ESP) </em>onsalt concentration and soil <em>ESP</em>. Furthermore, <em>Ks(C,ESP) </em>although decreasing the capillary flux, leaching flux, and evapotranspiration, it increases the magnitude of runoff. Also when <em>Ks(C,ESP)</em>affects both capillary and leaching flux under seasonal rainfall, the feedback effects are significant compared to the partial feedback (<em>Ks(C,ESP)</em>affects only leaching flux, but not capillary flux). In the second theme of this thesis, we have focused on optimizing irrigation water between two farms under water scarcity and salinity regimes. In arid and semi-arid regions, irrigation water is scarce and often saline. To reduce negative effects on crop yields, the irrigated amounts must include water for leaching and therefore exceed evapotranspiration. The leachate (drainage) water returns to water sources such as rivers or groundwater aquifers and increases their level of salinity and the leaching requirement for irrigation water of any sequential user. We develop a sequential (upstream-downstream) model of irrigation that predicts crop yields and water consumption and tracks the water flow and level of salinity along a river dependent on irrigation management decisions. The model incorporates an agro-physical model of plant response to environmental conditions including feedbacks. For a system with limited water resources, the model examines the impacts of water scarcity, salinity and inefficient application on yields for specific crop, soil, and climate conditions. As a general pattern we find that, as salinity level and technical inefficiency increase, the system benefits when upstream farms use less water than is available to them, to provide downstream farms with more and better quality water. We compute the marginal value of water, i.e. the price water that would command on a market, for different levels of water scarcity, salinity and levels of water loss. In summary this thesis aims to understand theoretically how soil salinity and sodicity develop under different climates, groundwater depths, soil types, root zone thicknesses, and different groundwater salinities. The developed salinity sodicity model can be applied in potential salt affected areas to predict the long term salinity, sodicity trends. Furthermore, quantification of feedback effects of reduction in saturated hydraulic conductivity (<em>Ks(C,ESP)) </em>on root zone fluxes, salinity, and sodicity guide us towards better management of soil, vegetation, and irrigation/groundwater.