Sistema de riego del Río Dulce, Santiago del Estero, Argentina. Brecha de rendimientos y productividad del agua en los cultivos de maíz y algodón

Seventy percent of the Planet Earth´surface is covered by water. However, only 2.5% is fresh water, the remaining 97.5% is salt water in oceans and seas. From 2.5% of freshwater, only 0.3% is available in rivers, lakes and reservoirs, 30% is located in underground aquifers and the rest is frozen in the poles and glaciers. It is clear that only a very small portion of the water of the Earth is currently available to meet the needs of mankind: drinking water, agriculture and industry. It is estimated that worldwide there are around 1,500 million hectares of farmland, of which 277 million are irrigated lands. Of the total freshwater delivered to different uses, irrigation is the main user of the planet: approximately 70% of the total freshwater is used to irrigate crops (Molden et al 2010; Kassam et al 2007). It is estimated that by year 2050 the Earth will be inhabited by 9,000 million people, which will demand more and better food. It is important to know that such a scenario must be addressed with a similar amount of arable land and fresh water. The agricultural sector should be able to provide the necessary answers, for example by making more efficient use of water and increasing their productivity. The concepts of "efficiency" and "productivity" of the water have been widely studied by the scientific and academic community. Both have different approaches, according to the scale of study, its objectives and the actors involved, which determines the framework to use. To characterize the use of water in irrigation systems is required to understand the characteristics of the cultures involved, identify the factors that restrict the efficient use of water, determine the prevailing farmers´ irrigation practices, and then be able to explore strategies of management within each particular technological, economic and social framework. One of the approaches to assess the water use and productivity in agriculture is based on the use of simulation models, whose main function is to estimate the production of crops according to climate, soil, and technological management. The model AquaCrop (Steduto et al 2009) focuses on the crops response to water availability. In the Río Dulce Irrigation System (SRRD), located in the province of Santiago del Estero, Argentina, crop production is lower than attainable. The low irrigation frequency and standard farmer´ practices expose crops to periods of water shortage, which have negative impact on yields, and also limit the efficient use of water. This thesis is aimed to characterize the use of irrigation water in maize and cotton crops and to analyze irrigation strategies allowing increasing crops production and improving water use in the SRRD. For that purspose, the following specific objectives have been raised: to calibrate and validate the AquaCrop model for maize and cotton crops; to evaluate the yields and water productivity gaps in those crops, and quantify the benefits of changing the system of fixed shifts to one more flexible. Calibration and validation of maize in the agro-ecological conditions of Santiago del Estero is dealt with in Chapter 2. To do this, experiments set at INTA between 2007 and 2012 were used. Their objectives were to determine the response of maize to various planting dates and deficit irrigation treatments. The model precisely simulated maize behavior. Statistical indicators gave a degree of adjustment that can be described as very good; by way of example, in yields, values were as follows: coefficient of determination (R2)=0,985; Willmott index (d) =0,995; Mean absolute error (MAE)=0, 259; Root of the mean square error (RMSE)=0, 361 and average normalized quadratic error (NRMSE) root=3,6. To close this chapter, a meta-analysis of AquaCrop ability to simulate the behavior of the corn in a wide range of climate, soil, genetic materials and agronomic management conditions was made, analyzing the existing literature. It showed the good response of the model in twelve places. The good behavior of the model in Santiago del Estero, Argentina, having different climate characteristics to the place where the x crop parameters were defined (Davis, California, USA), strengthens the viability of AquaCrop to simulate the production of maize. Calibration and validation of cotton is treated in Chapter 3. To this purpose, experiments set at INTA between 2007 and 2013 were used. Their objectives were to determine the cotton response to deficit irrigation treatments and to different soil water contents (without stress conditions). AquaCrop simulated very well the behavior of cotton. The values of the statistical indicators were: R2=0.940; d=0.974; MAE=0,317; RMSE=0,413 and NRMSE=10.1. The varieties used in the experiments are very different to those used in original crop parameterization, so some conservative parameters were changed, among others: canopy growth coefficient, time to maximum canopy cover, time to flowering and harvest index. It should be noted that the maximum yields observed in experiments and appropriately simulated by AquaCrop, quite exceed yields on which is based the original calibration. Finally, it is necessary to highlight the great adaptation of the model to different agronomic conditions (distance between lines, plant density and intensive use of growth regulators) with respect to which were used for their original parameterization. It is emphasized the simplicity of AquaCrop and its capacity of adaptation to different climates and agronomic management. The model can be used in several applications: yield estimation, analysis of production risks by impact of droughts, definition of irrigation strategies at farm level to improve water use and as a tool in assessing irrigation performance. This thesis is the first approach to assess, at system level, yield and water productivity gaps in the Río Dulce Irrigation System (SRRD), located in Santiago del Estero, Argentina. This is done in chapter 4 and the specific objectives were: characterize variations in maize and cotton water use in the San Martín District (SRRD); determine the gaps in yield and water productivity in those crops; assess the benefits of changing the actual irrigation rotational system to a more flexible water delivery, and evaluate how irrigation scheduling impact on water use. To achieve these objectives, AquaGIS (Lorite et al 2013) was used. AquaGIS is an interface that overcomes the current limitations of AquaCROP when it is necessary to do many simulations over large areas and for a large number of years. The use of AguaGIS was complemented with field surveys and questionnaires, to adjust the information of the current productive situation. Various scenarios have been set, whit the following factors: climate (series 1988-2013), soil (four), crop (maize and cotton), production levels (potential, water limited and actual), and planting date (two for each crop). The combination of variables resulted in 624 simulations per crop. Production levels were associated to irrigation strategies: potential production/on demand irrigation; water limited production/improved irrigation scheduling within the present rotational system; actual production/current farmers` irrigation management. For maize, the gap between potential and actual yield is estimated at 5900 kg ha-1 and the gap between potential yield and achievable (limited by the irrigation rotational system) yield was estimated at 1100 kg ha-1. Current water productivity (WP) is 17 kg ha-1.mm-1, considered low considering the potential of hybrids currently used. The WP in potential production is 25 kg ha-1.mm-1 and in water limited production, 23 kg ha-1.mm-1. The current irrigation water productivity (IWP) is very low, 4.5 kg ha-1.mmIW -1. The IWP in potential production is 27 kg ha-1.mmIW -1 and in water limited production 16 kg ha-1.mmIW -1. Simple changes in farmers` irrigation practices would lead to significant increases in maize WP and IWP, even within the constraints imposed by the present rotational irrigation system. In cotton, the gap between potential and actual yield is estimated at 2000 kg ha-1 and between potential...

La Tierra recibe, entre otros, el apelativo de "Planeta Azul" debido a que el 70% de su superficie está cubierta por agua. Sin embargo, sólo el 2.5% es agua dulce, el 97.5% restante son mares y océanos de agua salada. Del 2.5% de agua dulce, sólo el 0.3% está disponible en ríos, lagos y embalses, el 30% se encuentra en acuíferos subterráneos y el resto está congelada en los casquetes polares y glaciares. Resulta claro que sólo una muy pequeña porción del agua de la Tierra está actualmente disponible para cubrir las necesidades de la humanidad: agua potable, agricultura e industria. Se estima que en el mundo hay alrededor de 1,500 millones de hectáreas de tierras de cultivo, de las cuales 277 millones disponen de riego. Del total del agua dulce derivada para los distintos usos, el riego es el principal usuario del planeta, con aproximadamente el 70% del total se usa para el riego de los cultivos (Molden et al. 2010; Kassam et al 2007; FAO 2012). Se calcula que en el año 2050 el planeta Tierra estará habitado por 9,000 millones de seres humanos, que demandarán más y mejores alimentos. Se estima que los pilares permitirán hacer frente a tal desafío serán las técnicas agronómicas, la biotecnología y el manejo juicioso de los recursos naturales. Un aspecto clave a considerar es que la disponibilidad de tierras cultivables y agua será similar que en el siglo pasado. El sector agrícola deberá ser capaz de dar las respuestas necesarias, por ejemplo haciendo un uso más eficiente del agua e incrementando su productividad. Los conceptos de “uso eficiente” y “productividad” del agua han sido y son ampliamente estudiados por la comunidad científica y académica ligadas al uso del agua en la agricultura. Ambos tienen matices según la escala de estudio, sus objetivos y los actores involucrados, lo que determina el marco a utilizar. Para caracterizar el uso del agua en los sistemas de riego se requiere conocer las características de los cultivos involucrados, identificar los factores que restringen el uso eficiente del agua, determinar los hábitos predominantes de riego de los agricultores, para luego poder explorar estrategias de manejo en cada marco tecnológico, económico y social particular. Uno de los enfoques para evaluar el uso y la productividad del agua en la agricultura se basa en el uso de modelos de simulación, cuya función principal es estimar la producción de los cultivos en función del clima, el suelo y el manejo tecnológico. El modelo AquaCrop (Steduto et al 2009) se centra en la respuesta de los cultivos a la disponibilidad de agua. En el Sistema de Riego del Río Dulce-Los Quiroga, ubicado en la provincia de Santiago del Estero, Argentina, la producción de los cultivos es bastante menor que la que podría obtenerse. La baja frecuencia del turno de riego y las prácticas habituales de manejo exponen a los cultivos a períodos de déficit hídrico, que tienen impacto negativo en los rendimientos y, asimismo, limitan el uso eficiente del agua. La presente tesis tiene como finalidad, para el Sistema de Riego del Río Dulce-Los Quiroga, Santiago del Estero, Argentina, caracterizar el uso del agua de riego en los cultivos de maíz y algodón y analizar estrategias de riego que permitan aumentar su producción, mejorando el uso del agua. Para ello, se plantearon los siguientes objetivos específicos: calibrar y validar el modelo de simulación AquaCrop para los cultivos de maíz y algodón, evaluar la brecha de rendimientos en esos dos cultivos a nivel de productor y cuantificar los beneficios productivos de cambiar el sistema de turno fijo a uno más flexible, que permita acoplar el riego con las necesidades de agua de los cultivos. La calibración y validación del maíz para las condiciones agroecológicas de Santiago del Estero se aborda en el Capítulo 2. Para ello, se utilizaron experimentos realizados en INTA entre los años 2007 y 2012, que tuvieron como objetivos determinar la respuesta de híbridos de maíz a distintas fechas de siembra y a tratamientos de riego deficitario. El modelo simuló...