Optimum Design of Alternate and Conventional Furrow Fertigation to Minimize Nitrate Loss

Hamed Ebrahimian, Abdolmajid Liaghat, Masoud Parsinejad, Enrique Playán, Fariborz Abbasi, Maryam Navabian, Borja Lattore
2013 Journal of irrigation and drainage engineering  
1 Alternate furrow fertigation has shown potential to improve water and fertilizer application 2 efficiency in irrigated areas. The combination of simulation and optimization approaches permits 3 to identify optimum design and management practices in furrow fertigation, resulting in optimum 4 cost, irrigation performance or environmental impact. The objective of this paper is to apply 1D 5 surface and 2D subsurface simulation-optimization models to the minimization of nitrate losses 6 in two
more » ... es of alternate furrow fertigation: a) variable alternate furrow irrigation; and b) fixed 7 alternate furrow irrigation. For comparison purposes, optimizations are also reported for 8 conventional furrow irrigation. The model uses numerical surface fertigation and soil water 9 models to simulate water flow and nitrate transport in the soil surface and subsurface, 10 respectively. A genetic algorithm is used to solve the optimization problem. Four decision 11 variables (inflow discharge, cutoff time, start time and duration of fertilizer solution injection) 12 were optimized to minimize the selected objective function (nitrate loss) for two fertigation 13 events performed during a maize growing season. The simulation-optimization model succeeded 14 in substantially reducing the value of the objective function, as compared to the field conditions 15 for all irrigation treatments. In the experimental conditions, optimization led to decreased inflow 16 discharge and fertilizer injection during the first half of the irrigation event. This was due to the 17 high potential of the field experiment to lose water and nitrate via runoff. In the optimum 18 conditions, alternate furrow fertigation strongly reduced water and nitrate losses compared to 19 conventional furrow irrigation. The simulation-optimization model stands as a valuable tool for 20 the alleviation of the environmental impact of furrow irrigation. 21 22 30 Slatni et al. 2011). Therefore, using fertigation in alternate furrow irrigation could not only 31 conserve water, but also reduce fertilizer losses. 32 Using simulation models, different scenarios can be evaluated with minimum time and cost 33 to find convenient values of surface irrigation variables, such as inlet discharge and cutoff 34 time. Field experiments are costly and time consuming, and can not explore all values of 35 the relevant irrigation variables. Feinerman and Faakovitzo (1997) developed and applied a 36 mathematical model for identifying optimal scheduling of corn fertilization and irrigation, 37 resulting in maximum farmer's economic profit. These authors found that leaching was 38 much more sensitive to changes in fertilizer price than to changes in taxes on leached 39 nitrogen. Sabillon and Merkley (2004) developed a mathematical model of furrow 40 fertigation. After validating the model, they executed the model for about 50,000 times to 41 identify the start and end times of fertilizer injection leading to optimum fertilizer 42 application efficiency and uniformity. In their experimental conditions, the best injection 43 duration ranged from 5 to 15 % of cutoff time. 44 4 The relationship between surface irrigation and fertigation on one hand, and water and 45 solute transfer on the other, has been analyzed since the turn of the century. Popova et al. 46 (2005) reported the use of subsurface flow model HYDRUS-2D (Šimůnek et al. 1999) for 47 optimization of joint irrigation and fertilization practices in different climates and soil 48 contexts. Abbasi et al. (2004) and Crevoisier et al. (2008) proved that HYDRUS-2D could 49 successfully simulate water and solute transfer for conventional and alternate furrow 50 irrigation. Crevoisier et al. (2008) indicated that HYDRUS-2D performance was better than 51 HYDRUS-1D (Šimůnek et al. 1998) for simulating water content, nitrate concentration and 52 drainage. Zerihun et al. (2005) coupled a surface solute transport model with a subsurface 53 solute transport model (HYDRUS-1D) for simulating surface fertigation in borders and 54 basins. Adequate agreement was reported between field observed and model predicted 55 solute breakthrough curves in the surface stream. Wöhling and Schmitz (2007) also 56 presented a seasonal furrow irrigation model by coupling a 1D surface flow model (zero-57 inertia), HYDRUS-2D and a crop growth model. The coupled model could adequately 58 predict advance and recession times, soil moisture and crop yield (Wöhling and Mailhol, 59 2007). 60 Optimization methods, such as genetic algorithms (Goldberg, 1989), have proven useful for 61 optimizing design and management of irrigation systems for different purposes (economical 62 and environmental, among others). Genetic algorithms (GAs) have been used in the past 63 decade for irrigation project planning (Kuo et al. 2000), off-farm irrigation scheduling 64 (Nixon et al. 2001), flow and water quality predictions in watersheds (Preis and Ostfeld 65 2008) and for optimizing the cost of localized irrigation projects (Pais et al. 2010). 66 Navabian et al. (2010) presented a 1D model for optimizing fertigation in conventional 67 5 furrow irrigation. These authors found that optimization results depended on soil status 68 (bare vs. cropped). This approach could be effectively extended to the optimum design and 69 management of fertigation systems. 70 Alternate furrow fertigation has proved to have high potential to reduce water and fertilizer 71 losses. Simulating and optimizing the design and management of furrow fertigation will 72 therefore contribute to minimize the environmental pressure of agricultural irrigation on 73 water resources. Thus, the main objective of this study was to develop a 1D surface and 2D 74 subsurface simulation-optimization model for furrow fertigation. The model was applied to 75 two types of alternate furrow irrigation: a) variable alternate furrow irrigation (AFI); and b) 76 fixed alternate furrow irrigation (FFI), as well as for conventional furrow irrigation (CFI). 77 In all cases, the goal was to minimize nitrate losses. Optimization results were compared 78 with experimental results. 79 80 111 and U x is overland flow velocity at location x [L T -1 ]. 112 Model solutions permit to obtain the distribution along the furrow of infiltrated water and 113 nitrate. These values can be used to determine CU w and CU n , the Christiansen Uniformity 114 Coefficients for water and nitrate, respectively. 115 Model input data include furrow geometry, infiltration, roughness, discharge, and solute 116 properties. The upstream boundary condition is the irrigation discharge for water and the 117 applied nitrate concentration for fertilizer. The downstream boundary condition usually 118 includes uniform runoff flow or blocked-end runoff for water, and zero concentration 119
doi:10.1061/(asce)ir.1943-4774.0000635 fatcat:mjamm4xohvexhe57nnbuuztgkm