Optimal Reservoir Operations for Irrigation Using a Three Spatial Scales Approach

A three-step computational model for the optimal weekly interseasonal operation of a multipurpose (irrigation, environmental, domestic/industrial) reservoir is developed. Environmental and domestic/industrial uses are evaluated and considered as priority uses that induce deficit irrigation conditions. The spatiotemporal variability of the irrigation water demand at the basin level is accounted for. The objective is the maximization of the interseasonal agricultural profitability at the basin level. The optimal allocation process solves the competition for water on different temporal scales (weekly, seasonal, and interseasonal) and on different spatial scales (in basins among irrigation areas and in irrigation areas among crops). The three steps are simulation model operating at the soil-crop unit level, optimization model operating at the multicrop area level, and optimization model operating at the basin level. This consists of parametric dynamic programming for which an analytical objective function was defined and an analytical solution was determined. This solution replaces the iterative procedure, so that it is possible to account for all the variables without running into the “curse of dimensionality” problem. The environmental use allocation is expressed as a function of a parameter, the variations of which give different environmental protection levels. The validation case study emphasizes the importance of considering the spatiotemporal variability of the demand. This is consistent with the “computationally tractable” model algorithm.     

Multiobjective Differential Evolution with Application to Reservoir System Optimization

Many water resources systems are characterized by multiple objectives. For multiobjective optimization, typically there can be no single optimal solution which can simultaneously satisfy all the goals, but rather a set of technologically efficient noninferior or Pareto optimal solutions exists. Generating those Pareto optimal solutions is a challenging task and often difficulties arise in using the conventional methods. In the optimization of reservoir systems, most of the times there is interdependence among one or more decision variables. Recently, it is emphasized that the evolutionary operators used in differential evolution algorithms are very much suitable for problems having interdependence among the decision variables. This paper utilizes this aspect and presents an efficient and effective approach for multiobjective optimization, namely multiobjective differential evolution (MODE) algorithm with an application to a case study in reservoir system optimization. The developed MODE algorithm is first tested on a few benchmark test problems and validated with standard performance measures by comparing them with the nondominated sorting genetic algorithm-II. On achieving satisfactory performance for test problems, it is applied to generate Pareto optimal solutions to a multiobjective reservoir operation problem. It is found that MODE provides many alternative Pareto optimal solutions with uniform coverage and convergence to true Pareto optimal fronts. The results obtained show that the proposed MODE can be a viable alternative for generating optimal trade-offs in multiobjective optimization of water resources systems.     

Coupled Reservoir Operation-Irrigation Scheduling by Dynamic Programming

This paper develops a forward dynamic programming (FDP) model to solve the problem of reservoir operation and irrigation scheduling. The typical scenario for application of the model is composed of a system of two reservoirs in parallel supplying water to as many as three irrigation districts. Two models are coupled. The interseasonal model defines seasonal deliveries from the reservoir system. The intraseasonal model uses area and water allocations generated from the interseasonal model to produce an irrigation scheduling for the individual farms in one of the irrigation districts in the reservoir system. Crop evapotranspiration, reservoir evaporation, and inflows are forecasted. Upon availability of the current values, the forecast is updated and the model runs to generate a more precise irrigation schedule. This feature permits the application of the model for real-time operation of the irrigation district. At the end of the season, the intraseasonal model is updated. The FDP model is applied to a real watershed with a 2 year planning horizon for the interseasonal and 6 months for the intraseasonal model.     

Performance-Based Optimization of Land and Water Resources within Irrigation Schemes. I: Method

Optimum land and water allocation to different crops grown in different regions of an irrigation scheme is a complex process, especially when these irrigation schemes are characterized by different soils and environment and by a large network of canals. At the same time if the water supply in the irrigation schemes is limited, there is a need to allocate water both efficiently and equitably. This paper describes the approach to include both productivity (efficiency) and equity in the allocation process and to develop the allocation plans for optimum productivity and/or maximum equity for such irrigation schemes. The approach presented in this paper considers the different dimensions of equity such as water distribution over the season, water distribution during each irrigation, and benefits generated. It also includes distribution and conveyance losses while allocating water equitably to different allocation units. This paper explains the approach with the help of the area and water allocation model which uses the simulation–optimization technique for optimum allocation of land and water resources to different crops grown in different allocation units of the irrigation scheme.     

Irrigation Scheduling. II: Heuristics Approach

A sequential irrigation schedule that honors user demands of duration and minimizes earliness and tardiness is interpreted as a single-machine schedule with earliness and tardiness costs and a common deadline (or fixed interval). A heuristic solution is presented for this irrigation scheduling problem. Four models are presented to reflect the different methods in which an irrigation system at the tertiary unit level may be operated, the first model permits jobs to be noncontiguous, i.e., idle time between jobs is permitted, whereas the others permit contiguous jobs only. The heuristic is tested extensively and the solution quality is compared with either an optimum solution from an integer program or the best available solution obtained from an integer program within allocated computation time. The heuristic is computationally efficient for all models presented, however for schedules with noncontiguous jobs, or where idle time is inserted before and after a contiguous set of jobs, solution quality deteriorates. The work brings the science of single scheduling from operations research into irrigation scheduling and suggests areas for further development.     

Performance-Based Optimization of Land and Water Resources within Irrigation Schemes. II: Application

The area and water allocation model which uses simulation–optimization technique for optimum allocation of land and water resources to different crops cultivated in different allocation units of the irrigation scheme was modified to include both productivity and equity in the process of developing the allocation plans for optimum productivity and/or maximum equity. This paper illustrates the potential of this approach with the help of a case study on Nazare medium irrigation scheme in India. The allocation plans were developed for optimization of different performance parameters (productivity and equity) for different management strategies based on irrigation amount and irrigation interval and cropping distribution strategies of free and fixed cropping. The results indicated that the two performance objectives productivity and equity conflict with each other and in this case, equitable water distribution may be preferred over free water distribution at the cost of a small loss in productivity. Though these results relate to one case study, they show the value of the approach of incorporating productivity and equity in the allocation process with the help of the simulation-optimization model described in the companion paper.     

Estimation of Evapotranspiration of Different-Sized Navel-Orange Tree Orchards Using Energy Balance

Crop evapotranspiration (ETc) and crop coefficient (Kco) values of four clean-cultivated navel-orange orchards that were irrigated with microsprinklers, having different canopy features (e.g., age, height, and canopy cover) were evaluated. Half-hourly values of latent heat flux density were estimated as the residual of the energy balance equation using measured net radiation (Rn), soil heat flux density (G), and sensible heat flux density (H) estimated using the surface renewal method. Hourly means of latent heat flux density (LE) were calculated and were divided by the latent heat of vaporization (L) to obtain ETc. Crop coefficients were determined by calculating the ratio Kco = ETc/ETo, with reference evapotranspiration (ETo) determined using the hourly Penman–Monteith equation for short canopies. The estimated Kco values ranged from 0.45 to 0.93 for canopy covers having between 3.5 and 70% ground shading. The Kco values were compared with Kc values from FAO 24 (reported by Doorenbos and Pruitt in 1975) and FAO 56 (reported by Allen et al. in 1998) and with Kc values from research papers that estimated reference ET from pan evaporation data using the FAO 24 method. The observed Kco values were slightly higher than Kc values for clean-cultivated orchards with high-frequency drip irrigation in Arizona and were slightly lower than for nontilled orchards in Florida. The Kco values were considerably higher than Kc values from FAO 24 and FAO 56 and were higher than Kc values from border-irrigated orchards near Valencia, Spain.     

New Weighing Method to Measure Shoot Water Interception

The need exists to develop a method that can quantitatively measure water interception from plant shoots. This paper describes a new method for measuring canopy water interception. Corn (Zea Mays L.) was grown in 13?L buckets containing Valentine fine sand (Mixed, mesic Typic Ustipsamment) under a climate-controlled growth chamber. Plants were taken out of the growth chamber for 2–3?h periods for measurements of shoot water interception in a hydraulic laboratory equipped with an Accupulse system suspended from the ceiling that was used to wet the corn shoots at growth stages V7–V13. A lightweight, movable framework was placed around a balance, and buckets containing corn plants were placed on the scale one container at a time. Water was applied until all shoot surfaces were wet and runoff from the leaves and stalk surfaces could be observed. The weighing method for shoot water interception was tested by using the balance to instantaneously measure shoot water interception during application of water and after plant surface runoff ceased. The balance, bucket, and soil surface were covered with plastic to protect them from water, so only the shoots were wet. Interception by the shoot of corn ranged from 31?to?47?mL?shoot?1. These values were much smaller than previous values reported in the literature. The average coefficient of variation was 9.2% for two studies, which was much lower than previously accepted methods. This study suggests that the weighing method for shoot water interception can be used to quantitatively and more accurately measure water intercepted by corn shoots. The weighing intercepted method values presented in this paper are low and suggest that previous interception methods overestimated the interception values.     

Evaluation of a Genetic Algorithm for the Irrigation Scheduling Problem

A typical irrigation scheduling problem is one of preparing a schedule to service a group of outlets which may be serviced simultaneously. This problem has an analogy with the classical earliness/tardiness problem in operations research. In previously published work an integer program was used to solve this problem, however such scheduling problems belong to a class of combinatorial problems known to be computationally demanding (N-P hard). This is widely reported in operations research. Hence integer programs can only be used to solve relatively small problems usually in a research environment where considerable computational resources and time can be allocated to solve a single schedule. For practical applications metaheuristics such as genetic algorithms, simulated annealing, or tabu search methods need to be used. However as reported in the literature, these need to be formulated carefully and tested thoroughly. This paper demonstrates the importance of robust testing of one such genetic algorithm formulated to solve the irrigation scheduling problem with simultaneous outlets serviced against an integer program formulated to solve the same problem.     

Effect of Preharvest Deficit Irrigation on Second Crop Watermelon Grown in an Extremely Hot Climate

As a second crop, watermelons (Citrullus vulgaris, Crimson sweet) were grown in 2003 and 2004 in the Sanliurfa-Harran Plain in southeastern Turkey to determine the effect of preharvest water stress on fruit yield, quality (i.e., soluble solids contents and fruit size), leaf temperature, and some other physiological parameters. Preharvest drip irrigation treatments included (1) complete irrigation cutoff, dry (D); (2) full irrigation based on replenishment of soil water depleted from 0 to 90?cm soil profile (C); (3) 75% full irrigation (IR1); (4) 50% full irrigation (IR2); and (5) 25% full irrigation (IR3) with 3-day irrigation interval. Treatment plots received the same level of irrigation water until the fruit formation stage, except for Treatment D. Then, different water stress levels were imposed on treatment plots. Irrigation water applied to the five respective treatments were 636, 511, 395, 245, and 120?mm in 2003 and 648, 516, 403, 252, and 127?mm in 2004. Results indicated that fruit yield was significantly lowered by reduced water rates. The seasonal average yield response factor (ky) for both years was 1.0, but it was 0.97 for 2003 and 0.98 for 2004. The highest marketable fruit yield, obtained from treatment C, was 32.4?Mg?ha?1 in 2003 and 37.1?Mg?ha?1 in 2004. D, IR2, and IR3 treatments reduced most measured parameters, except for soluble solids contents (SSC). Both the fruit size and SSC were significantly affected by late-season irrigation management; individual fruit weights were significantly reduced, whereas SSC increased in the IR2 and IR3 treatments compared to the control values. The writers’ results clearly indicated that reduced preharvest irrigation was detrimental. Water use efficiency (WUE) was significantly affected by irrigation treatments. Even a 25% reduction in the irrigation amount caused a 15% reduction in marketable yield. This indicates that deficit irrigation in the ripening stage significantly increased water use efficiency. The study demonstrated that a moderate deficit irrigation, which is replenishment up to 50% of soil water depleted in the root zone, can be successfully used to improve WUE under semiarid climatic conditions.     

Modeling Reservoir Irrigation in Uncertain Hydrologic Environment

In a detailed model for reservoir irrigation taking into account the soil moisture dynamics in the root zone of the crops, the data set for reservoir inflow and rainfall in the command will usually be of sufficient length to enable their variations to be described by probability distributions. However, the potential evapotranspiration of the crop itself depends on the characteristics of the crop and the reference evaporation, the quantification of both being associated with a high degree of uncertainty. The main purpose of this paper is to propose a mathematical programming model to determine the annual relative yield of crops and to determine its reliability, for a single reservoir meant for irrigation of multiple crops, incorporating variations in inflow, rainfall in the command area, and crop consumptive use. The inflow to the reservoir and rainfall in the reservoir command area are treated as random variables, whereas potential evapotranspiration is modeled as a fuzzy set. The model’s application is illustrated with reference to an existing single-reservoir system in Southern India.     

Irrigation Scheduling with Genetic Algorithms

A typical irrigation scheduling problem is one of preparing a schedule to service a group of outlets that may be serviced simultaneously. This problem has an analogy with the classical multimachine earliness/tardiness scheduling problem in operations research (OR). In previously published work, integer programming was used to solve irrigation scheduling problems; however, such scheduling problems belong to a class of combinatorial optimization problems known to be computationally demanding. This is widely reported in OR literature. Hence integer programs (IPs) can be used only to solve relatively small problems typically in a research environment where considerable computational resources and time can be allocated to solve a single schedule. For practical applications, metaheuristics such as genetic algorithms, simulated annealing, or tabu search methods need to be used. However, these need to be formulated carefully and tested thoroughly. The current research explores the potential of genetic algorithms to solve the simultaneous irrigation scheduling problem. For this purpose, two models are presented: the stream tube model and the time block model. These are formulated as genetic algorithms, which are then tested extensively, and the solution quality is compared with solutions from an IP. The suitability of these models for the simultaneous irrigation scheduling problem is reported.     

Evaluation of Water Retention Functions and Computer Program “Rosetta” in Predicting Soil Water Characteristics of Seasonally Impounded Shrink–Swell Soils

Soil water retention is a critical factor influencing irrigation decisions and hence agricultural crop yields. However, information on soil water retention characteristics (SWRC) is seldom available for irrigation planning, crop yield modeling, or hydrological simulations, especially for problematic soils, such as seasonally impounded shrink-swell soils. As large scale direct measurement of SWRC is not viable due to a number of reasons, researchers have developed pedotransfer functions (PTFs) to estimate SWRC from easily measured soil properties, such as texture, organic matter content, bulk density, etc. However, PTF applicability in locations other than those of data collection has been rarely reported. One of the most recent PTFs that has shown overall reasonable predictions in evaluation studies is Rosetta, a numerical code for estimating soil hydraulic parameters with hierarchical pedotransfer functions. Relatively, the development of large databases makes it one of the widely used PTFs. If validated for spatial application, it has immense use potential in countries like India, where data on soil hydraulic properties are seldom available, a deficiency that hampers better simulations in processes, like partitioning runoff and infiltration, assessing evapotranspiration, irrigation scheduling, etc. Rosetta is also relatively flexible allowing estimation of hydraulic properties from easily available minimum input of textural fractions. This study was conducted to evaluate (1) an applicability of four widely used soil water retention functions to describe SWRC; and (2) the computer program Rosetta for its validity. Statistical indices, i.e., root mean square error (RMSE), mean absolute error, maximum absolute error, and degree of agreement (d) were computed to evaluate “goodness-of-fit” of the four functions to the measured SWRC data. These indices were also used to compare measured SWRC with estimates of SWRC by Rosetta. For soil samples collected from 41 profiles, 175 SWRC were measured in the laboratory. The van Genuchten function fitted relatively better (RMSE = 0.052?m3?m?3) to SWRC of clay soils, whereas the Brooks–Corey (BC) function was better in expressing SWRC of clay loam and sandy clay loam soils with RMSE = 0.06 and 0.07?m3?m?3, respectively. Campbell and Cass–Hutson (CH) functions were of intermediate value. Worst performing functions were BC (clay soils), Campbell (clay loam), and CH (sandy clay loam) with corresponding RMSE = 0.059, 0.065, and 0.077?m3?m?3. Estimates of two important points on the SWRC curve, i.e., field capacity and permanent wilting point were predicted with relatively better accuracy for clay and sandy clay loam soils by all the four functions. RMSE and d ranged from 0.027?to?0.043?m3?m?3 and from 0.73 to 0.88 for clay soils. Corresponding values for sandy clay loam soils were 0.008?–0.019?m3?m?3, and 0.92–0.98. However, in clay loam soils, only two functions were found suitable. Estimates of SWRC obtained by applying hierarchical rules in Rosetta were reliable (RMSE<0.05?m3?m?3). Magnitude of average RMSE increased progressively in clay loam, clay and sandy clay loam soils (0.028<0.035<0.042?m3?m?3). The study established that SWRC of the “Haveli” soils could be estimated using generic PTF and thus information that is prerequisite in simulating hydrological processes occurring in seasonally impounded soils could be acquired.     

Assessing Reference Evapotranspiration by the Hargreaves Method in Southern Spain

The Hargreaves method enables reference crop evapotranspiration (ET0) estimation in areas where meteorological information is scarce, as, for example, southern Spain. However, this method is known to produce considerable bias in this region, especially during the dry, hot summer months. An evaluation of the method is made by comparing daily estimates with those made by the more commonly recommended Penman–Monteith method at 16 meteorological stations. Computed ET0 values at the coastal stations are, on average, 0.69 mm?d?1 smaller than the Penman–Monteith estimates whereas at inland stations a small average overestimation of 0.13 mm?d?1 is shown. The adjusted Hargreaves coefficient (AHC), obtained through regression analysis, increases at the coastal stations, on average, to 0.0029, and decreases at the inland stations to 0.0022. Adjustment with the Samani method does generally not produce more accurate estimates in this region. Finally a linear relationship between the AHC and the rate of the average temperature to the average daily temperature range is proposed for the regional adjustment of the Hargreaves coefficient.     

Return Mapping Procedure for Frictional Force Calculation: Some Insights

The return mapping procedure is studied to gain insights for its numerical implementation to calculate the frictional force during contact analysis. A simple quasistatic truss–wall frictional contact analysis problem is used in the study. The problem has closed-form solutions which provide exact target solutions for a numerical algorithm. The penalty method and a true augmented Lagrangian method that automatically determines an accurate value of the penalty parameter are employed in the numerical study. It is determined that the return mapping procedure is not applicable unless the contacting node is constrained to stay at the initial contact point, and the total normal reaction force, the tangential reaction force and the friction limit have been determined. If these requirements are not met, inaccurate or even incorrect solutions are obtained. This characteristic of the procedure is studied by solving slip and stick cases with several different load increments. It is concluded that the return mapping procedure for friction force calculation should be implemented carefully to obtain accurate solutions for contact analysis problems.     

Automation of a Center Pivot Using the Temperature-Time-Threshold Method of Irrigation Scheduling

A center pivot was completely automated using the temperature-time-threshold method of irrigation scheduling. An array of infrared thermometers was mounted on the center pivot, and these were used to remotely determine the crop leaf temperature as an indicator of crop water stress. We describe methods used to automatically collect and analyze the canopy temperature data and control the moving irrigation system based on the data analysis. Automatic irrigation treatments were compared with manually scheduled irrigation treatments under the same center pivot during the growing seasons of 2004 and 2005. Manual irrigations were scheduled on a weekly basis using the neutron probe to determine the profile water content and the amount of water needed to replenish the profile to field capacity. In both years, there was no significant difference between manual and automatic treatments in soybean water use efficiency or irrigation water use efficiency. The automatic irrigation system has the potential to simplify management, while maintaining the yields of intensely managed irrigation.     

Sensitivity of the Food and Agriculture Organization Penman–Monteith Evapotranspiration Estimates to Alternative Procedures for Estimation of Parameters

Reference crop evapotranspiration (ETo) is a key variable in procedures established for estimating evapotranspiration rates of agricultural crops. As per internationally accepted procedures outlined in the United Nations Food and Agriculture Organization's Irrigation and Drainage Paper No. 56 (FAO-56), using the Penman–Monteith (PM) combination equation is the recommended approach to computing ETo from ground-based climatological observations. Applying of the PM equation requires converting input climate and site data into a number of parameters, and FAO-56 recommends exact procedures for estimating these parameters. However, a plethora of alternative procedures for estimating parameters exist in literature. As a consequence, it is likely that ambiguous results may be obtained from the FAO-56 PM equation because of the adoption of such alternative (nonrecommended) supporting equations. The purpose of the present study is to evaluate differences that could arise in FAO-56 ETo estimates if nonrecommended equations are used to compute the parameters. Using historical climate records from 1973 to 1992 of a station located in the humid tropical region of Karnataka State, India, monthly ETo estimates computed by FAO-56 recommended procedures were statistically compared with those obtained by introducing alternative procedures for estimating parameters. In all, 13 alternative algorithms for ETo estimation were formulated, involving modified procedures for parameters associated with weighting factors, net radiation, and vapor-pressure-deficit terms of the PM equation. For the 240-month period considered, nine of these algorithms yielded ETo estimates that were in close correspondence with FAO-56 estimates as indicated by mean absolute relative difference (AMEAN) values within 1% and maximum absolute relative difference (MAXE) values within 2%. The remaining four algorithms, involving nonrecommended procedures for the vapor-pressure-deficit and net-radiation parameters, yielded considerably different ETo estimates, giving rise to AMEAN values in the range of 2 to 8% and MAXE values ranging between 8 and 28%. The results of this study highlight the need for strict adherence to recommended procedures, especially for estimating of vapor-pressure-deficit and net-radiation parameters if consistent results are to be obtained by the FAO-56 approach.     

OCCASION: New Planning Tool for Optimal Climate Change Adaption Strategies in Irrigation

To sustain productive irrigated agriculture with limited water resources requires a high water use efficiency. This can be achieved by the precise scheduling of deficit irrigation systems taking into account the crops’ response to water stress at different stages of plant growth. Particularly in the light of climate change with rising population numbers and increasing water scarcity, an optimal solution for this task is of paramount importance. We solve the corresponding complex multidimensional and nonlinear optimization problem, i.e., finding the ideal schedule for maximum crop yield with a given water volume by a well tailored approach which offers straightforward application facilities. A global optimization technique allows, together with physically based modeling, for the risk assessment in yield reduction considering different sources of uncertainty (e.g., climate, soil conditions, and management). A new stochastic framework for decision support is developed which aims at optimal climate change adaption strategies in irrigation. It consists of: (1) a weather generator for simulating regional impacts of climate change; (2) a tailor-made evolutionary optimization algorithm for optimal irrigation scheduling with limited water supply; and (3) mechanistic models for rigorously simulating water transport and crop growth. The result, namely, stochastic crop-water production functions, allows to assess the impact of climate variability on potential yield and thus provides a valuable tool for estimating minimum water demands for irrigation in water resources planning and management, assisting furthermore in generating maps of yield uncertainty for specific crops and specific agricultural areas. The tool is successfully applied at an experimental site in southern France. The impacts of predicted climate variability on maize are discussed.     

Object-Oriented Framework for Genetic Algorithms with Application to Space Truss Optimization

Genetic algorithms have been shown to be very effective optimization tools for a number of engineering problems. Since the genetic processes typically operate independent of the actual problem, a core genetic algorithm library consisting of all the genetic operators having an interface to a generic objective function can serve as a very useful tool for learning as well as for solving a number of practical optimization problems. This paper discusses the object-oriented design and implementation of such a core library. Object-oriented design, apart from giving a more natural representation of information, also facilitates better memory management and code reusability. Next, it is shown how classes derived from the implemented libraries can be used for the practical size optimization of large space trusses, where several constructibility aspects have been incorporated to simulate real-world design constraints. Strategies are discussed to model the chromosome and to code genetic operators to handle such constraints. Strategies are also suggested for member grouping for reducing the problem size and implementing move-limit concepts for reducing the search space adaptively in a phased manner. The implemented libraries are tested on a number of large previously fabricated space trusses, and the results are compared with previously reported values. It is concluded that genetic algorithms implemented using efficient and flexible data structures can serve as a very useful tool in engineering design and optimization.     

SDI Dripline Spacing Effect on Corn and Soybean Yield in a Piedmont Clay Soil

A subsurface drip irrigation (SDI) system was installed in the Piedmont of North Carolina in a clay soil in the fall of 2001 to test the effect of dripline spacing on corn and soybean yield. The system was zoned into three sections; each section was cropped to either corn (Zea mays L.), full-season soybean [Glycine max (L.) Merr.], or winter wheat (Triticum aestivum) double cropped to soybean representing any year of a typical crop rotation in the region. Each section had four plots; two SDI plots with dripline spacing at either 1.52 or 2.28 m, an overhead sprinkler irrigated plot, and an unirrigated plot. There was no difference in average corn grain yield for 2002–2005 between dripline spacings or between either dripline spacing and sprinkler. Irrigation water use efficiency (IWUE) was greater for sprinkler irrigated corn than for either SDI treatment and there was no difference in IWUE in soybean. Water typically moved laterally from the driplines 0.38 to 0.50 m. SDI yield and IWUE increased relative to sprinkler yields and water use efficiency in the second and third year of the study. This may suggest that initial fracturing of the heavy clay soil during SDI system installation and subsequent settling of the soil affected water distribution.