Journal of Irrigation and Drainage Engineering

The crop coefficient ($K_c$) values of rice paddy are important for estimating accurate rice crop evapotranspiration ($E{T}_c$), water transfers planning, efficient irrigation management, and hydrological studies. In this study, $E{T}_c$ was measured and a generalized $K_c$ curve was calculated for paddy rice in the Sacramento Valley, California. Field experiments were conducted in three rice paddy fields during the 2011–2013 growing seasons. Surface renewal analysis, after calibration using eddy covariance method, was applied to obtain sensible heat flux values from high-frequency temperature readings; latent heat flux densities were characterized by the residual of the energy balance method. The results revealed that there is considerable variability in rice water use both spatially and temporally. The average 3-year measured seasonal $E{T}_c$ of the experimental fields ranged from 690 to 762 mm in Butte County and from 681 to 813 mm in the Colusa County. A mean daily seasonal $E{T}_c$ of $5.3\mathrm{mm}{\mathrm{d}}^{-1}$ and midseason $E{T}_c$ of $5.8\mathrm{mm}{\mathrm{d}}^{-1}$ was observed. The rice $K_c$ values were lower than those commonly used to estimate rice $E{T}_c$ during the midseason and were greater than expected during early growth before canopy closure. For a typical growing season of 145 days, the $K_c$ values were estimated as 1.10, 1.00, and 0.80 for the initial-growth, midseason, and late-season stages, respectively. The generalized $K_c$ curve is quite accurate for practical application and enables growers to determine rice crop water use in a reliable, usable, and affordable format. The proposed $K_c$ information can be used to refine the estimates of rice consumptive water use in the Sacramento Valley for the purpose of water transfers to water-short areas of the State. These $K_c$ values are likely applicable to other locations having similar climate to the Sacramento Valley in California.

The HadCM3 climate model is applied with greenhouse gas emission scenario A2 to predict climatic variables in the Aidoghmoush Basin (East Azerbaijan, Iran) over the period 2026–2039. The impacts of the predicted climate variables on reservoir inflow and irrigation water use are calculated with the IHACRES model and the Food and Agricultural Organization’s (FAO) evapotranspiration method, respectively. The presented results show that reservoir inflow and irrigation water use in the prediction period relative to the baseline period (1987–2000) would be decreased by 0.7% and increased by 6%, respectively. Therefore, the Aidoghmoush Basin is likely to face increasing water shortage under a changing climate. As a result, irrigation water use should be reduced by applying the following three adaptive strategies to climate change: delaying the planting date, applying irrigation deficit stress, and increasing the irrigation efficiency. Simulations are performed to evaluate the yield of the agricultural crops arising from the application of the three adaptive strategies. Optimal reservoir operating rules are calculated considering the adaptive strategies and the dual objectives of minimizing vulnerability and maximizing reliability of water supply by means of multiobjective genetic programming (MO-GP). The strategy of applying water stress would improve the vulnerability and reliability of the irrigation water supply. The strategy of increasing the irrigation efficiency would maintain crop yields in the study area under climatic change.

This paper develops an optimal, multistage operation of gated spillways for the Karkheh Reservoir, Iran. In each stage of the proposed method, the opening of the gates is proportional to the water level of the reservoir. Two novelties are introduced in this work. The first one is consideration of the absence of a spillway or the existence of blocked spillway gates as two operation scenarios. The second novelty is attributed to using improvement of dam safety as an objective function rather than a constraint of the optimization problem. A genetic algorithm (GA) was implemented for determining the optimal opening of gates to minimize downstream damages. The nondominated sorting genetic algorithm-II (NSGA-II) was applied to optimize the two objectives of minimizing downstream damages and reducing the probability of dam overtopping. This paper’s results reveal that increasing the number of stages of gate opening improved the value of the objective functions. On the other hand, the lack of spillways or blocked spillways increases the risk of dam overtopping for long return periods.

A 3-year study was conducted in the Catawba-Wateree River Basin during the spring and summer months of 2009–2011 to evaluate the water savings potential of three smart irrigation controllers compared with standard timer-based irrigation controllers. Thirty-six residential sites, 12 on each of three Duke Energy lakes (Lake Norman, Lake Hickory, and Lake Wylie), were selected for the study. Each site had an automated, underground irrigation system that directly drew water from the Duke Energy lake that the property bordered. None of the systems had previously been metered and there was no charge to homeowners for withdrawing water from the bordering lake. All sites were instrumented with water meters in 2009 to monitor irrigation water withdrawals, and irrigation system audits were performed before installation of smart technologies. After collecting baseline water-use data in 2009, existing controllers were replaced with one of three smart controllers at 27 of the study sites in 2010, and irrigation was monitored through the end of the 2011 irrigation season. Irrigation treatments included a standard irrigation controller with an add-on soil moisture sensor system (SMS), an evapotranspiration (ET) based controller that received daily reference evapotranspiration estimates from a third-party provider (ET1), an evapotranspiration-based controller with an on-site weather sensor (ET2), and a comparison group that used the existing irrigation controller with no intervention (COMP). Weekly water withdrawals were monitored with on-site data loggers, and turf quality was visually rated and characterized with a normalized difference vegetation index (NDVI) meter. The SMS treatment applied the most irrigation per week (mean of $27.4\mathrm{mm}{\mathrm{week}}^{-1}$), compared with $24.7\mathrm{mm}{\mathrm{week}}^{-1}$ for the ET1 treatment, and the COMP and ET2 treatments that applied 20.9 and $19.9\mathrm{mm}{\mathrm{week}}^{-1}$, respectively. Although applying the most irrigation, the SMS treatment had the greatest effect on water-use behavior for sites receiving technologies, reducing irrigation by $11.7\mathrm{mm}{\mathrm{week}}^{-1}$ compared with the baseline year in which the SMS group applied the most water. There was no evidence of a change in water-use behavior in the COMP treatment, but collectively the smart treatments reduced irrigation from 60% more than the gross irrigation requirement (GIR) in 2009 to only 10% more than GIR in 2010 and 2011, without adversely affecting the turf quality. Visual turf quality ratings in the COMP treatment were slightly less than the minimally acceptable level, which may have resulted from underirrigation during high water demand periods. Several challenges to retrofitting existing systems and issues with smart controllers were observed, including poorly designed and maintained existing irrigation systems, incompatibility between existing controllers and SMS sensor modules, failure of all nine soil-moisture sensors, communication failures in ET-based controllers, and manual overrides of smart technologies by study participants. This study emphasizes the importance of proper installation, programming, and maintenance of smart technologies on suitable irrigation systems if they are to function correctly. To maximize water savings, smart irrigation retrofits should be targeted toward systems that historically overirrigate.

To deal with the multiple uncertainties caused by natural conditions and human activities of the Shiyang River basin in arid northwest China, a simulation-based inexact fuzzy semi-infinite programming method is developed for agricultural cultivated area management. First, the variation tendencies of four key elements (including annual precipitation, temperature, total runoff of the six tributaries at mountain outlets, and the inflow into the Caiqi station) have been investigated to improve the precision of the stream-flow simulation. Secondly, considering that the inflow into the Caiqi station is comprehensively influenced by both natural variation and human activities, the runoff series of the Caiqi station from 1955 to 2013 can be divided into three periods (the first period: 1955–1974, the second period: 1975–2002, and the third period: 2003–2013) through comparing the observed values and the simulated results of multiple linear regression. Thirdly, based on the results of runoff simulation, the inexact fuzzy semi-infinite programming model is established under multiple uncertainties to generate optimal agricultural cultivated area plans. Optimal results indicate that economic benefits show an upward trend, with nearly 10–20% benefits in Period 3 increase compared with that in Period 2. The ratio of the areas planted with grain crops and the economic crops is approximately $7:3$ in Period 2 and close to $4:6$ in Period 3. With impacts of human activities varying from negativity to positivity, available water resources of Minqin County are gradually recovering and increasing, so that the agricultural areas are adjusted correspondingly by the authorities to make it more suitable for the local sustainable development. The results of both simulation and optimization are valuable for crop land planning considering the effects of nature and human.

In this study, a genetic algorithm optimization model is applied to two different scenarios of deficit irrigation in an attempt to develop an appropriate plan for the conjunctive use of surface and ground water resources in Nekouabad irrigation zone located in Zayandehrud Basin, Iran. The study aims to reduce water consumption, define an optimal cropping pattern, decrease the annual decline in the local water table, and to maximize the farmers’ net financial returns. Scenario I is based on normal conditions in terms of both water resources and weather parameters in the region while Scenario II is implemented to develop a 5-year irrigation plan for the study area. For the 5-year period in Scenario II, all possible weather (i.e., dry, normal, and wet) conditions are considered. Results show that water consumption would reduce by almost 24 million cubic meters and the net financial return would decline by 22% under Scenario I. Scenario II reveals that the decline in net revenues would be less under normal and wet conditions than it would be under drought conditions.

Knowledge of evapotranspiration (${\mathrm{ET}}_o$), which is the process of water loss from vegetated soils due to evaporation and transpiration, is important in real-time irrigation management and water-resource allocation, particularly in water-scarce regions. In this study, several methods used in estimating evapotranspiration, including the Blaney-Criddle, Hargreaves-Samani, Jensen-Haise, Linacre, and Turc methods were calibrated and validated against the Penman-Monteith model, which is considered as the standard method of estimating evapotranspiration. The paper utilizes data from the Doha International Airport meteorological station over a period of 30 years (January 1985–December 2014). ${\mathrm{ET}}_o$ values were estimated using the different methods. These values were then compared to those obtained by the Penman-Monteith method. Using appropriate indicators, the Turc method was found to be the best for estimating ${\mathrm{ET}}_o$ over Doha ($R^2=0.9519$, $\mathrm{RMSE}=1.4511\mathrm{mm}{\mathrm{d}\mathrm{a}\mathrm{y}}^{-1}$, and $\mathrm{MAE}=1.1633\mathrm{mm}{\mathrm{d}\mathrm{a}\mathrm{y}}^{-1}$). The Turc method comes in handy for estimating ${\mathrm{ET}}_o$ over Qatar as it utilizes only three meteorological parameters (mean temperature, relative humidity, and solar radiation), which are easily measurable over that area.

Multimachine scheduling problems with earliness/tardiness costs and sequence-dependent setup times are analogous to the simultaneous irrigation scheduling problem with water travel times between outlets in a canal irrigation system where all the farmers are supplied with identical discharges at their requested time, i.e., arranged demand irrigation scheduling. The multimachine scheduling problem with earliness/tardiness costs even without setup consideration is computationally very demanding and optimum solutions are not possible in practical time limits. The addition of the sequence-dependent setup time and the dual goal of minimizing earliness/tardiness and the number of machines makes it even more difficult, complicated, and novel. For practical applications, meta-heuristics such as genetic algorithms, simulated annealing, or tabu search methods need to be used. This study employs the genetic algorithm (GA) model. The model presented here is an improvement over earlier work as it considers travel time in a multimachine or simultaneous irrigation system and resolves the issue of computational time by using an approximate algorithm instead of an exact algorithm. However, no quantitative comparison can be done with earlier models as the current model accommodates travel time; hence, its objective function is numerically different than earlier models. The problem is successfully modeled using GA and its implementation is demonstrated. No comprehensive data set is available that completes the requirements of rigorous testing of the GA model. Therefore, to evaluate the performance of the GA model with travel time, instances were randomly generated from a uniform distribution, for three different values of travel times. The GA model was able to obtain feasible schedules for all the instances tested.