Journal of Irrigation and Drainage Engineering

Precision irrigation is crucial in decreasing water demands of irrigated agriculture, which account for the largest source of water withdrawal in most regions. One necessity of precision irrigation is the knowledge of soil moisture content in the root zone of irrigated crops. Dielectric soil moisture sensors can provide this information at a relatively low cost ($100) compared with other techniques; however, their accuracy cannot be guaranteed in all soils without site-specific calibration, the need for which has limited the use of these sensors. This study examined the Decagon 10HS soil moisture sensor in order to determine the accuracy of the manufacturer-supplied calibration equation on southwest Florida (SWFL) agricultural soils, and analyze whether it performs consistently in different soils found in SWFL. Laboratory calibration of the 10HS was done on four SWFL soils that represent the majority of agricultural land in the region and regression calibration equations were obtained. It was found that the manufacturer calibration equation does not accurately predict soil volumetric water content (VWC) within the $\pm 3\%$ accuracy range specified for the sensor, meriting the need for site-specific calibration. Additionally, it was found that all tested soils behave very similarly, allowing the development of a single calibration equation for use of the 10HS in all SWFL agricultural applications. The calibration equation developed during this research for SWFL could allow a more widespread use of the 10HS in the region.

A two-dimensional coupled model is proposed for the diffusion and channeling of both surface water and solute in all possible directions in basin fertigation. In the spatial discretization of the governing equation based on the unstructured triangle grid, the physical variables at grid interfaces are reconstructed by means of the grid center values. Then, the scalar-dissipation finite-volume method is used for the spatial discretization of the advection flux gradient vector. Meanwhile, the zero-dissipation finite-volume method is used to spatially discretize the water level gradient vector, diffusion vector, roughness vector, and infiltration vector. For the temporal scheme, the splitting method is implemented for the spatially discretized governing equation. The two-dimensional coupled model for surface water flow and solute transport in basin fertigation based on the scalar finite-volume method is proposed. The proposed model was validated based on observed data of three field experiments. Results show that the average relative errors between the simulated and observed data for the surface water advance and recession phases are from 3.8 to 4.7% and 10.4 to 12.7%, respectively. The water quantity conservation error is from 0.2 to 0.6%. The average relative error for the solute transport process between the simulated and observed data is less than 15%, and the solute quantity-conservation error is from 0.072 to 0.085%. At the same time, the convergence rate of simulation results is close to two-order. Therefore, the proposed model presents well-simulated performance.

Drip irrigation may need to be installed in situations in which there is not enough energy to operate the emitters under optimum performance conditions. This paper characterizes the head-discharge functional form of a pressure-compensated (PC) and non-pressure-compensating (NPC) drip emitter operating under low pressures. It also analyzes the statistical relationship between manufacturing and hydraulic variations of low-pressure operating emitters assuming that both are independent random variables. A hydraulic simulation methodology was implemented to predict the effect of emitter manufacturing and hydraulic variations on irrigation uniformity while working under a range of operating pressures. The manufacturing variation of PC and NPC emitters was evaluated with the Monte Carlo method for a single horizontal drip line with different pressure heads and different numbers of emitters evenly spaced. The coefficient of manufacturing variation (${\mathrm{CV}}_k$) observed in the laboratory was 9.2% for PC (Toro) emitters and 3.8% for NPC ($T$-tape) emitters. Laboratory test results showed that manufacturing and hydraulic factors were stochastic independent variables, and this independent relationship allowed the combination of statistics for estimating the coefficient of total variation. The manufacturing emitter variation has a considerable impact on low-quarter distribution uniformity (${\mathrm{DU}}_{\mathrm{l}\mathrm{q}}$), as the number of emitters in the drip line is increased. For instance, the ${\mathrm{DU}}_{\mathrm{l}\mathrm{q}}$ of the PC (Toro) emitter decreased considerably from 1.0 to 0.82 compared with the NPC (T-tape) emitter, which varied from 0.943 to 0.939 when the length was increased from 60 to 200 emitters long drip line. The PC emitter started compensating at 6.16 m of hydraulic head. Therefore, this type of emitter is not recommended for low operating pressure systems. It is recommended to design low-pressure irrigation systems with the right emitter type and operating pressure to maximize performance.

The ability to estimate the components of the hydrologic cycle is the key to the proper design and implementation of storm-water control measures (SCMs). Reuse, infiltration, and evapotranspiration (ET) are used by SCMs to reduce runoff. This study quantifies the ET component of a green-roof SCM water budget using weighing lysimeters for 3 years demonstrating its significance. Each of the years (2009–2011) was different climatologically. The measured ET was compared with potential ET calculated by several methods. Further, the measured ET is analyzed for its dependency on climatological factors and previous rain day. The average daily ET was observed to range from 1 to $10\mathrm{mm}/\mathrm{day}$ depending on season, temperature, relative humidity, solar radiation, and antecedent moisture condition.

Continuous simulation of storm-water control measures (SCM) requires consideration of the variable environmental and site factors that affect infiltration. This article reviews the theory of the Green and Ampt infiltration model and presents a unit process approach to how it can be applied to a bioinfiltration SCM. The paper discusses how to vary the soil moisture parameters using the soil water characteristic curve, how to modify the Green and Ampt equation for different basin cross-sections, and how to model or bound the infiltration rate when soil media layers are added. Both the standard Green and Ampt parameters and alternative moisture conditions were compared with data from a bioinfiltration SCM; these results confirm that the standard parameter tables often used to select Green and Ampt parameters represent conservative, wet conditions, and that a single value for the parameters will not be able to reproduce the range of infiltration rates observed in the field. When using variable soil moisture conditions, the Green and Ampt model adequately reproduces the infiltration rates observed at the bioinfiltration site.

The loss of topsoil in the United States has resulted in low crop yield, reduction of reservoir capacity, cost increase of water treatment, and detrimental effects on aquatic life and wildlife habitats. An initial step for taking conservation measures in any watershed is to identify locations where erosion protection measures are needed. Applying the Revised Universal Soil Loss Equation (RUSLE) and a geographic information system (GIS), this study attempts to locate the most erodible locations, namely erosion hotspots, for a watershed. Using GIS, the watershed is divided into $25\times 25\text{-}\mathrm{m}$ grids and the RUSLE, including rainfall-runoff erosivity factor, soil erodibility factor, combined slope length and slope steepness factor, cover management factor, and support practice factor, is applied for the estimation of soil erosion potential for each grid cell. By ranking these grid values of erosion potential in a descending order, the top 1% and corresponding locations are defined as the erosion hot spots, which can be expressed in an erosion hotspot map. Applying this method to the Charles Mill Lake Watershed in Ohio, it is found that the erosion hotspots for the watershed under investigation are generally located in the areas of strip mine and cropland/pasture.

Currently, time to drain and depth of flow are the two basic concepts for designing the drainage layer of highway, both of which are based on the saturated flow and cannot reasonably evaluate the drainage efficiency of drainage layer. A recent study demonstrated that capillary effects could become significant with unsaturated flow playing an important role in conducting water out of the drainage layer in some cases. In view of the disadvantages of the two methods, it is reasonable to combine them to guide the design of a drainage layer. Here, a model was established based on the one-dimensional (1D) transient Boussinesq equation with a capillary correction. The general solution to the governing equation was obtained using the method of separation variables. Furthermore, the presented model was validated against numerical simulations predicted by a finite-difference method (the MATLAB program) and the finite-element method (SUTRA program). A comparison was conducted to demonstrate the difference between the saturated drainage and unsaturated drainage processes. The results showed that the time to drain the same water volume based on the saturated model is much less than that based on the unsaturated model. The present model with capillary correction performs better than the saturated model in predicting time to drain in the drainage layer.

The canal delivery system in the Left Hemidelta area of the Ebro River in Spain consists of a tree-shaped net of open canals. The overall system can be quickly isolated in the case of an emergency by closing the upstream pool. Transients, in which the initial state is hydraulically far from the final state, are difficult to handle and cannot be made in only one gate movement in order to protect the canal lining. Therefore, they have to be as smooth as possible. GoRoSo is a feedforward control algorithm for irrigation canals based on sequential quadratic programming. With this tool, it is possible to calculate the gate trajectories that smoothly carry the canal from the initial state to the final state by keeping the water depth constant at checkpoints. The paper shows the efficient implementation of GoRoSo in both the closure and opening operations of the canal delivery system.

Due to water scarcity, providing reliable amounts of water is not possible for most of the irrigation districts in arid regions. In such conditions, equitable water delivery becomes one of the main concerns. Applying a centralized controller to achieve the high level of equity in water delivery along the main canal is not a practical solution in many irrigation districts due to high costs of the implementation, maintenance and inspections; and vandalism of equipment. In this paper, a water level difference control strategy is applied to keep the water level errors equal in adjacent reaches by using decentralized controllers. Local downstream and distant downstream configurations of the control method are designed for the Dez main canal, located in the southwest of Iran, consisting of 13 canal reaches. The designed controllers are tested on two extreme water off-taking scenarios for normal operation and operation under water shortage. The tests show the capability of the difference control strategy to deal with the mismatches between supply and demands under water shortage conditions. Also, acceptable operational performances in operational conditions with drastic increasing or decreasing of water off-taking in the system are achieved. As no telemetry system is required for the decentralized local downstream configuration, this difference control method is proposed as best option for implementation in the Dez canal.