Journal of Irrigation and Drainage Engineering

Water management of seedling cultivation in nurseries necessitates proper irrigation scheduling, especially in arid and semiarid regions that cope with water shortage. To this end, triggered irrigation scheduling of walnut seedlings in a field nursery was performed using the HYDRUS-1D model and Taguchi optimization approach. Soil moisture content variation and canopy cover (CC) development were monitored in the study field by a profile probe (PR2 sensor) and aerial-view images, respectively. Cross-validation technique was performed in order to calibrate and validate the HYDRUS-1D model. Two sets of combinations comprised of four and five factors with five levels for each were designed to represent 256 and 3,125 different scenarios, respectively, for optimizing triggered irrigation using the Taguchi method. An objective function equal to the sum of irrigation efficiency (IE) and normalized transpiration (NT) was considered to maximize using the Taguchi optimization process. Results of the cross-validation technique showed a reasonable agreement between measured and simulated data [root-mean-square error $(\mathrm{RMSE})=0.025{\mathrm{m}}^3{\mathrm{m}}^{-3}$, mean absolute error $(\mathrm{MAE})=0.018{\mathrm{m}}^3{\mathrm{m}}^{-3}$, and Nash-Sutcliff efficiency $(\mathrm{NSE})=0.79$ in both calibration and validation stages on average]. Running 25 combinations for each set resulted in the objective function values of 1.50 ($\mathrm{IE}=0.52$ and $\mathrm{NT}=0.98$) and 1.80 ($\mathrm{IE}=0.92$ and $\mathrm{NT}=0.88$) for the first and second data sets, respectively. The best scenario led to 74% less water consumption compared to actual irrigation practice. The results showed that the combination of the Taguchi optimization approach and HYDRUS-1D model is successfully applicable and provides a promising method for optimizing the triggered irrigation scheduling of walnut seedlings.

Automatic canal control is essential to improve water utilization efficiency in irrigation districts. Water level error in the model used by the control algorithm significantly impacts performance. The integrator delay model, which has a linear structure, is widely used in multi-input/multioutput control algorithms, including linear-quadratic control and model predictive control. However, the integrator delay model is suboptimal for large irrigation districts; the canals are long, the flows are large, there are many scattered offtakes, and the operating conditions vary. We developed an adaptive predictive control mode for the Qinhan Canal in Ningxia Province, China. We used a segment integrator delay model for this large irrigation district and employed a differential evolution algorithm for online system identification; this minimizes error. We analyzed and controlled for various uncertainties (i.e., flow observation noise, geometric and hydraulic parameter errors, and accidental interference with the automatic control system). Simulations indicated that the water level fitting errors were significantly reduced compared with the original model, and water level regulation was significantly improved, especially in terms of the maximum absolute error. Adaptive predictive control was better than linear-quadratic control and model predictive control approach for water level. Adaptive predictive control coped well with several uncertainties. The main factors affecting water level control were the flow observation and gate opening control accuracy, and the extent of real-time data coverage. These features require careful attention when automating and modernizing irrigation districts.

Quantifying evapotranspiration (ET) and infiltration from vegetated stormwater control measures (SCMs), such as rain gardens, is necessary to physically represent their volume reduction potential. Most states and regulatory entities utilize a design storm for rain garden design in which the rain garden’s capacity is considered using a static volume contribution. The static storage volume approach excludes the dynamic functions of infiltration during an event and ET between events. This work seeks to provide a method to incorporate the function of ET during interevent times into a design storm approach. The suggested method to do this is to estimate the void space recovery due to both ET and gravity drainage. An example is used to demonstrate the method for rain gardens in Pennsylvania where the void space recovery was estimated for 6 and 12 days between events. The void space recovery was estimated using a mathematical model based upon the 1D Richards equation coupled with the ASCE Penman-Monteith model. The mathematical model was validated using data from eight rain garden lysimeters in Villanova, Pennsylvania. This location is in the mid-Atlantic region with a Cfa climate in the Koppen-Geiger classification system. This void space recovery ranged from 15% to 40% (for both ET and gravity drainage) depending on soil type, drainage, rooting depth, crop coefficient and days between events. This void space recovery can be used to calculate a semidynamic recovery based on expected performance using the commonly employed design storm approach.

Abstract

Rainfall–runoff modeling is commonly employed by hydrologic engineers for the purpose of determining a runoff hydrograph. Historically, hydrologic methods utilizing unit hydrograph theory and basin averaged parameters have been used to obtain results. Of the available unit hydrograph methods, the Natural Resources Conservation Services (NRCS) procedure has gained widespread popularity and is commonly used by the professional community. Recent advancements in both computing power and accessibility to high-resolution digital elevation models (DEMs) have made 2D hydraulic rainfall–runoff modeling feasible. Version 6 of the Hydrologic Engineering Center’s River Analysis System (HEC-RAS) allows users to input a precipitation boundary condition used in combination with both curve number and Manning’s n layers for the purpose of modeling the rainfall–runoff relationship over a 2D terrain using hydrodynamic equations. The goal of this study was to simulate runoff hydrographs using both a hydrologic method (NRCS, via HEC-HMS) and a hydraulic method (via HEC-RAS) and then to compare results to measured values recorded at a USGS gauging station. A total of 29 watersheds were analyzed. Eleven of the watersheds had more than one historical storm, leading to a total of 45 rain events that were investigated. Analysis showed that neither technique produced consistently accurate results compared to historical data. The hydraulic (HEC-RAS) simulations produced average peak flow and runoff depth errors of 103% and 67% respectively, with 23 and 21 of 45 simulations overestimating the peak flow and total runoff depth, respectively. The hydrologic (HEC-HMS) simulations had average peak flow and runoff depth errors of 105% and 81% respectively, with 28 and 31 out of 45 simulations overestimating the peak flow and total runoff depth. In general, peak flow results for both models tended to improve for simulations involving larger watershed areas, larger flow return periods, and larger rainfall events. Total runoff depth results tended to improve only for larger rainfall events.

Practical Applications

This paper shows that rainfall–runoff modeling using a hydraulics-based approach yields better results with respect to both peak flow and direct runoff estimations. This is significant because if adopted by the professional community, it would lead to a fundamental shift in the approach that the practicing engineering community takes while performing this sort of analysis (peak flow prediction used for hydraulic structure design, volume estimation used for the reservoir, and stormwater management design). As higher resolution terrain information as well as various watershed features become more readily available, the results of the hydraulic modeling should theoretically improve. Ultimately, standard hydrologic methods involving unit hydrograph theory may become obsolete.