Journal of Irrigation and Drainage Engineering

Reference evapotranspiration and solar radiation estimation methods that use only measured temperature have been demonstrated to require regional calibration for acceptable performance. This study was conducted to evaluate regional calibration of incoming solar radiation and reference evapotranspiration in Florida, USA. Using satellite-derived estimates of solar radiation and reference evapotranspiration as the standard for comparison, regional calibration of the Hargreaves-Samani solar radiation model and the Turc temperature-based reference evapotranspiration equation were evaluated using monthly temperature data from 22 weather stations in Florida. Station-specific and regional calibration of the Hargreaves-Samani solar radiation model was found to produce comparable results, indicating the validity of this approach in regional calibration of radiation estimates. Using the regionally calibrated radiation estimates, the original Turc equation was found to underestimate reference evapotranspiration, indicating an overall negative bias for the equation in the region. Using the regionally calibrated radiation estimates with the calibrated Turc equation was found to produce comparable error but considerably lower bias compared to the original Turc equation and the original Hargreaves-Samani solar radiation model using coefficients recommended by the Food and Agricultural Organization.

The use of Penman-Monteith (PM)-type combination-based energy-balance equations to estimate reference evapotranspiration (${\mathrm{ET}}_{\mathrm{ref}}$) requires climatic data measured over well-watered and well-maintained reference grass or alfalfa vegetation surfaces. However, establishing and maintaining reference weather stations for a long period of time is a very formidable and expensive process. Thus, expansion of the microclimate data available for use in the PM equation for estimating ${\mathrm{ET}}_{\mathrm{ref}}$ is needed. In the absence of reference weather stations, one alternative is using microclimatic data measured over other well-watered vegetation surfaces as inputs to the PM equation. This study determines if weather data collected from a well-watered soybean surface in a semihumid climate can be used for this purpose. Measured and estimated microclimate variables, including net radiation ($R_n$), average air temperature ($T_{\mathrm{ave}}$), dew-point temperature ($T_d$), average relative humidity (${\mathrm{RH}}_{\mathrm{ave}}$), aerodynamic resistance ($r_a$), and wind speed at 3 m ($u_3$) of a soybean and a grass canopy in South Central Nebraska, were analyzed and compared. The aerodynamic resistances of the soybean and grass canopies showed the largest percent difference of any of the microclimate variables for both 2007 and 2008. Wind speed was the primary microclimate variable with the largest percent difference between the two fields. The average percent differences in $u_3$ between the soybean and grass field were 9.0 and 9.8% for 2007 and 2008. Although $T_{\mathrm{ave}}$, ${\mathrm{RH}}_{\mathrm{ave}}$, and $T_d$ percent differences were not that large, there were distinct seasonalities to the differences. Grass and alfalfa reference evapotranspiration (${\mathrm{ET}}_o$ and ${\mathrm{ET}}_r$, respectively) calculations using data from the soybean (${\mathrm{ET}}_{o\mathrm{\text{-}}s}$ and ${\mathrm{ET}}_{r\mathrm{\text{-}}s}$) and grass (${\mathrm{ET}}_{o\mathrm{\text{-}}g}$ and ${\mathrm{ET}}_{r\mathrm{\text{-}}g}$) canopies were compared daily and seasonally. Seasonal total ${\mathrm{ET}}_o$ and ${\mathrm{ET}}_r$ estimates using soybean and grass microclimate data were very close, and within 1 and 2% during 2007 (${\mathrm{ET}}_{o\mathrm{\text{-}}g}=583\mathrm{mm}$ and ${\mathrm{ET}}_{o\mathrm{\text{-}}s}=576\mathrm{mm}$; ${\mathrm{ET}}_{r\mathrm{\text{-}}g}=751\mathrm{mm}$ and ${\mathrm{ET}}_{r\mathrm{\text{-}}s}=733\mathrm{mm}$), and 4 and 5% during 2008 (${\mathrm{ET}}_{o\mathrm{\text{-}}g}=554\mathrm{mm}$ and ${\mathrm{ET}}_{o\mathrm{\text{-}}s}=531\mathrm{mm}$; ${\mathrm{ET}}_{r\mathrm{\text{-}}g}=707\mathrm{mm}$ and ${\mathrm{ET}}_{r\mathrm{\text{-}}s}=669\mathrm{mm}$). In 2007, differences in temperature variables were most correlated to differences in ${\mathrm{ET}}_{\mathrm{ref}}$ estimates. In 2008, the greatest correlations of differences in ${\mathrm{ET}}_o$ and ${\mathrm{ET}}_r$ were with differences in $T_{\mathrm{ave}}$, ${\mathrm{RH}}_{\mathrm{ave}}$, and $u_3$. The results indicated that the microclimate data measured over an irrigated soybean canopy during normal or wet years ($\mathrm{\text{rainfall}}\ge 300\mathrm{mm}$ during the growing season) can be used in place of measurements taken over a grass canopy to estimate ${\mathrm{ET}}_o$ and ${\mathrm{ET}}_r$ in climatic conditions similar to semihumid climatic conditions of South Central Nebraska when reference weather station data are not available to solve the standardized PM equation.

The objective of this project was to determine if automatic residential irrigation systems with soil moisture sensor irrigation controllers could reduce irrigation water application while maintaining acceptable turfgrass quality as successfully in homes as in plot studies. Research was conducted on-site at cooperating homes ($n=58$) in southwest Florida. Experimental treatments were (1) automatic timer with integration of bypass soil moisture sensor control system, (2) automatic timer with rain sensor and educational materials, (3) automatic timer with rain sensor, and (4) automatic timer only (typical for the region). Irrigation application, frequency, quarterly turf quality ratings, and weather data were collected over 26 months. Homes with soil moisture sensor irrigation controllers bypassed unneeded events during both rainy and dry periods, averaging 2 irrigation events per month; all other treatments averaged 4.5–6 events per month. Reduction in number of irrigation events by soil moisture sensor control systems resulted in significant savings, with 65% cumulative reduction compared to homes with typical timer irrigation control. Observed on-site savings were comparable to previous plot research, indicating that plot savings could be scaled up so long as soil moisture control systems are installed and set properly.

The canal command of the Hirakud multipurpose major irrigation project of Orissa in eastern India is under severe threat of waterlogging in the monsoon season and acute shortage of irrigation water in the nonmonsoon season. An irrigation scheduling model (ISM) and a linear-programming optimization model (LPM) under hydrologic uncertainty were developed with a view to manage the available land and water resources of the canal command effectively. The ISM was used to predict actual crop yield under different irrigation strategies—namely, full and deficit depths of irrigation. The crop yield obtained by the ISM under different irrigation management strategies was used in the LPM to optimize the land and water resources of the canal command at different probability of exceedances of net irrigation requirement and canal-water availability. The net annual return was found to decrease with the increase in the level of deficit with maximum return under full irrigation strategy. Hence, from the benefit point of view, crops must be provided with adequate irrigation. Most beneficial crops with comparatively lower water requirements, like pulses and vegetables, should be given priority, and the present practice of extensive rice cultivation should be limited to minimum possible extent for sustainable protection against global warming. It was also found that the uncertainty factor does not show any visible effect on the cropping pattern, which in turn is reflected in the overall water resources utilization pattern of the canal command. From the sensitivity point of view, cropping area should be given emphasis, followed by the market price and cost of cultivation of different crops during the course of further study.

Lysimetric experiments were conducted to investigate groundwater contributions made by safflower crop (Carthamus tinctorius L.). The plants were grown in 24 columns, each with a diameter of 0.40 m and packed with silty clay soil. The factorial manner experiments were carried out with four replicates using randomized complete block design with different treatment combinations. Six treatments were applied in each experiment by maintaining groundwater salinity with electric conductivity (EC) of $10\mathrm{dS}/\mathrm{m}$ and different water table levels of 0.60, 0.80, and 1.10 m, with and without irrigation strategies. The use of groundwater as a part of crop evapotranspiration was determined by taking daily measurements of the water levels in Mariotte tubes. Additional irrigation in the form of surface water was applied for each treatment with EC of $1\mathrm{dS}/\mathrm{m}$. The results of the experiments show that for different treatments, with and without irrigation and different water table levels of 0.60, 0.80, and 1.10 m, the average percentage of groundwater contributions achieved were 15, 19, and 25%; and 10, and 19% of the average annual safflower water requirement, respectively.

Friction head loss equations and friction correction factors were evaluated and compared to field observations collected from thirty center pivots with laterals made of PVCs. The friction head loss equations include Darcy-Weisbach (D-W), Hazen-Williams (H-W), and Scobey, in addition to a proposed equation valid for smooth and rough pipe types and for all turbulent flow types. The proposed equation was developed by combining the equations of D-W and H-W, along with the multiple nonlinear regression technique. The friction correction factors were computed by using the typical Christiansen, modified Christiansen, Anwar, and Alazba formulae. The evaluation has been based on statistical error techniques with observed values as a reference. With the combination of modified Christiansen, Anwar, and Alazba formulae, the results revealed that the magnitudes of friction head loss calculated by using the D-W, H-W, and proposed equations were in agreement with field observations. The root mean square deviation (RMSD) values ranged from 1.6 to 1.7 m. As expected, and when the typical Christiansen friction correction factor was used with the D-W, H-W, and proposed equations, the results showed poor agreement between observed and computed friction head loss values. This was clearly reflected by the high RMSD values that ranged from 5.4 to 5.9 m. On the other hand, agreement occurred between observed friction head loss values and those calculated by using the Scobey equation, invalid for PVC pipe type, when combined with the typical Christiansen formula. This interesting finding led to improved results of the Scobey equation through a developed $C_s$ coefficient suitably valid for PVC pipe type through analytically mathematical derivation; accordingly, the RMSD value dropped from approximately 8.6 to 1.6 m.

Irrigation management in vertisols is one of the major challenges to increase agricultural productivity in India and many developing countries. Unfortunately, information on hydraulic properties of these soils is very sparse. In an attempt to understand these soils for better management, 10 different functions were evaluated for their efficacy to describe soil-water retention characteristics (SWRC) of vertisols of India, and point pedotransfer functions (PTFs) were developed by using a nearest neighbor ($k$-NN) algorithm as an alternative to widely used artificial neural networks (ANN) for prediction of available water capacity (AWC). Soil profile information of 26 representative sites comprising 157 soil samples was used for analysis. The Campbell model fit to measured SWRC data better than any other model, with relatively lower root mean square error (RMSE) (0.0199), higher degree of agreement (0.9867), and lower absolute error on an average (0.0134). Three other functions, namely, modified Cass-Hutson, Brooks-Corey, and van Genuchten, also described the SWRC data with acceptable accuracy. Four levels of input information were used for point pedotransfer function (PTF) development: (1) textural data [data on sand, silt, and clay fraction (SSC)]; (2) Level $1+\mathrm{\text{bulk}}$ density data (SSCBD); (3) Level $2+\mathrm{\text{organic}}$ matter (SSCBDOM); and (4) Level $1+\mathrm{\text{organic}}$ matter (SSCOM). The RMSE in predictions by $k$-NN PTFs ranged from 0.0339 to $0.0450{\mathrm{m}}^3{\mathrm{m}}^{-3}$ with an average of $0.0403{\mathrm{m}}^3{\mathrm{m}}^{-3}$. The ANN PTFs performed with average RMSE $0.0426{\mathrm{m}}^3{\mathrm{m}}^{-3}$ and a range of 0.0395 to $0.0474{\mathrm{m}}^3{\mathrm{m}}^{-3}$. The $k$-NN algorithm provided a viable alternative to neural regression with marginally better performance and the benefit of flexibility in the appending reference database. The results are significant because SWRC data are still in the development stage in India, and $k$-NN PTFs would have a greater value because of the flexibility.

Indicator bacteria are a common source of impairment in surface waters in the United States. Urban storm water runoff has been identified as a contributor to elevated indicator bacteria concentrations. Six storm water control measures (SCMs) were monitored in Wilmington, North Carolina, for E. coli and enterococci. Monitored SCMs included two storm water wet ponds, two bioretention cells, and two storm water wetlands. Sandier watersheds in Wilmington potentially lead to differences in SCM performance for indicator bacteria compared to SCMs implemented in clayey watersheds. Results showed E. coli and enterococci concentration reductions between 70 and 98% for the two wet ponds and a bioretention cell with a 60-cm-deep fill media. Other SCMs showed poor removal of indicator bacteria, in some cases negative, with storm water wetlands performing the poorest overall for the three SCM types. Further analysis showed that SCMs with high concentration reductions tended to have geometric mean effluent concentrations lower than the U.S. EPA’s target surface-water concentration for E. coli. Conversely, no SCM had a geometric mean effluent enterococci concentration lower than the U.S. EPA target value. SCM geometric mean effluent concentrations were typically higher during North Carolina’s swimming season between the beginning of April and the end of October, although no statistically significant relationship could be found ($p<0.05$). Despite a lack of statistically significant relationships, the potential for higher effluent indicator bacteria concentrations from SCMs during the peak recreational season may have implications for both public health and watershed management and should be further evaluated by the scientific community.

Rain gardens have been recognized as an effective device for on-site runoff volume reduction and storm water quality enhancement. A rain garden is designed as an infiltration basin with a shallow, wide water storage volume. The subbase underneath the basin bottom is structured as a two-layered filtering medium. The upper sand layer provides the required filtering process and then the lower gravel layer provides a reservoir for a gradual release of the stored water through a perforated underdrain pipe. A newly constructed rain garden can have a high-infiltration capacity. After several years of service, the accumulation of solids intercepted in the filtering layer will develop a clogging effect or the infiltration capacity in the subbase continues decreasing. When the drain time exceeds the safety criteria, the rain garden needs to be replaced. Coping with a decaying infiltration rate, a rain garden needs to reduce its flow release in the early years. In this study, it is suggested that a cap-orifice be installed at the exit of the underdrain pipe. This cap-orifice is sized to reduce the initial infiltrating rate in the early years and then to work with the clogged infiltration rate throughout the life cycle of a rain garden. The design methodology presented in this paper allows the engineer to size this cap-orifice according to the decay of infiltration rate, required drain time, and allowable flow release.

Small drainage channels lined with rock riprap are often designed using Manning’s formula for uniform flow to calculate water depths and velocities. Resistance to flow is quantified by the Manning roughness coefficient $n$, which is strongly dependent on water depth when flows are shallow in comparison with the size of rock particles protecting the channel beds. To provide reliable formulas for evaluating $n$ in riprap-lined channels, particularly for relatively shallow flows, 142 measurements of flow resistance carried out by others in several experimental studies were assembled and analyzed. Two formulas founded on a theoretical assessment of flow resistance are presented in which roughness heights are calculated based on measures of the distribution of riprap particle diameters. The formulas are compared with other equations that have been developed specifically to predict $n$ in riprap-lined drainage channels, and they are shown to provide significantly more accurate estimates of flow resistance.