Simulation of moisture redistribution pattern on sloping lands under drip irrigation system

Abstract

Background and Objectives: The accurate estimation of the dimensions of the wetting pattern is one of the important parameters in the design of drip irrigation systems, which reduce deep water losses and additional costs of irrigation system design. The wetting dimensions of moisture bulb are affected by the pattern of moisture distribution in the two phases (distribution and redistribution). Various studies have been conducted on the pattern of moisture distribution in sloping lands, but few studies have been carried out or not reported on the pattern of moisture redistribution in these lands. Therefore, the main purpose of this study is to investigate and simulate the pattern of moisture redistribution in sloping lands. Also, in similar studies, the dimensions of the wetting pattern have been usually simulated on the soil surface, but in this research, in addition to the dimensions and wetting area, the full shape of the moisture bulb has been simulated.
Materials and Methods: In this research, two physical rectangular cubic models with dimensions of 1.2 * 1.2 * 0.6 m and 1.4 * 1.2 * 0.7 m were constructed to monitor the soil moisture advance front. These experiments were carried out for four different slopes (0, 10, 20 and 30%), three soil types with different textures (coarse, medium, fine) and three emitter discharges (2, 4, and 6 lit/hour). This study was aimed to simulate the wetting area in a drip irrigation system on slope land. The duration of irrigation was 4 hours and the redistribution wetting front was recorded for different times (e.g., 3, 6, 18, 42, and 66 h) on the Polycarbonate plate. Then, using the nonlinear regression analysis several equations were proposed to predict the redistribution pattern in slopping lands. In the proposed models, emitter flow rate, the volume of applied water, irrigation time, saturated hydraulic conductivity, the soil bulk density, the land slope, the percentage of sand, silt, and clay were utilized. Also, using an enhanced proposed model, the full shape of wetting bulb was estimated.
Results: The suggested models had the high accuracy in heavy soils with the average values of RMSE and MAE for the wetted radius equal to 0.34 and 0.28 cm, respectively. RMSE and MAE values for the wetted area were 0.0018 and 0.0014 m2, respectively. The suggested models had the low accuracy in light soils and RMSE and MAE statistical indices for wetting radius were 0.44 and 0.37 cm and for wetting area were 0.0029 and 0.0022, respectively. The values of calculated statistical indices for the wetted depth of the moisture redistribution front were similar for all the studied treatments and the values of RMSE and MAE varied between 0.43-0.5 and 0.31- 0.39 cm, respectively. Also, the CRM values of the models are mostly positive and their NS is about 0.99 for all the studied treatments.

Conclusion: The results of this research showed that the suggested models have higher precision in heavy soils than light soils. Also, the suggested models have acceptable ability to estimate the wetting radius, upstream and downstream area of the emitter, the wetting depth as well as the full shape of the moisture bulb. The prediction values of the models were mostly underestimated. Therefore, the use of these models recommended for determining the exact location of the emitters in sloping lands, to reduce deep percolation losses and optimal use of water via the plant.