شبیه‌سازی مساحت خیس شده پیاز رطوبتی در سیستم آبیاری قطره‌ای پالسی

Introduction: Among irrigation methods, a drip irrigation system (surface and subsurface) is more acceptable in arid and semi-arid regions due to high water use efficiency and potential crop yield. Pulse drip irrigation (with suitable management practices) is one of the drip irrigation methods (includes a set of cycles, each cycle consisting of the irrigation phase and a resting phase) that have high potential to improve the uniformity of soil moisture distribution. Suitable design and management of pulse or/and continuous drip irrigation systems substantially require a proper understanding of the moisture distribution pattern around the emitter. One of the critical parameters concerning the moisture distribution pattern, taking into account the wetted area of emitter. Important parameters of the wetted area include the down wetted area (Ad) for the surface and subsurface drip irrigation system as well as the up wetted area of an emitter (Aup) for the subsurface drip irrigation. Modeling the wetted area pattern and considering this parameter in design as one of the criteria for increasing water efficiency in surface and subsurface drip irrigation systems is critical and important.
Materials and Methods: In this research, experiments were carried out in a transparent rectangular cube with dimensions of (3 * 1 * 0.5 m) using three different soil textures (fine, heavy, and medium). The drippers were installed at three different soil depths (surface, 15cm, and 30cm). The emitter discharge was considered 2.4, 4, and 6 lit/hr. Also, these experiments were carried out for two continuous and pulse irrigation systems. In pulse irrigation, the pulse cycles were considered 30-30, 20-40, and 40-20 min. The first number refers to the irrigation time, and the second number refers to the resting time of the system in each cycle. In this research, using a nonlinear regression model, empirical models were developed to predict the wetted area of the moisture front. The input parameters of the suggested model include emitter discharge, saturated hydraulic conductivity, application time, soil bulk density, emitter installation depth, initial soil moisture content, pulse ratio (the ratio of irrigation time to complete period of each cycle) and the proportions of sand, silt and clay in the soil.
Results and Discussion: The results of this study show that the highest and the lowest down wetted area (for surface and subsurface drip irrigation systems) are related to sandy and clay soils, respectively. Also, the highest up wetted area in the subsurface irrigation system is related to loamy and clay soils. The results of the comparison between measured and simulated values of down and up wetted area indicated that these models have acceptable precision and accuracy in estimating the wetted area of the wetting front in surface and subsurface drip irrigation (with pulsed and continuous application). The comparison between the measured and simulated down wetted area of the emitter (for surface drip irrigation with pulsed application) showed that the R2, MAE and RMSE values varied between 0.98-0.99, 0.0027-0.0065 m2 and 0.0034-0.0082 m2, respectively. Concerning statistical values, it is evident that these models have excellent performance in estimation of down and up wetted area for subsurface drip irrigation. For subsurface drip irrigation with the pulsed application, the values of R2, MAE and RMSE for the down wetted area of emitter, ranged 0.91-0.99, 0.002-0.0077 and 0.0032-0.0098, respectively. These models also estimate up wetted areas with less error, and the values of R2, MAE, and RMSE for all treatments varied between 0.89-0.99, 0.0015-0.0067 m2, and 0.0019-0.0077 m2, respectively.
Conclusion: This paper was aimed at presenting relationships for estimating the up and down wetted area of emitter for surface and subsurface drip irrigation (with pulsed and continuous application). Regarding the importance and applicability of empirical models, in this research, nonlinear regression models (NLR, which are more widely used among researchers) were applied. For NLR method, different ten input variables (i.e., emitter discharge, saturated hydraulic conductivity, application time, soil bulk density, emitter installation depth, initial soil moisture content, pulse ratio (the ratio of irrigation time to complete period of each cycle) and the percentage of sand, silt and clay) were considered. The results of this study indicate that the NLR model can estimate the up and down wetted area, and the statistical indices values are within acceptable ranges. Considering these relations in designing surface and subsurface drip irrigation systems can improve the performance of these systems.