Simulation of open- and closed-end border irrigation systems using SIRMOD

TLDR

Surface irrigation dominates global irrigation, yet its low performance stems from poor design, operation, and maintenance, making mathematical simulation essential to reduce costs and accelerate analysis of efficiency and uniformity. The study aims to simulate open‑ and closed‑end border irrigation systems using the SIRMOD software package. The authors used 22 data sets of open‑ and closed‑end borders and applied the HD, ZI, and kinematic wave models, noting that the HD model’s complex Saint‑Venant equations introduce greater numerical uncertainty. The models predicted open‑end recession times more accurately than closed‑end, and both HD and ZI models estimated infiltrated water volumes that were equal to or less than observed volumes, though the HD model sometimes produced larger errors due to its more complex equations.

Abstract

In many regions in the world, more than 90% of areas equipped for irrigation apply surface methods. The major problem of the surface irrigation system is low performance due to poor design, operation, and maintenance. Use of the mathematical models for simulation of surface irrigation is necessary for reducing costs and decrease of time in analysis of indexes including application efficiency and distribution uniformity. This study aims to simulate border irrigation systems using the SIRMOD (surface irrigation simulation, evaluation and design, developed by Utah State University, Logan, UT, USA) software package under open- and closed-end conditions. For this purpose, 22 sets of data including four no-cultivated open-end borders, nine no-cultivated closed-end borders, and nine cultivated closed-end borders were used. The results showed that the models predicted open-end conditions better than closed-end for recession time. In addition, the hydrodynamic (HD) and the zero inertia (ZI) models estimated volume of infiltrated water, equal or less than volume of observed water in all the borders. Although the HD model uses the Saint-Venant equations without simplification, during numerical solution of them by the software, uncertainty is raised due to further calculations than the ZI and kinematic wave models. This leads to further error of the HD model than the other models in some cases.