A generalized Ross infiltration model and the application on hole irrigation based on the biconjugate gradient stabilized algorithm

Abstract

Soil water dynamics are very important for understanding and managing water movement in agricultural irrigation; Therefore, it is of great significance to establish an efficient and accurate infiltration model to simulate soil water dynamics. In this work, based on the finite volume method discrete format and principal variable transformation technology, the Taylor first-order expansion is used to solve the Richards equation, the adaptive time step is controlled by the change of soil maximum saturation, the CMS and BiCGSTAB are used to solve the large sparse matrix generated by the implicit format, and the generalized Ross infiltration model is generated. The validity of the model was verified by mass conservation and field experiment, and the soil water dynamic change law of hole irrigation with two irrigation methods (HSI and HAI) was explored by the model. The result shows that the generalized Ross infiltration model based on CMS and BiCGSTAB can significantly improve the computational efficiency of the model; the main variable transformation technique of the generalized Ross infiltration model improves the mass conservation of the model. Balancing the model accuracy and calculation efficiency, a scientific and reasonable time step can be determined by controlling the change in maximum effective saturation (∆Semax), the generalized Ross infiltration model is found to accurately simulate soil-water dynamics under hole irrigation (R2 > 0.9). It is found that HSI and HAI in four-hole irrigation provides more possibilities for further realizing the movement of soil nutrients with water without considering the evaporation of irrigation holes or small evaporation. If the time and infiltration volume fitting curve determined with the indoor single-hole infiltration test is used to control the water volume, HSI requires longer infiltration time to reach the same irrigation volume.