Rice Irrigation Schedule Optimization Based on the AquaCrop Model: Study of the Longtouqiao Irrigation District

Biying Zhai,Dong Liu,Qiang Fu,Yi Ji,Mo Li,Tianxiao Li,Song Cui

doi:10.3390/w11091799

Abstract

As a crop with high water consumption, rice is an important measure of efforts to improve agricultural irrigation efficiency and alleviate the contradiction between the supply and demand of agricultural water resources. This paper takes the Longtouqiao irrigation district, in the hinterland of the Sanjiang Plain, a major rice-producing area in northern China, as an example, and the AquaCrop crop growth model and entropy-cloud model are jointly used to develop a rice irrigation schedule optimization model based on three kinds of typical rainfall years. Different irrigation schemes are established and evaluated by using the model built based on images. The results showed that the yield’s normalized root mean square error (NRMSE) value of the AquaCrop model was 9.949% (< 10%) after calibration, and the our model results showed a good agreement with observed data, which indicated that the calibrated model was suitable for rice growth simulation in the research area. For the same irrigation water amount, rice was irrigated to a great extent at the tillering stage, and a small amount of irrigation water at the regreening stage of rice could improve rice yield. During irrigation, rice production can also be promoted by regulating the irrigation amount according to the rainfall in each growth period, and the optimal irrigation water amount can be controlled between 20 and 60 mm each time. Under the three typical annual scenarios of dry, normal and wet years, the respective optimal quantification results for the field capacity, total irrigation water amount and irrigation times in the rice growth period to attain the optimal irrigation effect were 25%, 425 mm, and 17 times, respectively; 25%, 450 mm, and 14 times; respectively, and 25%, 425 mm, and 17 times, respectively. The research results can provide a decision-making basis for water-saving measures and efficient rice irrigation water management.

Highlights

As a major irrigated country in the world, over 70% of China’s grain, 80% of its cotton and90% of its vegetables come from irrigation agriculture [1]
Since the irrigation scenarios set in this paper involve controlled irrigation, wet irrigation and basin irrigation methods, to improve the applicability of the AquaCrop model, the data for controlled irrigation during the whole growth period of rice in the field in 2010 were selected to calibrate the AquaCrop model
The results show that the AquaCrop model has an excellent simulation effect, which verifies that the model can be used to simulate rice growth

Summary

Introduction

As a major irrigated country in the world, over 70% of China’s grain, 80% of its cotton and90% of its vegetables come from irrigation agriculture [1]. Rice consumes 50% of the irrigation water and 63% of fertilizers [2,3]. Developing a reasonable rice irrigation schedule under the condition of insufficient irrigation is of great significance to improve the agricultural water use efficiency [7]. Dudley and De Lucia [8,9,10] adopted a dynamic programming model, setting runoff or rainfall as random variables to develop a crop irrigation schedule. A dynamic programming model was used to optimize the irrigation schedule at the rice growing stage [11,12,13]. Tian [14] constructed a decision system based on a geographic information system (GIS) and genetic algorithms to optimize the combined irrigation schedule. Based on the Jensen model, Yu Zhijing [15] applied the improved non-dominated sorting genetic algorithm (NSGA-II)

Methods

Results

Discussion

Conclusion