LAUSR

Abstract Severe water scarcity necessitates the development of a scientifically informed basis for efficient crop irrigation. Both irrigation and crop planting pattern (the planting dates and the planting ratio for male to female maize inbred lines) affect the yield components in hybrid maize production. A high kernel weight, which affects seed germination, is critical in maize seed production. Therefore, the aim of this study was to build a system that jointly optimizes irrigation and planting pattern (JOIPP) to guarantee a suitable kernel weight, maximizes yield, and saves water. The JOIPP was based on a model which could simulate flowering, predict kernel number and estimate the upper and lower limits of kernel weight based on the evapotranspiration and planting pattern. Parameters in JOIPP were calibrated and validated using data from experiments conducted from 2014 to 2016 in Hexi Corridor, northwestern China. A weighted sum method was used to maximize yield and minimize irrigation. A genetic algorithm with a modified initial population generation module was used to solve this optimization problem. Results showed that when the minimum lower limit of kernel weight (LowKWmin) increased, there was decreased irrigation at the flowering stage, increased irrigation at the grain-filling stage, less optimal yield and water use efficiency when the available water insufficient. However, the associated risk of seed germination failure decreased. The optimized planting pattern, compared with the crop planting pattern used by local farmers, increased yield by about 10% and water use efficiency by about 7.3% when LowKWmin was 260 mg/kernel. JOIPP is a useful tool that enables farmers and seed companies to save water and improve yield and seed quality when planting and irrigating maize for hybrid seed production.