Optimization of water and fertilizer regime for greenhouse radish using BBD response surface method coupled with NSGA-II algorithm

Pengrui Ai¹Yinjie Ma^1*Yurong Wei²

• ¹Xinjiang Agricultural University, Urumqi, China
• ²Engineering Research Centre of Cotton, Ministry of Education / College of Agriculture,, Xinjiang Agricultural University,, Urumqi, China

Experiments were conducted in summer and winter based on the Box - Behnken design. The interaction effects of water and fertilizers were analyzed through the response surface model, and the multi - objective collaborative optimization of yield, soluble sugar, and vitamin C was achieved by combining with the NSGA - II algorithm. There were significant seasonal differences. In summer, the available potassium concentration in the 10 - cm soil layer (206.98 mg/kg) under the W2K1N2 treatment was 83.4% higher than that in the deeper layer, and the surface accumulation rate of ammonium nitrogen reached 33.0% in winter. Under the coordinated regulation of water, potassium, and nitrogen, the above - ground dry matter mass under the W2K1N2 treatment in summer increased by 32.0% compared with the control, while the stem diameter in winter decreased by 8.6% due to low - temperature inhibition. The main effect of potassium fertilizer was significant in summer , and nitrogen was the yield - limiting factor in winter, which confirmed the rule of "potassium promotes quality, nitrogen promotes yield". Season - specific water and fertilizer schemes were optimized based on the BBD - NSGA - II algorithm. In summer (W: 19.96 mm, K: 123.17 kg/hm², N: 198 kg/hm²), the simulated yield reached 151.55 t/hm², an increase of 93.1% compared with the control, and the soluble sugar increased by 18.9%. In winter (W: 28.03 mm, K: 182.51 kg/hm², N: 296.91 kg/hm²), the yield was 153.21 t/hm², an increase of 65.5% compared with the control, and the soluble sugar and vitamin C increased by 69.4% and 314.4% respectively. This optimization scheme provides a scientific basis for the efficient production of greenhouse white radishes, which can significantly improve the physiological indicators, yield, and quality of crops, and optimize the resource utilization efficiency. The cross - season water and fertilizer optimization scheme based on BBD - NSGA - II can achieve the coordinated improvement of yield and quality of greenhouse white radishes and the efficient use of resources, providing a theoretical basis and an intelligent decision - making paradigm for the precise regulation of greenhouse vegetables in cold regions.