Strategies for improving crop comprehensive benefits via a decision-making system based on machine learning of rice-rape, rice-wheat and rice-garlic rotation systems in Southwest China

Abstract

Rice-rape, rice-wheat and rice-garlic rotations are common cropping systems in Southwest China and have played a significant role in ensuring ecological and economic benefits (EB) and addressing the challenges of China's food security in the region. But regionally, the crop yields in these rotation systems are 1.25%-14.73% lower than the national average. Intelligent decision-making with machine learning can analyze the key factors for obtaining better benefits, but it is rarely used to enhance the probability of obtaining such benefits of rotations in Southwest China. Thus, we used a data-intensive approach to construct intelligent decision-making with machine learning to provide strategies for improving the benefits of rice-rape, rice-wheat, and rice-garlic rotations in Southwest China. The results show that raising yield and partial fertilizer productivity (PFP) by increasing seed input under high fertilizer obtained optimal benefits with a 10% probability in rice-garlic system. Obtaining high yields and greenhouse gas (GHG) emissions by increasing N application and decreasing K application obtained suboptimal benefits with an 8% probability in rice-rape system. Reducing N and P to enhance PFP and yield obtained optimal benefits with the lowest probability (8%) in rice-wheat system. Based on the predictive analysis of a random forest model that decreases N by 25% and increases P and K by 8% and 74%, respectively, in rice-garlic system, that decreases N and K by 54% and by 36%, respectively, and increases P by 38% in rice-rape system, and that decreases N by 4% and increases P and K by 65% and 23% in rice-wheat system, these strategies could be further optimized by 17-34% for different benefits, all of these measures can improve the effectiveness of crop rotation systems to varying degrees. Overall, these findings provide insights into optimal agricultural inputs for higher benefits through an intelligent decision-making system with machine learning analysis in rice-rape, rice-wheat, and rice-garlic systems.