Temporal variability in soybean sowing and harvesting according to K-means and silhouette scores

Understanding the impact of seasonal variations on soybean productivity is crucial for optimizing agricultural practices. Given the influence of climatic factors such as temperature and rainfall on crop phenology, this study aims to analyze the effects of sowing and harvesting soybean cultivars in different seasons. This research investigates whether sowing soybeans in November and December leads to varying outcomes in terms of productivity and morphological characteristics, focusing on identifying the most stable cultivars across different climatic conditions. The study employed a comprehensive methodology, including data standardization, statistical tests (Levene, Shapiro-Wilk, ANOVA, Wilcoxon), and K-means clustering, to analyze 40 soybean cultivars across two seasons. Statistical preprocessing ensured data accuracy, while clustering helped identify cultivars with consistent responses to climatic changes. All computational analyses were performed using Python in the Google Colab environment. The findings revealed no significant difference in productivity between the two seasons, despite variations in temperature and rainfall. However, the moisture content of the grains (MTG) showed significant differences, influenced by higher rainfall in March and April and increased temperatures in December. K-means clustering highlighted SYN2282IPRO as the most stable cultivar and 77HO111I2X-GUAPORÉ as the most sensitive to climatic changes. The results emphasize the need for careful cultivar selection based on specific adaptability to seasonal variations. This study underscores the potential of computational tools like K-means clustering in agricultural optimization, offering a data-driven approach to selecting stable soybean cultivars. The adaptable methodology can be tailored to different geographical regions, soil types, and climate conditions, enhancing its relevance and applicability. These insights contribute to a better understanding of the complex interactions between climatic variables and soybean phenology, providing a foundation for improving agricultural practices in the face of climate change.