MACHINE LEARNING-BASED ANALYSIS OF CROP YIELD VARIABILITY IN THE PHILIPPINES UNDER IRRIGATED AND RAINFED CONDITIONS: THE ROLE OF NITROGEN, PHOSPHORUS, AND MAGNESIUM FERTILIZATION

This study explores the elements that impact the fluctuations in crop yield in the Philippines for both irrigation and rainfed agricultural systems, focusing on the effects of nitrogen, phosphorus, and magnesium fertilization on crop yield. Spearman's Rank Correlation determines the relationship between soil fertility, nutrient content, and crop yield. These correlations suggest that adequate water allows for the efficient use of nutrients. On the other hand, rainfed systems show a strong negative correlation with fertilization for nitrogen (r = -0.562, p \ < 0.001) and phosphorus (r = -0.565, p \< 0.001), suggesting that water limitations affect nitrogen use. We observed that irrigation has a strong positive correlation with nitrogen application (r = 0.773, p \< 0.001) and magnesium application (r = 0.346, p = 0.001), among other nutrients. Machine learning models such as Decision Tree, Random Forests, Support Vector Regression (SVR), and K-Nearest Neighbors (KNN) were used; regarding the model performance evaluation, the Random Forest model demonstrated strong consistency and robustness, regardless of whether it was an irrigated or rainfed area, with only slight increases in MAE (0.3107 to 0.3607), MSE (0.1790 to 0.2391), and RMSE (0.4230 to 0.4890), while maintaining high R² values (0.8661 to 0.8095). The study points out the need for tailored agricultural practices, emphasizing synchronized water and nutrient management in irrigated areas and water conservation in rainfed areas to enhance rice production and ensure food security.