This study evaluates the effectiveness of visible near-infrared (vis-NIR) spectroscopy in predicting soil quality indices (SQIs) using three methods: measured SQI (SQI_m), predicted SQI (SQI_p), and direct prediction of SQI (SQI_dp). The research focuses on seven soil indicators critical for soil fertility: pH, organic matter (OM), potassium (K), calcium (Ca), magnesium (Mg), available phosphorus (P), and total nitrogen (TN). These indicators influence nutrient availability, pH regulation, and soil structure, which are essential for soil fertility and health. The study uses the Cubist model to predict these soil properties with high accuracy (R² = 0.35–0.93), offering a cost-effective and environmentally friendly alternative to traditional laboratory analyses. Three approaches were used to calculate SQI: SQI_m, derived from laboratory-measured soil properties; SQI_p, calculated using predicted soil properties from spectral data; and SQI_dp, directly predicted using spectral data. The results showed that SQI_dp had the highest accuracy (R² = 0.90) in predicting soil quality compared to SQI_p (R² = 0.23). This indicates that direct prediction using vis-NIR spectroscopy is more accurate and efficient than other methods. The study analyzed 2830 soil profiles from Ontario, including samples from Peterborough, Ottawa, and Woodrill Limited. Soil samples were collected, processed, and analyzed using vis-NIR spectroscopy. The data were preprocessed to remove noise and enhance spectral quality. Three models—Partial Least Squares Regression (PLSR), Cubist, and Random Forest (RF)—were used to predict soil properties and SQI. The Cubist model performed best, with high R² values for most soil properties. The study highlights the potential of vis-NIR spectroscopy in soil quality assessment, providing a rapid, non-destructive, and cost-effective method for predicting soil properties and SQI. This approach is particularly useful for precision agriculture and environmental monitoring. The findings suggest that vis-NIR spectroscopy, combined with machine learning models like Cubist, can significantly improve the accuracy and efficiency of soil quality assessments, contributing to sustainable agricultural practices and environmental conservation.This study evaluates the effectiveness of visible near-infrared (vis-NIR) spectroscopy in predicting soil quality indices (SQIs) using three methods: measured SQI (SQI_m), predicted SQI (SQI_p), and direct prediction of SQI (SQI_dp). The research focuses on seven soil indicators critical for soil fertility: pH, organic matter (OM), potassium (K), calcium (Ca), magnesium (Mg), available phosphorus (P), and total nitrogen (TN). These indicators influence nutrient availability, pH regulation, and soil structure, which are essential for soil fertility and health. The study uses the Cubist model to predict these soil properties with high accuracy (R² = 0.35–0.93), offering a cost-effective and environmentally friendly alternative to traditional laboratory analyses. Three approaches were used to calculate SQI: SQI_m, derived from laboratory-measured soil properties; SQI_p, calculated using predicted soil properties from spectral data; and SQI_dp, directly predicted using spectral data. The results showed that SQI_dp had the highest accuracy (R² = 0.90) in predicting soil quality compared to SQI_p (R² = 0.23). This indicates that direct prediction using vis-NIR spectroscopy is more accurate and efficient than other methods. The study analyzed 2830 soil profiles from Ontario, including samples from Peterborough, Ottawa, and Woodrill Limited. Soil samples were collected, processed, and analyzed using vis-NIR spectroscopy. The data were preprocessed to remove noise and enhance spectral quality. Three models—Partial Least Squares Regression (PLSR), Cubist, and Random Forest (RF)—were used to predict soil properties and SQI. The Cubist model performed best, with high R² values for most soil properties. The study highlights the potential of vis-NIR spectroscopy in soil quality assessment, providing a rapid, non-destructive, and cost-effective method for predicting soil properties and SQI. This approach is particularly useful for precision agriculture and environmental monitoring. The findings suggest that vis-NIR spectroscopy, combined with machine learning models like Cubist, can significantly improve the accuracy and efficiency of soil quality assessments, contributing to sustainable agricultural practices and environmental conservation.