Appraisal of Mica-Based Potassium Mobilizing Bacterial Biofertilizers: Revolutionizing Soil Fertility and Plant Growth Employing Multi-Machine Learning Models

Muscovite and biotite are two forms of mica, potassium (K)-bearing minerals, that are waste products of the mica industry. This study aims to evaluate the pattern of K release from muscovite and biotite inoculated with K-mobilizing bacterial (KMB) strains such as <i>Bacillus cereus</i> K5B, <i>Bacillus cereus</i> K6, <i>Bacillus sp</i>. GG6 (2015) K12 and <i>Bacillus cereus</i> K15, as well as their effectiveness as K fertilizers using tomato plants (<i>Solanum Lycopersicum</i>) grown in Giridih soil (Alfisol). Applying muscovite and biotite (at 200 mg kg⁻¹ of soil) combined with <i>Bacillus cereus</i> K5B notably enhanced tomato plant biomass, boosted fruit quality and yield, and improved potassium availability in the soil. This approach also helped maintain soil quality, outperforming soils treated with synthetic potassium fertilizers and untreated soils in potassium availability. Statistical algorithms predict the free ion activity (FIAM) model to transport bioavailable forms of soil K to plant parts, forecasting KMBs’ effects on water-soluble K (WK) and exchangeable K (EK) mobilization from non-exchangeable K (NEK) form and uptake by soil-plant systems, enhancing plant growth, health, and yield. Multimodal machine learning techniques, such as Fuzzy-TOPSIS, Monte Carlo simulations and Sobol sensitivity analysis, have demonstrated the effectiveness of KMB fertilizers for K5B isolates. The Taylor diagram further aids in selecting the most accurate method for predicting outcomes within the dataset. The findings of this study suggest that bio-interventions using KMB with muscovite and biotite hold significant potential as a sustainable alternative to conventional potassium fertilizers. Machine learning and statistical modeling confirm the effectiveness of KMB strains, opening new pathways for advancing biofertilizer use in agriculture to boost soil health and crop productivity. In modern agriculture, these tools enable data-driven, precise, and sustainable farming practices that enhance yields reduce environmental impact, and help farmers adapt to changing conditions.