Optimizing Soil Microbial Dynamics and Environmental Factors for Crop Yield Prediction in Mixed Cropping Systems (Black Pepper, Arecanut, Banana)

Abstract

Mixed cropping is widely practiced in tropical agriculture to enhance land productivity, improve soil fertility, and reduce economic risks associated with monoculture. However, accurately predicting crop yield in mixed cropping environments remains challenging due to the complex interactions among soil properties, microbial activity, environmental conditions, and crop-specific growth characteristics. An artificial intelligence (AI)-enabled framework is presented in this research for predicting crop yield in mixed cropping systems comprising Black Pepper (Piper nigrum), Arecanut (Areca catechu), and Banana (Musa spp.). The suggested system combines crop-specific metrics, soil microbial data, and real-time environmental monitoring to enhance forecast accuracy and promote precision agriculture. An Internet of Things-based sensing platform continuously collects data on soil moisture, soil temperature, air temperature, relative humidity, rainfall, soil pH, and light intensity. Soil microbial information is periodically obtained through laboratory analysis and integrated with environmental observations to allow for yield prediction. The collected data are analysed using machine learning models to identify nonlinear correlations between agricultural productivity and environmental factors. The proposed system also enables continuous field monitoring and timely decision support for irrigation, nutrient management, and disease surveillance. Experimental observations obtained from the developed prototype demonstrate the feasibility of integrating microbial and environmental information for intelligent agricultural monitoring. The proposed framework offers a scalable and cost-effective solution for improving decision-making in mixed cropping systems and provides a foundation for future AI-driven precision farming applications.