Development of IoT-based Soil Moisture Monitoring System for Precision Agriculture

Thesis Abstract

In the pursuit of sustainable and efficient agricultural practices, soil moisture management remains a critical factor influencing crop yield and resource conservation. Traditional methods of soil moisture assessment are often labor-intensive, time-consuming, and lack real-time data capabilities, leading to suboptimal irrigation practices and potential resource wastage. Addressing this challenge, the study aims to develop a low-cost, scalable Internet of Things (IoT)-based soil moisture monitoring system specifically designed for precision agriculture applications. The primary objectives are to design and prototype a sensor network capable of real-time soil moisture detection, evaluate the system's accuracy and reliability, and assess its impact on irrigation efficiency and crop productivity. The research employed a mixed-methods approach, combining quantitative experimental design with qualitative usability assessment. A systematic review of existing IoT-based soil monitoring systems informed the development of a novel sensor array incorporating capacitance-based moisture sensors connected via Wi-Fi modules. The target population comprised smallholder farmers and irrigation managers within a peri-urban agricultural region, with a sample size of 50 participants selected through stratified random sampling to ensure diversity in farming practices. Data collection involved deploying ten sensor nodes across different soil types and moisture levels over a cropping season of six months, alongside structured interviews and focus group discussions. The quantitative data—soil moisture readings, crop yields, and water usage metrics—were analyzed using regression analysis to determine the system's predictive accuracy, while reliability was assessed through Cronbach’s alpha. Qualitative data from user feedback were examined through thematic analysis to evaluate system usability and acceptability. Key expected findings include that the IoT-based system will demonstrate high measurement accuracy (correlation coefficient above 0.9 compared to standard gravimetric methods), with an intraclass correlation coefficient indicating excellent reliability. The system is anticipated to detect soil moisture variations with a margin of error below 5%, facilitating precise irrigation adjustments. Furthermore, implementation of the system is expected to lead to at least a 20% reduction in water usage and a 15–20% increase in crop yield compared to traditional practices, substantiated through comparative analysis of pre- and post-deployment data. User feedback is projected to reveal high levels of system usability and acceptance, highlighting its potential for adoption among smallholder farmers. This research contributes novel insights into the integration of IoT technology into soil moisture management, filling existing gaps in affordable, user-friendly, real-time monitoring solutions tailored for resource-constrained agricultural settings. The theoretical framework draws on the Technology Acceptance Model (TAM) and Diffusion of Innovations theory to explain user adoption behavior, providing a basis for enhancing system interface design and dissemination strategies. By establishing a functional prototype and demonstrating its operational benefits, the study advances knowledge in precision agriculture technology and offers practical pathways for scalable deployment in similar contexts. In conclusion, the developed IoT-based soil moisture monitoring system has been shown to significantly improve irrigation efficiency and crop productivity, with high reliability and user acceptance. Recommendations include wider field trials across diverse agro-ecological zones, integration with decision-support systems for comprehensive farm management, and policy advocacy for subsidized deployment among smallholder farmers. Future research should explore the integration of additional environmental sensors and machine learning algorithms to further refine moisture prediction models, thereby maximizing the system's impact on sustainable agricultural practices.