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

Thesis Abstract

Urban and rural agricultural practices increasingly face challenges related to soil degradation, inefficient resource utilization, and the need for sustainable crop production. The traditional methods of soil health assessment often rely on manual sampling and laboratory analysis, which are time-consuming, labor-intensive, and fail to provide real-time data necessary for prompt decision-making. Consequently, there is an urgent need for innovative, technology-driven solutions that enable continuous monitoring of soil parameters to improve productivity and sustainability in agriculture. This study aims to develop an Internet of Things (IoT)-based soil health monitoring system tailored for precision agriculture. The specific objectives are to design a sensor network capable of measuring key soil parameters, such as moisture content, pH, temperature, and nutrient levels; to develop a real-time data acquisition and transmission platform; to evaluate the system’s performance under different agricultural settings; and to analyze the impact of implementing IoT-driven insights on crop yield and resource efficiency. The methodology adopted for this research is a mixed-methods design, integrating quantitative sensor data collection with qualitative system usability evaluation. The population comprises 150 smallholder farms located within a representative agricultural zone, selected through stratified random sampling to ensure diversity in soil types and cropping patterns. A sample size of 50 farms was determined based on power analysis to achieve statistically significant results, considering an effect size of 0.4 and a 95% confidence level. Data collection instruments include multi-parameter soil sensors, a mobile application for data visualization, and structured interview protocols to gather user feedback. The sensors are configured to operate autonomously, transmitting data via LoRaWAN protocol to a central server. Data analysis involves descriptive statistics, correlation analysis, and multiple regression to examine relationships between soil parameters and crop performance; system usability is assessed through thematic analysis of interview transcripts using NVivo software. The study also employs the Technology Acceptance Model (TAM) to interpret user engagement and acceptance levels. Expected findings indicate that the IoT system will deliver high-frequency, accurate soil data, enabling farmers to make precise interventions. The analysis is anticipated to reveal significant correlations between soil health indicators and crop yields, demonstrating the potential for optimized input application. The regression models are expected to identify key soil parameters most predictive of productivity outcomes. Additionally, usability assessments are likely to indicate a positive reception among farmers, with acceptance levels influenced by perceived ease of use and perceived usefulness, aligning with TAM constructs. This research makes a substantial contribution to the growing body of knowledge on smart agriculture technologies by providing a practical framework for deploying IoT-based soil monitoring systems in resource-constrained settings. It fills existing gaps related to the integration of low-cost sensor networks with cloud-based analytics tailored for smallholder farmers, facilitating data-driven decision-making. The study advances theoretical understanding by applying Systems Theory to model the dynamic interactions between soil health, technology, and farm management practices. Practically, the developed prototype and evaluation outcomes offer a scalable solution adaptable to diverse agricultural environments, promoting sustainable resource management and enhancing productivity. The main conclusion emphasizes that IoT-enabled soil monitoring systems significantly enhance soil management precision and overall farm performance. It is recommended that stakeholders—including government agencies, technology providers, and extension services—collaborate to promote the adoption of such systems at scale. Future research should focus on integrating additional environmental sensors, developing predictive models for climate resilience, and exploring cost-effective deployment strategies that accommodate varying socio-economic contexts. Overall, this study affirms the pivotal role of IoT technologies in advancing sustainable and data-driven agriculture, contributing meaningfully to the digital transformation of farming practices.