Reliability of Microclimate Data with IoT Technology on Coffee-Pine Farms in UB Forest

The Coffee-Pine fields in UB Forest play a crucial role in the local ecosystem, with a microclimate that supports optimal plant growth. However, challenges to the reliability of microclimate data often arise, resulting in uncertainty in agricultural decision-making. This research focuses on the application of Internet of Things (IoT) technology to improve the accuracy and consistency of microclimate data. By installing IoT sensors at strategic points on the land, data is collected over a year. This research aims to analyze microclimate reliability data from two plots, LC and BAU. Microclimate data were evaluated based on data quality, including checking for missing data and outliers. Descriptive statistical analysis and data visualization were carried out to gain deeper insight into microclimate variations at both locations. The LC data showed missing data 13.92% for each parameter while the BAU data had little missing data. Both data sets show the presence of outliers. After cleaning, the data shows more consistent and realistic variations. Visual analysis shows fluctuations over time for all parameters, with some significant differences between LC and BAU. Microclimate data from LC and BAU show variations over time and between locations. Data quality is an important consideration in analysis, and data cleaning increases the reliability of results. The differences between LC and BAU are due to different location plots. This research confirms the potential of IoT technology in overcoming the problem of microclimate data reliability, providing a strong basis for sustainable agricultural practices on Coffee-Pine land in UB Forest.