Skripsi
KARAKTERISASI KONSENTRASI KARBON MONOKSIDA PASCA KEBAKARAN HUTAN DAN LAHAN MENGGUNAKAN DATA SENSOR IOT SEBAGAI KONTRIBUSI PADA MATA KULIAH FISIKA LINGKUNGAN
This study aims to characterize the temporal variation of carbon monoxide (CO) concentration following forest and land fires (karhutla) in Ogan Ilir Regency using Internet of Things (IoT)-based sensor data, and to analyze the relationship between CO concentration and physical parameters, namely temperature and humidity. The data used were time-series data recorded at 5-minute intervals during the period of October 15-29, 2025, obtained from a single IoT sensor unit installed on the rooftop of Building D, Faculty of Teacher Training and Education, Universitas Sriwijaya. This research employed a quantitative descriptive approach using the Seasonal-Trend decomposition using LOESS (STL) method to separate trend, seasonal (diurnal), and residual components of CO concentration data. In addition, Cross-Correlation Function (CCF) analysis was applied to the residual data to identify temporal relationships and possible time lags between CO and temperature and humidity. The results show that CO concentrations ranged from approximately 190-280 ppb and exhibited consistent daily fluctuation patterns, with accumulation tendencies during nighttime to early morning. STL decomposition revealed the dominance of the diurnal component in data variability, while CCF analysis indicated statistically significant but weak correlations between CO and meteorological parameters at certain lags, suggesting the influence of atmospheric dynamics on pollutant dispersion. This study demonstrates that IoT-based sensors effectively support real-time air quality monitoring and contribute as instructional material in Environmental Physics courses.
No other version available