Tracking and Predicting CO2 Emission Transportation in Toba Regency

In 2018, a study published in Nature Climate Change highlighted that global tourism is responsible for 8 percent of worldwide emissions, with nearly half (49 percent) attributed to transportation services. In Indonesia, the government's priority tourism destination policy aims to enhance the quality of tourist experiences, which inadvertently affects global emissions. Lake Toba, one of these priority destinations, has seen a marked increase in transportation activity, contributing significantly to local CO2 emissions. This study aims to forecast CO2 emissions in Toba using vehicle data from UPT Samsat Balige, enriched by scraping emission information for each vehicle type to enhance data quality and aggregation. We employ Long Short-Term Memory (LSTM), AutoRegressive Integrated Moving Average (ARIMA), and Support Vector Machine (SVM) models to analyze and predict monthly and daily CO2 emission patterns. Our results indicate that the ARIMA model performs best for monthly predictions with a Mean Absolute Error (MAE) of 0.1822, while the SVM model excels in daily predictions, achieving the lowest error with an MAE of 0.0010.