Prediction of CO₂ Emissions in Sundarban Using Auto-ARIMA: A Comprehensive Analysis

Abstract

Climate change driven by greenhouse gas emissions poses a serious threat to the fragile ecosystem of the Indian Sundarban region. The objective of this study is to analyze the historical pattern of CO₂ emissions in the Indian Sundarbans and to forecast future emission trends using an automated time-series modelling approach. Annual CO₂ emission data for the period 1980–2008 are analyzed using the Auto-ARIMA framework which automatically identifies the optimal model parameters based on information criteria. Prior to model construction, stationarity of the series is examined using both Augmented Dickey–Fuller and KPSS tests. The selected Auto-ARIMA model effectively captures the underlying trend in the emission series and generates reliable short-term forecasts. Auto-ARIMA is used instead of conventional ARIMA to avoid subjective parameter selection and to ensure objective, reproducible model identification through information-criterion-based optimization. The results indicate a persistent increasing trend in CO₂ emissions in the Sundarban region and underscore the growing environmental risk. The findings of this study provide quantitative insight into emission dynamics and may assist policymakers and environmental planners in formulating informed mitigation and adaptation strategies to protect the Sundarban ecosystem.