Improving the data quality of CO2 continuous emissions monitoring systems: In the context of China's emissions trading scheme

Ensuring high-quality carbon emissions data is critical for effectively implementing environmental policies and achieving carbon neutrality goals, particularly in the context of China's emissions trading scheme (ETS) for decarbonizing energy-intensive sectors. Compared to traditional calculation-based methods, continuous emission monitoring systems (CEMS) offer significant advantages, including real-time monitoring, high precision, and reduced reliance on manual reporting. However, maintaining CEMS data quality remains challenging due to anomalies, incomplete records, and inconsistencies with other measurement approaches, all of which hinder its broader adoption. Existing studies on CEMS primarily focus on data quality concerns or provide general recommendations, but lack a systematic approach for data quality improvement. To bridge this gap, this study proposes a comprehensive framework for enhancing CEMS data quality by addressing accuracy, consistency, and completeness. The framework leverages data-driven techniques and expert knowledge to detect anomalies, calibrate emissions, and impute missing data. The effectiveness of the proposed method is demonstrated using high-frequency data collected from a thermal power plant in Shandong Province. These findings offer valuable insights for facilitating CEMS applications and provide practical policy implications for supporting its integration into China's ETS. The recommendations emphasize the importance of technical standards, quality control mechanisms, and pilot programs to improve the reliability of carbon emissions data and enhance policy enforcement.