A new satellite-based method developed by Chinese researchers improves the accuracy of carbon dioxide (CO₂) emission estimates from coal-fired power plants, according to a study published in Environmental Science & Technology.

The approach, developed by a research team led by Prof. CHENG Tianhai from the Aerospace Information Research Institute of the Chinese Academy of Sciences, enhances near-real-time tracking of greenhouse gas emissions—a capability critical to global climate change mitigation efforts and the achievement of national climate targets.

Accurate measurement of CO₂ emissions is key to understanding and managing the environmental impact of power generation. But current estimates rely heavily on inventories built from average emission factors and plant activity data, which often fail to reflect real-time changes driven by operational shifts or pollution controls. Compounding the challenge, carbon-monitoring satellites like NASA's Orbiting Carbon Observatory-2 (OCO-2) have limited observation frequencies, making it hard to capture daily and seasonal emission variations.

To address these gaps, the team developed the Pollution-Carbon Synergy Model (PCSM). The innovation uses nitrogen oxides (NO_x)—pollutants co-emitted with CO₂ during fossil fuel combustion—as a reliable proxy for estimating CO₂ emissions. Because the NO_x-to-CO₂ ratio remains relatively stable for each plant under consistent infrastructure, it serves as a valuable indicator of carbon output.

By combining near-simultaneous satellite data from Tropospheric Monitoring Instrument (TROPOMI) (which detects NO_x) and OCO-2 (which measures CO₂), the researchers derived plant-specific NO_x-to-CO₂ emission factors. These factors were then paired with daily NO_x emissions estimated from TROPOMI overpasses to calculate daily CO₂ emissions at a much higher temporal resolution.

Tests on 15 major U.S. coal-fired power plants showed the PCSM method outperformed OCO-2 alone, cutting average annual CO₂ emission estimation errors from 45.8% (5.02 million tons) to 13.0% (1.43 million tons). It also surpassed leading global emission inventories such as Open-Source Data Inventory for Anthropogenic CO₂ (ODIAC) and Emissions Database for Global Atmospheric Research (EDGAR), boosting correlation with actual data from the U.S. Continuous Emission Monitoring System (CEMS) by 0.16 to 0.35.

When applied to 38 power plants worldwide, the PCSM found that while overall annual CO₂ emissions aligned with inventory estimates, a breakdown by 10 MtCO₂/yr revealed systematic biases: smaller plants were overestimated, and larger ones were underestimated in existing inventories.

"This method offers a practical, cost-effective way to monitor CO₂ emissions from space with much higher accuracy," Prof. CHENG said. "It provides essential support for countries aiming to track their emissions and meet their climate commitments under the Paris Agreement."