Coal-fired power plants are among the largest contributors to global CO₂ emissions. However, accurately measuring their emissions has been challenging due to outdated inventories and the technological limitations of satellite-based methods. These challenges are particularly pronounced in regions that lack modern emissions data or ground-based monitoring infrastructure.

To tackle these issues, a research team led by Prof. SHI Yusheng from the Aerospace Information Research Institute of the Chinese Academy of Sciences has developed a novel solution. This approach provides high-precision CO₂ emission estimates from 14 large coal-fired power plants around the world, utilizing an enhanced Gaussian plume model and data from NASA's Orbiting Carbon Observatory-3 satellite.

This study was published in the Journal of Cleaner Production.

Key enhancements to the model include improved estimation of background CO₂ concentrations, a more realistic simulation of plume rise behavior, and dynamic optimization of wind direction parameters. These improvements have increased the accuracy of satellite-based CO₂ emission measurements. The updated model produced daily emission estimates ranging from 21.54 to 82.33 kilotonnes per day, demonstrating strong correlation coefficients (R values between 0.493 and 0.863), indicating reliable performance. Notably, optimizing wind direction and plume rise were identified as key factors contributing to this enhanced accuracy.

Verification of the results, by comparing them with existing emission inventories, showed a high degree of consistency overall. However, it also highlighted areas where existing databases underestimated actual emissions. For example, the Carbon Monitoring for Action (CARMA) database, developed by the Global Development Center, underestimated emissions from the Tuoketuo power plant in China, while the Global Coal Plant Tracker underestimated emissions from the James H. Miller Jr. plant in the United States. These discrepancies were primarily due to outdated emission factors and statistical methods.

Importantly, this study has established a framework for dynamic, high-resolution tracking of CO₂ emissions based on satellite data. While the current work focuses on coal-fired power plants, the model can also be applied to other major emission point sources, such as steel plants and oil and gas fields.

"This work demonstrates the powerful role that satellite remote sensing technology can play in independently verifying emissions," said Prof. SHI. "It not only supports more accurate regional carbon accounting but also helps identify abnormal emission events and evaluate the long-term effects of emission reduction measures."