Global Point Source CO₂ Emissions Monitoring Based on Hyperspectral Remote Sensing Imagery

This study developed a unified framework to compare carbon dioxide column observations from the DQ-1 active lidar and from passive optical satellites. The results show that DQ-1 is consistent with mature passive satellite products. It also provides observations where passive measurements are limited, including nighttime, high latitudes, and high-aerosol areas. Overall, DQ-1 can support future carbon dioxide monitoring and emission studies together with passive satellites.

This study presents a new way to estimate human-caused carbon dioxide emissions from cities using satellite observations. It combines measurements of nitrogen dioxide and carbon dioxide to better track emissions from urban areas. By applying this approach to several major cities with different environments, the study shows that combining multiple satellite observations can reduce uncertainty and provide more reliable city-level emission estimates.

Near-surface wind is crucial for weather and wind energy studies. This work uses physically constrained machine learning combined with satellite wind observations to generate a wind speed profile dataset within the global atmospheric boundary layer. The dataset accurately depicts the spatial distribution and variation of global wind speeds with improved accuracy, finer vertical detail, and reduced data gaps, supporting boundary-layer meteorology, climate studies, and wind energy applications.

The inventory overestimated emissions in Beijing and Riyadh by 10–20 % and underestimated emissions in Cairo by 10–30 %. The biosphere flux has a 20–40 % impact on emissions in Beijing during certain periods. Moreover, using night-time and day-time biosphere flux data separately can improve the simulation accuracy of the same orbit by 20–70 % compared to using daily average data.

CH₄ works as the second-most important greenhouse gas, its reported emission inventories being far less than CO₂. In this study, we developed a self-adjusted model to estimate the CH₄ emission rate from strong point sources by the UAV-based AirCore system. This model would reduce the uncertainty in CH₄ emission rate quantification accrued by errors in measurements of wind and concentration. Actual measurements on Pniówek coal demonstrate the high accuracy and stability of our developed model.