Quantifying and attributing CO emissions over Central Asia using TROPOMI and explicit observational uncertainty
- 1China University of Mining and Technology (Xuzhou, China)
Abstract
Carbon monoxide (CO) is an important atmospheric constituent affecting both air quality and climate. Using TROPOMI CO and HCHO column retrievals within the Model-Free Inversion Estimation Framework (MFIEF), we derived daily gridded CO emission estimates over Central Asia, including Xinjiang, Kazakhstan, Kyrgyzstan, and Uzbekistan, for 2019–2024. Regional mean CO emissions decreased by 37.9 % between 2019 and 2024, with a significant seasonally adjusted trend of -0.33µgm-2s-1yr-1. The identified emission hotspots also weakened over the study period. Higher emissions during winter and early spring are consistent with enhanced combustion associated with residential heating and industrial activity. A perturbation-based uncertainty analysis propagates assumed observational uncertainties through the inversion and indicates that approximately 69 % of the daily grid-cell emission estimates fall outside the adopted uncertainty bounds, highlighting the potential influence of atmospheric variability, dust loading, small sources, and observational uncertainty on the reliability of grid-scale emission estimates. The spatial distribution of satellite-derived grid-scale CO emission estimates reveals differences in urban emissions relative to bottom-up inventories, indicates a moderate positive association between coal consumption and inferred CO emissions at selected industrial locations, and highlights emissions associated with persistent underground coal fires, which are not explicitly represented as a separate source category in conventional anthropogenic emission inventories. This study demonstrates the potential of MFIEF as an observation-driven approach for characterizing regional CO emission patterns and temporal variability in data-sparse regions and provides complementary information for emission-inventory development and source characterization when production and location data are not available or are out of date.
Versions
Related
- Quantifying Uncertainty in Climate Sensitivity Metrics Using Large Ensemble Observational Datasets
- Investigating variability and uncertainty in upstream methane emissions intensity estimates of Western Canadian oil and gas production
- Quantifying AI data center nitrogen oxide (NOx) emissions from space
- Quantifying meteorological impacts on local landfill methane emissions by using field measurements and machine learning
- Kaya-guided hybrid deep learning for leave-country-out estimation of national CO2 emissions: a 112-country benchmark with country-clustered inference, conformal uncertainty, and explainable AI
BibTeX
@article{feng2026quantifying,
title = {{Quantifying and attributing CO emissions over Central Asia using TROPOMI and explicit observational uncertainty}},
author = {Ye Feng and Jason Blake Cohen and Xiaolu Li and Lingxiao Lu and Shuo Wang and Zhewen Liu and Lei Wang and Jian Liu and Kai Qin},
journal = {Atmospheric Chemistry and Physics},
year = {2026},
month = oct,
doi = {10.5194/acp-26-15139-2026},
url = {https://doi.org/10.5194/acp-26-15139-2026},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Quantifying and attributing CO emissions over Central Asia using TROPOMI and explicit observational uncertainty
AU - Ye Feng
AU - Jason Blake Cohen
AU - Xiaolu Li
AU - Lingxiao Lu
AU - Shuo Wang
AU - Zhewen Liu
AU - Lei Wang
AU - Jian Liu
AU - Kai Qin
PY - 2026
DA - 2026/10/28
JO - Atmospheric Chemistry and Physics
DO - 10.5194/acp-26-15139-2026
UR - https://doi.org/10.5194/acp-26-15139-2026
AB - Carbon monoxide (CO) is an important atmospheric constituent affecting both air quality and climate. Using TROPOMI CO and HCHO column retrievals within the Model-Free Inversion Estimation Framework (MFIEF), we derived daily gridded CO emission estimates over Central Asia, including Xinjiang, Kazakhstan, Kyrgyzstan, and Uzbekistan, for 2019–2024. Regional mean CO emissions decreased by 37.9 % between 2019 and 2024, with a significant seasonally adjusted trend of -0.33µgm-2s-1yr-1. The identified emission hotspots also weakened over the study period. Higher emissions during winter and early spring are consistent with enhanced combustion associated with residential heating and industrial activity. A perturbation-based uncertainty analysis propagates assumed observational uncertainties through the inversion and indicates that approximately 69 % of the daily grid-cell emission estimates fall outside the adopted uncertainty bounds, highlighting the potential influence of atmospheric variability, dust loading, small sources, and observational uncertainty on the reliability of grid-scale emission estimates. The spatial distribution of satellite-derived grid-scale CO emission estimates reveals differences in urban emissions relative to bottom-up inventories, indicates a moderate positive association between coal consumption and inferred CO emissions at selected industrial locations, and highlights emissions associated with persistent underground coal fires, which are not explicitly represented as a separate source category in conventional anthropogenic emission inventories. This study demonstrates the potential of MFIEF as an observation-driven approach for characterizing regional CO emission patterns and temporal variability in data-sparse regions and provides complementary information for emission-inventory development and source characterization when production and location data are not available or are out of date.
SN - 1680-7324
ER -