Analyzing urban and industrial carbon monoxide sources from space

Carbon monoxide (CO) is an important air pollutant that is a product of incomplete combustion. CO is not a direct greenhouse gas (GHG), but plays an important role in atmospheric chemistry since it is a precursor of tropospheric ozone (O3) and affects the lifetime of methane (CH4), both important greenhouse gases. Complete combustion of carbonaceous fuels does not produce carbon monoxide but only carbon dioxide (CO2) and water. Therefore, the ratio of CO over CO2 is an indicator of the completeness of combustion. As such, CO can be used to investigate combustion efficiencies, indicating where the combustion efficiency could be improved which could lead to less energy use and a reduction in air pollution. Additionally, as CO is co-emitted with CO2, CO also has the potential to improve our knowledge of CO2 emissions. Independent verification of reported anthropogenic emissions has been identified as a key aspect of reaching the climate goals set in the Paris agreement. The global coverage of satellites provides an important opportunity to support this effort. In this work, we quantify CO emission rates from large anthropogenic sources using the TROPOspheric Monitoring Instrument (TROPOMI) on board of ESA’s Sentinel-5 Precursor satellite. Our goal is to develop satellite-based emission quantification methods, test their performance, and use their emission estimates to improve our knowledge of anthropogenic CO emissions. Furthermore, we want to test if the extensive coverage of TROPOMI CO observations can be used to improve and support the quantification of CO2 emissions from large anthropogenic sources. Quantification of these emissions from space is currently limited by the sparse CO2 satellite observations. To this end, we develop a TROPOMI version of the fast, computationally-light, cross-sectional flux (CSF) method to estimate emissions from the 29 largest and highest-CO-emitting cities in Africa. We validate and calibrate our method against atmospheric simulations performed using the Weather Research and Forecasting (WRF) model, and show its reliability for emissions above 100 Gg yr−1. Additionally, we investigate CO emissions from integrated iron and steel plants in Europe to evaluate consistency between TROPOMI-observed CO concentrations and the European Pollutant Release and Transfer Register (E-PRTR), a high-quality emission reporting framework. Integrated iron and steel plants are the highest-CO-emitting point sources, yet most reported emission rates lie below 100 Gg yr−1. Therefore, instead of using the cross-sectional flux method, we perform annual analytical inversions, which are more accurate but also computationally more expensive as they model transport explicitly. Finally, we combine our satellite-based CO emission quantification method with CO2 measurements from the Orbiting Carbon Observatory-2 and 3 (OCO-2 and OCO-3) satellite instruments to determine combustion efficiencies using only satellite measurements. We focus on large cities, iron and steel plants, and power plants around the world, which are the largest ‘point’ sources of CO and CO2 as seen from space. The main challenge is the limited availability of CO2 observations which we try to overcome in part by using the daily global coverage of TROPOMI CO observations. We determine TROPOMI-based CO emission rates using the CSF method. Using the co-emission of CO and CO2, we then use the areas spanned by the CO CSF cross-sections to determine the corresponding CO2 enhancements measured with OCO. By leveraging the daily global coverage of TROPOMI CO observations, we can thus use a larger fraction of OCO CO2 observations, resulting in more days with CO2 emission quantifications. Together, the three papers investigate TROPOMI’s CO product, exploring its strengths and limitations. Most importantly, they demonstrate how a satellite can be used to improve our knowledge and understanding of anthropogenic emissions which is vital for efforts towards emission reduction.