09 Jun 2022
09 Jun 2022
Status: this preprint is currently under review for the journal AMT.

Estimation of NO2 emission strengths over Riyadh and Madrid from space from a combination of wind-assigned anomalies and machine learning technique

Qiansi Tu1,2, Frank Hase2, Zihan Chen3, Matthias Schneider2, Omaira García4, Farahnaz Khosrawi2, Thomas Blumenstock2, Fang Liu5, Kai Qin6, Song Lin1, Hongyan Jiang1,7, and Dianjun Fang1,8 Qiansi Tu et al.
  • 1Tongji University, School of Mechanical Engineering, Shanghai, China
  • 2Karlsruhe Institute of Technology (KIT), Institute of Meteorology and Climate Research (IMK-ASF), Karlsruhe, Germany
  • 3Karlsruhe Institute of Technology (KIT), Department of Informatics, Karlsruhe, Germany
  • 4Izaña Atmospheric Research Centre (IARC), Meteorological State Agency of Spain (AEMet), Tenerife, Spain
  • 5Beijing Chehejia Automobile Technology Co., Ltd., Beijing, China
  • 6China University of Mining and Technology, School of Environment and Spatial Informatics, Jiangsu, China
  • 7Jiangsu University of Science and Technology, School of Mechatronic and Power Engineering, Zhenjiang, China
  • 8Qingdao Sino-German Institute of Intelligent Technologies, Qingdao, China

Abstract. Nitrogen dioxide (NO2) air pollution provides valuable information for quantifying NOx emissions and exposures. This study presents a comprehensive method to estimate average tropospheric NO2 emission strengths derived from three-year (April 2018 – March 2021) TROPOMI observations by combining a wind-assigned anomaly approach and a Machine Learning (ML) method, the so-called Gradient Descent. This combined approach is firstly applied to the Saudi Arabian capital city Riyadh, as a test site, and yields a total emission rate of 1.04×1026 molec./s. The ML-trained anomalies fit very well with the wind-assigned anomalies with an R2 value of 1.0 and a slope of 0.99. Hotspots of NO2 emissions are apparent at several sites where the cement plant and power plants are located and over areas along the highways. Using the same approach, an emission rate of 1.80×1025 molec./s is estimated in the Madrid metropolitan area, Spain. Both the estimate and spatial pattern are comparable to the CAMS inventory.

Weekly variations of NO2 emission are highly related to anthropogenic activities, such as the transport sector. The NO2 emissions were reduced by 24 % at weekends in Riyadh, and high reductions are found near the city center and the areas along the highway. An average weekend reduction estimate of 30 % in Madrid is found. The regions with dominant sources are located in the east of Madrid, where the residential areas and the Madrid-Barajas airport are located. Additionally, the NO2 emissions decreased by 21 % in March–June 2020 compared to the same period in 2019 induced by the COVID-19 lockdowns in Riyadh. A much higher reduction (60 %) is estimated for Madrid where a very strict lockdown policy was implemented. The high emission strengths during lockdown only persist in the residential areas and cover smaller areas during weekdays than at weekends. The spatial patterns of NO2 emission strengths during lockdown are similar to those observed at weekends in both cities. Though our analysis is limited to two cities as testing examples, the method has proved to provide reliable and consistent results. Therefore, it is expected to be suitable for other trace gases and other target regions.

Qiansi Tu et al.

Status: final response (author comments only)

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
  • RC1: 'review of  "Estimation of NO2 emission strengths over Riyadh and Madrid from space from a combination of wind-assigned anomalies and machine learning technique" by Tu et al', Anonymous Referee #1, 15 Jul 2022
  • RC2: 'Comment on amt-2022-176', Anonymous Referee #2, 05 Aug 2022

Qiansi Tu et al.

Qiansi Tu et al.


Total article views: 414 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
319 86 9 414 1 5
  • HTML: 319
  • PDF: 86
  • XML: 9
  • Total: 414
  • BibTeX: 1
  • EndNote: 5
Views and downloads (calculated since 09 Jun 2022)
Cumulative views and downloads (calculated since 09 Jun 2022)

Viewed (geographical distribution)

Total article views: 382 (including HTML, PDF, and XML) Thereof 382 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
Latest update: 20 Nov 2022
Short summary
Three-year TROPOMI observations are used to derive tropospheric NO2 emissions in two mega(cities), where high anthropogenic activities exist. The wind-assigned anomalies are calculated, and the emission rates and spatial patterns are estimated based on a Machine Learning algorithm. The results are in reasonable agreement with the previous studies and the inventory. Our method is quite robust and can be served as a simple method to estimate the emissions of NO2 and other gases in other regions.