TROPOMI NO2 evaluation based on recurrent airborne campaigns

2023-2024
The ESA SVANTE project evaluates the TROPOMI tropospheric NO2 product through recurrent airborne campaigns using the SWING hyperspectral imager. Conducted over Berlin and Bucharest since 2021, these validation flights map the NO2 distribution, aligned with TROPOMI overpasses. The data captures various conditions, aiding in the assessment of TROPOMI's accuracy.

The project also develops harmonised tools and data formats for processing airborne data, ensuring consistent results and supporting future atmospheric missions like Sentinel-5 and Sentinel-4.

To monitor air quality, climate, and ozone

Launched on October 13, 2017, the Sentinel-5p/TROPOMI mission is a vital component of ESA’s Copernicus programme. Its mission is to monitor air quality, climate, and ozone. One of its primary data products is tropospheric nitrogen dioxide (NO2), a significant air pollutant that affects both human health and the environment.

Airborne campaigns for TROPOMI validation

To assess TROPOMI’s NO2 data, a series of 27 decentralised airborne campaigns have been conducted under the ESA SVANTE project since 2021. These campaigns collect independent measurements that are compared with the satellite data to ensure the product’s accuracy.

For these campaigns, BIRA-IASB mounted its SWING hyperspectral imager on different aircraft: a Cessna T207A from FUB flying over Berlin (Germany) and a BN-2 from INCAS flying over Bucharest (Romania). The flights typically cover 10 to 35 TROPOMI pixels, capturing the horizontal distribution of NO2 in these urban and industrial areas.

Comprehensive dataset and validation tools

A key feature of these validation efforts is the recurrent nature of the airborne campaigns. This provides a robust statistical dataset that spans varying geophysical and meteorological conditions, increasing the reliability of the validation process. 

To further ensure the consistency and accuracy of the validation results, a standardised airborne data format has been developed as well as a central processor and harmonised tools for processing and comparison. These help to ensure that the results are comparable across different datasets, improving the accuracy of the validation process.

Impact and future directions

The outcomes of these validation campaigns are not only crucial for assessing the current accuracy of TROPOMI's NO2 data products but also for future atmospheric missions. The validation strategies, tools, and datasets developed during these campaigns can be extended to upcoming missions like Sentinel-5 and Sentinel-4, and other imaging instruments.

The long-term goal is to provide a comprehensive, independent validation framework that can be used globally to monitor air pollution and improve environmental management. By assessing and enhancing the quality of satellite-based air quality products, these efforts contribute to a better understanding and mitigation of air pollution's impact on health and the environment.

Bucharest, Romania, map NO2 horizontal distribution
NO2 horizontal distribution during the overpass of TROPOMI on 28 March 2022.
Bucharest, Romania, map NO2 horizontal distribution
Figure 1. SWING flight over Bucharest, Romania, mapping the NO2 horizontal distribution during the overpass of TROPOMI on 28 March 2022. White dots indicate known point sources from the E-PRTR emission inventory. Their size scales to the amount of emissions reported.
SWING on INCAS BN-2 aircraft.
Figure 2. SWING on INCAS BN-2 aircraft.
Validation of the Sentinel-5P/TROPOMI NO2 product
Figure 3. Validation of the Sentinel-5P/TROPOMI NO2 product based on an ensemble of 16 SWING flights over Bucharest, Romania and 11 flights over Berlin, Germany. Every point corresponds to the comparison between one TROPOMI pixel and averaged SWING pixels, taking into account some well defined spatiotemporal constraints.