CE01 - Terre fluide et solide

PollutAnts and gReenhouse Gases emissiOns moNitoring from spAce at high ResolUTion – ARGONAUT

ARGONAUT : Pollutants and Greenhouse gases emissions monitoring from space at high resolution

The industrialization and urbanization of the society has led to the degradation of air quality (AQ) resulting in more than 7 million premature deaths per year and the advent of Climate Change (CC). The society faces major environmental challenges with the need to develop coordinated environmental policies at the level of territories as well as nations to limit the impact of pollutant and greenhouse gas emissions on AQ and CC.

High resolution French anthropogenic emissions estimates for the main pollutants (NOx, CO, NMVOCs) and carbon dioxide (CO2) by atmospheric inversion and satellite imagery

Coordinated monitoring and mitigation strategies from the territories scale to the country scale are needed to reduce pollutant emissions to non-impacting levels for human health and ecosystems and reduce GHG emissions to limit the CC effects. Monitoring the emissions at a high resolution are then required. The ARGONAUT project applies atmospheric inversion to provide a new estimate of anthropogenic emissions of the main AQ pollutants (NOx, CO and NMVOCs) and CO2 in France at 10 km resolution based on the co-assimilation of multiple species images from the last generation of satellite imaging (Sentinel-5P and CO2M). Upstream methodological investigation on the inversion of emission sources will provide strong basis to assess the benefit of optimized emission inventories constrained by the assimilation of satellite images in downstream operational products (AQ forecasts or exposure, local CO2 emission monitoring) and to provide recommendations for a possible future implementation of a national operational AQ and CO2 joint emission inversion system fed with satellite information in addition to national and local measurement networks complementary to the European Copernicus initiative and able to tackle the national specificities. The main ARGONAUT objectives are:<br />• to demonstrate the potential on the joint monitoring of AQ and CO2 emissions and strengthen the consistency between AQ and CO2 products.<br />• to evaluate the potential of the future CO2M mission to infer anthropogenic emissions.<br />• to demonstrate the potential of the city/plant scale resolution inversion of the anthropogenic emissions based on satellite images as a zoom or downstream application of the national scale inversion, which requires the improvement of the current inverse modeling techniques.

The methodology to address the objectives of the ARGONAUT project is based on atmospheric inversion and the last generation of satellite imagers to determine pollutants and CO2 emissions at politically relevant scales for France with a high resolution. Corrected emission inventories of NO2, CO, NMVOCs and CO2 will be derived at 10x10 km2 over France with a monthly temporal resolution or better covering the first two years of TROPOMI operation or specific periods such as the COVID pandemic. In addition, the potential of the future CO2M mission and its CO2 and NO2 data to infer anthropogenic CO2 and pollutant emissions will be evaluated based on Observing System Simulation Experiments (OSSEs, i.e., with synthetic data). The main challenge will be to develop joint assimilations of relevant pollutants and/or CO2, which include high-resolved images of co-observed species to increase the inversion capabilities of separating anthropogenic sources between them and from natural fluxes. Successful joint inversions will require a proper representation of the error statistics in the inversion system as well as the emission correlations across species. A specific attention will be paid to have the best estimate of uncertainties in the prior emissions for each species, of the correlations of uncertainties in the emissions, transport/chemistry and measurement between co-emitted species, and of the model and observation errors including specifically an account for their spatial correlations, which should be critical at high resolution.
In addition to this, ARGONAUT will address the question of inversions at the city/plant scale.
The main challenge to achieve such scales will be to adapt the inversion approach to overcome the transport error, which cannot be summarized as a statistical noise at such scales.

The new CIF (Community Inversion Framework) inversion system has been coupled with CHIMERE and adapted for reactive gases. Its sensitivity to inversion parameters has been evaluated and the system has been tested on a large scale for NOx inversion on the European domain from the OMI NO2 observations.
Preparatory work to explore the TROPOMI data was carried out. We mainly analyzed the NO2 data in 2020, with, in particular, a study of the lockdown impact on these data and their comparison with surface observations in France. Inversion tests of NOx emissions from NO2 images are in progress with a conventional inversion method and observation vector, but the quality of the TROPOMI product currently available does not always seem sufficient. A systematic analysis of the NO2 and CO images from TROPOMI (from January to April 2020) and the corresponding CHIMERE simulations revealed a fairly frequent discordance between the position and orientation of the observed and simulated NO2 plumes in addition to the quality of the data from CO insufficient to properly identify the plumes of this pollutant. Clustering approaches are being explored to improve information filtering and model / observation comparison. This approach should lead to the construction of an observation vector more suitable for inversion.
Regarding the detection and inversion of emissions from plumes in cities or large industrial sites, new metrics for comparing images of pollutant plumes have started to be tested on «toy models«. The continuation of this work involves the construction of a base of realistic images of plumes, the evaluation of new metrics on this base, their implementation in the project's inversion system and tests on real images of TROPOMI.

The European Commission is pushing for the development of operational services with a core satellite component through the Copernicus Program. In parallel, the French community also needs to develop its own national scale system fed with satellite information in complement to national and local measurement networks, and tackling some of the specificities of the national emissions. The ARGONAUT project with strong upstream methodological investigation on the inversion of emission sources from national to local scales will give the basis to the quantification of the added value of atmospheric inversion approaches to provide corrected emission inventories for downstream operational products such as AQ forecasts and local CO2 emission analysis. ARGONAUT will then contribute (i) to define the potential elementary components for such a future national operational system and/or downstream services, and (ii) to consolidate the visibility and position of the French scientific community of inverse modeling as key players for satellite-based European operational services.


Due to the large industrialization and urbanization, Air Quality (AQ) has been degraded worldwide, leading to more than 7 million premature deaths annually and Climate Change (CC) is becoming a reality with the five warmest years measured in the 2010s. The society is facing major environmental challenges: developing coordinated monitoring and mitigation strategies leading to optimal reduction of both AQ and CC impacts. Providing a response at the finer scale of territories is also critical to ensure efficient reduction policies. The primary objective of ARGONAUT is to provide new estimates of French anthropogenic emissions of the main AQ pollutants (nitrogen oxides - NOx, carbon monoxide - CO and non-methane volatile organic compounds - NMVOCs) and carbon dioxide (CO2) at high resolution, based on the atmospheric inversion and the last generation of satellites (Sentinel-5P/TROPOMI, CO2M). We will exploit the multiple-species high resolution imaging to exploit the local correlations between the various species and co-assimilate them to better constrain their emission estimates. Indeed, information on jointly observed pollutants, such as nitrogen dioxide (NO2) and CO, which have a long history of measurement from space, will add an additional valuable constraint on the emission inversion of co-emitted species that are more difficult to measure to useful levels, such as CO2 and NMVOCs. Highly resolved images are essential in this context to reveal the high correlation between the concentrations of different species locally, and then to clearly separate the anthropogenic sources, to quantify their emissions and to monitor their temporal evolution. Making a step forward in the joint assimilation of relevant pollutants and CO2 together, and addressing the correlation between them, will improve the emission inventories and their consistency across species and, more generally, should help addressing AQ and CC related emissions at the national to subnational scales. With the recent availability of TROPOMI colocalized images of NO2, CO and formaldehyde (HCHO), and in the future of CO2 and NO2 from CO2M, implementing such an approach becomes possible, as will be explored by ARGONAUT. ARGONAUT will put efforts to develop a highly sophisticated inverse modeling system, able to benefit from this high resolution imaging at the national level. A parallel objective of the project is to demonstrate the potential for addressing the city/plant scale with a zoom or dedicated systems downstream the national scale inversion. This bears additional challenges, because it requires the modeling and inversion systems to match the local fine scale atmospheric plumes from these sources despite a high and complex uncertainty in the meteorological simulation at this scale. Traditional atmospheric inversion techniques, which assume that transport modeling errors can be summarized as a statistical noise that has no temporal or spatial correlations are not adapted to tackle such a challenge. To overcome this issue and make the emission inversions much more reliable, we propose in the ARGONAUT project to move from classical statistical indicators to new non-local metrics. Finally, the project aims at quantifying the added value of atmospheric inversion approaches constrained by the assimilation of satellite images and of the derived optimized emission inventories in downstream operational products such as AQ forecasts or exposure assessments, and local CO2 emission monitoring. ARGONAUT will contribute (i) to define the potential elementary components for such a future national operational system and/or downstream services, and (ii) to consolidate the visibility and position of the French scientific community of inverse modeling as key players for satellite-based European operational services.

Project coordination

Gaëlle Dufour (Laboratoire inter-universitaire des systèmes atmosphèriques)

The author of this summary is the project coordinator, who is responsible for the content of this summary. The ANR declines any responsibility as for its contents.


CEREA Centre d'Enseignement et de Recherche en Environnement Atmosphérique
LSCE Laboratoire des Sciences du Climat et de l'Environnement
LISA Laboratoire inter-universitaire des systèmes atmosphèriques

Help of the ANR 601,596 euros
Beginning and duration of the scientific project: December 2019 - 48 Months

Useful links

Explorez notre base de projets financés



ANR makes available its datasets on funded projects, click here to find more.

Sign up for the latest news:
Subscribe to our newsletter