Retrieving the atmospheric concentrations of carbon dioxide and methane from the European Copernicus CO2M satellite mission using artificial neural networks

Carbon dioxide (CO2) and methane (CH4) are the most important anthropogenic greenhouse gases and the main drivers of climate change. Monitoring their concentrations from space helps detect and quantify anthropogenic emissions, supporting the mitigation efforts urgently needed to meet the primary objective of the Paris Agreement, adopted at the 21st Conference of the Parties to the United Nations Framework Convention on Climate Change (UNFCCC) in 2015, to limit the global average temperature increase to well below 2 degrees C above pre-industrial levels. In addition, satellite observations can be used to quantify natural sources and sinks, improving our understanding of the carbon cycle. Advancing these goals is one key motivation for the European Copernicus CO2 monitoring mission CO2M. The necessary accuracy and precision requirements for the measured quantities XCO2 and XCH4 (the column-averaged dry-air mole fractions of CO2 and CH4) are demanding. According to the CO2M mission requirements, the spatial and temporal variability of the systematic errors (or spatio-temporal systematic errors) of XCO2 and XCH4 must not exceed 0.5 ppm and 5 ppb, respectively. The stochastic errors due to instrument noise must not exceed 0.7 ppm for XCO2 and 10 ppb for XCH4. Conventional so-called full-physics algorithms for retrieving XCO2 and/or XCH4 from satellite-based measurements of reflected solar radiation are typically computationally intensive and still usually require empirical bias corrections based on supervised machine learning methods. Here we present the retrieval algorithm Neural networks for Remote sensing of Greenhouse gases from CO2M (NRG-CO2M), which derives XCO2 and XCH4 from CO2M radiance measurements with minimal computational effort using artificial neural networks (ANNs). In addition, NRG-CO2M also provides estimates of both the noise-driven uncertainties and the averaging kernels of XCO2 and XCH4 for each sounding. Since CO2M will not be launched until 2026, our study exploits simulated measurements over land surfaces from a comprehensive observing system simulation experiment (OSSE) that includes realistic meteorology, aerosols, surface bidirectional reflectance distribution function (BRDF), solar-induced chlorophyll fluorescence (SIF), and CO2 and CH4 concentrations. We created a novel hybrid learning approach that combines advantages of simulation-based and measurement-based training data to ensure coverage of a wide range of XCO2 and XCH4 values, making the training data representative of future concentrations as well. The algorithm's postprocessing is designed to achieve a high data yield of about 80 % of all cloud-free soundings. The spatio-temporal systematic errors of XCO2 and XCH4 are 0.44 ppm and 2.45 ppb, respectively. The average single sounding precision is 0.41 ppm for XCO2 and 2.74 ppb for XCH4. Therefore, the presented retrieval method has the potential to meet the demanding CO2M mission requirements for XCO2 and XCH4. While the presented results are a solid proof of concept, the actual achievable quality can only be determined once NRG-CO2M is trained on real data, where it is confronted, e.g., with unknown instrument effects and systematic errors in the training truth.