Tropospheric Formaldehyde Retrievals from GEMS within the GEO-RING Project

Formaldehyde (HCHO) is a short-lived product of volatile organic compound oxidation and a key precursor of tropospheric ozone, making it an essential proxy for surface emissions and air quality. Until recently, global HCHO monitoring relied on low-Earth-orbit (LEO) sensors, providing limited temporal coverage. The advent of geostationary UV–visible spectrometers such as GEMS, TEMPO, and Sentinel-4 now enables continuous daytime observations with high spatial and temporal resolution respectively over Asia, North America and Europe.

Within the GEO-RING project, we evaluate the GEMS operational HCHO product and explore methods to improve retrieval accuracy. Performance is assessed against ground-based networks (PGN Pandora, NDACC FTIR and MAX-DOAS) and LEO sensors (TROPOMI, GOME2-B,C), focusing on intermediate quantities such as slant columns that determine the information content from the instrument. To enhance the DOAS inversion, we test two approaches: (1) pseudo-cross-sections derived from principal component analysis of fit residuals, and (2) a CODOAS approach exploiting the covariance of the fit residuals. These methods are designed to be applicable consistently across GEMS, TEMPO, and Sentinel-4 to mitigate instrumental artifacts and scene inhomogeneity.

We also revisit background correction and air mass factor calculations for GEO observations, ensuring compatibility with LEO-based HCHO climate data records (ESA CCI). Finally, GEMS data are analysed to characterise diurnal variability over selected regions and compared with combined morning and afternoon LEO observations, demonstrating the added value of GEO sensors for future long-term atmospheric composition monitoring.