Extension of the S5P/TROPOMI CCD tropospheric ozone retrieval to middle latitudes

Abstract

Tropospheric ozone, a critical pollutant and greenhouse gas, exhibits spatio-temporal variability, challenging satellite observations. Existing methods like the Convective Cloud Differential (CCD) and Cloud Slicing Algorithms (CSA) are standard for Tropospheric Column Ozone (TCO) retrieval but are limited to the tropics (20°S-20°N). Notably, the CCD approach has proven successful with satellite sensors like Aura OMI, MetOp GOME-2, and Sentinel-5 Precursor TROPOMI. In this study, we present the first successful application of CCD retrieval outside the tropical region. We introduce CHORA-CCD (Cloud Height Ozone Reference Algorithm-CCD) for retrieving TCO from TROPOMI in middle latitudes. It utilises a local cloud reference sector (CLCD, CHORA-CCD Local Cloud Decision) to determine the stratospheric (above cloud) column (ACCO) ozone.  This ACCO is later subtracted from the total column in clear-sky scenes to determine the TCO. The new approach minimises the impact of variances in stratospheric ozone. An iterative approach is used to automatically select an optimal local cloud reference sector around each retrieval grid point, varying the radius from 60 to a maximum of 600 km around the grid box, for which a mean TCO is determined until a sufficient number of ground pixels with nearly fill cloud cover are found. Due to the prevalence of low-level clouds in middle latitudes, the estimation of TCO is constrained to the column up to a  reference altitude of 450 hPa.  An alternative method is introduced to directly estimate the ACCO down to 450 hPa by Theil-Sen regression in cases where the cloud-top heights in the local cloud sector are variable. The algorithm dynamically decides between CCD and Theil-Sen method for ACCO estimation by analysing the cloud characteristics. The CLCD algorithm is further refined by introducing a homogeneity criterion for total ozone to overcome inhomogeneities in stratospheric ozone. Monthly averaged CLCD-TCOs have been determined over the middle latitudes (60◦S-60◦N) from TROPOMI for the time period from 2018 to 2022. The accuracy of the method was investigated by comparisons with spatially collocated SHADOZ/WOUDC/NDACC ozonesondes from thirty-one stations. The validation results reveal that TCO retrievals at 450 hPa using the CLCD algorithm exhibit good agreement with ozonesondes at most stations. At the tropical station Natal (5.4°S, 35.4°W), there is an outstanding agreement between CLCD and ozonesondes, showcasing minimal bias and scatter (0.3 ± 1.0 DU). Similarly, in the subtropics over Irene (25.9°S, 28.2°E), CLCD exhibits a significantly lower bias and scatter (0.0 ± 1.4 DU). Specifically, at one of the northernmost stations, Legionowo (52.4°N, 21°E), bias and dispersion are minimal (0.4 ± 2.2 DU). Across all stations, the maximum observed bias and dispersion are below around 5 DU and 4 DU, respectively.  In this presentation, a detailed validation of the new local CCD retrievals will be given, underlining the advantage of using the local cloud reference sector in the middle latitudes, providing an important basis for subsequent systematic applications in current and future missions of geostationary satellites.